Unnamed: 0
int64 0
192k
| title
stringlengths 1
200
| text
stringlengths 10
100k
| url
stringlengths 32
885
| authors
stringlengths 2
392
| timestamp
stringlengths 19
32
⌀ | tags
stringlengths 6
263
| info
stringlengths 45
90.4k
|
---|---|---|---|---|---|---|---|
700 | Run C# Machine Learning Code Directly In A Jupyter Notebook | Artist impression of a Jupyter notebook running C# ;)
Jupyter notebooks have become the standard tool for hosting advanced machine learning code online.
A Jupyter notebooks is a fully interactive document that allows mixing of content, live source code, and program output, all in a single document. Notebooks are living dynamic documents and their embedded source code can be interactively run and modified.
This makes notebooks ideal for AI research or for teaching complex machine learning skills to IT students. But unfortunately they only support Python and R code. There isn’t any support for C#.
Until today that is!
Thanks to Microsoft’s tireless efforts we now have a fully functional C# kernel that allows us to run C# machine learning code directly in a Jupyter notebook.
And getting everything set up is a piece of cake. Here’s a quick guide on how you can run your own notebook server locally with full C# support.
Let’s start by installing Python. Yes, we do need python because the Jupyter server itself is a python application.
A great Python distribution for AI that comes bundled with Jupyter is called Anaconda. Here’s how to install it on Windows:
$ choco install anaconda3
I’m using Chocolatey, a nice package management system for Windows that runs on Powershell and makes installing new software a breeze.
Next you’ll need the NET Core SDK version 3. You can download it here.
You’ll also need dotnet try , an interactive version of the NET runtime environment that can run code interactively and which forms the core of the Jupyter C# kernel.
Type the following on the command line:
dotnet tool install -g dotnet-try
This will install dotnet try as a global tool.
Almost done! The final step is to check your start menu for the new Anaconda3 folder. Open it and run the Anaconda Powershell Prompt app. Then run the following command to add C# support to Jupyter:
dotnet try jupyter install
That’s all!
Now open the Anaconda3 menu from the start menu and click the Jupyter Notebook app.
You should see a Jupyter session appearing: | https://medium.com/machinelearningadvantage/run-c-machine-learning-code-directly-in-a-jupyter-notebook-a32e13e40b9c | ['Mark Farragher'] | 2019-11-07 20:34:23.472000+00:00 | ['Machine Learning', 'Artificial Intelligence', 'Jupyter Notebook', 'Csharp', 'Programming'] | Title Run C Machine Learning Code Directly Jupyter NotebookContent Artist impression Jupyter notebook running C Jupyter notebook become standard tool hosting advanced machine learning code online Jupyter notebook fully interactive document allows mixing content live source code program output single document Notebooks living dynamic document embedded source code interactively run modified make notebook ideal AI research teaching complex machine learning skill student unfortunately support Python R code isn’t support C today Thanks Microsoft’s tireless effort fully functional C kernel allows u run C machine learning code directly Jupyter notebook getting everything set piece cake Here’s quick guide run notebook server locally full C support Let’s start installing Python Yes need python Jupyter server python application great Python distribution AI come bundled Jupyter called Anaconda Here’s install Windows choco install anaconda3 I’m using Chocolatey nice package management system Windows run Powershell make installing new software breeze Next you’ll need NET Core SDK version 3 download You’ll also need dotnet try interactive version NET runtime environment run code interactively form core Jupyter C kernel Type following command line dotnet tool install g dotnettry install dotnet try global tool Almost done final step check start menu new Anaconda3 folder Open run Anaconda Powershell Prompt app run following command add C support Jupyter dotnet try jupyter install That’s open Anaconda3 menu start menu click Jupyter Notebook app see Jupyter session appearingTags Machine Learning Artificial Intelligence Jupyter Notebook Csharp Programming |
701 | Building a Team as a Model for Cooperative Work | I’ve known for about a year that I’d taken my business, Ninja Writers, about as far as it could go as a mostly one-person show. Since 2015 it’s just been me and Zach J. Payne, who has always been my right hand behind the scenes.
I was at capacity for how many students I could effectively teach and how many members there could be in the Ninja Writers Club and still provide the kind of experience I wanted the community to have.
I had a decision to make. Did I keep things the way they were — which, to be honest, was a pretty incredible place to be — or did I build a team so that Ninja Writers could continue to grow.
It started with my daughter.
Adrienne was in her last year of graduate school last year. She did the thing that appeals the most to me: She came to me with ideas. That’s the way to my heart and she knows me well enough to know it.
With another person on board, I was able to expand into small group workshops for writers. We kept it small, but over the last six months the workshops have blossomed. We’ve had several students in them who have finished writing their books or found real success in blogging.
We doubled Ninja Writers as a business in the last year. The Ninja Writers Club went from two weekly Zoom co-working calls, where I teach and we all workshop each other, to four. With Adrienne’s help, Ninja Writers Club was able to double in size.
And by the winter, we found ourselves at capacity again.
We could stay here. It’s a good place. The energy in Ninja Writers is so incredible. It’s infectious. But there’s no more of us to go around. We had ideas that we just had to keep, maybe permanently, on the back burner.
Growth kept happening though, super organically.
I didn’t know this was the way it would happen. I thought that if I was going to have a Ninja Writers team, I’d have to hire people I didn’t know and try to teach them how to be Ninja Writers.
That didn’t appeal to me, so I never did it.
But what happened was that I started to notice Ninja Writers who were doing that thing I love — remember? They were bringing ideas to the table. And they were implementing them on their own, just because they were really good ideas that made Ninja Writers better.
First, Meg Stewart started teaching a core group of Ninja Writers how to do very basic tech stuff. No one else was teaching them the things they needed to know, so she just started doing it and it started to become a business. My students were so enthusiastic about what she was doing that I finally contacted her to ask if she’d like to join the team to teach within Ninja Writers officially.
This was where the idea of cooperative work really took hold for me. I didn’t want to absorb Meg’s business. I wanted to help her grow it, while bringing her into Ninja Writers so that her work could benefit our community, too. She filled a gap in my business beautifully.
Then Ashley Shannon started teaching other Ninja Writers how to be better bloggers. They ate it up. More importantly, it worked. Her students were killing it. She came to me with a plan for making the program she’d developed part of Ninja Writers. More than that, she had a plan for making our marketing better and brought an energy that was infectious.
There was another gap and another person filling it. Ashley is building her own empire, and her being part of Ninja Writers means that I get to benefit from her and be a part of her growth, too.
Finally, I started to notice that there was a group of Ninja Writers who all of a sudden were very focused. They were finishing their books. They were getting their work done in a way that was clearly obvious to me, and to other people in the community. I realized that it was Juneta Key who had made the difference. She started hosting write-ins and showing up to encourage Ninja Writers in our Slack channel and on Facebook. Making her our Accountability Guru was an easy decision.
One more gap filled. And being part of Juneta’s growth as an entrepreneur is an absolute joy.
The key to growing my business was to grow it with people who already loved it. And who I already loved.
Meg, Ashley, and Juneta aren’t just employees. They are Ninja Writers. They understand the community and they are invested in making it as awesome as it can be.
The members already know them. The shift to having them onboard has been completely organic and exciting.
More than that, though, they are talented people who have their own fantastic ideas. We’re a model of how cooperative work can be. As they help me grow Ninja Writers, they grow their own businesses as well and I get to be part of that.
They aren’t just employees. They’re family. To me, but also to the whole Ninja Writers community.
One of the new things we’re doing is a podcast.
I kind of halfway started one on my own a few months ago, but it’s one of those things that I knew would be good and I wanted to do, but I was so close to my capacity that I didn’t have the time or energy to do it really well.
But now Meg, Ashley and I are doing it together and — whoa. It’s going to be incredible. We recorded our first weekly episode, which introduces you to the whole team and talks about what it means to all three of us to be working writers.
You can check that out here.
We’re starting up a Ninja Writers publication, to showcase the talented writers in our community. We’re adding new classes. We’re doing things to make the whole experience of being a Ninja Writer better — like adding onboarding calls so that the sheer volume of what we offer isn’t overwhelming to new members.
We’re on a wild ride. I couldn’t be more excited.
Sometimes things just happen right when they’re supposed to.
The right people showed up right when I was in a position to hire them. Talk about serendipity.
I knew that I needed to pivot Ninja Writers toward supporting the community as so many people found themselves needing to earn a living with their writing thanks to Covid. Adrienne, Meg, Ashley, and Juneta are like having a bunch of freaking superheroes show up to give your slow-moving ship a shove in the right direction.
And then other things just happened. Like after five years, Teachable finally implemented a way for me to offer new Ninja Writers a way to try out the Ninja Writers Club for free. Yay for free trials!
And, halfway through the first full year of the Ninja Writers Academy (those small group workshops), a bunch of students started having real success. They finished their first drafts. They were gaining traction as bloggers. People were getting attention from agents and publishers.
Seriously. We’re on the verge of really big things. This is going to be so much fun. | https://medium.com/the-write-brain/building-a-team-as-a-model-for-cooperative-work-1b167b9f638a | ['Shaunta Grimes'] | 2020-06-21 14:52:03.570000+00:00 | ['Creativity', 'Business', 'Entrepreneurship', 'Productivity', 'Work'] | Title Building Team Model Cooperative WorkContent I’ve known year I’d taken business Ninja Writers far could go mostly oneperson show Since 2015 it’s Zach J Payne always right hand behind scene capacity many student could effectively teach many member could Ninja Writers Club still provide kind experience wanted community decision make keep thing way — honest pretty incredible place — build team Ninja Writers could continue grow started daughter Adrienne last year graduate school last year thing appeal came idea That’s way heart know well enough know another person board able expand small group workshop writer kept small last six month workshop blossomed We’ve several student finished writing book found real success blogging doubled Ninja Writers business last year Ninja Writers Club went two weekly Zoom coworking call teach workshop four Adrienne’s help Ninja Writers Club able double size winter found capacity could stay It’s good place energy Ninja Writers incredible It’s infectious there’s u go around idea keep maybe permanently back burner Growth kept happening though super organically didn’t know way would happen thought going Ninja Writers team I’d hire people didn’t know try teach Ninja Writers didn’t appeal never happened started notice Ninja Writers thing love — remember bringing idea table implementing really good idea made Ninja Writers better First Meg Stewart started teaching core group Ninja Writers basic tech stuff one else teaching thing needed know started started become business student enthusiastic finally contacted ask she’d like join team teach within Ninja Writers officially idea cooperative work really took hold didn’t want absorb Meg’s business wanted help grow bringing Ninja Writers work could benefit community filled gap business beautifully Ashley Shannon started teaching Ninja Writers better blogger ate importantly worked student killing came plan making program she’d developed part Ninja Writers plan making marketing better brought energy infectious another gap another person filling Ashley building empire part Ninja Writers mean get benefit part growth Finally started notice group Ninja Writers sudden focused finishing book getting work done way clearly obvious people community realized Juneta Key made difference started hosting writeins showing encourage Ninja Writers Slack channel Facebook Making Accountability Guru easy decision One gap filled part Juneta’s growth entrepreneur absolute joy key growing business grow people already loved already loved Meg Ashley Juneta aren’t employee Ninja Writers understand community invested making awesome member already know shift onboard completely organic exciting though talented people fantastic idea We’re model cooperative work help grow Ninja Writers grow business well get part aren’t employee They’re family also whole Ninja Writers community One new thing we’re podcast kind halfway started one month ago it’s one thing knew would good wanted close capacity didn’t time energy really well Meg Ashley together — whoa It’s going incredible recorded first weekly episode introduces whole team talk mean three u working writer check We’re starting Ninja Writers publication showcase talented writer community We’re adding new class We’re thing make whole experience Ninja Writer better — like adding onboarding call sheer volume offer isn’t overwhelming new member We’re wild ride couldn’t excited Sometimes thing happen right they’re supposed right people showed right position hire Talk serendipity knew needed pivot Ninja Writers toward supporting community many people found needing earn living writing thanks Covid Adrienne Meg Ashley Juneta like bunch freaking superheroes show give slowmoving ship shove right direction thing happened Like five year Teachable finally implemented way offer new Ninja Writers way try Ninja Writers Club free Yay free trial halfway first full year Ninja Writers Academy small group workshop bunch student started real success finished first draft gaining traction blogger People getting attention agent publisher Seriously We’re verge really big thing going much funTags Creativity Business Entrepreneurship Productivity Work |
702 | How to Make Your Work Stand Out Online | The Internet is a crowded place. Every day someone writes a new blog post in your niche or publishes a new book on Amazon in your genre. In a sea of content, how can you get people to pay attention to your work?
I remember a shoe store my mom used to take me to when I was a little kid. It was called Merry-Go-Round Shoes and was quite unlike any store I’d ever been to. There was a merry-go-round in the center of the store that children could ride on.
The store also had a funny vending machine with a mechanical chicken. If you inserted a quarter in the vending machine, you’d receive a plastic egg with a special prize inside. (Yes, I begged my mom for quarters.)
What kid wouldn’t love a shoe store like that? Whenever my brother and I needed new shoes, we’d always ask our mom to take us there. Sometimes we’d even ask her to take us there when we didn’t need new shoes.
I’m sure that there were lots of other shoe stores near where we lived that carried the exact same shoes as Merry-Go-Round. But the owners of Merry-Go-Round had flexed their creative muscle and found a way to make their store unique and unforgettable.
In order to capture people’s attention on the Internet, we have to do the same thing. We have to look for ways that we can make our writing and marketing strategies stand out so that we don’t get lost in the crowd. Seth Godin coined the term “purple cow” to describe this principle.
What is the Purple Cow Principle?
In a 2003 TED talk, Godin explained,
“…My parable here is you’re driving down the road and you see a cow, and you keep driving because you’ve seen cows before. Cows are invisible. Cows are boring. Who’s going to stop and pull over and say — ‘Oh, look, a cow.’ Nobody. But if the cow was purple…you’d notice it for a while. I mean, if all cows were purple you’d get bored with those, too. The thing that’s going to decide what gets talked about, what gets done, what gets changed, what gets purchased, what gets built, is: ‘Is it remarkable?’ And ‘remarkable’ is a really cool word, because we think it just means ‘neat,’ but it also means ‘worth making a remark about.’”
Now, the purple cow principle doesn’t mean you have to do something crazy to get attention. In an article in The Guardian, Godin clarified,
“Being noticed is not the same as being remarkable. Running down the street naked will get you noticed, but it won’t accomplish as much. It’s easy to pull off a stunt, but not useful.”
Rather, implementing the purple cow principle means looking for ways that you can be remarkable in order to ensure long-term success, not just fifteen minutes of Internet fame.
This starts, of course, with crafting excellent content that you can be proud of. Excellent content will always stand out in a sea of mediocrity.
Second, it means looking for ways that you can draw on your unique skills and strengths to innovate on standard marketing strategies when you package your content and promote it.
For example, blog posts with a lead photo are usually read and shared more than blog posts that don’t have a lead photo. Thus, nearly every blog post nowadays has a lead photo. But some bloggers, who have an artistic bent, take this to the next level by using their own photos or hand-drawn illustrations rather than stock photos. That’s one way they make their content memorable.
You can also make your work stand out in more subtle ways, like simply changing a few lines of copy on your website. Read on to find out what happened when I did that earlier this month.
How I Used the Purple Cow Principle to Skyrocket My Email Conversion Rate
This past year, I wanted to update the email pop-up form on my website. I’d checked my analytics and noticed it wasn’t performing very well.
A pop-up form is one of those email sign-up forms that literally “pop-up” on the screen when you visit a website. Usually, you have to enter your email or close the form to continue scrolling.
Pop-ups are incredibly effective at reminding a visitor to subscribe, but they can also be annoying and intrusive. You don’t want to have multiple pop-ups on your site or create a pop-up that a visitor can’t click out of.
On my website, I have a slide-in pop-up form. It appears discreetly in the bottom right-hand corner of the website and doesn’t prevent the visitor from continuing to scroll and read a blog post. It’s been a few years since I edited the pop-up so I knew it was time for an update.
First, I decided to change the pop-up’s timing. Previously, I’d had the pop-up appear after about 30 seconds. This is what most websites do, but I wanted to try something different. I felt that this wasn’t giving a visitor enough time to look around my website, read an article, get to know me, and reach the point where they were ready to subscribe.
After all, who wants to be virtually yelled at with a pop-up when they first land on a website?
I checked my Google analytics account to see what was the average length of time that someone visited my website. It was about four minutes so I timed my pop-up to appear after two minutes.
And then it was time to tackle the copy. This is what my original pop-up looked like.
Yes, that copy definitely needed to be stronger. I thought about writing something like, “Want to boost your writing productivity? Get the free eBook, Famous Writers’ Productivity Hacks”. That’s the type of copy that most Internet marketers use on their sign-up forms.
But then I realized that people probably see hundreds of pop-ups like that every day. We’re getting bombarded with all kinds of freebies when we visit websites. How could I make mine stand out?
In order for my new pop-up to be unlike most pop-ups you see on the web, I decided to try injecting a little warmth and personality. My goal was for it not to look like an ad.
Here’s what my new pop-up looks like:
Yes, it’s a little wordy. But it captures my personality so much better than the first pop-up. It creates a much more human connection with my visitors and hopefully makes them smile when they read it.
The Results?
My original pop-up had a 0.6% conversion rate. The new pop-up right now is hovering around 5%. Wow, that’s a huge increase!
Of course, it’s a little difficult to track because someone who visits the site and sees a pop-up might already be a subscriber or subscribe through a different form. And this pop-up also hasn’t yet been shown to as many people as the first one has, but so far the conversion rate seems to be holding steady.
To be honest, I was afraid that no one would subscribe through the new pop-up since the copy was so different compared to what usually “works” on the web. Thus, any sign-ups through that form I’m considering a success.
I’m planning to continue experimenting and split test it (create two pop-ups with slightly different copy and see which one performs better), but right now I’m very happy with the results.
Best of all, I believe that the copy on this new pop-up will attract people who genuinely want to be on my email list and read my blog posts and aren’t only after the freebie.
Ultimately, when you use the purple cow principle in your marketing, you attract the people who truly resonate with your personality, writing style, and message while filtering out those who don’t.
The Takeaway
How will you implement the purple cow principle in your marketing efforts and writing this year? What steps can you take to make your work stand out on the Internet?
Maybe you can put a different spin on a topic everyone is writing about. Or maybe you can try updating your blog’s design. Or rewriting your bio or about page. Or astounding your readers with your generosity and creating incredible free resources for them.
Let me know in the comments how you will experiment with the purple cow principle this week.
As you craft your content and marketing strategy this year, keep the words of the famous adman Leo Burnett in mind,
“Make it simple. Make it memorable. Make it inviting to look at. Make it fun to read.”
Then you will truly be on your way to creating a purple cow that will delight and captivate your audience. | https://medium.com/copywriting-secrets/how-to-make-your-work-stand-out-online-7c61c3b223bc | ['Nicole Bianchi'] | 2019-10-23 05:48:36.367000+00:00 | ['Life Lessons', 'Writing', 'Business', 'Marketing', 'Creativity'] | Title Make Work Stand OnlineContent Internet crowded place Every day someone writes new blog post niche publishes new book Amazon genre sea content get people pay attention work remember shoe store mom used take little kid called MerryGoRound Shoes quite unlike store I’d ever merrygoround center store child could ride store also funny vending machine mechanical chicken inserted quarter vending machine you’d receive plastic egg special prize inside Yes begged mom quarter kid wouldn’t love shoe store like Whenever brother needed new shoe we’d always ask mom take u Sometimes we’d even ask take u didn’t need new shoe I’m sure lot shoe store near lived carried exact shoe MerryGoRound owner MerryGoRound flexed creative muscle found way make store unique unforgettable order capture people’s attention Internet thing look way make writing marketing strategy stand don’t get lost crowd Seth Godin coined term “purple cow” describe principle Purple Cow Principle 2003 TED talk Godin explained “…My parable you’re driving road see cow keep driving you’ve seen cow Cows invisible Cows boring Who’s going stop pull say — ‘Oh look cow’ Nobody cow purple…you’d notice mean cow purple you’d get bored thing that’s going decide get talked get done get changed get purchased get built ‘Is remarkable’ ‘remarkable’ really cool word think mean ‘neat’ also mean ‘worth making remark about’” purple cow principle doesn’t mean something crazy get attention article Guardian Godin clarified “Being noticed remarkable Running street naked get noticed won’t accomplish much It’s easy pull stunt useful” Rather implementing purple cow principle mean looking way remarkable order ensure longterm success fifteen minute Internet fame start course crafting excellent content proud Excellent content always stand sea mediocrity Second mean looking way draw unique skill strength innovate standard marketing strategy package content promote example blog post lead photo usually read shared blog post don’t lead photo Thus nearly every blog post nowadays lead photo blogger artistic bent take next level using photo handdrawn illustration rather stock photo That’s one way make content memorable also make work stand subtle way like simply changing line copy website Read find happened earlier month Used Purple Cow Principle Skyrocket Email Conversion Rate past year wanted update email popup form website I’d checked analytics noticed wasn’t performing well popup form one email signup form literally “popup” screen visit website Usually enter email close form continue scrolling Popups incredibly effective reminding visitor subscribe also annoying intrusive don’t want multiple popups site create popup visitor can’t click website slidein popup form appears discreetly bottom righthand corner website doesn’t prevent visitor continuing scroll read blog post It’s year since edited popup knew time update First decided change popup’s timing Previously I’d popup appear 30 second website wanted try something different felt wasn’t giving visitor enough time look around website read article get know reach point ready subscribe want virtually yelled popup first land website checked Google analytics account see average length time someone visited website four minute timed popup appear two minute time tackle copy original popup looked like Yes copy definitely needed stronger thought writing something like “Want boost writing productivity Get free eBook Famous Writers’ Productivity Hacks” That’s type copy Internet marketer use signup form realized people probably see hundred popups like every day We’re getting bombarded kind freebie visit website could make mine stand order new popup unlike popups see web decided try injecting little warmth personality goal look like ad Here’s new popup look like Yes it’s little wordy capture personality much better first popup creates much human connection visitor hopefully make smile read Results original popup 06 conversion rate new popup right hovering around 5 Wow that’s huge increase course it’s little difficult track someone visit site see popup might already subscriber subscribe different form popup also hasn’t yet shown many people first one far conversion rate seems holding steady honest afraid one would subscribe new popup since copy different compared usually “works” web Thus signups form I’m considering success I’m planning continue experimenting split test create two popups slightly different copy see one performs better right I’m happy result Best believe copy new popup attract people genuinely want email list read blog post aren’t freebie Ultimately use purple cow principle marketing attract people truly resonate personality writing style message filtering don’t Takeaway implement purple cow principle marketing effort writing year step take make work stand Internet Maybe put different spin topic everyone writing maybe try updating blog’s design rewriting bio page astounding reader generosity creating incredible free resource Let know comment experiment purple cow principle week craft content marketing strategy year keep word famous adman Leo Burnett mind “Make simple Make memorable Make inviting look Make fun read” truly way creating purple cow delight captivate audienceTags Life Lessons Writing Business Marketing Creativity |
703 | A Case for Writing on Your Phone | A Case for Writing on Your Phone
Turn your mobile distraction device into a productive writing resource
Photo by Chad Madden on Unsplash
How often do you use that little rectangle in your pocket for distractions? Scrolling social media, staying up to date with friends, wasting time while you’re waiting in line… We now live in an age where our phones satisfy that ‘boredom’ twitch. It fills the empty space between moments in our lives.
But it can provide a lot more than just a distraction from boredom (this is not the article where I make a case for embracing boredom).
Have you ever been out in public and been struck with the next greatest novel idea only to find that you have no notebook within reach? Or maybe you’re standing in line and only have a few seconds. What about when you’re on your dog walk in the morning and aren’t carrying writing supplies?
As much negative impact we claim that our phones create for us, they also provide us with a wonderful resource to improve our writing productivity. Why aren’t we taking advantage of our endless connectivity to produce something wonderful or improve our skills?
There are so many ways our phone can help us stay productive and improve our writing. Here are some ways you can set your phone up to stay writing and learning, even when you aren’t near a keyboard.
Listen to podcasts / audiobooks
Writing isn’t always about writing. In fact, many authors would argue it’s equally important to be reading and learning in addition to writing. Podcasts and audiobooks are right at your fingertips, providing you with free resources to improve your craft and learn from the best.
There are plenty of podcasts that cover the business, process, and even behind the scenes of writing. Look up interview podcasts and learn from veterans. Find a podcast that teaches you the business of writing and how to be successful. Find inspiration within a serial fiction podcast that teaches you storytelling.
On top of that, audiobooks are a great way to learn the craft and even study the storytelling techniques of experts. Libraries are adopting online catalogs between ebooks and audiobooks, which provide the perfect reason to grab a library card. Download audiobooks and take them wherever you go, whether you study via non-fiction or immerse yourself in the storytelling adventure of your favorite authors.
Learn to dictate
As crazy as it might feel to walk around speaking into your phone without talking to anyone, we’re writers in the end. We’re already crazy.
Dictation might feel foreign to you, but it’s a great way to increase your word count without much effort on your part. If you’re walking the dog or doing chores around the house, you might not be able to write things down. Dictation solves all of your problems. Even if you don’t have an app that transcribes your words, recording them and transcribing it later is a great way to improve your skills. Now, whenever you have that strike of inspiration or solve your plot hole and you’re not near your writing document, you can make a note.
Talking out your scenes is also a great opportunity to find your way through a blockage. If you don’t have a sounding board for new ideas or a way to work through tough scenes, dictation and talking through the possibilities can help you work out your problems.
Find small, wasted moments
Especially during NaNoWriMo, writers are constantly searching for time in their day to get the words written. Think about the small, wasted moments that could be better served writing and find a way to incorporate writing into them. Are you bored and wanting to scroll social media? Write in your notes app instead. Standing in line at the grocery store waiting to check out? Write a few sentences. Sitting at a restaurant waiting for your food? You guessed it. Little moments like that might not feel like they are very productive, but small bits do add up over time.
Break a sweat
This is by far my favorite writing habit — writing while working out. In 2018, I wrote about 8K words of my NaNo novel while sitting on the exercise bike and typing away at my phone. I didn’t quite have the guts to dictate my novel then, and also there were people around, but this year I’m looking forward to dictating while I’m in the privacy of my living room and spinning away. Not only are you getting a workout and staying healthy, you’re also getting those words in. Whether you’re doing NaNo or not, that’s a great chance to have thirty minutes of uninterrupted writing time.
Utilize commute times
When I was driving an hour to and from work every day, I could have passed that as wasted time that I’d never get back. Instead, I used that time to either listen to podcasts or work through novels via dictation. Commute times take up a lot of our work week, and there’s no rules against turning them into productive writing sprints. Figure out how you can best use that time, either by learning or writing, and use that to your advantage. | https://medium.com/swlh/a-case-for-writing-on-your-phone-5f7c597020cb | ['Laura Winter'] | 2020-10-28 23:04:25.470000+00:00 | ['Writers On Writing', 'Writing', 'Creativity', 'Writing Tips', 'Productivity'] | Title Case Writing PhoneContent Case Writing Phone Turn mobile distraction device productive writing resource Photo Chad Madden Unsplash often use little rectangle pocket distraction Scrolling social medium staying date friend wasting time you’re waiting line… live age phone satisfy ‘boredom’ twitch fill empty space moment life provide lot distraction boredom article make case embracing boredom ever public struck next greatest novel idea find notebook within reach maybe you’re standing line second you’re dog walk morning aren’t carrying writing supply much negative impact claim phone create u also provide u wonderful resource improve writing productivity aren’t taking advantage endless connectivity produce something wonderful improve skill many way phone help u stay productive improve writing way set phone stay writing learning even aren’t near keyboard Listen podcasts audiobooks Writing isn’t always writing fact many author would argue it’s equally important reading learning addition writing Podcasts audiobooks right fingertip providing free resource improve craft learn best plenty podcasts cover business process even behind scene writing Look interview podcasts learn veteran Find podcast teach business writing successful Find inspiration within serial fiction podcast teach storytelling top audiobooks great way learn craft even study storytelling technique expert Libraries adopting online catalog ebooks audiobooks provide perfect reason grab library card Download audiobooks take wherever go whether study via nonfiction immerse storytelling adventure favorite author Learn dictate crazy might feel walk around speaking phone without talking anyone we’re writer end We’re already crazy Dictation might feel foreign it’s great way increase word count without much effort part you’re walking dog chore around house might able write thing Dictation solves problem Even don’t app transcribes word recording transcribing later great way improve skill whenever strike inspiration solve plot hole you’re near writing document make note Talking scene also great opportunity find way blockage don’t sounding board new idea way work tough scene dictation talking possibility help work problem Find small wasted moment Especially NaNoWriMo writer constantly searching time day get word written Think small wasted moment could better served writing find way incorporate writing bored wanting scroll social medium Write note app instead Standing line grocery store waiting check Write sentence Sitting restaurant waiting food guessed Little moment like might feel like productive small bit add time Break sweat far favorite writing habit — writing working 2018 wrote 8K word NaNo novel sitting exercise bike typing away phone didn’t quite gut dictate novel also people around year I’m looking forward dictating I’m privacy living room spinning away getting workout staying healthy you’re also getting word Whether you’re NaNo that’s great chance thirty minute uninterrupted writing time Utilize commute time driving hour work every day could passed wasted time I’d never get back Instead used time either listen podcasts work novel via dictation Commute time take lot work week there’s rule turning productive writing sprint Figure best use time either learning writing use advantageTags Writers Writing Writing Creativity Writing Tips Productivity |
704 | 5 Quotes for Mental Balance and Clarity | In our quest of trying to have the most productive day, things can quickly spiral out of control. We need to reply to emails, we’ve got deadlines to meet, our family needs our attention, on and on our day unfolds, packed full of responsibilities.
Before we know it, things become too complicated. We wish we could just shut everything down and disappear into a quiet universe where we can just be alone; but for most of us, this is not a feasible thing to hope for.
What we can do, however, is put our minds in the right place. Since most of our suffering originates from our minds, maybe what we need the most is some mental clarity — an ability to enjoy life and to create a balance between life activities and efforts.
Here are five quotes about finding balance and clarity that have proven helpful for me and can also help you too if you apply them. | https://medium.com/age-of-awareness/5-quotes-for-mental-balance-and-clarity-2259bc66d5e5 | ['Destiny Femi'] | 2020-08-20 15:02:10.184000+00:00 | ['Self-awareness', 'Self', 'Mental Health', 'Work Life Balance', 'Productivity'] | Title 5 Quotes Mental Balance ClarityContent quest trying productive day thing quickly spiral control need reply email we’ve got deadline meet family need attention day unfolds packed full responsibility know thing become complicated wish could shut everything disappear quiet universe alone u feasible thing hope however put mind right place Since suffering originates mind maybe need mental clarity — ability enjoy life create balance life activity effort five quote finding balance clarity proven helpful also help apply themTags Selfawareness Self Mental Health Work Life Balance Productivity |
705 | How Boredom Can Be Beneficial to You, According to Science | What if I tell you that we have been thinking or taught about the feeling of boredom ALL WRONG!!!?
Well, there is no scientific definition of boredom. We experience that feeling very often but still, it’s difficult to define it. Think about it as if a person is least interested in its surroundings or as if he is indulged in somewhat repetitive tasks, then he will start catching the feeling of boredom.
Many of us think about that unknown feeling as not so beneficial but its a really important emotion that we tend to ignore very often.
The Apple co-founder once quoted,
“I’m a big believer in boredom. Boredom allows one to indulge in curiosity, and out of curiosity comes everything.” — Steve Jobs
Curiosity is curative to boredom. It keeps us alive by stimulating our brains.
Have not you come up with creative and purposeful ideas while taking shower or washing dishes?
“I’ve noticed that my best ideas always bubble up when the outside world fails in its primary job of frightening, wounding, or entertaining me.” — Scott Adams
In our daily routines when we are knocked by that feeling we tend to avoid it by counter-productive things like we put out our phones, television, music, we start spending time on social sites or other social distractions.
These smartphones are real enemies of positive and useful boredom. We indulge in them because we are afraid of boredom. The more we are tapping on our phones the less we are having time to tap on the power of our brain.
Our brain is offering us that strength for free and we are not utilizing it. Try avoiding mobile phones, especially on public transport or while waiting for someone. Give space to your thoughts to be heard out loud.
“Boredom is your window… Once this window opens, don’t try to shut it; on the contrary, throw it wide open.” — Joseph Brodsky
Scientific researchers at the University of South California had done a research survey on teenagers who use social media while being with their friends or while doing homework. They found out them to be less imaginative, thoughtful, and creative as compared to others in the span of two years.
This is the reason why this generation is not able to focus on big problems or even about their own future.
To tell you the truth, that social media connectivity thing won’t remain cool in a couple of years. | https://medium.com/an-idea/how-boredom-can-be-beneficial-to-you-according-to-science-57168c67c1d7 | ['Areeba Merriam'] | 2020-10-12 22:59:38.767000+00:00 | ['Life Lessons', 'Boredom', 'Creativity', 'Psychology', 'Science'] | Title Boredom Beneficial According ScienceContent tell thinking taught feeling boredom WRONG Well scientific definition boredom experience feeling often still it’s difficult define Think person least interested surroundings indulged somewhat repetitive task start catching feeling boredom Many u think unknown feeling beneficial really important emotion tend ignore often Apple cofounder quoted “I’m big believer boredom Boredom allows one indulge curiosity curiosity come everything” — Steve Jobs Curiosity curative boredom keep u alive stimulating brain come creative purposeful idea taking shower washing dish “I’ve noticed best idea always bubble outside world fails primary job frightening wounding entertaining me” — Scott Adams daily routine knocked feeling tend avoid counterproductive thing like put phone television music start spending time social site social distraction smartphones real enemy positive useful boredom indulge afraid boredom tapping phone le time tap power brain brain offering u strength free utilizing Try avoiding mobile phone especially public transport waiting someone Give space thought heard loud “Boredom window… window open don’t try shut contrary throw wide open” — Joseph Brodsky Scientific researcher University South California done research survey teenager use social medium friend homework found le imaginative thoughtful creative compared others span two year reason generation able focus big problem even future tell truth social medium connectivity thing won’t remain cool couple yearsTags Life Lessons Boredom Creativity Psychology Science |
706 | Serverless at Comic Relief | We built APIs on top of various databases, simplified our contact service, built a step counter service for UK schools participating in our billion steps challenge for Sport Relief, run a UK schools lookup service, and are currently converting our main donation platform to a serverless application.
From microservices to functions
Let’s take a step back and see what has happened. As we mentioned previously we’ve embarked on a journey into containers and microservices. Two years in, we are still big fans of microservices and service-oriented architecture. However, we also learned a couple of things on the way. For example, we do spend considerable time making our services scale to accommodate expected traffic peaks. As Comic Relief is in the business of fundraising, it is often really hard for us to predict traffic levels beforehand. As a result, we often over-scale and have infrastructure left idle.
Comic Relief runs some of the biggest fundraising events in the UK — Sport Relief and Red Nose Day — and due to the nature of those events, we have incredibly spiky traffic. We traditionally upscale in anticipation of those traffic spikes, but it is impractical or simply impossible to auto-scale our containers in very short timeframes. Working with functions provide us with a platform that does not require us to prepare for those traffic spikes, nor do we have infrastructure left unused. Instead, we can spend our time building logical components that execute business logic and not worry much about infrastructure nor scaling.
For us, serverless really is all about producing value faster, and not having to worry too much about everything else that comes along the developer’s path and often is nothing more than a distraction. One of the greatest things about invoking functions — and a real paradigm shifter — is that you pay for execution time, rather than for running the underlying infrastructure of your codebase. That direct relationship gives you an instant incentive to keep things performant and forces you to keep things simple (functions can typically not run longer than 15 minutes). The combination of hands-off management and a direct focus on functions that map on business logic proves to be a really powerful combination, and we’re not the only ones to think that.
Function-oriented architecture
Let’s illustrate that thinking by looking at how we recently rebuilt our contact service. Previously, we’d have our contact forms as an integral part of our CMS software (Drupal). We factored this out into a separate decoupled service: a front-end (built in ReactJS) that captures user data, and a back-end that receives that data and adds it to a message queue. We then pick up that data item from the queue and pass it on to our CRM. We also drop a message in another queue that is consumed by a service that will send back a confirmation mail to our user. In our microservices world, we’d build the back-end bits in Symfony or Slim (a micro-framework in PHP), and host it on self-managed infrastructure.
In our new serverless world, we structure that logic around functions:
One function to receives user data (via POST) and drops a message into a CRM and an email queue
One function that consumes the CRM queue and passes data to our CRM
One function that consumes the email queue and sends back a welcome email via our email service (another example of a serverless service).
None of those requires us to manage underlying infrastructure nor scale up for moments in which we expect lots of traffic through our contact platform.
Establishing our development patterns in a world of functions
We detected a pattern we now use for all our serverless services.
A node module for each functional component or service. We really like Javascript, but are looking at Go as well, simply because it runs faster and thus will save us money. Using Javascript both in the front-end and the back-end has the additional advantage that we can reduce the amount of languages our engineering team specialises in.
A set of functions that map to functional parts of our node module. Those can be anything from API callbacks (via API Gateway) to cron-based functions.
The incredible serverless framework to handle provisioning of our functions to our function provider. We currently prefer AWS Lambda, simply because it ties in well with the rest of our stack.
Concourse CI to deploy functions to a staging environment, run tests and follow up with a production deployment. Whilst the serverless framework does the heavy lifting for you (packaging your code, uploading to S3, creating cloud formation stacks), we prefer to automate the entire deployment process so we can really just focus on writing our logic into functions.
If we need to handle persistent data (we prefer queues where possible to decouple and remove the direct dependency on a database layer), we rely on Elastic Search or DynamoDB, both managed services provided by AWS.
For all our front-end needs, we follow a pretty standard component-based architecture powered by our pattern lab and ReactJS storybook.
Looking forward
Good examples of serverless implementations within an enterprise setting are unfortunately still hard to come by. However, as we see this new exciting technology being adopted, we expect to see many more use cases being published. We hope to be at the forefront of this and will be sharing as much as we can through our public Github.
Cross-posted from the Comic Relief Technology Blog | https://medium.com/comic-relief/serverless-at-comic-relief-eaf3ff25cbfe | ['Peter Vanhee'] | 2019-04-03 11:06:20.895000+00:00 | ['Programming', 'Serverless', 'Cloud Computing', 'AWS', 'Engineering'] | Title Serverless Comic ReliefContent built APIs top various database simplified contact service built step counter service UK school participating billion step challenge Sport Relief run UK school lookup service currently converting main donation platform serverless application microservices function Let’s take step back see happened mentioned previously we’ve embarked journey container microservices Two year still big fan microservices serviceoriented architecture However also learned couple thing way example spend considerable time making service scale accommodate expected traffic peak Comic Relief business fundraising often really hard u predict traffic level beforehand result often overscale infrastructure left idle Comic Relief run biggest fundraising event UK — Sport Relief Red Nose Day — due nature event incredibly spiky traffic traditionally upscale anticipation traffic spike impractical simply impossible autoscale container short timeframes Working function provide u platform require u prepare traffic spike infrastructure left unused Instead spend time building logical component execute business logic worry much infrastructure scaling u serverless really producing value faster worry much everything else come along developer’s path often nothing distraction One greatest thing invoking function — real paradigm shifter — pay execution time rather running underlying infrastructure codebase direct relationship give instant incentive keep thing performant force keep thing simple function typically run longer 15 minute combination handsoff management direct focus function map business logic prof really powerful combination we’re one think Functionoriented architecture Let’s illustrate thinking looking recently rebuilt contact service Previously we’d contact form integral part CMS software Drupal factored separate decoupled service frontend built ReactJS capture user data backend receives data add message queue pick data item queue pas CRM also drop message another queue consumed service send back confirmation mail user microservices world we’d build backend bit Symfony Slim microframework PHP host selfmanaged infrastructure new serverless world structure logic around function One function receives user data via POST drop message CRM email queue One function consumes CRM queue pass data CRM One function consumes email queue sends back welcome email via email service another example serverless service None requires u manage underlying infrastructure scale moment expect lot traffic contact platform Establishing development pattern world function detected pattern use serverless service node module functional component service really like Javascript looking Go well simply run faster thus save u money Using Javascript frontend backend additional advantage reduce amount language engineering team specialises set function map functional part node module anything API callback via API Gateway cronbased function incredible serverless framework handle provisioning function function provider currently prefer AWS Lambda simply tie well rest stack Concourse CI deploy function staging environment run test follow production deployment Whilst serverless framework heavy lifting packaging code uploading S3 creating cloud formation stack prefer automate entire deployment process really focus writing logic function need handle persistent data prefer queue possible decouple remove direct dependency database layer rely Elastic Search DynamoDB managed service provided AWS frontend need follow pretty standard componentbased architecture powered pattern lab ReactJS storybook Looking forward Good example serverless implementation within enterprise setting unfortunately still hard come However see new exciting technology adopted expect see many use case published hope forefront sharing much public Github Crossposted Comic Relief Technology BlogTags Programming Serverless Cloud Computing AWS Engineering |
707 | Why Java Is Dying | Why Java Is Dying
What does the future hold for Java?
Photo by Jonas Jacobsson on Unsplash.
To better align with the tech stack at my new job, I’ve been reacquainting myself over the last two weeks with an old friend: Java. Not too long ago, it kickstarted my software career with unmatched zeal and vigor. This journey lasted for roughly two-and-a-half years but soon diminished with the advent of containers and microservices. Fast-forward to today, and it’s been three years now since I last wrote any serious Java code. And in all honesty, I never expected it to see it again — especially in the microservices landscape.
So, what happened? The answer is simple: The pervasive wave of microservices engulfed us.
Ease of scaling
High availability
Simpler code base from not having to worry about concurrency and multi-threading
Portability benefits from containerization
All of these factors prompted us to question the efficacy of Java (more specifically, JVM), not to mention its most notorious framework, Spring.
At times, with hands immersed in tech like Kubernetes, it felt like Java’s time was almost over and that it didn’t hold up well in the ecosystem of containers and microservices (which was key to software scalability and high availability). But as a once-staunch supporter of Java — and despite having been swayed by the simplicity and elegance of languages like Python (which has now grown to become my go-to language) — I still continued to hold a reserve for some of Java’s undeniable merits.
For instance, I was well aware of Java’s powerful threading capabilities, having used them firsthand early in my career for critical banking applications. While it’s not fair to compare performance metrics of a compiled language with scripted languages, Java’s rock-solid performance was simply unmatched.
But in the world of horizontal scalability and microservices architecture, this innate performance of a language mattered too little, for one could simply spawn more containers to achieve far superior performance. Clearly, these scripting languages, combined with their ability to instantly scale up or down in the land of containers, were giving Java a good run for its money. I was fully convinced, once and for all, that Java was done (at least in the land of microservices). And I was right!
At my new job, these convictions have only further strengthened and painfully made me realize how irksome, irritable, and convoluted the language has become — partly due to anachronistic ceremonious frameworks like Spring. | https://medium.com/better-programming/why-java-is-dying-b02b5fd44db9 | ['Komal Venkatesh Ganesan'] | 2020-12-18 01:54:03.802000+00:00 | ['Java', 'Programming', 'Software Development', 'Software Engineering', 'Microservices'] | Title Java DyingContent Java Dying future hold Java Photo Jonas Jacobsson Unsplash better align tech stack new job I’ve reacquainting last two week old friend Java long ago kickstarted software career unmatched zeal vigor journey lasted roughly twoandahalf year soon diminished advent container microservices Fastforward today it’s three year since last wrote serious Java code honesty never expected see — especially microservices landscape happened answer simple pervasive wave microservices engulfed u Ease scaling High availability Simpler code base worry concurrency multithreading Portability benefit containerization factor prompted u question efficacy Java specifically JVM mention notorious framework Spring time hand immersed tech like Kubernetes felt like Java’s time almost didn’t hold well ecosystem container microservices key software scalability high availability oncestaunch supporter Java — despite swayed simplicity elegance language like Python grown become goto language — still continued hold reserve Java’s undeniable merit instance well aware Java’s powerful threading capability used firsthand early career critical banking application it’s fair compare performance metric compiled language scripted language Java’s rocksolid performance simply unmatched world horizontal scalability microservices architecture innate performance language mattered little one could simply spawn container achieve far superior performance Clearly scripting language combined ability instantly scale land container giving Java good run money fully convinced Java done least land microservices right new job conviction strengthened painfully made realize irksome irritable convoluted language become — partly due anachronistic ceremonious framework like SpringTags Java Programming Software Development Software Engineering Microservices |
708 | Your Brain On Chaos: Debunking The Myth Of Tortured Artist | It’s no surprise that the words tortured and artist are forever connected. Many of history’s favorite artists are as well-known for their inner turmoil as they are for their work. Jackson Pollock’s paint-drip technique is forever attached to his alcoholism and rage. Sylvia Plath is remembered as the “tormented poet who stuck her head in the oven.” And you can’t say the words tortured artist without Vincent van Gogh coming to mind — a visionary so wracked with paranoia he chopped off his own ear.
The list of artists, writers, and musicians who landed in an early grave is so long and haunting that it’s no wonder people believe suffering is the best fuel for a creative fire. It’s true that a creative practice can help untangle some of the existential dread that comes with existence. However, the notion that turmoil is a prerequisite for being artistic is a pathetic myth.
Our ancient human brains still interpret stress and anxiety as danger. When mental anguish arrives, our stress hormones tell the brain to redirect resources toward basic survival. The limbic system (the part of the brain responsible for creative thinking) slows down and our reptilian default of fight-or-flight takes over. And since our hardwiring hasn’t updated with all the ways of the modern world, these response systems have a hard time distinguishing the difference between a hungry leopard pouncing out of a tree and an angry spouse shouting up the stairs.
Ever notice how your heart beats faster when anxiety arrives? That’s an increase in blood pressure. It’s not there to help you write your screenplay, prepare the presentation, or cook a nice dinner — it’s there to help you outrun the leopard.
Just because creativity is inherently human doesn’t make it a necessity for survival. If the brain is constantly being toggled between fight-or-flight mode, creativity is going to get pushed to the backseat. Self-actualization is at the top of Maslow’s Hierarchy of Needs for a reason. If you don’t know where your next meal is coming from, “staying inspired” isn’t going to register as an evolutionary imperative. If you’re out drinking every night, living the “life of an artist,” when will you have the energy to bring your inspiration to life? Hangovers, emotional roller coasters, physical exhaustion — these are not “tools of the artist.” These are the things that will almost always stand between an artist and their best work.
The filmmaker, David Lynch, says of van Gogh in particular, “People might bring up Vincent van Gogh as an example of a painter who did great work in spite of, or because of his suffering. I like to think that van Gogh would have been even more prolific and even greater if he wasn’t so restricted by the things tormenting him. I don’t think it was pain that made him so great, I think painting brought him whatever happiness he had.”
Slicing off an ear or sticking a head in the oven are dramatic examples. But how often do we stick our head so deep in our work that we forget to breathe? How many hours of tunnel vision does it take before we lose sight of what we’re doing and why we’re doing it? Overwork and burnout torment the mind and restrict great work from being made just as often as these extreme cases of chaos.
We have to ask ourselves: Is our life’s purpose to create things that leave us worn down and weary? Is that what life is asking of us? To become martyrs to our own productivity?
We don’t know how many more beautiful paintings or classic novels the world might have if some of our greatest creatives had been able to outrun their chaos. But I think it’s safe to say that their sons and daughters, husbands and wives, would gladly trade any piece of art for one more day with their favorite artist. | https://medium.com/publishous/your-brain-on-chaos-debunking-the-myth-of-tortured-artist-c75502c014ce | ['Corey Mccomb'] | 2020-12-01 02:41:23.050000+00:00 | ['Writing', 'Psychology', 'Self', 'Personal Development', 'Creativity'] | Title Brain Chaos Debunking Myth Tortured ArtistContent It’s surprise word tortured artist forever connected Many history’s favorite artist wellknown inner turmoil work Jackson Pollock’s paintdrip technique forever attached alcoholism rage Sylvia Plath remembered “tormented poet stuck head oven” can’t say word tortured artist without Vincent van Gogh coming mind — visionary wracked paranoia chopped ear list artist writer musician landed early grave long haunting it’s wonder people believe suffering best fuel creative fire It’s true creative practice help untangle existential dread come existence However notion turmoil prerequisite artistic pathetic myth ancient human brain still interpret stress anxiety danger mental anguish arrives stress hormone tell brain redirect resource toward basic survival limbic system part brain responsible creative thinking slows reptilian default fightorflight take since hardwiring hasn’t updated way modern world response system hard time distinguishing difference hungry leopard pouncing tree angry spouse shouting stair Ever notice heart beat faster anxiety arrives That’s increase blood pressure It’s help write screenplay prepare presentation cook nice dinner — it’s help outrun leopard creativity inherently human doesn’t make necessity survival brain constantly toggled fightorflight mode creativity going get pushed backseat Selfactualization top Maslow’s Hierarchy Needs reason don’t know next meal coming “staying inspired” isn’t going register evolutionary imperative you’re drinking every night living “life artist” energy bring inspiration life Hangovers emotional roller coaster physical exhaustion — “tools artist” thing almost always stand artist best work filmmaker David Lynch say van Gogh particular “People might bring Vincent van Gogh example painter great work spite suffering like think van Gogh would even prolific even greater wasn’t restricted thing tormenting don’t think pain made great think painting brought whatever happiness had” Slicing ear sticking head oven dramatic example often stick head deep work forget breathe many hour tunnel vision take lose sight we’re we’re Overwork burnout torment mind restrict great work made often extreme case chaos ask life’s purpose create thing leave u worn weary life asking u become martyr productivity don’t know many beautiful painting classic novel world might greatest creatives able outrun chaos think it’s safe say son daughter husband wife would gladly trade piece art one day favorite artistTags Writing Psychology Self Personal Development Creativity |
709 | Use the Decoy Effect to Sell More | How to Apply the Decoy Effect
You can apply the effect to your marketing — it could help you increase your profits and sales of some products. You would first determine what thing you want to sell more of. This is the “target” product. Then, you would create an option that is worse than that option, but not worse than all the options.
An example for SaaS companies
For example, imagine you are a SaaS company. You sell have three subscription options:
Basic: for $20 a month, the customer gets four product features.
Professional: for $100 a month, the customer gets eight product features.
Enterprise: for $200 a month, the customer gets 12 product features.
Imagine you want to increase the proportion of your customers choosing the Professional option instead of the Basic option. The target product is the Professional subscription option. You could add a decoy to the product spread that makes the Professional subscription look more valuable. So, for example, you might add a “Professional lite” subscription:
Basic: for $20 a month, the customer gets four product features.
Professional Lite: For $95 a month, the customer gets five product features.
Professional: For $100 a month, the customer gets eight product features.
Enterprise: for $200 a month, the customer gets 12 product features.
The Professional Lite option is a decoy because it’s inferior to the Professional option. It doesn’t give you many extra features, but it costs almost the same as the Professional. Customers would be more likely to go for the Professional option because they can see that it adds quite a bit more value for not much extra money.
Now imagine your target is the Enterprise subscription. You could change the options such that it appears to be more valuable in comparison:
Basic: for $20 a month, the customer gets four product features.
Professional: For $100 a month, the customer gets eight product features.
Enterprise Lite: For $195 a month, the customer gets nine product features.
Enterprise: for $200 a month, the customer gets 12 product features.
Adding the Enterprise Lite option highlights the value of the Enterprise option because you can get many more features for only $5 extra a month. It makes that option much more attractive to customers. Including the Enterprise Lite option would likely increase the proportion of customers who would purchase the Enterprise option.
Other applications
There are tons of other contexts in which you could use this effect.
Coffee. You sell three sizes: small, medium, and large. Your target is the large size option. Make the price of the medium size close to the large size, and you will probably start selling more of the large size.
Real estate. You are a realtor and you want to highlight the value of a particular house to a client. Emphasize the value by creating comparisons to two options that are not as good; perhaps two houses that are the same price or more expensive but with fewer amenities or worse locations.
Freelance writers. You sell blog posts, email copy, and other content marketing. Your target is your “SEO Content Strategy” bundle. Choose pricing bundles such that you include a “decoy” or an unattractive option. It could look something like this:
Single Blog Post: Get one blog article for $250.
Blog Post Bundle: Get five blog articles for $1,200.
SEO Content Strategy: Get a full SEO site audit, content strategy, calendar, and five articles for $1,300.
In this case, you would likely sell more of the “SEO Content Strategy” option because it’s not much more than the “Blog Post Bundle” and comes with much more content. | https://medium.com/better-marketing/use-the-decoy-effect-to-sell-more-2d475559c7f1 | ['Ramsay Lewis'] | 2020-05-28 17:46:10.535000+00:00 | ['Marketing', 'Business', 'Startups', 'Psychology', 'Science'] | Title Use Decoy Effect Sell MoreContent Apply Decoy Effect apply effect marketing — could help increase profit sale product would first determine thing want sell “target” product would create option worse option worse option example SaaS company example imagine SaaS company sell three subscription option Basic 20 month customer get four product feature Professional 100 month customer get eight product feature Enterprise 200 month customer get 12 product feature Imagine want increase proportion customer choosing Professional option instead Basic option target product Professional subscription option could add decoy product spread make Professional subscription look valuable example might add “Professional lite” subscription Basic 20 month customer get four product feature Professional Lite 95 month customer get five product feature Professional 100 month customer get eight product feature Enterprise 200 month customer get 12 product feature Professional Lite option decoy it’s inferior Professional option doesn’t give many extra feature cost almost Professional Customers would likely go Professional option see add quite bit value much extra money imagine target Enterprise subscription could change option appears valuable comparison Basic 20 month customer get four product feature Professional 100 month customer get eight product feature Enterprise Lite 195 month customer get nine product feature Enterprise 200 month customer get 12 product feature Adding Enterprise Lite option highlight value Enterprise option get many feature 5 extra month make option much attractive customer Including Enterprise Lite option would likely increase proportion customer would purchase Enterprise option application ton context could use effect Coffee sell three size small medium large target large size option Make price medium size close large size probably start selling large size Real estate realtor want highlight value particular house client Emphasize value creating comparison two option good perhaps two house price expensive fewer amenity worse location Freelance writer sell blog post email copy content marketing target “SEO Content Strategy” bundle Choose pricing bundle include “decoy” unattractive option could look something like Single Blog Post Get one blog article 250 Blog Post Bundle Get five blog article 1200 SEO Content Strategy Get full SEO site audit content strategy calendar five article 1300 case would likely sell “SEO Content Strategy” option it’s much “Blog Post Bundle” come much contentTags Marketing Business Startups Psychology Science |
710 | How to build an MP3 Music Player with Python | Part I: Set the environment and import modules
The first step is to create a directory called environments where to store all your virtual Environments and install the python modules. In this way the program is available only within the set environment and not throughout my computer.
So, open your terminal and create a directory as below.
mkdir Environments
And be sure to be into the directory ‘Environments’.
cd C:\Users\moryb\Environments
The second step is to use a module virtualenv to create isolated virtual environments.
pip install virtualenv
Now, we can create a virtual environment named myenv. If you have installed python 3 you can easily digit:
virtualenv myenv
Inside the directory myenv there is a copy of Python interpreter, the standard library and the various supporting files.
We can activate the environment with the following command:
myenv\Scripts\activate.bat
out:
(myenv) C:\Users\moryb\Environments>
Then, we need to install Pygame. Pygame is a module that is used for creating videogames. For creating our MP3 Music Player we will use the sound component.
pip install pygame
In this second phase, I will use Atom as python editor.
So I will import the following modules. | https://towardsdatascience.com/how-to-build-an-mp3-music-player-with-python-619e0c0dcee2 | ['Papa Moryba Kouate'] | 2020-12-26 19:22:34.600000+00:00 | ['Programming', 'Music', 'Data Science', 'Python', 'Development'] | Title build MP3 Music Player PythonContent Part Set environment import module first step create directory called environment store virtual Environments install python module way program available within set environment throughout computer open terminal create directory mkdir Environments sure directory ‘Environments’ cd CUsersmorybEnvironments second step use module virtualenv create isolated virtual environment pip install virtualenv create virtual environment named myenv installed python 3 easily digit virtualenv myenv Inside directory myenv copy Python interpreter standard library various supporting file activate environment following command myenvScriptsactivatebat myenv CUsersmorybEnvironments need install Pygame Pygame module used creating videogames creating MP3 Music Player use sound component pip install pygame second phase use Atom python editor import following modulesTags Programming Music Data Science Python Development |
711 | Creating your own Chaos Monkey with AWS Systems Manager Automation | In a recent post, I explained how to use AWS SSM Run Command to inject failures on EC2 instances. SSM Run Command is well-suited to execute custom scripts on EC2 instances, especially to inject latency or blackouts on the network interface, do resource exhaustion of CPUs, memory, and IO.
However, we need more than that. Failure injection should target resources, network characteristics and dependencies, applications, processes and service, and also the infrastructure.
We also need to have a broad set of controls and capabilities to perform chaos experiments safely. We might want to:
Execute commands and scripts directly into EC2 instances.
directly into EC2 instances. Invoke Lambda functions to run custom scripts.
Orchestrate several failure injections to form chaos scenarios.
several failure injections to form chaos scenarios. Schedule them for execution at specific times.
them for execution at specific times. Have automatic cancellations if errors are detected.
if errors are detected. Have safety measures in places with approvals.
in places with approvals. Apply velocity controls to limit the blast radius of experiments.
That is where AWS System Manager Automation (SSM** Automation) comes in. So, let’s take a look!
** Note: AWS Systems Manager was formerly known as Amazon Simple Systems Manager (SSM). The original abbreviated name of the service, SSM, is still used and reflected in various AWS resources. | https://medium.com/the-cloud-architect/creating-your-own-chaos-monkey-with-aws-systems-manager-automation-6ad2b06acf20 | ['Adrian Hornsby'] | 2020-06-15 09:59:45.584000+00:00 | ['Cloud Computing', 'Software Engineering', 'Fault Tolerance', 'Chaos Engineering', 'AWS'] | Title Creating Chaos Monkey AWS Systems Manager AutomationContent recent post explained use AWS SSM Run Command inject failure EC2 instance SSM Run Command wellsuited execute custom script EC2 instance especially inject latency blackout network interface resource exhaustion CPUs memory IO However need Failure injection target resource network characteristic dependency application process service also infrastructure also need broad set control capability perform chaos experiment safely might want Execute command script directly EC2 instance directly EC2 instance Invoke Lambda function run custom script Orchestrate several failure injection form chaos scenario several failure injection form chaos scenario Schedule execution specific time execution specific time automatic cancellation error detected error detected safety measure place approval place approval Apply velocity control limit blast radius experiment AWS System Manager Automation SSM Automation come let’s take look Note AWS Systems Manager formerly known Amazon Simple Systems Manager SSM original abbreviated name service SSM still used reflected various AWS resourcesTags Cloud Computing Software Engineering Fault Tolerance Chaos Engineering AWS |
712 | Just a Gen X Girl in a Millennial World | Something occurred to me the other day that made my head spin.
I’m pretty sure that I’ve entered some kind of zone of invisibility.
No, wait, I’m serious. Stay with me here.
Okay, first of all, I was born in 1971.
That makes me solidly part of Generation X.
Everyone knows that Gen X is the forgotten generation. That’s so known that it’s a freaking meme, right?
Like one minute it’s all about the Baby Boomers and then all of a sudden it’s all about Millennials.
And meanwhile, somehow I got stuck in the middle taking care of everyone.
I’ll be 48 in October, which seems to be like some kind of magical invisibility age. I’ll be standing right there and people just look through me, I swear. I feel like I need to pinch them by the cheeks and turn their face to me to make them look at me.
I’m at the door of being middle-aged and I’m a woman, which seems to make things even worse.
I looked up from working the other day and realized something. There are almost no women my age doing what I do.Very few women in their 40s or 50s write about productivity or about writing in general or are in the online business space.
There are some Millennial women. There are men my age, especially if they started younger. And there are most definitely Millennial men. In fact, this space is dominated by men in their late 20s and early 30s.
But Gen X women? Not so much.
Is that because there aren’t many who want to do this work?
Or is it because most just got tired of not being listened to?
Or are they out there and just actually invisible, so I can’t see them?
Oh. And I’m fat.
Everyone knows that being fat makes you invisible.
Taking up more space should make you more visible, but it doesn’t. I have a good friend who is much heavier than I am. I go to the doctor with him for support sometimes and I’m not joking, his doctors often ignore him and talk to me.
It’s strange and uncomfortable.
So, I’m a fat newly-middle-aged woman. And that’s given me the dubious superpower (I guess) of being able to walk through the world without being detected.
If I want to.
And if I don’t want to — I have to figure out how not to. This is not an easy superpower to turn off.
When I was twelve, my dad took me to see The Tap Dance Kid in Los Angeles.
When the kid’s sister, Emma, sang Four Strikes Against Me, I was absolutely riveted. I identified with her and this song so strongly. I begged my dad to buy me the album and I played the song over and over again.
That memory has stuck with me for more than 30 years.
Emma is young, female, black, and fat — her four strikes. She’s an older sister who nobody pays attention to and she feels unseen. I felt her when I was twelve, all the way down to my bones.
Today, I’m middle-aged, female, and fat. And I still feel Emma.
I belong to an informal group of writers who mostly all write in the same space that I do.
I hadn’t noticed before that I am the only person in the group who is a Gen X woman. Most of the group is Millennial men. There are a couple of other women, but they’re younger than me by at least ten years.
So, I posted something in this group about being frustrated about not getting the results I wanted lately. It seems like I have been writing into a kind of void.
My own followers, the people I’ve cultivated and nurtured and who have been with me a long time — they see me. But something has shifted lately. I can feel it.
I just couldn’t figure it out. So I asked this group: what am I doing wrong? Why is your work getting so much more attention than mine?
And then it hit me.
I’m not doing anything wrong.
I’ve just entered this middle-aged woman invisibility zone in a place that’s mostly populated by writers who could be my sons.
I’m not complaining. I know it might sound like I am, but I swear I’m not.
I’m kind of excited. Because now that I’ve figured this shit out, I can work with it.
You know the movie Ghost? (You might have to be a middle-aged woman to get the reference . . . )
Patrick Swayze’s character has died, but his spirit is still around and he wants to connect with Demi Moore’s character, because she’s his wife and someone’s trying to kill her.
He jumps around like a lunatic. Screams. Begs her to see him. None of it works. Because he’s invisible. She can’t see him, so none of that stuff will ever work, no matter how much of it he does.
He could spend eternity doing the things that people do to get attention and it would never, ever work.
He can’t get her to see him until he figures out a way to make his wife see him for what he is. He has to learn how to get her to notice him as a ghost.
And that’s what I realized this week.
Doing things the same way these Millennial guys do them will never work for me. Not really. Not the way I want them to. And trying to make that happen is threatening my sanity.
But I can figure out how to be seen just the way I am.
Emma Sheridan in The Tap Dance kid ends Four Strikes Against Me by belting out — “No one’s going to stop me. No one’s going to top me. You’ll just have to wait and see what tomorrow brings. The game isn’t over until the fat lady sings.”
That about sums it up.
So, now what?
Patrick Swayze goes out and finds Whoopie Goldberg in Ghost— a fake medium who turns out to be a real medium who can hear him. She talks to his wife for him.
I’m not sure what the answer is for me.
Maybe being louder. A little more boisterous; a little less ladylike. It’s hard to admit, but I’m sure that there are ways that I comport myself that speak to my upbringing and not wanting to stand out.
How much of being invisible is my own fault?
The result of a lifetime of education designed to hammer me down and turn me into a standard adult that doesn’t stand out from all the other adults.
The result of nearly five decades of living in a world that has not turned out very many women who are speaking about the things I want to talk about.
I have legit considered an experiment that involves inventing a Millennial hipster man persona to write under. But . . . eh. I like being me.
I’ve changed my life in the last three years. I’ve started a business that took me from a $9 an hour day job to a six-figure income. I’ve sold two novels to a major publisher. I’ve even lost more than 100 pounds.
I’ve got something to say. I’ll figure out a way to be heard.
Yeah. This is going to be fun. | https://medium.com/the-write-brain/just-a-gen-x-girl-in-a-millennial-world-cc5dc34b196a | ['Shaunta Grimes'] | 2019-09-08 02:17:31.673000+00:00 | ['Self', 'Creativity', 'Writing', 'Life', 'Marketing'] | Title Gen X Girl Millennial WorldContent Something occurred day made head spin I’m pretty sure I’ve entered kind zone invisibility wait I’m serious Stay Okay first born 1971 make solidly part Generation X Everyone know Gen X forgotten generation That’s known it’s freaking meme right Like one minute it’s Baby Boomers sudden it’s Millennials meanwhile somehow got stuck middle taking care everyone I’ll 48 October seems like kind magical invisibility age I’ll standing right people look swear feel like need pinch cheek turn face make look I’m door middleaged I’m woman seems make thing even worse looked working day realized something almost woman age doVery woman 40 50 write productivity writing general online business space Millennial woman men age especially started younger definitely Millennial men fact space dominated men late 20 early 30 Gen X woman much aren’t many want work got tired listened actually invisible can’t see Oh I’m fat Everyone know fat make invisible Taking space make visible doesn’t good friend much heavier go doctor support sometimes I’m joking doctor often ignore talk It’s strange uncomfortable I’m fat newlymiddleaged woman that’s given dubious superpower guess able walk world without detected want don’t want — figure easy superpower turn twelve dad took see Tap Dance Kid Los Angeles kid’s sister Emma sang Four Strikes absolutely riveted identified song strongly begged dad buy album played song memory stuck 30 year Emma young female black fat — four strike She’s older sister nobody pay attention feel unseen felt twelve way bone Today I’m middleaged female fat still feel Emma belong informal group writer mostly write space hadn’t noticed person group Gen X woman group Millennial men couple woman they’re younger least ten year posted something group frustrated getting result wanted lately seems like writing kind void follower people I’ve cultivated nurtured long time — see something shifted lately feel couldn’t figure asked group wrong work getting much attention mine hit I’m anything wrong I’ve entered middleaged woman invisibility zone place that’s mostly populated writer could son I’m complaining know might sound like swear I’m I’m kind excited I’ve figured shit work know movie Ghost might middleaged woman get reference Patrick Swayze’s character died spirit still around want connect Demi Moore’s character she’s wife someone’s trying kill jump around like lunatic Screams Begs see None work he’s invisible can’t see none stuff ever work matter much could spend eternity thing people get attention would never ever work can’t get see figure way make wife see learn get notice ghost that’s realized week thing way Millennial guy never work really way want trying make happen threatening sanity figure seen way Emma Sheridan Tap Dance kid end Four Strikes belting — “No one’s going stop one’s going top You’ll wait see tomorrow brings game isn’t fat lady sings” sum Patrick Swayze go find Whoopie Goldberg Ghost— fake medium turn real medium hear talk wife I’m sure answer Maybe louder little boisterous little le ladylike It’s hard admit I’m sure way comport speak upbringing wanting stand much invisible fault result lifetime education designed hammer turn standard adult doesn’t stand adult result nearly five decade living world turned many woman speaking thing want talk legit considered experiment involves inventing Millennial hipster man persona write eh like I’ve changed life last three year I’ve started business took 9 hour day job sixfigure income I’ve sold two novel major publisher I’ve even lost 100 pound I’ve got something say I’ll figure way heard Yeah going funTags Self Creativity Writing Life Marketing |
713 | When Tics, OCD, or Anorexia Develop Following Strep Infection | Infection | Brain
When Tics, OCD, or Anorexia Develop Following Strep Infection
It’s called PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections).
In the late 1900s, physicians reported that children started showing tics following a strep sore throat outbreak. The tics also disappeared when strep is killed by antibiotics.
A 2007 Case Report
An 8-year-old boy with recurrent strep sore throat developed a set of rituals before eating. He nodded, flapped his arms, tapped his mount, opened and closed his fists, etc. before taking a bite. He said the rituals helped him cope with intrusive thoughts that the food is poisoned. He became severely conscious with healthy eating and body fat, avoiding all ‘bad’ foods and eating only 200 calories/day — eventually losing 8kg. After receiving treatment for his recurrent sore throat, all his behavioural anomalies vanished.
The boy had PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections), obsessive-compulsive disorder (OCD), tics and anorexia nervosa.
PANDAS: What you need to know
Diagnosis: It’s diagnosed when a child develops tics or OCD right after a strep infection. Or when symptoms of tics/OCD worsen following a strep infection. The DSM-5 put PANDAS under “Obsessive-Compulsive and Related Disorders Due to Another Medical Condition.” While PANDAS is more closely associated with tic and OCD, it may include anorexia in some cases.
Cause: Nearly all children experienced strep throat infection at least once in life. The average schoolchild gets it 2–3x per year. It’s caused by a contagious bacterium called Streptococcus pyogenes. A strep sore throat infection can clear up on its own or with simple antibiotics. But in a small subset of children, perhaps with genetics and cultural predispositions, it can lead to PANDAS.
Statistics: Due to the lack of diagnostic guidelines, its precise prevalence is still unknown. Scientists, however, estimated that PANDAS contribute to at least 10% of childhood OCD and tic disorders.
Treatment: Usually with antibiotics and cognitive behavioural therapy. Steroids and antidepressants are prescribed if deemed necessary, though antidepressants often cause side effects in children with PANDAS. The best treatment is aggressive antibiotics — with about 50% cure rate according to a 2017 population study.
Prognosis: Long-term outcomes of PANDAS vary between cases due to the variability of PANDAS subtypes and treatment protocols. “Whether children with a PANDAS subtype typically will go on to remission or progress to a more chronic course of illness is not known,” says Tanya Murphy, MD, professor and director at the Rothman Center for Pediatric Neuropsychiatry, Florida.
If Strep infection is recurrent, for example, long-term antibiotics until adulthood is necessary. Even if Strep is eliminated, some may still suffer neuropsychiatric symptoms as the brain (or more specifically, the basal ganglia) is sufficiently compromised. Some children recover naturally too as the immune system matures and can distinguish Strep from the brain.
Basal Ganglia
Tic is viewed as the most common form of movement disorder in children. Tics are sudden and repetitive movement (e.g., shoulder shrugging, blinking, or head jerking) or vocalizations (e.g., sniffing or grunting). Tics can be short-term (11–20% prevalence) or long-tern (0.5–0.7% prevalence). Whereas OCD affects 1–3% in the general population. It’s characterized by persistent unwanted thoughts (obsessions) and/or repetitive behaviours driven by strong urges (compulsions).
What PANDAS, tic and OCD have in common is that the basal ganglia are involved. It’s a set of brain structures deep within the brain that is enriched with dopamine neurons responsible for movement coordination. It also has connections to the outer cortex that controls voluntary movement.
The dopamine-powered basal ganglia — motor cortex circuitry “control and select goal-directed motor, cognitive and motivational behaviour,…inhibitory control and habit formation” psychiatrists wrote.
The Trickster Streptococci
About 40% of children and adolescents with tics/OCD have higher levels of Strep-specific antibodies in their blood. This means that their immune system was actively fighting Strep bacteria.
Intriguingly, their Strep-specific antibodies reacted with brain dopamine receptors in a lab plate. Give Strep-specific antibodies to rodents and the antibodies deposit in the basal ganglia, causing motoric disorders. Immunize mice for Strep infection and the same thing happened.
This shows that antibodies generated by the immune system to neutralize Strep can invade the brain and attack the basal ganglia (where dopamine neurons are abundant). In a sense, the immune system mistook the basal ganglia for Strep bacteria. Or that Strep bacteria trick the immune system into attacking the basal ganglia.
This is a form of autoimmunity — as the ‘A’ implies in PANDAS — when the immune system attacks the host own cells.
What’s the Trick?
Molecular mimicry is currently the best explanation for any autoimmunity initiated by microbial infections. In molecular mimicry, the structure of the pathogen is similar to some tissues of the body. Antibodies that bind to the pathogen might also unintentionally bind to some other tissues by chance.
Chances of molecular mimicry happening are unknown. Scientists think that people with genetics prone to developing autoimmune diseases might be at a higher risk for PANDAS.
“Maybe you have more of a genetic risk for OCD or anorexia, and the infection then unmasks that vulnerability. That’s one possibility,” opines Kyle Williams, the director of the pediatric-neuropsychiatry-and-immunology program at Massachusetts General Hospital for Children.
PANDAS Anorexia
Symptoms of anorexia can be viewed as OCD, such as food and body weight obsessions and compulsive exercise or purging. Anorexia is also a neuropsychiatric disorder with multifactorial causes that include sociocultural, psychological and biological aspects.
In 2000, Mae Sokol, MD and professor of psychiatry at Creighton University in Omaha, reported 4 cases of “infection-triggered anorexia nervosa in children.” These children were also diagnosed with PANDAS and OCD. They successfully regained their weight (and OCD disappeared) after taking antibiotics.
“Maybe we are seeing the same thing in children with PANDAS anorexia and children with PANDAS OCD — only in the PANDAS anorexia, the obsessions are about food and weight, whereas in PANDAS OCD they are about other things,” Prof. Dr. Sokol said.
“Since the basal ganglia are also involved with emotion, we think this area of the brain may be affected slightly differently with each condition,” the psychiatrist professor added. Besides the basal ganglia, researchers have found higher levels of antibodies that target appetite-regulating hormones in the bloodstream of anorexics. They also suggested that molecular mimicry is at play but has not pinpointed a specific microbial agent. More details here:
It seems that Strep could very well be the microbial agent playing molecular mimicry tricks on the immune system to target the basal ganglia (and appetite hormones).
To Close
“We say disease is due to biological factors, social factors, and psychological factors all interacting together,” remarks Jim Morris, a professor at Lancaster University, UK, in response to PANDAS. “Well, it works with psychiatric disease as well.”
Researchers have focused on genetics and other medical comorbidities for decades. But now the microbial angle has started to gain recognition as part of the biological aspect in psychiatric diseases. Though scientists still have a long way to go, as evidenced by the lack of clinical studies, diagnostic and treatment guidelines, statistical prevalence, etc.
“When a child suddenly develops food rituals, becomes unusually picky about food and loses weight, you might not think of a bacterial, viral, or other types of infection,” says the psychotherapist Sharon K. Farber, PhD, at New York University.
This article was inspired by Caelyn H. | https://medium.com/microbial-instincts/when-tics-ocd-or-anorexia-develop-following-strep-infection-6a35c47519b0 | ['Shin Jie Yong'] | 2020-04-06 05:50:28.525000+00:00 | ['Mental Health', 'Technology', 'Health', 'Psychology', 'Ideas'] | Title Tics OCD Anorexia Develop Following Strep InfectionContent Infection Brain Tics OCD Anorexia Develop Following Strep Infection It’s called PANDAS Pediatric Autoimmune Neuropsychiatric Disorders Associated Streptococcal infection late 1900s physician reported child started showing tic following strep sore throat outbreak tic also disappeared strep killed antibiotic 2007 Case Report 8yearold boy recurrent strep sore throat developed set ritual eating nodded flapped arm tapped mount opened closed fist etc taking bite said ritual helped cope intrusive thought food poisoned became severely conscious healthy eating body fat avoiding ‘bad’ food eating 200 caloriesday — eventually losing 8kg receiving treatment recurrent sore throat behavioural anomaly vanished boy PANDAS Pediatric Autoimmune Neuropsychiatric Disorders Associated Streptococcal infection obsessivecompulsive disorder OCD tic anorexia nervosa PANDAS need know Diagnosis It’s diagnosed child develops tic OCD right strep infection symptom ticsOCD worsen following strep infection DSM5 put PANDAS “ObsessiveCompulsive Related Disorders Due Another Medical Condition” PANDAS closely associated tic OCD may include anorexia case Cause Nearly child experienced strep throat infection least life average schoolchild get 2–3x per year It’s caused contagious bacterium called Streptococcus pyogenes strep sore throat infection clear simple antibiotic small subset child perhaps genetics cultural predisposition lead PANDAS Statistics Due lack diagnostic guideline precise prevalence still unknown Scientists however estimated PANDAS contribute least 10 childhood OCD tic disorder Treatment Usually antibiotic cognitive behavioural therapy Steroids antidepressant prescribed deemed necessary though antidepressant often cause side effect child PANDAS best treatment aggressive antibiotic — 50 cure rate according 2017 population study Prognosis Longterm outcome PANDAS vary case due variability PANDAS subtypes treatment protocol “Whether child PANDAS subtype typically go remission progress chronic course illness known” say Tanya Murphy MD professor director Rothman Center Pediatric Neuropsychiatry Florida Strep infection recurrent example longterm antibiotic adulthood necessary Even Strep eliminated may still suffer neuropsychiatric symptom brain specifically basal ganglion sufficiently compromised child recover naturally immune system matures distinguish Strep brain Basal Ganglia Tic viewed common form movement disorder child Tics sudden repetitive movement eg shoulder shrugging blinking head jerking vocalization eg sniffing grunting Tics shortterm 11–20 prevalence longtern 05–07 prevalence Whereas OCD affect 1–3 general population It’s characterized persistent unwanted thought obsession andor repetitive behaviour driven strong urge compulsion PANDAS tic OCD common basal ganglion involved It’s set brain structure deep within brain enriched dopamine neuron responsible movement coordination also connection outer cortex control voluntary movement dopaminepowered basal ganglion — motor cortex circuitry “control select goaldirected motor cognitive motivational behaviour…inhibitory control habit formation” psychiatrist wrote Trickster Streptococci 40 child adolescent ticsOCD higher level Strepspecific antibody blood mean immune system actively fighting Strep bacteria Intriguingly Strepspecific antibody reacted brain dopamine receptor lab plate Give Strepspecific antibody rodent antibody deposit basal ganglion causing motoric disorder Immunize mouse Strep infection thing happened show antibody generated immune system neutralize Strep invade brain attack basal ganglion dopamine neuron abundant sense immune system mistook basal ganglion Strep bacteria Strep bacteria trick immune system attacking basal ganglion form autoimmunity — ‘A’ implies PANDAS — immune system attack host cell What’s Trick Molecular mimicry currently best explanation autoimmunity initiated microbial infection molecular mimicry structure pathogen similar tissue body Antibodies bind pathogen might also unintentionally bind tissue chance Chances molecular mimicry happening unknown Scientists think people genetics prone developing autoimmune disease might higher risk PANDAS “Maybe genetic risk OCD anorexia infection unmasks vulnerability That’s one possibility” opines Kyle Williams director pediatricneuropsychiatryandimmunology program Massachusetts General Hospital Children PANDAS Anorexia Symptoms anorexia viewed OCD food body weight obsession compulsive exercise purging Anorexia also neuropsychiatric disorder multifactorial cause include sociocultural psychological biological aspect 2000 Mae Sokol MD professor psychiatry Creighton University Omaha reported 4 case “infectiontriggered anorexia nervosa children” child also diagnosed PANDAS OCD successfully regained weight OCD disappeared taking antibiotic “Maybe seeing thing child PANDAS anorexia child PANDAS OCD — PANDAS anorexia obsession food weight whereas PANDAS OCD things” Prof Dr Sokol said “Since basal ganglion also involved emotion think area brain may affected slightly differently condition” psychiatrist professor added Besides basal ganglion researcher found higher level antibody target appetiteregulating hormone bloodstream anorexic also suggested molecular mimicry play pinpointed specific microbial agent detail seems Strep could well microbial agent playing molecular mimicry trick immune system target basal ganglion appetite hormone Close “We say disease due biological factor social factor psychological factor interacting together” remark Jim Morris professor Lancaster University UK response PANDAS “Well work psychiatric disease well” Researchers focused genetics medical comorbidities decade microbial angle started gain recognition part biological aspect psychiatric disease Though scientist still long way go evidenced lack clinical study diagnostic treatment guideline statistical prevalence etc “When child suddenly develops food ritual becomes unusually picky food loses weight might think bacterial viral type infection” say psychotherapist Sharon K Farber PhD New York University article inspired Caelyn HTags Mental Health Technology Health Psychology Ideas |
714 | Data Science, Alexander of the Times Ahead | Raise your hands if you ever heard about Data Science! It has been the buzz word in the recent past. This field can retain and stay in the marketplace for a long time and provide numerous opportunities to people who add value by utilizing data. I strongly believe Data science is the place to be going forward in the next couple of years or even decades.
Firstly, What is Data Science?
Data science is a field that uses algorithms, processes, and models to extract knowledge and generate insights from data. To know more about Data Science, you can read my previous article here.
There’s lots of data in the world. And ultimately, data is the resource that our scientists work with. If there is more data, there’s more work, and if there’s less data, there’s less work. Let’s jump straight into it. Since the dawn of time up until 2005, humans had used 130 exabytes of data. Well, this tells us absolutely nothing.
What is an Exabyte? Why is this impressive?
Well, let’s start from the ground up. Let’s start with one byte here we’ve got the letter A. It takes up exactly one byte of space.
Now, assume we have a page of letters. Each page of letters can have between two thousand to five thousand letters. So we’ll say that half a page of text is about right for one kilobyte.
Let us zoom out a thousand times, and you’ll get a book. So a book of about 500 pages that’ll take up about a megabyte.
Designed using Canva
Now, let’s zoom out another thousand times. And what we’ll get is a gigabyte, and a gigabyte is sufficient to fit the whole human genome. So a human genome, once it is coded (in an ideal world) can fit into a gigabyte.
Source — By Elymas via Shutter Stock
So you can fit an entire person onto a gigabyte if you think of it that way. But here you can counter-argument by telling that the genome of a person is not sufficient to tell you everything about that person. A person has life experiences. Different people live their lives differently. And there’s lots more to a person than just a genome.
Well then, let’s zoom out another thousand times, and we’ll get to a terabyte. If you take an HD camera and you follow a person for every single day of their life for every single hour, minute, and second and you film everything that they’re doing for 70 or 80 years, you can fit all of that material onto one terabyte.
Source — By Donald Tong via Pexels
How about that?
This seems to be pretty impressive already. So one terabyte can fit all of your life experiences filmed throughout your life for every single second that you live.
And what next? This is getting better.
Source — By Tom Fisk via Pexels
Consider the Amazon rain forest that takes up about 1.4 billion acres. That’s 1.4 billion acres of trees. There are about 500 trees per acre, making it about 700 billion trees(Estimated). Now, if you take all of these trees and you chop them down, and you turn them into paper, and you fill that paper with letters entirely on both sides of the sheet, then that will amount to approximately some petabytes of data.
Well now, if you zoom out another thousand times, you will get an exabyte, and an Exabyte is a thousand petabytes.
I guess you got an idea of the abundance of data we have used.
Wait! I mentioned 130 exabytes until 2005 only. What about today? How much data have we consumed until today?
Designed using Canva
So you can see that data in the world is growing very very quickly. Data is being generated in huge numbers as time moves on. | https://medium.com/towards-artificial-intelligence/data-science-alexander-of-the-times-ahead-1f9caa8f9c4d | ['Saiteja Kura'] | 2020-05-06 13:05:50.337000+00:00 | ['Big Data', 'Artificial Intelligence', 'Future', 'Data Science', 'Technology'] | Title Data Science Alexander Times AheadContent Raise hand ever heard Data Science buzz word recent past field retain stay marketplace long time provide numerous opportunity people add value utilizing data strongly believe Data science place going forward next couple year even decade Firstly Data Science Data science field us algorithm process model extract knowledge generate insight data know Data Science read previous article There’s lot data world ultimately data resource scientist work data there’s work there’s le data there’s le work Let’s jump straight Since dawn time 2005 human used 130 exabyte data Well tell u absolutely nothing Exabyte impressive Well let’s start ground Let’s start one byte we’ve got letter take exactly one byte space assume page letter page letter two thousand five thousand letter we’ll say half page text right one kilobyte Let u zoom thousand time you’ll get book book 500 page that’ll take megabyte Designed using Canva let’s zoom another thousand time we’ll get gigabyte gigabyte sufficient fit whole human genome human genome coded ideal world fit gigabyte Source — Elymas via Shutter Stock fit entire person onto gigabyte think way counterargument telling genome person sufficient tell everything person person life experience Different people live life differently there’s lot person genome Well let’s zoom another thousand time we’ll get terabyte take HD camera follow person every single day life every single hour minute second film everything they’re 70 80 year fit material onto one terabyte Source — Donald Tong via Pexels seems pretty impressive already one terabyte fit life experience filmed throughout life every single second live next getting better Source — Tom Fisk via Pexels Consider Amazon rain forest take 14 billion acre That’s 14 billion acre tree 500 tree per acre making 700 billion treesEstimated take tree chop turn paper fill paper letter entirely side sheet amount approximately petabyte data Well zoom another thousand time get exabyte Exabyte thousand petabyte guess got idea abundance data used Wait mentioned 130 exabyte 2005 today much data consumed today Designed using Canva see data world growing quickly Data generated huge number time move onTags Big Data Artificial Intelligence Future Data Science Technology |
715 | How to Find Inspiration for Your Business Even in Toughest Times | Getting your business up and running is a tall task. It means countless hours of your precious time and hundreds of sleepless nights invested in something unknown. Although it’s very fulfilling to work for yourself, this moon has its dark side — you’re left alone with the problems and have to go through the rough times by yourself. When an employee can weather the storm by simply changing the workplace, you have to find the mental force to cope with turmoil and put your business back on the rails.
If you’ve clicked on this article, I assume you’re a grown-up who is well aware of how to deal with problems. Such advice as “analyze what went wrong and try again” is surely what you’ve tried and is probably not worth noting in the first place.
But what can really be done if the project has hit a dead-end and there’s no feasible way to escape this trap? Here are some of the more or less obvious things you should try out to get your mojo back on track and drop the weight of the world from your shoulders: these won’t fix the bugs in your code but will give you a look at how to get an impulse for coping with such things.
No mentors, do it yourself
Well, the Number 1 rule is to avoid “mentors” whose services cost real money but have no real value. Trust me, you can get all these platitudes for free on the Internet: paying for the “get over it” phrases is not the way to prosperity. There are almost no real psychologists among “trainers,” but their paycheck implies they’re Oxford graduates.
Ironically, the person who has probably inspired the most people, Shia LaBeouf, who made an arbitrary “Just do it” with a big portion of humor, can’t even fix his own life now. Each person is unique, so don’t expect someone to come up with a ready decision for you, seek the inspiration yourself!
Physical activity
It’s not about building muscles — getting up and moving proves to be a great way to alleviate stress.
Physical exercise is the clear victor here: it’s a great way to get your blood going, which will stimulate the brain to work. This will also help your mental state: sports are a great way to fight depression. And I’m talking about casual stuff like walking, swimming, or ping-pong — something to keep you relaxed and engaged at the same time.
Meditation and yoga are no longer exotic and require a minimum commitment unless you want to become a pro, which is not the case. While yoga is more about making you balanced and patient, meditation can help you focus on the things you need: figuring out the main issues is easy when nothing else distracts you.
Music and ambience
Although Chopin is arguably the best composer to listen to while working (according to the TV series “Dexter”), you can pick any of the music greats to become energized. If all you need to rock is to hear a motivational speech, then maybe choose… rock? It’s a less stressful way to hear the words you need, plus it will rhyme on top of having great instrumentals. The human brain is more susceptible to soaking up info when it is neatly packaged, so it’s okay to avoid monotonous speakers in favor of a lovely melody with a life-affirming message.
Some people don’t even need the words: like long-time family members, they can be content with silence. So, some classical pieces are often what could make your brain come up with stellar ideas — Sherlock and his violin will not lie.
Some active and rhythmic music can also be a motivator; its fast tempo can set the pace for your mood and ideas, hopefully becoming the driving force towards accomplishing your goals.
Setting up the ambiance is an important thing in general: music is an essential part of this and it can be complemented nicely by cozy furniture, relaxing colors, and even maybe some pictures that will motivate you — that’s where it won’t hurt to throw in some inspirational quotes.
Balanced eating
Forget about tons of food. Seriously, it’s not about fat-shaming or something similar; eating is a great short-term solution for the anxiety but it’s also one of the signs of depression. Not only does emotional eating distract from the real problems but it is also time-consuming, and unless you prefer consuming lots of junk food, this is not the way to go either.
Your brainstorming sessions may also suffer from those who feast before the event — try to keep it modest prior to such meetings.
It’s a proven fact that it’s better to make intuitive decisions when not being well-fed, especially when you need to come up with an original concept.
But we all like to eat and it often makes us happy, so just eat the food you could benefit from. Although broccoli, being the champ of useful food, isn’t the tastiest of things, fish can become your choice — even the fattier species are well-refined and stimulate the brain to work. Olive oil will be a great assistant in making your dishes fancy, and nuts will help you keep fed on top of being great for neural activity.
Get enough sleep
Sleeping is one of the best ways to fill yourself with energy. An entrepreneur’s life is dense and hard to predict, but sleeping for 7–8 hours a day is what will keep your morale high. Your brain needs to rest, so let it chill for a while and refresh in order to solve problems and escape the dead-end status of your project.
A lack of sleep is the main cause of all sorts of angst and depression, so don’t worry about going to bed instead of watching another mindless show or newscast.
Imagine that your PC or laptop has been turned on for several weeks and is lagging behind at all sorts of tasks. But when you press the “reset” button, it starts working just fine. This is pretty much how our brain works; just be kind to it and give a break — it will pay you back with more processing power!
Mastermind group
Here is some harsh advice: be the dumbest person around. This sounds bad for your self-esteem, but hold on — it’s a fine way to become inspired. If you surround yourself with people who are smarter, more active, agile, and resourceful than you, it will give you a nice impetus to move forward and overcome the tough times.
Although it may look like an AA group, having a group of people who can listen to you and hold you accountable for further activities is a good tool in your well-being arsenal. This can be a group of your peers who you meet every 2 weeks and are around 6–8 people, so it’s not too crowded or too small at the same time. Sure, you’re a strong person if you decided to be an entrepreneur, but having someone to lift you up is simply too good to ignore.
Combination of cool things
Like a generous Walmart discount, try to get 2 things for the price of one. For example, take a long walk while listening to “Dust in the Wind” or go to the beach if there’s a seashore — a river bank is a nice option, too.
Or just go crazy and visit a foreign country with some extra side trips. Whether it’s Nordic fjords or mid-Europe canals, there’s a ton of places that will reinvigorate you to unwind and get inspired for new triumphs.
The pioneers of the past used to draw inspiration from their environment — even during the mid-project stages, they took long walks, granted there was no Internet. At its basic level, the human brain hasn’t changed much since; just keep your eyes fresh so the mind can capture ideas that may become a solution for you later. | https://medium.com/hackernoon/how-to-find-inspiration-for-your-business-even-in-toughest-times-da702a200eec | ['Valeriy Bykanov'] | 2018-12-17 09:22:06.320000+00:00 | ['Productivity', 'Inspiration', 'Motivation', 'Startup', 'Business'] | Title Find Inspiration Business Even Toughest TimesContent Getting business running tall task mean countless hour precious time hundred sleepless night invested something unknown Although it’s fulfilling work moon dark side — you’re left alone problem go rough time employee weather storm simply changing workplace find mental force cope turmoil put business back rail you’ve clicked article assume you’re grownup well aware deal problem advice “analyze went wrong try again” surely you’ve tried probably worth noting first place really done project hit deadend there’s feasible way escape trap le obvious thing try get mojo back track drop weight world shoulder won’t fix bug code give look get impulse coping thing mentor Well Number 1 rule avoid “mentors” whose service cost real money real value Trust get platitude free Internet paying “get it” phrase way prosperity almost real psychologist among “trainers” paycheck implies they’re Oxford graduate Ironically person probably inspired people Shia LaBeouf made arbitrary “Just it” big portion humor can’t even fix life person unique don’t expect someone come ready decision seek inspiration Physical activity It’s building muscle — getting moving prof great way alleviate stress Physical exercise clear victor it’s great way get blood going stimulate brain work also help mental state sport great way fight depression I’m talking casual stuff like walking swimming pingpong — something keep relaxed engaged time Meditation yoga longer exotic require minimum commitment unless want become pro case yoga making balanced patient meditation help focus thing need figuring main issue easy nothing else distracts Music ambience Although Chopin arguably best composer listen working according TV series “Dexter” pick music great become energized need rock hear motivational speech maybe choose… rock It’s le stressful way hear word need plus rhyme top great instrumentals human brain susceptible soaking info neatly packaged it’s okay avoid monotonous speaker favor lovely melody lifeaffirming message people don’t even need word like longtime family member content silence classical piece often could make brain come stellar idea — Sherlock violin lie active rhythmic music also motivator fast tempo set pace mood idea hopefully becoming driving force towards accomplishing goal Setting ambiance important thing general music essential part complemented nicely cozy furniture relaxing color even maybe picture motivate — that’s won’t hurt throw inspirational quote Balanced eating Forget ton food Seriously it’s fatshaming something similar eating great shortterm solution anxiety it’s also one sign depression emotional eating distract real problem also timeconsuming unless prefer consuming lot junk food way go either brainstorming session may also suffer feast event — try keep modest prior meeting It’s proven fact it’s better make intuitive decision wellfed especially need come original concept like eat often make u happy eat food could benefit Although broccoli champ useful food isn’t tastiest thing fish become choice — even fattier specie wellrefined stimulate brain work Olive oil great assistant making dish fancy nut help keep fed top great neural activity Get enough sleep Sleeping one best way fill energy entrepreneur’s life dense hard predict sleeping 7–8 hour day keep morale high brain need rest let chill refresh order solve problem escape deadend status project lack sleep main cause sort angst depression don’t worry going bed instead watching another mindless show newscast Imagine PC laptop turned several week lagging behind sort task press “reset” button start working fine pretty much brain work kind give break — pay back processing power Mastermind group harsh advice dumbest person around sound bad selfesteem hold — it’s fine way become inspired surround people smarter active agile resourceful give nice impetus move forward overcome tough time Although may look like AA group group people listen hold accountable activity good tool wellbeing arsenal group peer meet every 2 week around 6–8 people it’s crowded small time Sure you’re strong person decided entrepreneur someone lift simply good ignore Combination cool thing Like generous Walmart discount try get 2 thing price one example take long walk listening “Dust Wind” go beach there’s seashore — river bank nice option go crazy visit foreign country extra side trip Whether it’s Nordic fjord midEurope canal there’s ton place reinvigorate unwind get inspired new triumph pioneer past used draw inspiration environment — even midproject stage took long walk granted Internet basic level human brain hasn’t changed much since keep eye fresh mind capture idea may become solution laterTags Productivity Inspiration Motivation Startup Business |
716 | The Evidence That Luke Skywalker Is Snoke In The Last Jedi Novelization | The Missing Third Lesson
In Chapter 11, Luke tells Rey he’ll train her. “Three lessons. I will teach you the ways of the Jedi — and why they need to end.”
Lesson One comes in Chapter 13 when Luke says, “And this is the lesson — that Force does not belong to the Jedi. It’s so much bigger. To say that if the Jedi die the light dies is vanity.”
Lesson Two comes in Chapter 17 when Luke says, “the legacy of the Jedi is failure. Hypocrisy. Hubris”
Luke never gives Rey the third lesson in the movie, and nowhere in the novelization does Luke say “Lesson Three”.
Will we find out the third lesson in Episode IX? Or was the third lesson implied somewhere in The Last Jedi and we just missed it?
One possibility for the missing lesson is the moment right before Rey first senses “the dark place” beneath the island, to which Luke responds, “Balance. Powerful light, powerful darkness.” That echoes Luke’s sentiment about young Ben, as well as the nature of the Prime Jedi mosaic in the Jedi Temple.
But then Rey says, “That place can show me something. It was trying to.” Was the third lesson what the dark mirror beneath Ahch-To teaches?
In Chapter 21, Rey falls through the dark portal and has her vision in mirror cave.
“Rey knew she had to go deeper — that the world inside the stone only seemed to go on forever. It was leading somewhere, and if she only had the courage to follow, that secret place would show her what she had come to see — and what she was most afraid to know.”
Rey begged to see her parents. “She saw two dark figures beneath its surface. As her heartbeat hammered in her ears, the two became one.”
This imagery evokes Jungian symbolism. That they merge into her own reflection speak to Rey’s spiritual journey to an wholly integrated Self.
The significance of the Dark Portal on Ahch-To is underscored in Chapter 23. Rey is brought before Snoke and finds his eyes, “piercing and hungry.”
“Their pull was akin to what she’d felt near the pit on Ahch-To — whispering of secrets that had been reserved for her, that belonged to her. Ancient, hidden knowledge that would destroy the weak but elevate the strong. The Worthy.”
In Chapter 26, Rey describes the feeling of Snoke entering her mind to find Luke’s location.
“Snoke’s presence felt like a pit, empty and cold and dark — as if the dark-side cave beneath Ahch-To had gone on forever.”
There is an obvious parallel to the first time Rey sensed the pit on Ahch-To during Lesson One, when it pulled her in and tried to give her something she needed. Luke was at her side, horrified that she didn’t try to fight it.
Meanwhile, Rey thought she sensed nothing from Luke. But maybe her sense of the pit and sensing Luke, were one in the same. | https://medium.com/jung-skywalker/star-wars-symbology-the-evidence-in-the-last-jedi-novelization-that-luke-skywalker-is-snoke-ec94833bedd | ['Brian Deines'] | 2020-05-12 18:01:48.245000+00:00 | ['Star Wars', 'Self-awareness', 'Psychology', 'Film', 'Writing'] | Title Evidence Luke Skywalker Snoke Last Jedi NovelizationContent Missing Third Lesson Chapter 11 Luke tell Rey he’ll train “Three lesson teach way Jedi — need end” Lesson One come Chapter 13 Luke say “And lesson — Force belong Jedi It’s much bigger say Jedi die light dy vanity” Lesson Two come Chapter 17 Luke say “the legacy Jedi failure Hypocrisy Hubris” Luke never give Rey third lesson movie nowhere novelization Luke say “Lesson Three” find third lesson Episode IX third lesson implied somewhere Last Jedi missed One possibility missing lesson moment right Rey first sens “the dark place” beneath island Luke responds “Balance Powerful light powerful darkness” echo Luke’s sentiment young Ben well nature Prime Jedi mosaic Jedi Temple Rey say “That place show something trying to” third lesson dark mirror beneath AhchTo teach Chapter 21 Rey fall dark portal vision mirror cave “Rey knew go deeper — world inside stone seemed go forever leading somewhere courage follow secret place would show come see — afraid know” Rey begged see parent “She saw two dark figure beneath surface heartbeat hammered ear two became one” imagery evokes Jungian symbolism merge reflection speak Rey’s spiritual journey wholly integrated Self significance Dark Portal AhchTo underscored Chapter 23 Rey brought Snoke find eye “piercing hungry” “Their pull akin she’d felt near pit AhchTo — whispering secret reserved belonged Ancient hidden knowledge would destroy weak elevate strong Worthy” Chapter 26 Rey describes feeling Snoke entering mind find Luke’s location “Snoke’s presence felt like pit empty cold dark — darkside cave beneath AhchTo gone forever” obvious parallel first time Rey sensed pit AhchTo Lesson One pulled tried give something needed Luke side horrified didn’t try fight Meanwhile Rey thought sensed nothing Luke maybe sense pit sensing Luke one sameTags Star Wars Selfawareness Psychology Film Writing |
717 | Think Ahead: Boost Your BI with AI | An analyst’s primary responsibility is to observe patterns and locate opportunities. Yet even experts can overlook trends or misinterpret results, particularly if they have to sift through a lot of information. Consequently, more businesses are enhancing their data analysis with ML. In addition to detecting anomalies, ML models can detect meaningful event correlations — some of which businesses can exploit for big wins. For instance, a brand’s emails could have the strongest open rates when they go out on Wednesday mornings and their subject lines contain emojis. This detail is useful to marketers because it can help them strategically plan their campaigns. Nonetheless, it could’ve been missed with ordinary data analysis. ML can return sharper insights in less time, and it’s not as likely to exhibit bias as a human analyst.
4: NATURAL LANGUAGE PROCESSING
Photo by Kelsey Curtis on Unsplash
Google Analytics has a built-in AI function that can correlate events and auto-generate insights. Enterprise BI tools like Yellowfin and ThoughtSpot use similar technology to uncover any hidden wisdom in their customers’ data. Interestingly, all of these products harness Natural Language Processing (NLP) to communicate their findings and make data conversational. They can display insights in plain English, as in “25.33% of your site traffic is from the Referral channel.” Additionally, some of these NLP programs let users submit questions and receive answers.
A demo of how ThoughtSpot implements NLP. A user types in a specific question and gets an answer.
NLP can transform a dashboard into a virtual analyst; it lowers the barrier of entry to BI so anyone can stay in the know about their business. Users don’t have to deal with spreadsheets, charts, or SQL queries to get the information they need. Instead, they can interact with software that analyzes the data for them.
5: PREDICTIVE ANALYTICS
Photo by dylan nolte on Unsplash
AI can illuminate the past and present, but it can also offer a vision of things to come. Predictive analytics leverages ML to find regular patterns in data and carry them into the future. These projections are common in financial and weather forecasting, and they’re increasingly relevant to BI. Case in point, any retailer who can anticipate product demand will promote higher rates of customer satisfaction. In other words, they’ll have enough of an item in stock when consumers want it the most. Predictive capability comes with a competitive advantage, and more businesses can acquire it via platforms like IBM Watson or DataRobot. The popular BI tool Tableau includes a forecasting feature too.
IT’S NEVER TOO EARLY TO START
AI products will get even better at delivering warnings, correlations, and predictions. While the best is yet to come, you should embrace this technology now. Like many businesses, your competition may depend on humans to conduct their entire analytics operation. You immediately stand out from them once you begin automating your BI. The benefit of being an early adopter is that you’re already proficient by the time most people are just starting out. Prepare for the next wave of innovation, otherwise it’ll sweep you away rather than lift you up.
Always remember that AI isn’t perfect, it’s only as reliable as the data it’s based on. Back in 2018, Waze directed a user to drive his car into a lake and he complied. He assumed the app was correct, despite common sense and his obvious surroundings. In this case, the GPS navigation didn’t give the right instructions because it didn’t have the right information. Moral of the story: never follow AI blindly. The results from your ML models should enhance your judgment, not replace it altogether.
Similarly, AI isn’t a substitute for data analysts or data scientists. We still need specialists to oversee and validate research, except now they can do this work faster and more effectively with AI. | https://medium.com/eyeful/think-ahead-boost-your-bi-with-ai-d8df8f64851f | ['Therese Moriarty'] | 2020-04-03 20:30:05.808000+00:00 | ['Machine Learning', 'AI', 'Eyeful Trends', 'Artificial Intelligence', 'Business Intelligence'] | Title Think Ahead Boost BI AIContent analyst’s primary responsibility observe pattern locate opportunity Yet even expert overlook trend misinterpret result particularly sift lot information Consequently business enhancing data analysis ML addition detecting anomaly ML model detect meaningful event correlation — business exploit big win instance brand’s email could strongest open rate go Wednesday morning subject line contain emojis detail useful marketer help strategically plan campaign Nonetheless could’ve missed ordinary data analysis ML return sharper insight le time it’s likely exhibit bias human analyst 4 NATURAL LANGUAGE PROCESSING Photo Kelsey Curtis Unsplash Google Analytics builtin AI function correlate event autogenerate insight Enterprise BI tool like Yellowfin ThoughtSpot use similar technology uncover hidden wisdom customers’ data Interestingly product harness Natural Language Processing NLP communicate finding make data conversational display insight plain English “2533 site traffic Referral channel” Additionally NLP program let user submit question receive answer demo ThoughtSpot implement NLP user type specific question get answer NLP transform dashboard virtual analyst lower barrier entry BI anyone stay know business Users don’t deal spreadsheet chart SQL query get information need Instead interact software analyzes data 5 PREDICTIVE ANALYTICS Photo dylan nolte Unsplash AI illuminate past present also offer vision thing come Predictive analytics leverage ML find regular pattern data carry future projection common financial weather forecasting they’re increasingly relevant BI Case point retailer anticipate product demand promote higher rate customer satisfaction word they’ll enough item stock consumer want Predictive capability come competitive advantage business acquire via platform like IBM Watson DataRobot popular BI tool Tableau includes forecasting feature IT’S NEVER EARLY START AI product get even better delivering warning correlation prediction best yet come embrace technology Like many business competition may depend human conduct entire analytics operation immediately stand begin automating BI benefit early adopter you’re already proficient time people starting Prepare next wave innovation otherwise it’ll sweep away rather lift Always remember AI isn’t perfect it’s reliable data it’s based Back 2018 Waze directed user drive car lake complied assumed app correct despite common sense obvious surroundings case GPS navigation didn’t give right instruction didn’t right information Moral story never follow AI blindly result ML model enhance judgment replace altogether Similarly AI isn’t substitute data analyst data scientist still need specialist oversee validate research except work faster effectively AITags Machine Learning AI Eyeful Trends Artificial Intelligence Business Intelligence |
718 | Maddie Stone published a generous, thoughtful, and mind-expanding piece exploring the implications… | Maddie Stone published a generous, thoughtful, and mind-expanding essay exploring the implications of the near future extrapolated in my latest novel:
Geoengineering, or hacking the planet to cool it down, is either a maniacal plan dreamt up by foolhardy scientists or a useful tool for staving off climate catastrophe — maybe both. It raises hard questions about what sorts of sacrifices humanity may have to make for the greater good and who gets to decide; questions that beg for nuanced conversations about the social, environmental, and political risks and rewards.
Yet in science fiction, geoengineering tends to get treated with all the nuance of Thor’s hammer striking a rock monster. Which is why Eliot Peper’s recent novel Veil, set on a near future Earth beset by climate crises, is such a refreshing read. This book gets geoengineering right by showing that there are no obvious right answers. | https://eliotpeper.medium.com/maddie-stone-published-a-generous-thoughtful-and-mind-expanding-piece-exploring-the-implications-4456da8dfa42 | ['Eliot Peper'] | 2020-12-09 04:37:08.468000+00:00 | ['Technology', 'Climate Change', 'Future', 'Science', 'Books'] | Title Maddie Stone published generous thoughtful mindexpanding piece exploring implications…Content Maddie Stone published generous thoughtful mindexpanding essay exploring implication near future extrapolated latest novel Geoengineering hacking planet cool either maniacal plan dreamt foolhardy scientist useful tool staving climate catastrophe — maybe raise hard question sort sacrifice humanity may make greater good get decide question beg nuanced conversation social environmental political risk reward Yet science fiction geoengineering tends get treated nuance Thor’s hammer striking rock monster Eliot Peper’s recent novel Veil set near future Earth beset climate crisis refreshing read book get geoengineering right showing obvious right answersTags Technology Climate Change Future Science Books |
719 | 5 minutes to understand data storytelling! | Data-visualization, or Dataviz… all of this overlaps with one and the same concept: the graphical exploitation of data. But what does this mean in practice?
The advent of digital also coincides with the explosion of Big Data. Through their digital activities, companies today aggregate hundreds of thousands of data… and to analyze them and draw conclusions: they can transform tables of figures into visual representations thanks to Dataviz! Next comes data storytelling…
A simple definition
“Do you want me to draw you a picture?”. Indeed, when a concept is difficult to understand, presenting it visually makes it much more affordable. The human brain is made up like this and processes images much faster than text!
Take the example of a 3,000-line Excel file. It contains vital information, but how do you get its intelligence from it? The Dataviz then makes it possible to aggregate millions of data (from companies or organizations such as INSEE) and visualize them through graphs, diagrams, maps, chronologies, pictograms, infographs or even unpublished graphic creations or photos. Data storytelling then succeeds in narrating this data in images, with a view to analysing it.
A bit of history
Several stages: :
As early as the end of the 18th century, the Scottish engineer and economist William Playfair created the histogram, the pie chart and the time series: visualizations that are still used today! Concrete examples arrive as early as the 19th century on subjects related to mortality and war…
Birth of a graphic language as early as 1967 thanks to Jacques Bertin, who was digging into ways of presenting information on graphs (colours, sizes, shapes).
The massive arrival of digital technology makes data storytelling a good thing. The amount of information to be analysed is high, and new ways of presenting information thanks to digital techniques optimise Dataviz (3D for example).
Today, data collection has been generalized to the whole company, and the analysis and transmission of the results remains a practice accessible to a simple minority of employees. Data storytelling thus makes it possible to democratize access to complex information in the company (and no longer only to the Business Intelligence department!).
Concrete applications
Data-storytelling is present in all organizations. There are many of them! Use of service, sales of a product, number of visitors to a site, fluctuations in a price… each sector of activity generates some, and the simple fact of creating an Excel table to order the data and draw conclusions is already part of Dataviz!
It will be useful to the trades related to these datas, which will then practice the “piloting by the data”. This allows a more precise decision making, which goes beyond human reasoning. It is thus possible to follow many key indicators simultaneously and to keep a broad vision of one’s activity.
But it also serves the general public! Generally to highlight data and educate consciences on subjects such as ecology, economy, synthesis of financial results?
The anecdote to know
The best-known data visualization dates back to the time of Napoleon. It denounces the absurdity of war. In 1812 Napoleon led the famous Russian campaign, more than 400,000 men set out to conquer Moscow. Only a quarter of the men reached the capital. Napoleon was forced to retreat, and a second trap closed in on the troops: the cold. Only a few thousand men crossed the border again… Charles Joseph Minard, a contemporary of the campaign, produced the map below several decades later. | https://medium.com/toucan-toco/5-minutes-pour-comprendre-le-data-storytelling-4a9e546b663 | ['Charles Miglietti'] | 2020-02-06 14:35:00.622000+00:00 | ['Dashboard', 'Dataviz', 'Business Intelligence', 'Data Visualization', 'Productivité'] | Title 5 minute understand data storytellingContent Datavisualization Dataviz… overlap one concept graphical exploitation data mean practice advent digital also coincides explosion Big Data digital activity company today aggregate hundred thousand data… analyze draw conclusion transform table figure visual representation thanks Dataviz Next come data storytelling… simple definition “Do want draw picture” Indeed concept difficult understand presenting visually make much affordable human brain made like process image much faster text Take example 3000line Excel file contains vital information get intelligence Dataviz make possible aggregate million data company organization INSEE visualize graph diagram map chronology pictograms infographs even unpublished graphic creation photo Data storytelling succeeds narrating data image view analysing bit history Several stage early end 18th century Scottish engineer economist William Playfair created histogram pie chart time series visualization still used today Concrete example arrive early 19th century subject related mortality war… Birth graphic language early 1967 thanks Jacques Bertin digging way presenting information graph colour size shape massive arrival digital technology make data storytelling good thing amount information analysed high new way presenting information thanks digital technique optimise Dataviz 3D example Today data collection generalized whole company analysis transmission result remains practice accessible simple minority employee Data storytelling thus make possible democratize access complex information company longer Business Intelligence department Concrete application Datastorytelling present organization many Use service sale product number visitor site fluctuation price… sector activity generates simple fact creating Excel table order data draw conclusion already part Dataviz useful trade related data practice “piloting data” allows precise decision making go beyond human reasoning thus possible follow many key indicator simultaneously keep broad vision one’s activity also serf general public Generally highlight data educate conscience subject ecology economy synthesis financial result anecdote know bestknown data visualization date back time Napoleon denounces absurdity war 1812 Napoleon led famous Russian campaign 400000 men set conquer Moscow quarter men reached capital Napoleon forced retreat second trap closed troop cold thousand men crossed border again… Charles Joseph Minard contemporary campaign produced map several decade laterTags Dashboard Dataviz Business Intelligence Data Visualization Productivité |
720 | In 2020, Human-Made Objects Tipped the Earth Out of Balance | In 2020, Human-Made Objects Tipped the Earth Out of Balance
A new study redefines our impact on the natural world
Siak River, Pekanbaru, Indonesia. Photo: Barcroft Media/Getty Images
There is considerably more plastic on our planet than there is living animal mass. Buildings and roads account for more mass than trees and shrubs. As of 2020, the weight of humankind’s creations is on track to surpass that of all the living biomass on Earth, a remarkable development that gives us a new way of understanding our impact on the natural world.
Researchers at the Weizmann Institute of Science in Israel detailed this claim in a study published in Nature last week. The authors call 2020 the “crossover point.” After this, human-produced mass, or anthropogenic mass, will outweigh the planet’s natural biomass. There will be more people stuff than other stuff. This has happened quickly: The study explains that anthropogenic mass has doubled every 20 years since the turn of the 20th century.
(Note that the researchers’ calculations are complex and have enough give to swing a few years in either direction. Perhaps we have already passed the crossover point, or are just about to.)
There is a related website that explains the research through a series of infographics: The mass of the Eiffel Tower is roughly equal to the mass represented by all the white rhinos left in the world, and the human-made mass in New York City is roughly equivalent to that of all the fish. It’s like a bedtime story that fades into an unsettling dream.
“We hope that once we all have these somewhat shocking figures before our eyes, we can, as a species, take responsibility.”
You might awake feeling restless. Though you could imagine someone looking at this research and seeing a triumph of humanity — all the world harvested for our buildings and roads and technological progress — a realist will understand that it may portend disaster.
“We hope that once we all have these somewhat shocking figures before our eyes, we can, as a species, take responsibility,” Ron Milo, PhD, an author of the study, said in a press release.
This level of production, referred to as our “socioeconomic metabolism” in the paper, cannot continue forever. It has already changed our planet in dramatic ways, namely through climate change and all that it inflicts: the poisoning of our air, the transformation or outright destruction of natural ecosystems, the intensification of natural disasters, and so much else.
With the natural world ravaged, our socioeconomic metabolism may need to be oriented away from luxuries and the pursuit of unabated economic growth and toward imperfect solutions like geoengineering, or building refugee camps to replace the underwater cities.
The paper also discusses mass waste at length. “If current trends continue, anthropogenic mass, including waste, is expected to exceed 3 [teratonnes] by 2040 — almost triple the dry biomass on Earth,” it says. (“Dry biomass” means water is excluded.)
The term “waste” may seem to suggest that this mass is discarded, able to be forgotten. This is not the case. We currently produce around 55 million tons of electronic waste every year, for instance — “5 million elephants or 110 Burj Khalifas piled on top of each other,” as Josh Lepawsky put it in a story for OneZero. The creation of technology is destructive, and so is its aftermath. Silicon Valley is home to more toxic Superfund sites than any other county in the United States, where chemicals from electronics manufacturing seep into the earth and groundwater. Dead technology becomes junk, often omitting toxic chemicals itself in its journey around the planet. Objects do not necessarily become inert when they are no longer useful; while most anthropogenic mass may be material like concrete and gravel, the study nonetheless invites us to consider the impact of all our creations.
Future humans will have a lot to deal with. This study is a useful document for quantifying the magnitude of our challenge. It is not a call to despair because despair by its very definition can accomplish nothing. In fact, this research is a reminder that our problems can be measured and defined — with effort, they can begin to be solved. | https://futurehuman.medium.com/in-2020-human-made-objects-tipped-the-earth-out-of-balance-1f0c4d7acf42 | ['Damon Beres'] | 2020-12-18 19:13:40.930000+00:00 | ['Society', 'Waste', 'Climate Change', 'Future', 'Environment'] | Title 2020 HumanMade Objects Tipped Earth BalanceContent 2020 HumanMade Objects Tipped Earth Balance new study redefines impact natural world Siak River Pekanbaru Indonesia Photo Barcroft MediaGetty Images considerably plastic planet living animal mass Buildings road account mass tree shrub 2020 weight humankind’s creation track surpass living biomass Earth remarkable development give u new way understanding impact natural world Researchers Weizmann Institute Science Israel detailed claim study published Nature last week author call 2020 “crossover point” humanproduced mass anthropogenic mass outweigh planet’s natural biomass people stuff stuff happened quickly study explains anthropogenic mass doubled every 20 year since turn 20th century Note researchers’ calculation complex enough give swing year either direction Perhaps already passed crossover point related website explains research series infographics mass Eiffel Tower roughly equal mass represented white rhino left world humanmade mass New York City roughly equivalent fish It’s like bedtime story fade unsettling dream “We hope somewhat shocking figure eye specie take responsibility” might awake feeling restless Though could imagine someone looking research seeing triumph humanity — world harvested building road technological progress — realist understand may portend disaster “We hope somewhat shocking figure eye specie take responsibility” Ron Milo PhD author study said press release level production referred “socioeconomic metabolism” paper cannot continue forever already changed planet dramatic way namely climate change inflicts poisoning air transformation outright destruction natural ecosystem intensification natural disaster much else natural world ravaged socioeconomic metabolism may need oriented away luxury pursuit unabated economic growth toward imperfect solution like geoengineering building refugee camp replace underwater city paper also discus mass waste length “If current trend continue anthropogenic mass including waste expected exceed 3 teratonnes 2040 — almost triple dry biomass Earth” say “Dry biomass” mean water excluded term “waste” may seem suggest mass discarded able forgotten case currently produce around 55 million ton electronic waste every year instance — “5 million elephant 110 Burj Khalifas piled top other” Josh Lepawsky put story OneZero creation technology destructive aftermath Silicon Valley home toxic Superfund site county United States chemical electronics manufacturing seep earth groundwater Dead technology becomes junk often omitting toxic chemical journey around planet Objects necessarily become inert longer useful anthropogenic mass may material like concrete gravel study nonetheless invite u consider impact creation Future human lot deal study useful document quantifying magnitude challenge call despair despair definition accomplish nothing fact research reminder problem measured defined — effort begin solvedTags Society Waste Climate Change Future Environment |
721 | How to Solve These 5 Underrated Writer’s Problems | We are talking about the usual problems all the time. I myself have typed imposter syndrome so many times that it comes up in my predictive text and autocorrect. If you google writer’s block you get over 23 million results in a split second. Not to mention the tons of ideas to come up with tons of ideas, the exercises to finetune your writing voice and the productivity hacks to beat procrastination, to build a better writing routine, to get the fiction writing done before you start your day.
They are so typical and frequent that while we are solving the most significant problems in our writing lives we forget about some of the underrated issues that can stop us from thriving.
They are not a lesser form of hindering factors, they are there, but while the list of the above comes up on a daily basis, these occur only occasionally and therefore they seem less important. When you need to deal with putting out the fire of your burning house it is easy to ignore a splinter in your thumb — but you need to deal with that too eventually because it’s uncomfortable, painful and it can get nasty only too quickly.
Here are the few problems I find underrated and untackled — and here’s how to manage them.
The finished-project depression
When you struggle with starting something or finishing that draft you have been working at forever it seems very unreal that you would ever get depressed by having a finished product. But it happens. Especially if you work on something for a long time and it becomes a part of your routine, a part of your life, or if it is an exceptionally vulnerable piece that takes up a huge emotional space in your days.
You submit your draft, you hit publish — and then what? You are empty. Your hands feel idle, your brain feels vacant, your ideas are missing and now you have no idea what to do about it. You were looking forward to the moment when it’s finally done, but then it hits you with full force: what if I can’t replicate it ever again? What if I never do anything good again? What if?
This is the time to celebrate and recharge your batteries. Give yourself some credit and pat yourself on the back. Brag about what you have achieved and collect all the praise you can get — if your own validation feels insufficient. Take a conscious break from writing — decide on the time and stick to your decision. Give yourself 24 hours, a week or a month — without guilt and second thoughts. Distract yourself from the words and indulge in another hobby. If your writing became more than a hobby then you need a new hobby, you know that, right? Look for one and spend time with it. Spend time with your friends and family you ignored during the creation process. Move, exercise and relax. You deserve it. Don’t let your inner slave-master push you onto another project before you had your resting time.
The I-always-write-about-the-same-things distress
You have a look at your last articles, last drafts and there is a whole list of recurring themes and characters all the time. It’s not coincidental, you keep repeating yourself. First, it’s sort of cute, but then you start wondering whether others notice it too. Imposter syndrome hits: you are a fraud. Okay, you might be quite good at writing about yoga, but nothing else works. Okay, you might create a great male character, but he appears in each and every draft you write — it’s only that his eye colour of blue not brown this time. You duck, and brace for impact — they will find it out. You almost want to be found out.
Relax, it’s all okay. You have an audience who reads you for a very good reason: they are interested in your topic, they like your stories and characters and they like your writing voice. No one is that critical with you as you are with yourself. You have your niche for a reason, but to give you a bit of peace of mind you can do this: experiment and diversify. Exaggerate and turn it around.
If you always write in the same niche find another topic that can be important for the same target audience. Know them, learn about them, talk to them. Find out what else they are interested in.
If you find yourself writing the same character all the time, take one trait of him, exaggerate it to the extreme and watch how that alters his position in your plot. Or alternatively take your protagonist and twist him to become your villain. At the end of the process, you wouldn’t recognise the character anymore, let alone your readers.
The someone-took-my-idea-dammit rage
I am not talking about plagiarism, but the occasion when you have the most brilliant book idea and the most exceptional approach to an essay — and shock-horror-scandal it has been already written! Someone had the same idea and they had it before you!
First, there is literally nothing new under the sun — when it comes to ideas. Second, okay, this might hurt… if someone has the exact same idea as you, maybe you need to take up your ideation process a notch, you need to evaluate your own ideas better and kill the very first, very obvious ones at once. This way you can assure that you are not spending time with something that might occur to every other writer on earth. Third, just because the idea is the same it doesn’t mean that your approach or your execution needs to be the same.
According to a formula that I created based on my years of evaluating advertising ideas, the best ideas can be described by three variables: originality of the idea, novelty of the approach and uniqueness of the execution. You can read about it here:
You can check your ideation process, or you can work on your novel approach and unique execution (this is your writing voice).
The too-many-ideas anxiety
We are talking about writer’s block a lot. Staring at the empty page, not being able to start, being paralysed by our own inhibitions — it’s very common. But what happens, when you need to face the opposite? When you don’t get to do anything because you are working on too many things at the same time? When one idea leads to another and your draft folder is overflowing with half sentences, one-paragraph wonders and never finished pieces?
You need to have discipline — to get through this anxiety. It might sound easier said than done, but there is only one thing to do: pick a topic and work on it. If it doesn’t work out — throw it away. Give it your best but if it’s not really yours, then file it in a different folder and revisit it at a later point in life.
Take this advice from Neil Gaiman:
This is how you do it: you sit down at the keyboard and you put one word after another until it’s done. It’s that easy and that hard.
The I-want-my-comfort-zone-back jitters
We all have topics we are comfortable with. We might say it is a difficult topic, but we are even comfortable with the difficulty of it — it’s a calculated risk. And it still feels safe and it still feels to be your choice. But sometimes you wander outside your comfort zone. You might encounter someone who inspires you to try out a different form, to venture into a very personal vulnerable confession, to introduce an unusual item to your routine.
It can easily unbalance you and make you squirm. You get anxious and jittery — this is not what you have signed up for.
Maybe it’s not the time to leave your comfort zone if it makes you more anxious than it excites you, then stop it, get back in there and choose another time.
If you can feel that this is exciting and good and might be only slightly bigger than you expected you can try to conquer it. Go for it, attack it with what you have, experiment, go all-in. Remember — this is not a fight, you don’t have enemies here, the only one you can conquer is yourself. You don’t need to prove anything to anyone — but you. | https://zitafontaine.medium.com/how-to-solve-these-5-underrated-writers-problems-f09309abffe5 | ['Zita Fontaine'] | 2019-09-01 09:29:17.006000+00:00 | ['Mental Health', 'Self', 'Productivity', 'Advice', 'Writing'] | Title Solve 5 Underrated Writer’s ProblemsContent talking usual problem time typed imposter syndrome many time come predictive text autocorrect google writer’s block get 23 million result split second mention ton idea come ton idea exercise finetune writing voice productivity hack beat procrastination build better writing routine get fiction writing done start day typical frequent solving significant problem writing life forget underrated issue stop u thriving lesser form hindering factor list come daily basis occur occasionally therefore seem le important need deal putting fire burning house easy ignore splinter thumb — need deal eventually it’s uncomfortable painful get nasty quickly problem find underrated untackled — here’s manage finishedproject depression struggle starting something finishing draft working forever seems unreal would ever get depressed finished product happens Especially work something long time becomes part routine part life exceptionally vulnerable piece take huge emotional space day submit draft hit publish — empty hand feel idle brain feel vacant idea missing idea looking forward moment it’s finally done hit full force can’t replicate ever never anything good time celebrate recharge battery Give credit pat back Brag achieved collect praise get — validation feel insufficient Take conscious break writing — decide time stick decision Give 24 hour week month — without guilt second thought Distract word indulge another hobby writing became hobby need new hobby know right Look one spend time Spend time friend family ignored creation process Move exercise relax deserve Don’t let inner slavemaster push onto another project resting time Ialwayswriteaboutthesamethings distress look last article last draft whole list recurring theme character time It’s coincidental keep repeating First it’s sort cute start wondering whether others notice Imposter syndrome hit fraud Okay might quite good writing yoga nothing else work Okay might create great male character appears every draft write — it’s eye colour blue brown time duck brace impact — find almost want found Relax it’s okay audience read good reason interested topic like story character like writing voice one critical niche reason give bit peace mind experiment diversify Exaggerate turn around always write niche find another topic important target audience Know learn talk Find else interested find writing character time take one trait exaggerate extreme watch alters position plot alternatively take protagonist twist become villain end process wouldn’t recognise character anymore let alone reader someonetookmyideadammit rage talking plagiarism occasion brilliant book idea exceptional approach essay — shockhorrorscandal already written Someone idea First literally nothing new sun — come idea Second okay might hurt… someone exact idea maybe need take ideation process notch need evaluate idea better kill first obvious one way assure spending time something might occur every writer earth Third idea doesn’t mean approach execution need According formula created based year evaluating advertising idea best idea described three variable originality idea novelty approach uniqueness execution read check ideation process work novel approach unique execution writing voice toomanyideas anxiety talking writer’s block lot Staring empty page able start paralysed inhibition — it’s common happens need face opposite don’t get anything working many thing time one idea lead another draft folder overflowing half sentence oneparagraph wonder never finished piece need discipline — get anxiety might sound easier said done one thing pick topic work doesn’t work — throw away Give best it’s really file different folder revisit later point life Take advice Neil Gaiman sit keyboard put one word another it’s done It’s easy hard Iwantmycomfortzoneback jitter topic comfortable might say difficult topic even comfortable difficulty — it’s calculated risk still feel safe still feel choice sometimes wander outside comfort zone might encounter someone inspires try different form venture personal vulnerable confession introduce unusual item routine easily unbalance make squirm get anxious jittery — signed Maybe it’s time leave comfort zone make anxious excites stop get back choose another time feel exciting good might slightly bigger expected try conquer Go attack experiment go allin Remember — fight don’t enemy one conquer don’t need prove anything anyone — youTags Mental Health Self Productivity Advice Writing |
722 | How to Stop Getting Annoyed at Yourself and Actually Get Motivated this Winter | 1. Stay close to your ‘why’
It’s easy to crawl back into bed when it’s cold and dark. It’s even easier when you tell yourself there is no need to get up. But there is a reason that you told yourself you needed to get up early in the first place. There is something you want to do, something you want to achieve by getting up this early.
When it’s cold and dark it’s easy to lose sight of that reason or make that reason feel less important. But it’s not less important now than it was in the summer. Arguably, working on your dream is one of the most important things in the world.
When you’ve just woken up make sure you stay clear on that reason, remember why you are doing this and what you want to achieve. The one thing you don’t want to be doing is having this same conversation with yourself in a year’s time because you didn’t commit.
To create a habit of getting up early, in the hard times, you need to remember why you are doing it. What’s your dream? Why do you need to get up at this time? Why does it matter?
Actionable tips:
Ask yourself why you wanted to get up early in the first place? What are you trying to achieve?
Ensure the first thing you look at in the morning is related to your goal and you're why.
2. Drink a big glass of water
Hydration gets the body started in the morning. It’s easy to fall onto the sofa, foggy eyed and drift off back to sleep. Instead, create a habit of drinking a big glass of water as the way to start your day. When you think about it, you haven’t drunk a sip of anything for the last 8 hours. Your body needs refreshment.
You’ll find that you feel instantly more awake and more alert for the day. There is really no better way to start the day than with a glass of water.
Part of getting up successfully in the winter is ensuring the first 5 minutes aren’t wasted. It’s about making sure that you don’t get back to bed and you don’t go back on what you promised yourself.
So it makes sense to feel as awake as possible as quickly as possible. That includes drinking a big glass of water as the first step. But it’s not limited to this. Everyone is different. Naturally, there will be things that make me feel more awake in the morning that might make you feel sleepy and vice versa. Experiment with different approaches, here are some ideas:
Meditation
Showering first thing
An early breakfast
Writing
Going outside and getting some fresh air
Having a coffee
Actionable tip:
Work on understanding the habits in the morning that make you feel most awake, experiment with different things.
3. Get dressed and do it early
There is something about being in your pyjamas that tells your body ‘this is rest time’. Of course, when you first wake up it’s easy to feel like staying in bed, you’ve got your pyjamas on, it’s cold, it’s dark. Everything is stacked against you getting back into bed and getting some more rest.
If you want to combat that, then get dressed. If you want to level up even further, get your clothes out the night before and put them in the bathroom. That way when you go to brush your teeth they’ll be there ready and waiting and it will be easy to get changed.
Getting changed changes your mindset and your outlook. ‘Dress for the occasion’ is said for a reason. If you dress for work you’re more likely to do the work. If you are sat in your pyjamas and cosy clothes your more likely to curl up on the sofa and go back to sleep.
Actionable tips:
Dress for the occasion, wake up and get changed.
Put your clothes in the bathroom the night before, brush your teeth and then get changed.
4. Create a morning routine that works
Part of the reason we get stuck for motivation is lack of structure. We wake up, sit down and then think ‘now what?’ That’s an issue because then you are forced to spend the next 30–40 minutes thinking about how to spend your morning and what to do. And if you don’t know what to do, the easiest thing is to fall back to sleep.
Which is quite frankly, a waste of time. In the morning, fresh out of bed, you are sleepy and barely functioning. It’s not the best time to decide on what to do for the day. The result is you overthinking how to spend your morning and ending up not being very productive. Instead, what works better is to decide the night before what you need to do with your morning or, if you prefer, stick to a morning routine that works.
Now I’m not saying you need to a drink of a cocktail of questionably sounding ingredients followed by a 50-mile run and a cold shower all whilst telling yourself you are invincible and will achieve your dreams. But what is quite a good idea is using the time in the morning to do the things that are important to you but that often get lost in the day.
Things like stretching, meditation, writing down any thoughts that have been bugging you. If you find a routine that works for you, try to stick to it the best you can, you’ll find that you’re spending less time than ever trying to decide on what to do and more spend more time doing.
Actionable tips:
Do not use the time in the morning to work out what to do with the day, you’ll be too foggy to think very clearly.
Plan the night before, work out exactly what you are going to do with the morning once you wake up.
Take it a step further and get specific. For example, if you are thinking of ‘writing’ in the morning, get clear on exactly what article you are writing.
5. Do something, anything
Action creates motivation. People are often mistaken in thinking that motivation creates action but it doesn’t. The result is that you wait and wait for motivation to fall into your lap so you can start with your day. What always happens is that the waiting continues and motivation never turns up.
Motivation is a lousy friend.
Instead, start by doing. It doesn’t have to be your hardest task, it just needs to be a task. It needs to be something that you have on your to-do list that can be ticked off. There are loads of things that fit into that category, things like:
Putting the wash away
Putting the washing in the washing machine
Hoovering the downstairs
Washing the pots
Cleaning the kitchen sides
Making a phone call
There are lots of things that take less than five minutes you can tick off your to-do list to make you feel more productive. Once you’ve ticked a few of these off you will find that you are ready to tackle the bigger things.
Actionable tips:
Create momentum by doing. Tick off some small, mundane tasks and you’ll notice you are ready for the bigger tasks.
Write a to-do list that you can get through quickly to get you motivated.
6. Schedule breaks that excite you
Taking a break will male you more productive and more motivated. Often you can fall into the trap of sitting, staring at the desk knowing that you are not doing anything. Yet you trick yourself into thinking you are being productive. By simply sitting at the desk, somehow it makes you feel like you are chipping away at your goals.
You’re not.
You’re wasting time. If you are not doing you have three options, sit there and do nothing, sit there and do something, or get up and take a break. It might sound counterintuitive to pick the third option but it’s not. After hours spent staring at the screen, taking a break and getting outside can only be a good thing. Your brain gets tired of the same old scenery. It gets stuck in that track of thinking.
Instead, get up and get outside. The different sights and smells change your train of thought and give you some rest bite for a little while. Changing the space will mean that when you come back to your work you feel refreshed and ready to go.
Not only that though, but you should also schedule breaks that excite you. Is there a Christmas film you’re dying to watch or Christmas decorations you want to put up? Use that as a motivation to get your work done.
Actionable tips:
Use the season to motivate you, schedule Christmas related breaks that excite you.
Understand that breaks actually make you more productive in the long run.
7. Schedule your time
Scheduled time is time that you don’t need to think about too much. It’s time that is already in the bag and assigned to a certain task. Now whilst you can dip into overscheduling and making life a little too regimented, scheduling time to work on the things that are important to you is key.
I find that if I schedule my important tasks on a Sunday it makes it clear what I need to achieve in the week. It makes it simple to follow because it’s broken down by day and by the task.
It means I can feel confident that if I stick to the plan I will be able to achieve all the things I want to in the time I’ve got. Having this plan takes the weight off and it gives you less of a reason to procrastinate. When you have unstructured time you can sometimes feel lost with how to spend it.
Actionable tips: | https://medium.com/curious/how-to-stop-getting-annoyed-at-yourself-and-actually-get-motivated-this-winter-5983d1a16544 | ['Eve Arnold'] | 2020-12-09 17:12:48.936000+00:00 | ['Motivation', 'Self Improvement', 'Self', 'Psychology', 'Productivity'] | Title Stop Getting Annoyed Actually Get Motivated WinterContent 1 Stay close ‘why’ It’s easy crawl back bed it’s cold dark It’s even easier tell need get reason told needed get early first place something want something want achieve getting early it’s cold dark it’s easy lose sight reason make reason feel le important it’s le important summer Arguably working dream one important thing world you’ve woken make sure stay clear reason remember want achieve one thing don’t want conversation year’s time didn’t commit create habit getting early hard time need remember What’s dream need get time matter Actionable tip Ask wanted get early first place trying achieve Ensure first thing look morning related goal youre 2 Drink big glass water Hydration get body started morning It’s easy fall onto sofa foggy eyed drift back sleep Instead create habit drinking big glass water way start day think haven’t drunk sip anything last 8 hour body need refreshment You’ll find feel instantly awake alert day really better way start day glass water Part getting successfully winter ensuring first 5 minute aren’t wasted It’s making sure don’t get back bed don’t go back promised make sense feel awake possible quickly possible includes drinking big glass water first step it’s limited Everyone different Naturally thing make feel awake morning might make feel sleepy vice versa Experiment different approach idea Meditation Showering first thing early breakfast Writing Going outside getting fresh air coffee Actionable tip Work understanding habit morning make feel awake experiment different thing 3 Get dressed early something pyjama tell body ‘this rest time’ course first wake it’s easy feel like staying bed you’ve got pyjama it’s cold it’s dark Everything stacked getting back bed getting rest want combat get dressed want level even get clothes night put bathroom way go brush teeth they’ll ready waiting easy get changed Getting changed change mindset outlook ‘Dress occasion’ said reason dress work you’re likely work sat pyjama cosy clothes likely curl sofa go back sleep Actionable tip Dress occasion wake get changed Put clothes bathroom night brush teeth get changed 4 Create morning routine work Part reason get stuck motivation lack structure wake sit think ‘now what’ That’s issue forced spend next 30–40 minute thinking spend morning don’t know easiest thing fall back sleep quite frankly waste time morning fresh bed sleepy barely functioning It’s best time decide day result overthinking spend morning ending productive Instead work better decide night need morning prefer stick morning routine work I’m saying need drink cocktail questionably sounding ingredient followed 50mile run cold shower whilst telling invincible achieve dream quite good idea using time morning thing important often get lost day Things like stretching meditation writing thought bugging find routine work try stick best you’ll find you’re spending le time ever trying decide spend time Actionable tip use time morning work day you’ll foggy think clearly Plan night work exactly going morning wake Take step get specific example thinking ‘writing’ morning get clear exactly article writing 5 something anything Action creates motivation People often mistaken thinking motivation creates action doesn’t result wait wait motivation fall lap start day always happens waiting continues motivation never turn Motivation lousy friend Instead start doesn’t hardest task need task need something todo list ticked load thing fit category thing like Putting wash away Putting washing washing machine Hoovering downstairs Washing pot Cleaning kitchen side Making phone call lot thing take le five minute tick todo list make feel productive you’ve ticked find ready tackle bigger thing Actionable tip Create momentum Tick small mundane task you’ll notice ready bigger task Write todo list get quickly get motivated 6 Schedule break excite Taking break male productive motivated Often fall trap sitting staring desk knowing anything Yet trick thinking productive simply sitting desk somehow make feel like chipping away goal You’re You’re wasting time three option sit nothing sit something get take break might sound counterintuitive pick third option it’s hour spent staring screen taking break getting outside good thing brain get tired old scenery get stuck track thinking Instead get get outside different sight smell change train thought give rest bite little Changing space mean come back work feel refreshed ready go though also schedule break excite Christmas film you’re dying watch Christmas decoration want put Use motivation get work done Actionable tip Use season motivate schedule Christmas related break excite Understand break actually make productive long run 7 Schedule time Scheduled time time don’t need think much It’s time already bag assigned certain task whilst dip overscheduling making life little regimented scheduling time work thing important key find schedule important task Sunday make clear need achieve week make simple follow it’s broken day task mean feel confident stick plan able achieve thing want time I’ve got plan take weight give le reason procrastinate unstructured time sometimes feel lost spend Actionable tipsTags Motivation Self Improvement Self Psychology Productivity |
723 | Designed Intelligence: Achieving the full potential of human and artificial intelligence | By Connor Upton, Group Design Director, Fjord at The Dock, Accenture’s flagship R&D and global innovation center in Dublin, Ireland.
Artificial intelligence (AI) is already firmly enmeshed in our lives and businesses the world over. It sits behind many of our apps and services, making our busy lives manageable through prediction, personalisation, and automation. We once asked how can we apply artificial intelligence, but the question now is how do we use AI to its full potential? It’s a big question — not just for businesses aiming to improve services but for each of us as individuals.
As we grow ever more reliant upon it, there is a growing critique of its pervasiveness in our lives and its unintended consequences. At times, AI systems have failed to make good decisions because they can’t quite fathom the complexity and context of the real-world.
We’ve seen evidence of algorithmic discrimination with gender bias in recruitment recommendations and credit scoring, and racial bias in healthcare and judicial sentencing decision support systems.
There are also issues around AI dependency when people become over-reliant on technology. The most dramatic examples of this are autonomous vehicle accidents, where drivers don’t have full awareness when the system needs them to step in.
Failures like these deeply affect the lives of individuals, they can result in reputational damage for businesses, and at a societal level they reduce trust in technology and data. They stem from a mindset that perceives AI as automation technology and a business case that seeks to replace human activity to improve efficiency. In reality, most things that can be automated will be automated. After all, we don’t want to be hand weaving our clothes or managing the spam in our inboxes. But as AI becomes broadly adopted and is applied to more complex problems we must be careful to understand the unique value that people bring. People tend to not just define useful processes but are also remarkably flexible, for example developing communications and hacks that keep organisations running. If we don’t learn from these, we run the risk of making systems less resilient and increasingly risky. By the same token, paying close attention to the contribution of the human in AI systems allows us to better confront and account for our own harmful biases.
“When designing artificial intelligence it’s critical that we also design for human intelligence and the interactions between these agents.”
Beyond the risks, taking a more human- and systems-oriented approach to AI brings huge opportunities. What if we looked beyond cost reduction and adopted a growth mindset? Organisations are driven by people and their ingenuity. So, the fundamental question should be: how might we use AI to amplify people’s abilities to create net new value? Achieving this requires a new systematic approach.
We call this approach Designed Intelligence. It encapsulates three aspects of how businesses can harness AI to generate value responsibly. First is the application of AI to envision new strategies, second is harnessing it to empower people in increasingly complex and dynamic systems, and third is using it to enhance the human experience.
The three E’s of Designed Intelligence.
Envisioning new strategies
AI-inspired Tapas from the Dock
Artificial Intelligence is not like human intelligence. It sees the world in a very different, data-oriented way. This alternative perspective can help people break out of their orthodoxies and envision new strategies, services, and products that they would never be able to imagine on their own. We use this approach in projects and design thinking sessions to help businesses blend human expertise with analytical insight, giving them what we call “informed intuition”. We’ve applied this to help food manufacturers develop new flavour combinations, to help banks imagine new personal services for financial wellness, and to help hotels redesign how they measure quality.
Empowering people within systems
While digital services have made our lives more convenient, they can also make systems difficult to understand and less transparent. For customers, this includes understanding why you’re getting that product recommendation, or not getting that loan approval. For workers, it might be figuring out why your caseload is changing or deciphering your performance report. As AI becomes more prevalent in digital services, their logic could become even less clear as algorithms are typically designed for outcomes, not explanations.
This poses a challenge for businesses that want to apply AI. While AI has the potential to bring great benefits, it will only succeed if people are willing to adopt it. In our research, we find that trust between people and AI tools requires transparency, control, and relevant interactions. To get the best out of an AI investment, we help businesses understand how their people are currently working, what pain points they encounter and how AI will change their practices.
For example, in our work on the Accenture Logistics Platform, we helped postal dispatchers harness AI to predict service demand and optimise delivery routes. We also designed a mobile app to guide delivery workers in the field. Workers can make adjustments to the AI’s suggested routes directly in the application and even override them if needed. This means workers can build up trust in the service over time while remaining empowered. Crucially this ability to act independently is also a source of new training data that allows the AI models to get feedback, adapt to changes and improve.
Interface from the Accenture Logistics Platform. Click to watch video.
As the world evolves, businesses will continue to need people long into the future. People can deal with ill-defined problems, interpret the real world with all its messiness and bring context to complex issues like social biases and discrimination. At the same time, AI can radically scale our human expertise by speeding up mundane tasks, monitoring dynamic systems and discovering relationships in large complex data sets.
Enhancing the human experience
At the experience level, we’re applying AI to extend human capabilities. Computer vision and voice technologies are forms of AI that provide more natural ways to interact with information. Applied correctly they can create automagical experiences that truly engage customers. Working with VELUX, for example, we built a mobile app that lets people visualise what new roof windows would look like in their home. This intuitive service gives homeowners the ability to envisage 3D space like an architect and provides them with the confidence to make dramatic changes to their homes. | https://medium.com/design-voices/designed-intelligence-achieving-the-full-potential-of-human-and-artificial-intelligence-e84ca3920255 | [] | 2020-09-11 10:58:14.042000+00:00 | ['AI', 'Artificial Intelligence', 'Innovation', 'Design', 'Designed Intelligence'] | Title Designed Intelligence Achieving full potential human artificial intelligenceContent Connor Upton Group Design Director Fjord Dock Accenture’s flagship RD global innovation center Dublin Ireland Artificial intelligence AI already firmly enmeshed life business world sits behind many apps service making busy life manageable prediction personalisation automation asked apply artificial intelligence question use AI full potential It’s big question — business aiming improve service u individual grow ever reliant upon growing critique pervasiveness life unintended consequence time AI system failed make good decision can’t quite fathom complexity context realworld We’ve seen evidence algorithmic discrimination gender bias recruitment recommendation credit scoring racial bias healthcare judicial sentencing decision support system also issue around AI dependency people become overreliant technology dramatic example autonomous vehicle accident driver don’t full awareness system need step Failures like deeply affect life individual result reputational damage business societal level reduce trust technology data stem mindset perceives AI automation technology business case seek replace human activity improve efficiency reality thing automated automated don’t want hand weaving clothes managing spam inboxes AI becomes broadly adopted applied complex problem must careful understand unique value people bring People tend define useful process also remarkably flexible example developing communication hack keep organisation running don’t learn run risk making system le resilient increasingly risky token paying close attention contribution human AI system allows u better confront account harmful bias “When designing artificial intelligence it’s critical also design human intelligence interaction agents” Beyond risk taking human systemsoriented approach AI brings huge opportunity looked beyond cost reduction adopted growth mindset Organisations driven people ingenuity fundamental question might use AI amplify people’s ability create net new value Achieving requires new systematic approach call approach Designed Intelligence encapsulates three aspect business harness AI generate value responsibly First application AI envision new strategy second harnessing empower people increasingly complex dynamic system third using enhance human experience three E’s Designed Intelligence Envisioning new strategy AIinspired Tapas Dock Artificial Intelligence like human intelligence see world different dataoriented way alternative perspective help people break orthodoxy envision new strategy service product would never able imagine use approach project design thinking session help business blend human expertise analytical insight giving call “informed intuition” We’ve applied help food manufacturer develop new flavour combination help bank imagine new personal service financial wellness help hotel redesign measure quality Empowering people within system digital service made life convenient also make system difficult understand le transparent customer includes understanding you’re getting product recommendation getting loan approval worker might figuring caseload changing deciphering performance report AI becomes prevalent digital service logic could become even le clear algorithm typically designed outcome explanation pose challenge business want apply AI AI potential bring great benefit succeed people willing adopt research find trust people AI tool requires transparency control relevant interaction get best AI investment help business understand people currently working pain point encounter AI change practice example work Accenture Logistics Platform helped postal dispatcher harness AI predict service demand optimise delivery route also designed mobile app guide delivery worker field Workers make adjustment AI’s suggested route directly application even override needed mean worker build trust service time remaining empowered Crucially ability act independently also source new training data allows AI model get feedback adapt change improve Interface Accenture Logistics Platform Click watch video world evolves business continue need people long future People deal illdefined problem interpret real world messiness bring context complex issue like social bias discrimination time AI radically scale human expertise speeding mundane task monitoring dynamic system discovering relationship large complex data set Enhancing human experience experience level we’re applying AI extend human capability Computer vision voice technology form AI provide natural way interact information Applied correctly create automagical experience truly engage customer Working VELUX example built mobile app let people visualise new roof window would look like home intuitive service give homeowner ability envisage 3D space like architect provides confidence make dramatic change homesTags AI Artificial Intelligence Innovation Design Designed Intelligence |
724 | Permissionless Professors #2: Sam Parr | We believe monetization is a hidden secret. Secrets hiding in plain sight. From the psychology of anchoring, the mathematics of power-law pricing tables, the application of demand elasticity, and the market positioning of your price — your path to better monetization awaits.
Follow | https://medium.com/monetization-manifesto/permissionless-professors-2-sam-parr-54a603be925b | ['Gary Bailey - Monetization Manifesto'] | 2020-12-17 14:47:49.428000+00:00 | ['Growth', 'Marketing', 'Monetization', 'Startup', 'Entrepreneurship'] | Title Permissionless Professors 2 Sam ParrContent believe monetization hidden secret Secrets hiding plain sight psychology anchoring mathematics powerlaw pricing table application demand elasticity market positioning price — path better monetization awaits FollowTags Growth Marketing Monetization Startup Entrepreneurship |
725 | AWS — Deploying React App With Java Backend On EKS | AWS — Deploying React App With Java Backend On EKS
A step by step guide with an example project
AWS provides more than 100 services and it’s very important to know which service you should select for your needs. Amazon Elastic Kubernetes Service (Amazon EKS) is a managed service that makes it easy for you to run Kubernetes on AWS without needing to stand up or maintain your own Kubernetes control plane. Kubernetes is an open-source system for automating the deployment, scaling, and management of containerized applications.
In this post, we are going to deploy the React application with the Java environment. First, we dockerize our app and push that image to Amazon ECR and run that app on Amazon EKS.
Example Project
Prerequisites
Dockerize the Project
Pushing Docker Image To ECR
Create a Cluster and Worker Nodes
Configure kubectl to use Cluster
Deploy Kubernetes Objects On AWS EKS Cluster
Summary
Conclusion
Example Project
This is a simple project which demonstrates developing and running React application with Java. We have a simple app in which we can add users, count, and display them at the side, and retrieve them whenever you want.
Example Project
If you want to practice your own here is a Github link to this project. You can clone it and run it on your machine as well. | https://medium.com/bb-tutorials-and-thoughts/aws-deploying-react-app-with-java-backend-on-eks-2cc396888944 | ['Bhargav Bachina'] | 2020-11-15 06:02:54.844000+00:00 | ['AWS', 'Programming', 'Software Development', 'Kubernetes', 'Cloud Computing'] | Title AWS — Deploying React App Java Backend EKSContent AWS — Deploying React App Java Backend EKS step step guide example project AWS provides 100 service it’s important know service select need Amazon Elastic Kubernetes Service Amazon EKS managed service make easy run Kubernetes AWS without needing stand maintain Kubernetes control plane Kubernetes opensource system automating deployment scaling management containerized application post going deploy React application Java environment First dockerize app push image Amazon ECR run app Amazon EKS Example Project Prerequisites Dockerize Project Pushing Docker Image ECR Create Cluster Worker Nodes Configure kubectl use Cluster Deploy Kubernetes Objects AWS EKS Cluster Summary Conclusion Example Project simple project demonstrates developing running React application Java simple app add user count display side retrieve whenever want Example Project want practice Github link project clone run machine wellTags AWS Programming Software Development Kubernetes Cloud Computing |
726 | Understanding Facebook Reactions using Google Sentiment Analysis | Facebook Angry Reactions — Sentiment Analysis
We can see for the posts that generated a strong Angry Response, the majority of them are classified by google cloud as texts with a negative or neutral score.
There are angry posts that have a positive score, but the majority of posts that generated a angry response, have a text description that got a negative score.
If we look also at the magnitude, we see that it`s getting harder to find angry posts in the positive score part (right part of the image), after the magnitude of the text gets bigger then 1.5.
Facebook Love Reactions — Sentiment Analysis
Love is positive,hate is negative. Proven with data.
If we look were is the love, we see a different distribution. Articles that have a negative score tend to get less love. At least if the magnitude of the text is over 1
Wows posts are shallow and superficial
I never understood the wows. What are they ? It`s not a haha, is not a like, is not love. I consider them something shallow . I don`t get wows.
Why i`m saying this story ?
Because this play good with my confirmation bias that wow is superficial,shallow. Thanks Ray Alez for the confirmation biased article.
And now with the data that google had offered, it helps me demonstrate this theory. Look at the image and then scroll down for a pseudo-explanation.
The majority of the posts that got a big percentage of the total number of reactions the reaction wow, had a magnitude of the text lower then 1.
So we have 2 possible explanations. One, the posts that get a big percent of overall reactions as wow does not convey any emotional content. (shallow)
Second, the posts that get a big percent of overall reactions as wow are short in text.
The magnitude of a document’s sentiment indicates how much emotional content is present within the document, and this value is often proportional to the length of the document. A document with a neutral score (around 0.0 ) may indicate a low-emotion document, or may indicate mixed emotions, with both high positive and negative values which cancel each out.
This is a list with posts that got at least 1000 reactions, and between 75 and 100% of that reactions were WOW
A distribution per FB page looks like this. I love it. Using WOW`s seems like a good way to find “sensational posts” posted by the alternative media or other categories.
What makes a user sad ? Facebook Sad Reactions.
The top pages that generate the sad emotion are :
Posts were 80 to 100% of the emotions are sad.
- The Detroit Free Press lost a beloved family member today. Former restaurant critic Sylvia Rector died at the age of 66. She will be missed, and remembered fondly. - Alan Thicke, best known as the dad on ‘Growing Pains’, dies at 69 after reportedly suffering fatal heart attack. - A tribute to the iconic world leaders, athletes and celebrities who passed away in 2016.
Almost all articles are related with the death of somebody.
The interesting thing here is combining the information's from the percent of people that are sad to the percent of people that are angry.
If somebody dies because he was old, people will give a sad reaction, leave a comment and they are done.
But if somebody dies because a bomb exploded, then you have a mix of sad and angry reactions, with some weird wow in the middle.
Posts were 60 to 80% of the emotions are sad and 20 to 40% of the emotions are angry.
Here we can mixed posts like
- “I glanced down…and saw my 12-week old fetus. Its head was smashed, and it’s eyes were on either side of its head. Its body was attached to the head, and only the right arm and hand remained. I was shocked and horrified by what I saw.” — source facebook.com/liveaction - Investigators are trying to piece together the final moments of three children who were killed when they arrived home with their mother to find a gunman waiting inside their Albuquerque, New Mexico home, officials said. — FoxNews The chilling words … Hundreds of Aleppo men are reported “missing.” Reminds me of Srebrenica. Turned out 8,000 Muslim men and boys who had gone “missing” were in fact slaughtered. — facebook.com/camanpour/
Posts were the number of sad and angry reactions are almost the same.
- At least 9 people are dead after truck slams into a Christmas market in Berlin and the attacker is on the run. — BretBaierSR - We can’t forget about Flint.
2017 will mark the three-year anniversary of the water crisis that poisoned thousands of residents in Flint. Still now, many of the city’s residents are still using bottled water to drink, to cook, to bathe.
This Sunday, watch a special 17-minute documentary video on life in Flint created by Ryan Garza and Brian Kaufman of the Detroit Free Press. — facebook.com/detroitfreepress/
I`m curious if we can see a connection between the number of sad vs angry percentage for parts of the world. We take 1000 articles were somebody was killed in the US, 1000 were somebody was killed in Europe, 1000 were somebody was killed in the middle east and we compare the reactions percentage.
Do you Haha ? — Facebook Hahas reactions
This following pages seems specialized in creating Haha moments with their Facebook users.
Posts that have at more then 80% Hahas
- Here’s a video that puts the “choke” in artichoke. — FoxNews - A lone protester outside the Trump victory tour embarrasses herself and threatens to sue Infowars for lawfully filming in a public place. — AlexanderEmerickJones - Trump is a living parody of himself. This is remarkably ridiculous. — usuncut - It’s fair to say that Trump supporters aren’t exactly the brightest Cran in the box… — OccupyDemocrats
Posts that have between 60–80% Hahas and more then 10% Angry.
I would rather ________ than read the leftist Washington Post. — Breitbart Andrea Mitchell’s anti-Trump fretting has officially reached fever pitch! Now she’s hand-wringing that Obama’s climate agenda might dare be called into question under Trump. — newsbusters Whoopi Goldberg slams Trump! — Breitbart During his last press conference of 2016, President Barack Obama shared his thoughts on Hillary Clinton’s treatment in this year’s election.
Do you agree or disagree with his statement? — FoxNews
About Me
In the last 3 years i collaborate with the Rise Project, par of the OCCRP investigative Journalism network, were i do data analysis and pattern recognition to uncover patterns of corruption in unstructured datasets.
In September 2016 i have moved to San Francisco, to start a new life. Searching for a Job were i can apply my expertise and pay the rent in SFO.
Currently Building a tool that detects possible fake viral news, before they go viral.
You can find me online on Medium Florin Badita, AngelList, Twitter , Linkedin, Openstreetmap, Github, Quora, Facebook
Sometimes i write on my blog http://florinbadita.com/ | https://medium.com/google-cloud/understanding-facebook-reactions-using-sentiment-analysis-f17b6e561ff3 | ['Florin Badita'] | 2019-02-13 07:31:05.407000+00:00 | ['Social Media', 'Facebook', 'Data Visualization', 'Big Data', 'Journalism'] | Title Understanding Facebook Reactions using Google Sentiment AnalysisContent Facebook Angry Reactions — Sentiment Analysis see post generated strong Angry Response majority classified google cloud text negative neutral score angry post positive score majority post generated angry response text description got negative score look also magnitude see getting harder find angry post positive score part right part image magnitude text get bigger 15 Facebook Love Reactions — Sentiment Analysis Love positivehate negative Proven data look love see different distribution Articles negative score tend get le love least magnitude text 1 Wows post shallow superficial never understood wow haha like love consider something shallow dont get wow im saying story play good confirmation bias wow superficialshallow Thanks Ray Alez confirmation biased article data google offered help demonstrate theory Look image scroll pseudoexplanation majority post got big percentage total number reaction reaction wow magnitude text lower 1 2 possible explanation One post get big percent overall reaction wow convey emotional content shallow Second post get big percent overall reaction wow short text magnitude document’s sentiment indicates much emotional content present within document value often proportional length document document neutral score around 00 may indicate lowemotion document may indicate mixed emotion high positive negative value cancel list post got least 1000 reaction 75 100 reaction WOW distribution per FB page look like love Using WOWs seems like good way find “sensational posts” posted alternative medium category make user sad Facebook Sad Reactions top page generate sad emotion Posts 80 100 emotion sad Detroit Free Press lost beloved family member today Former restaurant critic Sylvia Rector died age 66 missed remembered fondly Alan Thicke best known dad ‘Growing Pains’ dy 69 reportedly suffering fatal heart attack tribute iconic world leader athlete celebrity passed away 2016 Almost article related death somebody interesting thing combining information percent people sad percent people angry somebody dy old people give sad reaction leave comment done somebody dy bomb exploded mix sad angry reaction weird wow middle Posts 60 80 emotion sad 20 40 emotion angry mixed post like “I glanced down…and saw 12week old fetus head smashed it’s eye either side head body attached head right arm hand remained shocked horrified saw” — source facebookcomliveaction Investigators trying piece together final moment three child killed arrived home mother find gunman waiting inside Albuquerque New Mexico home official said — FoxNews chilling word … Hundreds Aleppo men reported “missing” Reminds Srebrenica Turned 8000 Muslim men boy gone “missing” fact slaughtered — facebookcomcamanpour Posts number sad angry reaction almost least 9 people dead truck slam Christmas market Berlin attacker run — BretBaierSR can’t forget Flint 2017 mark threeyear anniversary water crisis poisoned thousand resident Flint Still many city’s resident still using bottled water drink cook bathe Sunday watch special 17minute documentary video life Flint created Ryan Garza Brian Kaufman Detroit Free Press — facebookcomdetroitfreepress Im curious see connection number sad v angry percentage part world take 1000 article somebody killed US 1000 somebody killed Europe 1000 somebody killed middle east compare reaction percentage Haha — Facebook Hahas reaction following page seems specialized creating Haha moment Facebook user Posts 80 Hahas Here’s video put “choke” artichoke — FoxNews lone protester outside Trump victory tour embarrasses threatens sue Infowars lawfully filming public place — AlexanderEmerickJones Trump living parody remarkably ridiculous — usuncut It’s fair say Trump supporter aren’t exactly brightest Cran box… — OccupyDemocrats Posts 60–80 Hahas 10 Angry would rather read leftist Washington Post — Breitbart Andrea Mitchell’s antiTrump fretting officially reached fever pitch she’s handwringing Obama’s climate agenda might dare called question Trump — newsbusters Whoopi Goldberg slam Trump — Breitbart last press conference 2016 President Barack Obama shared thought Hillary Clinton’s treatment year’s election agree disagree statement — FoxNews last 3 year collaborate Rise Project par OCCRP investigative Journalism network data analysis pattern recognition uncover pattern corruption unstructured datasets September 2016 moved San Francisco start new life Searching Job apply expertise pay rent SFO Currently Building tool detects possible fake viral news go viral find online Medium Florin Badita AngelList Twitter Linkedin Openstreetmap Github Quora Facebook Sometimes write blog httpflorinbaditacomTags Social Media Facebook Data Visualization Big Data Journalism |
727 | You Can Live Without Social Media | I used to be the woman who worked a room. If invited, I would’ve gone to the opening of an envelope. Social media didn’t exist in my twenties, so you had to occupy physical spaces. A breathing body was proof of life. Emails wouldn’t suffice since we were suspicious of the kind of communication that could disappear into a black hole. There was no did you get my email when someone’s breathing two feet in front of your face.
Back when I was playing with my abacus and learning how to start fires in my cave, I worked in book publishing. Friends used to introduce me as someone who was good to know. The concept of FOMO was non-existent, but I had a severe case of it. I thought the more I was seen, the more I put myself out there, the more people I accumulated in my contact list, something would happen. Although I wasn’t entirely certain what that something was.
I suppose it worked because when I published my first book in 2008, it was everywhere. Blogs covered it. Podcast hosts and website editors interviewed me. My book was featured in USA Today, Vanity Fair, Elle, and Entertainment Weekly. While the world was smaller back then — we didn’t have the tweet swarm, Pavlovian status updates, fingers habituated to doomsday scrolling and swiping— technology would soon make me feel claustrophobic and alone.
Although my book wasn’t a bestseller, it earned out its advance and was well-received. All that hustling, networking, and work paid off — but at what cost? I was the clown living in a hurt circus, unable to strip the mask from my face. I swallowed all the best parts of me so I could make adults sporting ironic eyewear laugh at my expense. Drinking softened the edges until the room fell too quiet and I could forget I was there.
I knew some people laughed at me behind my back. There was a now-disgraced Paris Review editor who sized me up at a book party and walked away. There was the guy who wrote self-absorbed stories in my Columbia workshop who pretended to not know me. When he published a novel about a self-absorbed guy suffering the tortures of the damned, he experienced some minor celebrity. Every time we were introduced, he would act as if it was the first time we’d met. And then there those who issued barbs that weren’t so thinly-veiled.
Still, I’d smile and show up. Still, I’d ride the subway home in tears.
The mask I’d been wearing became the shape of my face.
In 2009, I torched the joint. Closed shop on my revered literary magazine and resigned from my lucrative book publishing job. I burned all the bridges and devoured the ashes. I took a senior role at a digital media agency and threw myself into my work.
The internet exploded. Everyone was on Facebook, YouTube, and Twitter. Instagram would launch the following year. All the physical spaces morphed into digital ones. We hid behind our shiny devices. Everyone preened for the screen where we became the architects of our personal brands. The new world order dictated if you were nobody if you weren’t online.
It was like a job interview with cocktails only you were stone sober and you knew you were never getting the job. Still, you had to stomach the small talk.
While I signed up for all the accounts, I didn’t know how to be on social media when I realized digital masks had replaced physical ones. Nobody wanted your present-tense sadness. They preferred you not to kick up a fuss or make a scene. Your pain was awkward and uncomfortable and no one wanted to bear the weight of it. You were expected to deliver a life in sepia where everything was beautiful and nothing hurt.
Otherwise, the motley lot unfollowed, muted, and blocked you. We got good at bleaching our feeds clean of hurt. Of all the things that make us human.
While I build brands for a living, I struggled with mine. My personality is polarizing. I have opinions and they’re almost always unpopular. I write short, clipped emails. Unsolicited advice frustrates me — I must’ve blacked out when I asked for your opinion. Yet, people still give it and I have a hard time hiding my irritation. I’m guarded because I’ve been hurt and used and laughed at and I’ve much to protect. People are rarely pure — they’ll shake you down for what you can give. They’ll dig through your contacts list and virtual pockets. They’ll pick you clean.
I know this because it’s happened time and time and time again. How do I sell congeniality when I’m a walking wound?
But I thought if I endured it, spent enough time online, I would find an agent to replace the one I’d resigned and get that third book off the ground. I’d blow up my consulting business. I’d have million-dollar course launches so I can spend my twilight years wearing muumuus and saving puppies.
Instead, I had an Ativan and Sancerre-fueled public online breakdown where my “friends” issued fatwas and virtual restraining orders. They reported my tweets, blocked and unfollowed me, and dodged my calls. They’re mental health advocates until depression fills the rooms of their house.
Instead of pulling a 2009 and falling out of the frame, I set boundaries and made a plan. I read every book on returning to the life I’d grown up living — the irony of which doesn’t escape me. I devoured Cal Newport’s Digital Minimalism and saw its plain, simple truths. Here’s what I did.
Evaluate the purpose of every digital platform you use
People are apoplectic when they learn I’m not on every social network. They tell me I’ll never get business leads until I remind them social media has never delivered me clients — my writing has. Don’t listen to the fear-mongering and herd mentality. Most people follow the herd without questioning why they’re moving in the first place and in what direction.
I stepped back and defined how I want to spend my days. I evaluated each platform and defined its value, use, and purpose. This is where you need to get surgical and honest with yourself and perform that old cost-benefit analysis. Facebook and Twitter only made me angry, and I can curate my news sources instead of reacting in the moment. Instagram made me feel like a loser. Pinterest, I never understood. YouTube frightens me. I know I’d snort TikTok like the coke I did at Rodman’s house when I was twenty-six.
I settled on LinkedIn and Medium. I have a tiny, private Instagram account where I mostly look at animal rescue videos when I’m feeling blue. There’s a veneer, a wall to LinkedIn I like, which keeps everything professional and on an even keel. I’ve been on Medium since 2013, and I see it as a place where I can showcase my smarts and experiment with fiction no one reads.
I don’t get caught up in the dramas and politicking because I had enough of that in book publishing, and it’s the same, tired show only a different channel. I don’t care about claps, income, and popularity because I’m allergic to people and I’d rather assemble a tiny, mighty community than get bombarded by the masses. I’ve been writing dark literary fiction for decades — I’m used to being unpopular and not widely read.
Medium is a place where I understand the geography and social order and know where I fit.
Now, I spend time writing and sharing work that means something to me and moves and educates people. I oscillate between my two storytelling worlds — story writing and marketing — and I love the feeling of switching off one side of my brain and firing up another. I buy cookbooks to feel the weight of the paper and binding in my hands and make food without posting pictures of it.
Remember when a room was clouded with flour?
Return to the life I knew as a kid
We took the things we held in our hands for granted. I spend much of my days tethered to a laptop and a phone, so in my off-hours, I strive for balance. My hobbies are tactile and multi-sensory. I walk, hike, run, write longhand, write letters, read books, participate in book exchanges, play the clarinet, and cook. Now that I rarely see people or have the opportunity to hold them close, touching, smelling, and being around that which is real is even more important to me.
I’m a better artist who’s constantly curious because I reach for a world beyond the digital borders we’ve created. I don’t read how-to books on writing, I diagram novels, mark up the pages, and vivisect authors who challenge me. I’m an artist and professional who’s evolving and growing because I exercise my senses beyond a keyboard and a screen.
Every morning, I walk for five miles while listening to podcasts. Sometimes, I walk in quiet, absorbing the streets, smells, and sounds.
I see humans instead of their names on screens
I set up monthly Zoom dates with old peers and friends. There’s no agenda other than to talk to a human. Sometimes, we don’t even use the video because we’re fatigued and have no interest in putting on pants. But it’s good to hear the voices of people I love and respect. I miss that. I miss cues and decoding body language. I miss vibing off of someone’s physical energy and breath. We talk about our careers, lives, and children (my cat’s my kid, so there’s that). | https://medium.com/the-gathering-kind/you-can-live-without-social-media-59d4a5d8d6e5 | ['Felicia C. Sullivan'] | 2020-12-13 19:04:12.180000+00:00 | ['Social Media', 'Writing', 'Mental Health', 'Relationships', 'Creativity'] | Title Live Without Social MediaContent used woman worked room invited would’ve gone opening envelope Social medium didn’t exist twenty occupy physical space breathing body proof life Emails wouldn’t suffice since suspicious kind communication could disappear black hole get email someone’s breathing two foot front face Back playing abacus learning start fire cave worked book publishing Friends used introduce someone good know concept FOMO nonexistent severe case thought seen put people accumulated contact list something would happen Although wasn’t entirely certain something suppose worked published first book 2008 everywhere Blogs covered Podcast host website editor interviewed book featured USA Today Vanity Fair Elle Entertainment Weekly world smaller back — didn’t tweet swarm Pavlovian status update finger habituated doomsday scrolling swiping— technology would soon make feel claustrophobic alone Although book wasn’t bestseller earned advance wellreceived hustling networking work paid — cost clown living hurt circus unable strip mask face swallowed best part could make adult sporting ironic eyewear laugh expense Drinking softened edge room fell quiet could forget knew people laughed behind back nowdisgraced Paris Review editor sized book party walked away guy wrote selfabsorbed story Columbia workshop pretended know published novel selfabsorbed guy suffering torture damned experienced minor celebrity Every time introduced would act first time we’d met issued barb weren’t thinlyveiled Still I’d smile show Still I’d ride subway home tear mask I’d wearing became shape face 2009 torched joint Closed shop revered literary magazine resigned lucrative book publishing job burned bridge devoured ash took senior role digital medium agency threw work internet exploded Everyone Facebook YouTube Twitter Instagram would launch following year physical space morphed digital one hid behind shiny device Everyone preened screen became architect personal brand new world order dictated nobody weren’t online like job interview cocktail stone sober knew never getting job Still stomach small talk signed account didn’t know social medium realized digital mask replaced physical one Nobody wanted presenttense sadness preferred kick fuss make scene pain awkward uncomfortable one wanted bear weight expected deliver life sepia everything beautiful nothing hurt Otherwise motley lot unfollowed muted blocked got good bleaching feed clean hurt thing make u human build brand living struggled mine personality polarizing opinion they’re almost always unpopular write short clipped email Unsolicited advice frustrates — must’ve blacked asked opinion Yet people still give hard time hiding irritation I’m guarded I’ve hurt used laughed I’ve much protect People rarely pure — they’ll shake give They’ll dig contact list virtual pocket They’ll pick clean know it’s happened time time time sell congeniality I’m walking wound thought endured spent enough time online would find agent replace one I’d resigned get third book ground I’d blow consulting business I’d milliondollar course launch spend twilight year wearing muumuu saving puppy Instead Ativan Sancerrefueled public online breakdown “friends” issued fatwa virtual restraining order reported tweet blocked unfollowed dodged call They’re mental health advocate depression fill room house Instead pulling 2009 falling frame set boundary made plan read every book returning life I’d grown living — irony doesn’t escape devoured Cal Newport’s Digital Minimalism saw plain simple truth Here’s Evaluate purpose every digital platform use People apoplectic learn I’m every social network tell I’ll never get business lead remind social medium never delivered client — writing Don’t listen fearmongering herd mentality people follow herd without questioning they’re moving first place direction stepped back defined want spend day evaluated platform defined value use purpose need get surgical honest perform old costbenefit analysis Facebook Twitter made angry curate news source instead reacting moment Instagram made feel like loser Pinterest never understood YouTube frightens know I’d snort TikTok like coke Rodman’s house twentysix settled LinkedIn Medium tiny private Instagram account mostly look animal rescue video I’m feeling blue There’s veneer wall LinkedIn like keep everything professional even keel I’ve Medium since 2013 see place showcase smart experiment fiction one read don’t get caught drama politicking enough book publishing it’s tired show different channel don’t care clap income popularity I’m allergic people I’d rather assemble tiny mighty community get bombarded mass I’ve writing dark literary fiction decade — I’m used unpopular widely read Medium place understand geography social order know fit spend time writing sharing work mean something move educates people oscillate two storytelling world — story writing marketing — love feeling switching one side brain firing another buy cookbook feel weight paper binding hand make food without posting picture Remember room clouded flour Return life knew kid took thing held hand granted spend much day tethered laptop phone offhours strive balance hobby tactile multisensory walk hike run write longhand write letter read book participate book exchange play clarinet cook rarely see people opportunity hold close touching smelling around real even important I’m better artist who’s constantly curious reach world beyond digital border we’ve created don’t read howto book writing diagram novel mark page vivisect author challenge I’m artist professional who’s evolving growing exercise sens beyond keyboard screen Every morning walk five mile listening podcasts Sometimes walk quiet absorbing street smell sound see human instead name screen set monthly Zoom date old peer friend There’s agenda talk human Sometimes don’t even use video we’re fatigued interest putting pant it’s good hear voice people love respect miss miss cue decoding body language miss vibing someone’s physical energy breath talk career life child cat’s kid there’s thatTags Social Media Writing Mental Health Relationships Creativity |
728 | Hyperkops: Scaleable Hyperparameter Tuning of Machine Learning Models Using Bayesian optimisation in Kubernetes | TLDR:
Hyperkops is a technology which enables the use of Hyperopt (a Bayesian optimisation library in Python) in Kubernetes. This provides a scaleable platform in which Data Scientists and Engineers can share infrastructure to estimate the optimal set of hyperparameters for their machine learning models.
Introduction
We introduce technology that enables Bayesian optimisation of hyperparameters of machine learning models using the Python based Hyperopt Bayesian optimisation library.
Bayesian optimisation is a sequential learning approach where iterative fitting of models is combined with Bayesian inference to arrive at an estimate of best possible hyperparameters. Hyperkops uses the Python library Hyperopt to execute this type of optimisation, and scale the parallel execution of these calculations by deploying them on Kubernetes. It exploits Hyperopt’s ability to run calculations asynchronously by storing state in MongoDB alongside Kubernetes’ ability to automatically scale our hardware provisions to match our current requirements. Thus, by running these calculations within Kubernetes, we can create a system which autoscales to our needs and can be used by multiple tenants at the same time.
The Kubernetes environment also allows us to connect to this architecture from within the Kubernetes cluster as well as from our local development machines. This means this architecture can provide mass parallelism for our scheduled fitting of production models at the same time as providing a tool to our data scientists for exploration from their local machines.
Motivation
At hipages we find ourselves creating lots of models, and lots of different versions of those models with slightly different applications. We’re also finding that choosing the correct hyperparameters for these models can have a significant effect on performance.
Performance-wise we tried a few different ways of searching our hyperparameter space:
Grid-search — Iterating through a fixed range of hyperparameters. This is very computationally expensive and time consuming, with significant time wasted on fitting models in part of the hyperparameter space that is likely to yield underperforming candidates
— Iterating through a fixed range of hyperparameters. This is very computationally expensive and time consuming, with significant time wasted on fitting models in part of the hyperparameter space that is likely to yield underperforming candidates Random-search — Randomly sampling hyperparameters from a bounded space. We found this to be effective, but because of the stochastic nature of the hyperparameter selection in each fit it’s likely that the hyperparameter selection is sub-optimal, and similar to Grid-Search there is lots of computational time and effort spent on low performant models.
— Randomly sampling hyperparameters from a bounded space. We found this to be effective, but because of the stochastic nature of the hyperparameter selection in each fit it’s likely that the hyperparameter selection is sub-optimal, and similar to Grid-Search there is lots of computational time and effort spent on low performant models. Bayesian Search: An iterative method where each fit is targeted at improving the model’s performance, with the selection of the updated hyperparameters based on information from previous fits. We found this to improve performance significantly but it also takes longer due to the sequential nature of this approach.
In our case model performance was key so we decided to forge ahead using Bayesian search methods. For this task we selected the Python library Hyperopt as our Bayesian optimisation tool.
To speed up our fits we needed to parallelise the work-flow, and have multiple trials fitting at once. Hyperopt parallelises its workload by storing state within a MongoDB instance, and having the actual fits done by workers. The workers poll the MongoDB instance and, when a trial (a trial here would be fitting a model with the provided hyperparameters) is available, execute the workload and return the scoring metric for that set of hyperparameters. This separation of workloads lends itself well to deployment on Kubernetes.
Deployment on Kubernetes also comes with lots of other benefits:
Autoscaling : We can automatically scale up and down the number of workers we have available to us based on specific metrics (e.g. CPU usage). So when we have multiple hyperparameter optimisations running in parallel the cluster will automatically scale up to meet our needs, and then scale back when we’re done.
: We can automatically scale up and down the number of workers we have available to us based on specific metrics (e.g. CPU usage). So when we have multiple hyperparameter optimisations running in parallel the cluster will automatically scale up to meet our needs, and then scale back when we’re done. Production Fitting: Once we’ve deployed our fitting infrastructure we can connect from both inside and outside the Kubernetes cluster. In our case this means we can automate and schedule the fitting and export of our final model from within Kubernetes alongside our usual ETL workflows.
Once we’ve deployed our fitting infrastructure we can connect from both inside and outside the Kubernetes cluster. In our case this means we can automate and schedule the fitting and export of our final model from within Kubernetes alongside our usual ETL workflows. Use in Discovery Phases: During discovery phases a data scientist might want to fit multiple models to get a feel for the model’s performance. By deploying our system on Kubernetes they can dial up mass parallelisation of the models by connecting to the cluster and submitting jobs from their local machines. This means there is no need to closely manage expensive cloud-compute instances. The reduced costs keep Dev-ops happy, and can significantly reduce our time-to-insight during discovery.
Hyperkops
With these benefits in mind we set out to deploy Hyperopt within our own Kubernetes, but found that we need to add a few extra components to get our hyperparameter searches to finish. The result of this work was our Hyperkops architecture, as depicted below:
Hyperkops Architecture
The Hyperkops architecture is comprised of three main components:
Hyperkops Worker: Each of these Pods run a single hyperopt-worker. These Pods therefore poll the MongoDB for trials to execute, and return the score they generate with that trials’ hyperparameters. In our context a trial would be fitting a single machine learning model with the hyperparameters provided, and returning the performance metric.
Hyperkops Monitor: Identifies and updates Hyperopt trials as failed due to Pod failure or rotation
MongoDB: MongoDB Instance that holds the state of all the trials
Hyperopt Fitting Master: These are the processes which launch the Hyperopt optimisations.
Why is the Hyperkops Monitor required?
In Kubernetes the Pods which execute the hyperopt-workers can be significantly shorter lived than some optimisation jobs and are expected to get rotated on a regular basis. If a Pod is deleted whilst executing an experiment the hyperopt-worker will be killed before it can emit an error signal and jobs remain in MongoDB indefinitely in a JOB_RUNNING_STATE. Hyperopt waits until all of its running jobs have finished before selecting the best possible set of hyperparameters, and therefore jobs which are trapped in a JOB_RUNNING_STATE, with no way of concluding, leave our optimisation jobs running in perpetuity.
We therefore introduced an extra component (the Hyperkops Monitor) to monitor our deployment, and update relevant MongoDB entries for trials we know to have been running on failed or deleted Pods. This component thus allows our hyperparameter optimisation to conclude and return the optimal set of hyperparameters it discovered.
Future Work
We’re just at the start of this project, but here’s a flavour of what we have in mind for the future:
Automate the installation of required libraries : When a worker Pod receives a method to execute it requires the Pod to either already have the required libraries installed, or for the method itself to to ensure it installs the libraries on the fly. A next step will be to produce a class that allows us to automate these installation processes.
: When a worker Pod receives a method to execute it requires the Pod to either already have the required libraries installed, or for the method itself to to ensure it installs the libraries on the fly. A next step will be to produce a class that allows us to automate these installation processes. Add Monitoring UI: MongoDB and the Kubernetes API provide us with sufficient information to monitor the system from a hardware level through to metrics around the current optimisation processes. One advancement of this system would be to provide a single UI which allows to monitor all of these metrics in real-time.
MongoDB and the Kubernetes API provide us with sufficient information to monitor the system from a hardware level through to metrics around the current optimisation processes. One advancement of this system would be to provide a single UI which allows to monitor all of these metrics in real-time. Add high-availability to MongoDB: In the example workload we’ve provided, MongoDB is deployed as a single Pod, with no persistence of the data or resilience to failure. This means that if the Pod is rotated, or fails, then all work done by the cluster will be lost and restarted from scratch. In order to guard against this failure mode we need to concentrate on bringing up MongoDB in high-availability mode, which will help protect the system from these forms of failure.
Conclusion
We’re really excited to open-source this bit of software, and hope people find it as useful as we do. We’d really like to thank the developers of the Hyperopt library for creating something great! Don’t hesitate to nip over the Github Repo if you have any feedback or want to contribute to the library. | https://medium.com/hipages-engineering/hyperkops-scaleable-tuning-of-machine-learning-models-using-bayesian-optimisation-in-kubernetes-bd0ce480b02b | ['Chris Coulson'] | 2019-06-25 05:34:40.349000+00:00 | ['AI', 'Kubernetes', 'Python', 'Data Science', 'Machine Learning'] | Title Hyperkops Scaleable Hyperparameter Tuning Machine Learning Models Using Bayesian optimisation KubernetesContent TLDR Hyperkops technology enables use Hyperopt Bayesian optimisation library Python Kubernetes provides scaleable platform Data Scientists Engineers share infrastructure estimate optimal set hyperparameters machine learning model Introduction introduce technology enables Bayesian optimisation hyperparameters machine learning model using Python based Hyperopt Bayesian optimisation library Bayesian optimisation sequential learning approach iterative fitting model combined Bayesian inference arrive estimate best possible hyperparameters Hyperkops us Python library Hyperopt execute type optimisation scale parallel execution calculation deploying Kubernetes exploit Hyperopt’s ability run calculation asynchronously storing state MongoDB alongside Kubernetes’ ability automatically scale hardware provision match current requirement Thus running calculation within Kubernetes create system autoscales need used multiple tenant time Kubernetes environment also allows u connect architecture within Kubernetes cluster well local development machine mean architecture provide mass parallelism scheduled fitting production model time providing tool data scientist exploration local machine Motivation hipages find creating lot model lot different version model slightly different application We’re also finding choosing correct hyperparameters model significant effect performance Performancewise tried different way searching hyperparameter space Gridsearch — Iterating fixed range hyperparameters computationally expensive time consuming significant time wasted fitting model part hyperparameter space likely yield underperforming candidate — Iterating fixed range hyperparameters computationally expensive time consuming significant time wasted fitting model part hyperparameter space likely yield underperforming candidate Randomsearch — Randomly sampling hyperparameters bounded space found effective stochastic nature hyperparameter selection fit it’s likely hyperparameter selection suboptimal similar GridSearch lot computational time effort spent low performant model — Randomly sampling hyperparameters bounded space found effective stochastic nature hyperparameter selection fit it’s likely hyperparameter selection suboptimal similar GridSearch lot computational time effort spent low performant model Bayesian Search iterative method fit targeted improving model’s performance selection updated hyperparameters based information previous fit found improve performance significantly also take longer due sequential nature approach case model performance key decided forge ahead using Bayesian search method task selected Python library Hyperopt Bayesian optimisation tool speed fit needed parallelise workflow multiple trial fitting Hyperopt parallelises workload storing state within MongoDB instance actual fit done worker worker poll MongoDB instance trial trial would fitting model provided hyperparameters available execute workload return scoring metric set hyperparameters separation workload lends well deployment Kubernetes Deployment Kubernetes also come lot benefit Autoscaling automatically scale number worker available u based specific metric eg CPU usage multiple hyperparameter optimisation running parallel cluster automatically scale meet need scale back we’re done automatically scale number worker available u based specific metric eg CPU usage multiple hyperparameter optimisation running parallel cluster automatically scale meet need scale back we’re done Production Fitting we’ve deployed fitting infrastructure connect inside outside Kubernetes cluster case mean automate schedule fitting export final model within Kubernetes alongside usual ETL workflow we’ve deployed fitting infrastructure connect inside outside Kubernetes cluster case mean automate schedule fitting export final model within Kubernetes alongside usual ETL workflow Use Discovery Phases discovery phase data scientist might want fit multiple model get feel model’s performance deploying system Kubernetes dial mass parallelisation model connecting cluster submitting job local machine mean need closely manage expensive cloudcompute instance reduced cost keep Devops happy significantly reduce timetoinsight discovery Hyperkops benefit mind set deploy Hyperopt within Kubernetes found need add extra component get hyperparameter search finish result work Hyperkops architecture depicted Hyperkops Architecture Hyperkops architecture comprised three main component Hyperkops Worker Pods run single hyperoptworker Pods therefore poll MongoDB trial execute return score generate trials’ hyperparameters context trial would fitting single machine learning model hyperparameters provided returning performance metric Hyperkops Monitor Identifies update Hyperopt trial failed due Pod failure rotation MongoDB MongoDB Instance hold state trial Hyperopt Fitting Master process launch Hyperopt optimisation Hyperkops Monitor required Kubernetes Pods execute hyperoptworkers significantly shorter lived optimisation job expected get rotated regular basis Pod deleted whilst executing experiment hyperoptworker killed emit error signal job remain MongoDB indefinitely JOBRUNNINGSTATE Hyperopt wait running job finished selecting best possible set hyperparameters therefore job trapped JOBRUNNINGSTATE way concluding leave optimisation job running perpetuity therefore introduced extra component Hyperkops Monitor monitor deployment update relevant MongoDB entry trial know running failed deleted Pods component thus allows hyperparameter optimisation conclude return optimal set hyperparameters discovered Future Work We’re start project here’s flavour mind future Automate installation required library worker Pod receives method execute requires Pod either already required library installed method ensure installs library fly next step produce class allows u automate installation process worker Pod receives method execute requires Pod either already required library installed method ensure installs library fly next step produce class allows u automate installation process Add Monitoring UI MongoDB Kubernetes API provide u sufficient information monitor system hardware level metric around current optimisation process One advancement system would provide single UI allows monitor metric realtime MongoDB Kubernetes API provide u sufficient information monitor system hardware level metric around current optimisation process One advancement system would provide single UI allows monitor metric realtime Add highavailability MongoDB example workload we’ve provided MongoDB deployed single Pod persistence data resilience failure mean Pod rotated fails work done cluster lost restarted scratch order guard failure mode need concentrate bringing MongoDB highavailability mode help protect system form failure Conclusion We’re really excited opensource bit software hope people find useful We’d really like thank developer Hyperopt library creating something great Don’t hesitate nip Github Repo feedback want contribute libraryTags AI Kubernetes Python Data Science Machine Learning |
729 | How To Quit Your Job and Start a Business That Lasts | How To Quit Your Job and Start a Business That Lasts
You’ll need to avoid some common mistakes to survive
If you’re going to ditch your day job and step out on your own, you’re going to want you to make sure that what you’re stepping into evolves into a sustainable, profitable, and growing company.
Look, I get it. It’s every employee’s dream — You can do what you do better on your own, with higher quality, lower overhead, and you can do it for more people and for people who need it more. Plus you’ll enjoy a greater return for your efforts.
That may be true — for a while. But if you want to build a lasting business, not just starting a side gig, not just transitioning from comfortable employee to scratch-and-claw contractor, but growing a business that turns into a startup that turns into a company, you’re going to have to serve two markets, probably more.
Here’s what those markets are and how to build and grow them.
First: Take In All You Can From the Existing Market
The top mistake most solos make is leaving their day job too soon. Going solo takes months, even years of preparation. I’m not just talking about the experience you’ll need to deliver an expert product or service in your field, but the situational knowledge required to build a sustainable company around delivering that product or service.
There are reasons why your former company is leaking efficiency, quality, affordability, customer satisfaction — all those reasons why you left. Yeah, your former employer might have gotten lazy, but it’s more likely they got bloated when they built a bunch of infrastructure to allow them to grow large enough to hire smart, talented people like you. You don’t want to make the same mistakes.
So while you’re still at your day job, take in everything you can before you leave. I’m not talking about contacts or customers. I’m talking about that situational knowledge and the infrastructure that developed out of it.
Look around and understand what your employer is doing around every aspect of your job that doesn’t directly involve you delivering the product or service itself. What makes your job easier? What will you miss the most? What does your employer have that you won’t? Do you need it? How can you recreate it? How can you make it more efficient?
Your First Second Job: Develop a New Market
When you go solo, your first job is domain expert: You are the lead on developing and delivering the product or service you’re now offering. The one additional job I can guarantee you’ll need to take on is the CEO role of growing your business.
But the mistake most solos make when they start their own business is splitting their duties between domain expert and sales generator for their existing market.
Don’t get me wrong, you’ll constantly need to be spending your time developing a sales pipeline and closing new deals. But if you’re doing that the way it’s always been done, maybe even the way your former employer did it, you’ll just wind up being weak competition in your existing market.
If you want your business to last, you’ll have to revolutionize not only how the product or service gets delivered, but also how it’s sold and who it’s sold to. You’ll need to carve out new markets, which will allow you to generate business both immediately and into the future.
So how do you develop a new market?
Serve Old and New Markets Alike
What you need to do first is break down the components of your product or service and figure out which components can serve your existing market and new markets at the same time.
I did this myself when I went solo after a run of successful (and a couple of unsuccessful) VC-backed and self-funded startups. I founded a consulting firm that worked mainly with startups, and I did the things for them that had made my own startups successful. But I also carved out a new niche for larger, established companies, who would be my new market.
I know. This is kind of backwards. Most solos fund themselves with larger, established companies and then try to establish new markets with smaller companies or startups. You can do that too. The idea is the same. Serve the people who are familiar with your product or service, and then carve out parts and components of that product or service to bring to new markets.
In my case, I found a number of large, established companies that wanted to operate with the speed and efficiency of a startup. So I became a kind of special ops within this new market and developed a new kind of consulting business to serve both the old and new markets.
Now you’ve got some choices to make.
Select the Best Targets in Both the Old and New Markets
You’ve got a market on either end of your business. On one end are the established players who look and act a lot like your former employer’s existing market. On the other end you’ve got a whole new group of players who may or may not be able to understand, find value in, or afford your product or service.
You need to pick the best prospects from both markets and focus on those prospects.
From the existing market, you target the most forward-thinking customers or those who are most closely aligned with your version of the product or service. With these customers, you act in your primary role, domain expert, and you do your thing.
From the new market, you target the ones who best understand your product or service, or who are most likely to find value in it, or can most likely afford it. With them, you’re in the CEO role, figuring out how you need to transform your product or service and your sales strategy to conform to the needs of this new market.
There are competitive benefits to this strategy. You’re not going after every customer in the existing market — leave the shitty ones for your former employer. You’re also generating new business from a market your former employer (and others like them) aren’t interested in or maybe even aware of. If you can turn those prospects into customers, you’ll gain a huge competitive edge.
Now you can think about scaling.
Find the Commonalities and Build on Them
Another mistake solos make is they tend to try to scale by bringing in the same business they used to do when they worked for someone else. The solos that do survive using this strategy end up becoming exactly what they left behind in the first place.
The main reason you want to work with two (or more) markets is to figure out which components of your product or service are common across those markets. You’ll use this knowledge to scale the business.
If you’re a service company, your inner CEO starts automating these common components so they can be done more quickly and by other people, especially by people who aren’t domain experts.
If you’re a product company, your inner CEO emphasizes and upgrades these components so you can intelligently broaden your audience and go get more customers at better margins.
Now you have to handle that growth.
Hire Equally to Handle Both Markets
If you want to split markets correctly, it will take a ton of time. That time gets taken away from you as a domain expert and also from you as the CEO. So you’ll need to backfill your time equally.
Hires on the domain expert side should provide short term help to your company. If you’re a service company, more domain experts means more revenue. If you’re a product company, more domain experts means immediate increases in speed to market, quality of product, and effectiveness of sales strategy.
One mistake solos make is hiring only domain experts to build the business person by person, which is something service solos do all the time. The result is the same bloat that made your former employer unattractive.
Hires on the new market side should help your company realize and execute your vision of bringing an evolved product or service to a new market. In both service and product companies, you’re bringing in technologists and product engineers and market strategists to help you find and cover new ground.
Another mistake solos make is hiring too much new market help too quickly. They wind up building amazing tech and marketing materials for concepts that they haven’t even proved viable yet.
The main thing to remember here is don’t hire more of you. Hire for gaps, and when you do, hire fractionally at first if you can. Because when you’re a new unique company serving a new unique market, all you need to do is get over that first survival hurdle. It’s not easy, but if you do it, you’ve got the best chance at building a business that will last a long, long time.
Hey! If you found this post actionable or insightful, please consider signing up for my weekly newsletter at joeprocopio.com so you don’t miss any new posts. It’s short and to the point. | https://jproco.medium.com/how-to-quit-your-job-and-start-a-business-that-lasts-63d7645ac516 | ['Joe Procopio'] | 2019-12-15 15:09:36.073000+00:00 | ['Leadership', 'Entrepreneurship', 'Business', 'Startup', 'Productivity'] | Title Quit Job Start Business LastsContent Quit Job Start Business Lasts You’ll need avoid common mistake survive you’re going ditch day job step you’re going want make sure you’re stepping evolves sustainable profitable growing company Look get It’s every employee’s dream — better higher quality lower overhead people people need Plus you’ll enjoy greater return effort may true — want build lasting business starting side gig transitioning comfortable employee scratchandclaw contractor growing business turn startup turn company you’re going serve two market probably Here’s market build grow First Take Existing Market top mistake solo make leaving day job soon Going solo take month even year preparation I’m talking experience you’ll need deliver expert product service field situational knowledge required build sustainable company around delivering product service reason former company leaking efficiency quality affordability customer satisfaction — reason left Yeah former employer might gotten lazy it’s likely got bloated built bunch infrastructure allow grow large enough hire smart talented people like don’t want make mistake you’re still day job take everything leave I’m talking contact customer I’m talking situational knowledge infrastructure developed Look around understand employer around every aspect job doesn’t directly involve delivering product service make job easier miss employer won’t need recreate make efficient First Second Job Develop New Market go solo first job domain expert lead developing delivering product service you’re offering one additional job guarantee you’ll need take CEO role growing business mistake solo make start business splitting duty domain expert sale generator existing market Don’t get wrong you’ll constantly need spending time developing sale pipeline closing new deal you’re way it’s always done maybe even way former employer you’ll wind weak competition existing market want business last you’ll revolutionize product service get delivered also it’s sold it’s sold You’ll need carve new market allow generate business immediately future develop new market Serve Old New Markets Alike need first break component product service figure component serve existing market new market time went solo run successful couple unsuccessful VCbacked selffunded startup founded consulting firm worked mainly startup thing made startup successful also carved new niche larger established company would new market know kind backwards solo fund larger established company try establish new market smaller company startup idea Serve people familiar product service carve part component product service bring new market case found number large established company wanted operate speed efficiency startup became kind special ops within new market developed new kind consulting business serve old new market you’ve got choice make Select Best Targets Old New Markets You’ve got market either end business one end established player look act lot like former employer’s existing market end you’ve got whole new group player may may able understand find value afford product service need pick best prospect market focus prospect existing market target forwardthinking customer closely aligned version product service customer act primary role domain expert thing new market target one best understand product service likely find value likely afford you’re CEO role figuring need transform product service sale strategy conform need new market competitive benefit strategy You’re going every customer existing market — leave shitty one former employer You’re also generating new business market former employer others like aren’t interested maybe even aware turn prospect customer you’ll gain huge competitive edge think scaling Find Commonalities Build Another mistake solo make tend try scale bringing business used worked someone else solo survive using strategy end becoming exactly left behind first place main reason want work two market figure component product service common across market You’ll use knowledge scale business you’re service company inner CEO start automating common component done quickly people especially people aren’t domain expert you’re product company inner CEO emphasizes upgrade component intelligently broaden audience go get customer better margin handle growth Hire Equally Handle Markets want split market correctly take ton time time get taken away domain expert also CEO you’ll need backfill time equally Hires domain expert side provide short term help company you’re service company domain expert mean revenue you’re product company domain expert mean immediate increase speed market quality product effectiveness sale strategy One mistake solo make hiring domain expert build business person person something service solo time result bloat made former employer unattractive Hires new market side help company realize execute vision bringing evolved product service new market service product company you’re bringing technologist product engineer market strategist help find cover new ground Another mistake solo make hiring much new market help quickly wind building amazing tech marketing material concept haven’t even proved viable yet main thing remember don’t hire Hire gap hire fractionally first you’re new unique company serving new unique market need get first survival hurdle It’s easy you’ve got best chance building business last long long time Hey found post actionable insightful please consider signing weekly newsletter joeprocopiocom don’t miss new post It’s short pointTags Leadership Entrepreneurship Business Startup Productivity |
730 | Are You A Night Owl? We Are Closer to Knowing Why | Health
Are You A Night Owl? We Are Closer to Knowing Why
A surprisingly common mutation causes this sleep disorder by changing a vital component of the biological clock that maintains the body’s daily rhythms.
Photo by Mpho Mojapelo on Unsplash
Sleep shows up in several well-known films. In the 2010 Christopher Nolan heist film Inception, Leonardo DiCaprio stars as a thief who steals information by infiltrating individuals’ dreams and subconscious minds. Nolan based the movie on lucid dreaming, a state in which you are aware of your dreams and can exercise a modicum of control over them.
Have you seen the sci-fi film Eternal Sunshine of the Spotless Mind? The 2004 movie features Jim Carrey and Kate Winslet as star-crossed lovers. The two erase each other from their memory using the services of a fictional company. The procedure happens during sleep; Joel Barish, the Carrey character, spends much of the movie in a dream state as he has his memories erased.
Our inner clock: A brief history
During the 1700s, the astronomer Jean Jacques d’Ortous de Mairan studied mimosa plants. He found that the plant’s leaves would open towards the sun during the daytime but closed at dusk. But what if the plant lived only in darkness? The leaves continued to follow their regular daily oscillation, seeming to possess an internal biological clock.
Similar to the plants with which we live, we humans have a biological clock that adjusts our physiology throughout the day. You probably know this adaptation as the circadian rhythm, a descriptor derived from the Latin circa (“around”) and dies (“day”).
How does our internal circadian biological clock function? We have recently made additional strides in answering this question.
Our body has clock proteins
Earth’s life has adapted to the planet’s rotation. As is the case with other organisms, humans have an internal biological clock. This time regulator helps us adjust to the daily rhythms of living. But how does such a clock function?
Today, we turn to our body’s clock proteins involved in regulating our internal clock. First, a bit of history. In the 1970s, Seymour Benzer and his protege Ronald Konopka wondered if they could find the genes that regulate the circadian rhythms of fruit flies. They found success, demonstrating that mutations in a previously unknown gene messed up the flies’ circadian rhythm. They designated this gene period. | https://drmichaelhunter.medium.com/are-you-a-night-owl-we-are-closer-to-knowing-why-90d0a7a2ece1 | ['Michael Hunter Md'] | 2020-11-05 17:20:41.116000+00:00 | ['Sleep', 'Wellness', 'Health', 'Lifestyle', 'Science'] | Title Night Owl Closer Knowing WhyContent Health Night Owl Closer Knowing surprisingly common mutation cause sleep disorder changing vital component biological clock maintains body’s daily rhythm Photo Mpho Mojapelo Unsplash Sleep show several wellknown film 2010 Christopher Nolan heist film Inception Leonardo DiCaprio star thief steal information infiltrating individuals’ dream subconscious mind Nolan based movie lucid dreaming state aware dream exercise modicum control seen scifi film Eternal Sunshine Spotless Mind 2004 movie feature Jim Carrey Kate Winslet starcrossed lover two erase memory using service fictional company procedure happens sleep Joel Barish Carrey character spends much movie dream state memory erased inner clock brief history 1700s astronomer Jean Jacques d’Ortous de Mairan studied mimosa plant found plant’s leaf would open towards sun daytime closed dusk plant lived darkness leaf continued follow regular daily oscillation seeming posse internal biological clock Similar plant live human biological clock adjusts physiology throughout day probably know adaptation circadian rhythm descriptor derived Latin circa “around” dy “day” internal circadian biological clock function recently made additional stride answering question body clock protein Earth’s life adapted planet’s rotation case organism human internal biological clock time regulator help u adjust daily rhythm living clock function Today turn body’s clock protein involved regulating internal clock First bit history 1970s Seymour Benzer protege Ronald Konopka wondered could find gene regulate circadian rhythm fruit fly found success demonstrating mutation previously unknown gene messed flies’ circadian rhythm designated gene periodTags Sleep Wellness Health Lifestyle Science |
731 | Taste 2.0 is Here | Taste 2.0 is Here
The discovery of a new taste cell deepens our understanding of this complex and adaptive sensory system.
By Jason Socrates Bardi
Years ago, my father took a huge slurping bite of what he thought was a fat navel orange and immediately gagged. Spitting it out, he later recalled, it was the most poisonous piece of fruit he’d ever tasted — or so he thought. What’s wrong with this orange? he asked his companion. It’s actually a grapefruit, she said. And with that, the horrible taste of the rotten fruit turned delicious in his mouth, almost like the flick of a switch.
For some time after I heard this story, I wondered whether it would it be possible to do the same thing deliberately. If the brain can be fooled into shaping our experience of taste by accident, could you also trick it on purpose-by taking a pill or painting a few drops of solution on your tongue before a meal? Can science turn the experience of eating the worst Soylent Green into one of the best crème brûlées ever?
This anecdote is germane to a group of researchers at the University of Buffalo in upstate New York. They recently discovered a new type of taste cell -a discovery that has somewhat upended what we thought we knew about the human taste system.
A taste of evolution
The taste system helps accomplish one of life’s oldest needs: analyzing chemicals on the tongue and detecting flavors that indicate whether something is food or foul. Even the most ancient prokaryotic organisms to emerge from the primordial ooze on the tangled banks of wherever needed to find food, and the prevailing idea in biology today is that the modern mammalian taste system evolved to serve that most fundamental of needs.
We are adept at detecting certain flavors of things in our environment that contain the nutrients we need and at detecting more abhorrent flavors of things that would make us sick. We detect sweet things quite well because sugars are a key chemical by-product of carbohydrates, which we typically consume as a basic food in our diet. We need sodium and chloride to maintain cardiovascular homeostasis, so we detect salt quite well.
And in genomic tribute to our forgotten, foraging past, humans are able to taste bitter flavors with finely honed accuracy. The human genome has no fewer than 25 different subtypes of Tas2r genes to detect bitter tastes-less than frogs, which have 50 such genes, or mice, which have 35, but far more than chickens, which have only three bitter taste receptors.
“The taste system is just absolutely required for all organisms to be able to consume the nutrients that they need and avoid the things that would make them sick,” says Kathryn Medler, a biologist at the University of Buffalo in New York who led the team that discovered the new taste cell.
As a fundamental tool for survival today, taste seems almost unnecessary for the modern human — at least anywhere one can drive past a thousand restaurants to a 24-hour grocery store bursting with food choices. We did not evolve with such abundance — but rather in a wild, dangerously fluctuating world full of frequent famine and strange, bitter things that may be toxic.
As a result, says Paul Breslin, a nutritional scientist at Rutgers University and researcher at the Monell Chemical Senses Center in Philadelphia, “we’re a food chemistry analytical machine.”
Cats, cows, koalas, and dolphins
Consider the cat. Cats are carnivores, and they have no need in their diet for carbohydrates or sweet things, except perhaps the occasional non-critter bon-bon. This isn’t merely a question of preference — it’s evolution. Cats actually don’t even have sugar receptor genes.
“They don’t even taste it — they’re taste-blind to sweet,” says University of Miami biologist Steve Roper, another expert in the field who published a review on the current state of knowledge on taste buds a few years ago. If a cat licks something sweet, Roper added, it’s because they detect the fat in that food and not the sugar.
What would it be like to eat meat with a cat’s tongue?
Cats do have a highly developed sense of savory taste — dubbed umami in Japanese (delicious savory taste) — and this makes sense because they are consuming protein all the time. What would it be like to eat meat with a cat’s tongue? It’s impossible for science to say for certain, but one can imagine for a sensory system specifically tuned to taste savory flesh, it must be delicious.
On the other hand, you would never want to eat a salad with a cow’s tongue because cows eat grass all day — something they would never be able to do if they were actually able to taste it the way we do. Unlike humans, with our dozens of bitter receptors, cows are highly insensitive to bitter foods. They still detect it, but apparently not so well.
“In some respects, the taste system is like a mirror of [our] environment,” says Maik Behrens, a biologist at the Leibniz Institute for Systems Biology in Freising, Germany. He has studied the taste systems of multiple animals, publishing a paper a few years ago that explored the ability of chickens and frogs to taste bitter flavors.
Lots of creatures in the animal kingdom have reached the point over time where they have more or less one thing in their diet. Koalas are known to mostly eat eucalyptus, pandas are primarily bamboo eaters, and sea lions are bona fide pescatarians. According to experts, you see the same trend in lots of animals like these: a refinement of their taste systems to suit their niche.
“That’s why we taste bitter better than bottlenose dolphins,” which actually have no taste receptors at all, Behrens says. “Apparently they simply swallow their fish without tasting a lot.”
If you want to enjoy some sushi, don’t try it with a dolphin’s tongue
That ape is alive and well within us
If an animal’s sense of taste is informed by what’s to eat in the environment in which they live, how they find that food in the first place matters. The more an animal forages, the more receptors it will need to detect potentially dangerous bitters.
“If you don’t know what you’re going to eat,” says Breslin, “you have to engage in this process of evaluation.”
Foodies often tout the learned aspect of taste, a sense of refinement-through-experience. But fundamentally, our taste preferences are hard coded from the beginning. There is evidence that fetuses taste foods in utero, and newborn infants will coo at sweet and pucker at sour flavors placed in their mouths.
“We love sugar for the same reason the chimpanzees do — that other apes do,” Breslin says. “Just like the gorillas, chimpanzees, bonobos, and orangutans, which are all largely fruit eaters, we were meant to be sugar eaters.”
“That ape is alive and well within us,” he added.
Reached for comment on the discovery of the new type of taste cell, Medler explained how it did not behave the way we expect taste cells to behave.
Decades ago, there was thought to be a single type of taste cell that responded to all tastes. But over the years, that concept was slowly refined as we discovered different taste cells. Scientists then classified taste cells into different types, but the sense was still that a single cell would detect a single taste.
“That’s how it has stood for many years,” says Roper, the biologist from the University of Miami.
But what Medler and her colleagues recently discovered was a subset of these cells that were actually broadly responsive to four of the five basic taste qualities and seem to be just as important for the perception of those tastes.
“It’s a very elegant study,” says Julie Mennella, a biologist at Monell Chemical Senses Center who studies taste also but was not involved in Medler’s research. The study involved characterizing isolated batches of broadly receptive taste cells for their electrical signaling as well as extensive behavioral experiments in which mice placed in special cages with different bottles of solution were trained to seek out certain bottles that offered tastes they like. Medler’s team showed that when mice lose their broadly responsive taste cells, the animals basically behave like they can’t taste bitter, sweet, and umami.
The quest for new tastes
Not everyone agrees with the current breakdown of taste into five key flavors, and some scientists have proposed that additional tastes could exist, and some of these potential new tastes have been identified.
“There are quite a number of additional candidates out there,” says Purdue University nutritional scientist Richard Mattes, who coined the word oleogustus (fatty taste) to describe one of these proposed new sensory flavors. Calcium, carbon dioxide, starch, and short-chain carbohydrates also have all been variously proposed as possible new tastes. “But they are by no means accepted broadly in the scientific community,” Mattes says.
The evidence for fat as a taste is overwhelming, Mattes says. There is at least one receptor found on human taste cells, called CD36, that can detect fat. There is also a mechanism for it to transduce signals to the brain, and these signals seem to be tuned specifically for fat. There are also sensory experiments where people can sensitively detect fat on their tongues.
“It’s not just sour, salty, bitter, or umami — people can, under the right testing conditions, say, ‘Oh — no! This is not like any of those others,” Mattes says. “In my mind, it’s a pretty strong case.”
The evolutionary argument is that humans have the ability to detect a specific type of fat known as free fatty acids, which are present in rancid food. The food industry has been aware of this for decades and goes to great lengths to eliminate free fatty acids from food. The reason why olive oil is cold pressed, for instance, is to reduce the temperature-dependent oxidation of the oil, which produces these free fatty acids and fouls the taste of the oil.
The last time a new flavor was added to the list was 20 years ago, when umami was added after the discovery of its receptor — but that was almost a century after Japanese chemist Kikunae Ikeda first proposed it in 1909. The discovery of the new type of taste cell by Medler and her colleagues is not quite as profound as Ikeda’s, but experts say it’s not all about new tastes. Identifying these new cells is an important step in our efforts to understand the basic features of human taste because it reveals new details, like how taste cells communicate with each other-and ultimately how our sense of taste works.
“The hope is that this study is going to help us understand better how taste signals are sent back to the brain,” Medler says.
COVID, chemo, shingles, and spit
Ultimately, in human physiology, it always seems to come back to the brain, and that’s certainly the case with taste. The mind exerts a huge influence over how we perceive foods because it combines taste cues with signals from our other senses and integrates them with memories and other things.
When people taste white wine that’s been dyed red but otherwise left unaltered, they will describe its taste using more robust adjectives as if it were a red.
For that reason lots of the tricks you can do with taste have to do more with how taste signals are processed than with flavors themselves. If you hold your nose, close your eyes, and eat a jellybean, you won’t be able to tell if it’s strawberry, lemon, or grape because you won’t have the right olfactory input. Color, which can have nothing at all to do with taste, is nevertheless a strong driver of flavor perception. When people taste white wine that’s been dyed red but otherwise left unaltered, they will describe its taste using more robust adjectives as if it were a red.
The experience my father had with the grapefruit/orange was another example of this — and a dramatic example of an abrupt change in flavor brought about by new information.
But taste is not all in your mind. It’s highly physically adaptive as well. The proteins in saliva have been seen to change according to the diet of rats, and when rats are fed a diet of bitter foods, their saliva upregulates proteins that start interacting with bitter compounds, blocking them from interacting with their bitter taste receptors. The same thing is seen in people. Giving almond milk to people will, over time, change their saliva profile, and they come to crave it more.
Other things can also physically affect taste. Shingles can impact the gustatory nerves that connect taste cells to the brain and cause people to lose their sense of taste entirely. A common side effect of COVID-19 is the loss of smell and taste, and a group at Harvard showed earlier this summer that this was probably due to coronavirus genes interfering with the human olfactory mucosa.
Aging can lead to the loss of taste because as people age, their cell turnover process may become impaired. The result can be profound. Many older people who experience taste cell degradation suffer from loss of appetite.
“It’s a big problem with old people becoming malnourished,” Medler says. “When the taste system is out of whack, it can really affect our consumption.”
Head injuries and radiation therapy can also interfere with the taste system and change the taste of foods — as can ear surgeries and tooth extractions. And the loss or distortion of taste is often cited as an unwanted side effect of chemotherapy and other medicines, which can impact both the quality of life as well as treatment outcomes.
“The number one reason for non-compliance for a medicine is because of how that medicine tastes,” Mennella says.
Hacking your tongue
So the million-dollar question in the end is: If the human sense of taste can be abruptly changed through infection or injury — or modulated over time by things like almond milk — can it also be hacked?
There are lots of potential health applications for such a hack: restoring temporary taste loss, formulating oral drugs so they don’t taste so bad, altering the chemistry of food to make it more palatable, reducing salt content without affecting taste, and sweetening foods with lower amounts of sugar and calories. Then there is the purely culinary question of changing taste for artistic or hedonic effect. Could we find ways to modulate flavors and enhance the experience of eating by hacking the tongue or the food? Could a chef tinker with all new options to delight the senses and direct a dining experience?
“Dr. Medler’s work does not directly lead to an ‘A-ha!’ moment that way,” Roper says. The work is significant, and it gives us more information, he says, but we still have a long way to go to understand the taste system. Asked whether it will be possible to deliberately, practically alter the taste of a food in this way, the experts seemed skeptical it will happen any time soon.
“Will people on the ship on its way to Mars licking a tube of some toothpaste-looking stuff and then eating a Soylent say, ‘Oooh, this tenderloin is extremely tasty?” Breslin asked. “That’s going to be a really, really hard thing to do.”
Roper echoed that sentiment in equally whimsical terms by comparing our understanding of the taste system to the ability to hear but a few notes of a musical composition — it won’t be until we hear whole chords that we’ll be able to say whether it’s Beethoven or Stravinsky, he says.
Or for that matter whether it’s an orange or grapefruit. | https://medium.com/neodotlife/taste-2-0-is-here-68ca8a3911ef | [] | 2020-11-28 20:47:54.082000+00:00 | ['Food', 'Coronavirus', 'Science', 'Health', 'Research'] | Title Taste 20 HereContent Taste 20 discovery new taste cell deepens understanding complex adaptive sensory system Jason Socrates Bardi Years ago father took huge slurping bite thought fat navel orange immediately gagged Spitting later recalled poisonous piece fruit he’d ever tasted — thought What’s wrong orange asked companion It’s actually grapefruit said horrible taste rotten fruit turned delicious mouth almost like flick switch time heard story wondered whether would possible thing deliberately brain fooled shaping experience taste accident could also trick purposeby taking pill painting drop solution tongue meal science turn experience eating worst Soylent Green one best crème brûlées ever anecdote germane group researcher University Buffalo upstate New York recently discovered new type taste cell discovery somewhat upended thought knew human taste system taste evolution taste system help accomplish one life’s oldest need analyzing chemical tongue detecting flavor indicate whether something food foul Even ancient prokaryotic organism emerge primordial ooze tangled bank wherever needed find food prevailing idea biology today modern mammalian taste system evolved serve fundamental need adept detecting certain flavor thing environment contain nutrient need detecting abhorrent flavor thing would make u sick detect sweet thing quite well sugar key chemical byproduct carbohydrate typically consume basic food diet need sodium chloride maintain cardiovascular homeostasis detect salt quite well genomic tribute forgotten foraging past human able taste bitter flavor finely honed accuracy human genome fewer 25 different subtypes Tas2r gene detect bitter tastesless frog 50 gene mouse 35 far chicken three bitter taste receptor “The taste system absolutely required organism able consume nutrient need avoid thing would make sick” say Kathryn Medler biologist University Buffalo New York led team discovered new taste cell fundamental tool survival today taste seems almost unnecessary modern human — least anywhere one drive past thousand restaurant 24hour grocery store bursting food choice evolve abundance — rather wild dangerously fluctuating world full frequent famine strange bitter thing may toxic result say Paul Breslin nutritional scientist Rutgers University researcher Monell Chemical Senses Center Philadelphia “we’re food chemistry analytical machine” Cats cow koala dolphin Consider cat Cats carnivore need diet carbohydrate sweet thing except perhaps occasional noncritter bonbon isn’t merely question preference — it’s evolution Cats actually don’t even sugar receptor gene “They don’t even taste — they’re tasteblind sweet” say University Miami biologist Steve Roper another expert field published review current state knowledge taste bud year ago cat lick something sweet Roper added it’s detect fat food sugar would like eat meat cat’s tongue Cats highly developed sense savory taste — dubbed umami Japanese delicious savory taste — make sense consuming protein time would like eat meat cat’s tongue It’s impossible science say certain one imagine sensory system specifically tuned taste savory flesh must delicious hand would never want eat salad cow’s tongue cow eat grass day — something would never able actually able taste way Unlike human dozen bitter receptor cow highly insensitive bitter food still detect apparently well “In respect taste system like mirror environment” say Maik Behrens biologist Leibniz Institute Systems Biology Freising Germany studied taste system multiple animal publishing paper year ago explored ability chicken frog taste bitter flavor Lots creature animal kingdom reached point time le one thing diet Koalas known mostly eat eucalyptus panda primarily bamboo eater sea lion bona fide pescatarians According expert see trend lot animal like refinement taste system suit niche “That’s taste bitter better bottlenose dolphins” actually taste receptor Behrens say “Apparently simply swallow fish without tasting lot” want enjoy sushi don’t try dolphin’s tongue ape alive well within u animal’s sense taste informed what’s eat environment live find food first place matter animal forage receptor need detect potentially dangerous bitter “If don’t know you’re going eat” say Breslin “you engage process evaluation” Foodies often tout learned aspect taste sense refinementthroughexperience fundamentally taste preference hard coded beginning evidence fetus taste food utero newborn infant coo sweet pucker sour flavor placed mouth “We love sugar reason chimpanzee — ape do” Breslin say “Just like gorilla chimpanzee bonobo orangutan largely fruit eater meant sugar eaters” “That ape alive well within us” added Reached comment discovery new type taste cell Medler explained behave way expect taste cell behave Decades ago thought single type taste cell responded taste year concept slowly refined discovered different taste cell Scientists classified taste cell different type sense still single cell would detect single taste “That’s stood many years” say Roper biologist University Miami Medler colleague recently discovered subset cell actually broadly responsive four five basic taste quality seem important perception taste “It’s elegant study” say Julie Mennella biologist Monell Chemical Senses Center study taste also involved Medler’s research study involved characterizing isolated batch broadly receptive taste cell electrical signaling well extensive behavioral experiment mouse placed special cage different bottle solution trained seek certain bottle offered taste like Medler’s team showed mouse lose broadly responsive taste cell animal basically behave like can’t taste bitter sweet umami quest new taste everyone agrees current breakdown taste five key flavor scientist proposed additional taste could exist potential new taste identified “There quite number additional candidate there” say Purdue University nutritional scientist Richard Mattes coined word oleogustus fatty taste describe one proposed new sensory flavor Calcium carbon dioxide starch shortchain carbohydrate also variously proposed possible new taste “But mean accepted broadly scientific community” Mattes say evidence fat taste overwhelming Mattes say least one receptor found human taste cell called CD36 detect fat also mechanism transduce signal brain signal seem tuned specifically fat also sensory experiment people sensitively detect fat tongue “It’s sour salty bitter umami — people right testing condition say ‘Oh — like others” Mattes say “In mind it’s pretty strong case” evolutionary argument human ability detect specific type fat known free fatty acid present rancid food food industry aware decade go great length eliminate free fatty acid food reason olive oil cold pressed instance reduce temperaturedependent oxidation oil produce free fatty acid foul taste oil last time new flavor added list 20 year ago umami added discovery receptor — almost century Japanese chemist Kikunae Ikeda first proposed 1909 discovery new type taste cell Medler colleague quite profound Ikeda’s expert say it’s new taste Identifying new cell important step effort understand basic feature human taste reveals new detail like taste cell communicate otherand ultimately sense taste work “The hope study going help u understand better taste signal sent back brain” Medler say COVID chemo shingle spit Ultimately human physiology always seems come back brain that’s certainly case taste mind exerts huge influence perceive food combine taste cue signal sens integrates memory thing people taste white wine that’s dyed red otherwise left unaltered describe taste using robust adjective red reason lot trick taste taste signal processed flavor hold nose close eye eat jellybean won’t able tell it’s strawberry lemon grape won’t right olfactory input Color nothing taste nevertheless strong driver flavor perception people taste white wine that’s dyed red otherwise left unaltered describe taste using robust adjective red experience father grapefruitorange another example — dramatic example abrupt change flavor brought new information taste mind It’s highly physically adaptive well protein saliva seen change according diet rat rat fed diet bitter food saliva upregulates protein start interacting bitter compound blocking interacting bitter taste receptor thing seen people Giving almond milk people time change saliva profile come crave thing also physically affect taste Shingles impact gustatory nerve connect taste cell brain cause people lose sense taste entirely common side effect COVID19 loss smell taste group Harvard showed earlier summer probably due coronavirus gene interfering human olfactory mucosa Aging lead loss taste people age cell turnover process may become impaired result profound Many older people experience taste cell degradation suffer loss appetite “It’s big problem old people becoming malnourished” Medler say “When taste system whack really affect consumption” Head injury radiation therapy also interfere taste system change taste food — ear surgery tooth extraction loss distortion taste often cited unwanted side effect chemotherapy medicine impact quality life well treatment outcome “The number one reason noncompliance medicine medicine tastes” Mennella say Hacking tongue milliondollar question end human sense taste abruptly changed infection injury — modulated time thing like almond milk — also hacked lot potential health application hack restoring temporary taste loss formulating oral drug don’t taste bad altering chemistry food make palatable reducing salt content without affecting taste sweetening food lower amount sugar calorie purely culinary question changing taste artistic hedonic effect Could find way modulate flavor enhance experience eating hacking tongue food Could chef tinker new option delight sens direct dining experience “Dr Medler’s work directly lead ‘Aha’ moment way” Roper say work significant give u information say still long way go understand taste system Asked whether possible deliberately practically alter taste food way expert seemed skeptical happen time soon “Will people ship way Mars licking tube toothpastelooking stuff eating Soylent say ‘Oooh tenderloin extremely tasty” Breslin asked “That’s going really really hard thing do” Roper echoed sentiment equally whimsical term comparing understanding taste system ability hear note musical composition — won’t hear whole chord we’ll able say whether it’s Beethoven Stravinsky say matter whether it’s orange grapefruitTags Food Coronavirus Science Health Research |
732 | How To Create An Easy Blog Calendar and Stick To It | You want to blog regularly. You need to. Yet, you can’t get it together.
Why? Oh, several reasons:
I have so many topics running around in my mind. I can’t settle on one.
I can’t find the time.
I’m completely blocked. I’ve already written about everything.
I started blogging in 2008, so I’ve experienced all of the above. I coach clients as well on this and I hear these statements a lot. We are all busy, so the best way to handle all of the above is to get yourself organized. How? It’s a simple approach to what is honestly a simple problem: have a system.
Let’s discuss a few options.
Keywords/KeyPhrases
One of the biggest mistakes I see writers and bloggers make is not identifying their author branding, which is the very foundation of everything they will do on their social media and in blogging.
They just want to talk about and sell sell sell their book. We all know these authors. We all avoid these authors. Maybe you’re one of them. Let’s change that.
By identifying what interests you, what you’re an expert in, what you’re passionate about, what you want to learn more about (aka, your keywords and keyphrases), you are creating a roadmap for yourself. A plan to follow. A system.
These topics are what you will blog about.
I suggest 5–6 main branding keywords/keyphrases, and 3–4 “back-ups” — topics that are fun, lighter, yet still make you, you.
Remember, we’re talking personal (aka, author) branding here. We brand the author, not (only) the book. Why? You’ll be writing more than one book. If you talk only about the book you’ve written, what happens when the next book comes along? You have to create all new accounts. Save yourself the trouble.
People connect with people, readers follow authors, so build relationships with readers by being the person you are. You’ve been doing it your whole life. Be authentically who you are. That’s what personal branding is all about.
Want to learn more about personal branding? Read here:
Example: For my writing account, Rachel Thompson, Author (aka, @RachelintheOC) my keywords/keyphrases are the following:
surviving childhood sexual abuse
relationships
love and loss
women and children’s health
mental health
writing
My back-up keywords are:
my cats
my lack of cooking skills
my love of Nutella, reading, and champagne
You won’t see me writing about champagne or my silly cats on my blog, however, I will share photos of them on my social media. This is how the keywords/keyphrases work across our entire platform. This also solves the issue of having too many topics.
Narrow your focus to help your readers know what to expect from you, yet don’t make it so incredibly restricted that you’re only writing about one thing all the time because, boring.
First Assignment: Write out a list of what interests you, as I mentioned above. Pick your 5–6 major topics and your 3–4 back-ups. Keep it handy.
Creating Your Topic Lists
Now that you know your 5–6 main keywords/keyphrases, create topic lists. I do this in my business blogging idea notebook in columns. Do what works for you. I spend about 10 minutes each week coming up with blogging ideas and just keep adding to my columns.
If it’s a blank-idea day, that’s okay. I know I’ll come back to it.
Here’s a super high-tech image of my super high-tech blogging idea notebook:
Courtesy of Rachel Thompson, Author
No need to try to figure out what I wrote. Point is: get it out of your head and onto something, whether that’s a computer program, a Notes app on your phone, or an actual notebook.
So in the example above, I have four blog post ideas under the heading of BLOGGING, and four blog post ideas under the heading of SOCIAL MEDA.
This takes a few minutes out of my week. Surely you have the time to take a few minutes to write down blog topics instead of futzing around on social media.
This is your second step in getting organized.
Once I’ve written down the blog ideas under the main topics, I transcribe them into my blogging calendar (see more below), and start again with more ideas. Stuck for ideas? Even when you think you’ve written everything there is to write, I guarantee you haven’t. I’ve been consistently blogging for 12 years.
Where to find ideas, beyond what I’ve mentioned above? Social media is a great place. I asked my @BadRedheadMedia Twitter which of these four topics held the most interest for them (all on my blogging calendar already), and “creating a blogging calendar” was by far the most popular, so this is what I’m bringing to you this week.
Courtesy of Rachel Thompson, Author
Honestly, this is my favorite go-to way to assess interest. I received over 25 responses and then went over to Facebook with a more specifically-worded question: do you have a blogging calendar and if so, what do you use and if not, why not? Those responses helped me create this post as well.
Some folks go to Quora, a Q&A platform. See what topics interest you and dive in.
I also find great topics here on Medium when I read what someone wrote and feel inspired, or, more often, read the comments and completely disagree!
Inspiration is everywhere, y’all.
Create Your Blogging Calendar
Step number three. There are many, many ways of going about this. Fancy, simple, and somewhere in-between. I’m a practical girl, so I go with something that works for me and isn’t difficult to learn: Google Calendar.
It’s free, easy to use, and easy to share, in case you have an assistant or accountability partner.
Take the lists you’ve created in your blogging notebook (or wherever you stashed it), and now add your topics to an actual calendar, which takes only a few minutes. I keep it short and sweet:
Research and outline on Mondays
Fill in on Wednesdays
Finish and submit on Fridays
Here’s what my October calendar looks like: | https://medium.com/ninja-writers/how-to-create-an-easy-blog-calendar-and-stick-to-it-22dd871239b3 | ['Rachel Thompson'] | 2020-10-17 16:45:47.311000+00:00 | ['Marketing', 'Blogging', 'Social Media', 'Productivity', 'Writing'] | Title Create Easy Blog Calendar Stick ItContent want blog regularly need Yet can’t get together Oh several reason many topic running around mind can’t settle one can’t find time I’m completely blocked I’ve already written everything started blogging 2008 I’ve experienced coach client well hear statement lot busy best way handle get organized It’s simple approach honestly simple problem system Let’s discus option KeywordsKeyPhrases One biggest mistake see writer blogger make identifying author branding foundation everything social medium blogging want talk sell sell sell book know author avoid author Maybe you’re one Let’s change identifying interest you’re expert you’re passionate want learn aka keywords keyphrases creating roadmap plan follow system topic blog suggest 5–6 main branding keywordskeyphrases 3–4 “backups” — topic fun lighter yet still make Remember we’re talking personal aka author branding brand author book You’ll writing one book talk book you’ve written happens next book come along create new account Save trouble People connect people reader follow author build relationship reader person You’ve whole life authentically That’s personal branding Want learn personal branding Read Example writing account Rachel Thompson Author aka RachelintheOC keywordskeyphrases following surviving childhood sexual abuse relationship love loss woman children’s health mental health writing backup keywords cat lack cooking skill love Nutella reading champagne won’t see writing champagne silly cat blog however share photo social medium keywordskeyphrases work across entire platform also solves issue many topic Narrow focus help reader know expect yet don’t make incredibly restricted you’re writing one thing time boring First Assignment Write list interest mentioned Pick 5–6 major topic 3–4 backup Keep handy Creating Topic Lists know 5–6 main keywordskeyphrases create topic list business blogging idea notebook column work spend 10 minute week coming blogging idea keep adding column it’s blankidea day that’s okay know I’ll come back Here’s super hightech image super hightech blogging idea notebook Courtesy Rachel Thompson Author need try figure wrote Point get head onto something whether that’s computer program Notes app phone actual notebook example four blog post idea heading BLOGGING four blog post idea heading SOCIAL MEDA take minute week Surely time take minute write blog topic instead futzing around social medium second step getting organized I’ve written blog idea main topic transcribe blogging calendar see start idea Stuck idea Even think you’ve written everything write guarantee haven’t I’ve consistently blogging 12 year find idea beyond I’ve mentioned Social medium great place asked BadRedheadMedia Twitter four topic held interest blogging calendar already “creating blogging calendar” far popular I’m bringing week Courtesy Rachel Thompson Author Honestly favorite goto way ass interest received 25 response went Facebook specificallyworded question blogging calendar use response helped create post well folk go Quora QA platform See topic interest dive also find great topic Medium read someone wrote feel inspired often read comment completely disagree Inspiration everywhere y’all Create Blogging Calendar Step number three many many way going Fancy simple somewhere inbetween I’m practical girl go something work isn’t difficult learn Google Calendar It’s free easy use easy share case assistant accountability partner Take list you’ve created blogging notebook wherever stashed add topic actual calendar take minute keep short sweet Research outline Mondays Fill Wednesdays Finish submit Fridays Here’s October calendar look likeTags Marketing Blogging Social Media Productivity Writing |
733 | Feed Your Brain to Make it Stronger | You can improve growth and function at ANY age
Image by Chenspec from Pixabay
It is the repository of all we know and have ever experienced; it has the capability to solve problems and create them; it is the uninhibited fountain of dreams and ideas and the channel of disciplined action. It is also the largest sex organ in our bodies (but size doesn’t matter).
Of course, I am referring to our brains. In particular, I am referring to our aging brains.
First, let’s dispel the popular myths. These are the things holding us back, inhibiting us from enjoying the fullness of our aging brains.
Misconception: As we age, our brains decline, resulting in an inability to recall information and master new things.
Certainly, at some point, we have all found it a little difficult to remember a name, misplaced our keys, or taken a little longer than usual to perform a mental task. Those “senior moments” are not necessarily signs of mental decline. They are signs of mental change.
In some ways, a healthy older brain has been shown to function better than healthy younger brains.
“Studies have shown that older people have better judgment, are better at making rational decisions, and are better able to screen out negativity than their juniors are,” according to an article from the Harvard Medical School’s Harvard Medical Publishing.
Writing in the Harvard Business Review, Roderick Gilkey and Clint Kilts note that work began in the 1990s has “yielded a broad front of research and training that has upended some deeply held beliefs about the brain. One such belief is that the brain necessarily diminishes with age. It turns out that neurons, the basic cells that allow information transfer to support the brain’s computing power, do not have to die off as we get older. In fact, a number of regions of the brain important to functions such as motor behavior and memory can actually expand their complement of neurons as we age. This process, called neurogenesis, used to be unthinkable in mainstream neuroscience.”
In short, we can physically improve our brains and brain function. How? The same way we improve our physical function — through a good workout. But not just a physical workout — a regular brain workout.
Brain imaging studies have shown you can make changes in your brain by learning a new skill, such as studying a new language, playing a musical instrument, taking up a new hobby, or doing almost anything you’ve never done before.
The simple use of will power can also improve your brain’s capabilities. Further brain imaging studies have shown that participants were able to control the sensation of pain through concentration. Accomplished practitioners of meditation have long been able to control physical pain or irritation.
“These advances in neuroscience suggest that there is no reason why your brain at 60 can’t be as competent as it was at 25,” according to Gilkey and Kilts. “That would not have been news to thinkers such as Socrates, Copernicus, and Galileo, who were all still at the peak of their intellectual powers in their sixties and seventies. Nor would it surprise business leaders such as Alan Greenspan, Warren Buffett, and Sumner Redstone. These icons and others like them have intuitively understood that the brain’s alertness is the result of what we call cognitive fitness — a state of optimized ability to reason, remember, learn, plan, and adapt that is enhanced by certain attitudes, lifestyle choices, and exercises.”
Many researchers say that new tricks or variations on old habits are the prescription for maintaining and improving cognitive function.
“Exercising, keeping your brain active, and maintaining creativity can actually help to prevent cognitive decline and memory problems.” Melinda Smith, M.A., Jeanne Segal, PhD., and Monika White, Ph.D. wrote in the nonprofit publication Getting Help. “The more active and social you are and the more you use and sharpen your brain, the more benefits you will get.”
Smith, Segal, and White recommend three actions for maintaining and improving mental acuity. | https://medium.com/crows-feet/feed-your-brain-to-make-it-stronger-61bcb87d31f6 | ['Max K. Erkiletian'] | 2020-11-18 16:57:59.864000+00:00 | ['Aging', 'Brain', 'Health', 'Seniors', 'Neuroscience'] | Title Feed Brain Make StrongerContent improve growth function age Image Chenspec Pixabay repository know ever experienced capability solve problem create uninhibited fountain dream idea channel disciplined action also largest sex organ body size doesn’t matter course referring brain particular referring aging brain First let’s dispel popular myth thing holding u back inhibiting u enjoying fullness aging brain Misconception age brain decline resulting inability recall information master new thing Certainly point found little difficult remember name misplaced key taken little longer usual perform mental task “senior moments” necessarily sign mental decline sign mental change way healthy older brain shown function better healthy younger brain “Studies shown older people better judgment better making rational decision better able screen negativity junior are” according article Harvard Medical School’s Harvard Medical Publishing Writing Harvard Business Review Roderick Gilkey Clint Kilts note work began 1990s “yielded broad front research training upended deeply held belief brain One belief brain necessarily diminishes age turn neuron basic cell allow information transfer support brain’s computing power die get older fact number region brain important function motor behavior memory actually expand complement neuron age process called neurogenesis used unthinkable mainstream neuroscience” short physically improve brain brain function way improve physical function — good workout physical workout — regular brain workout Brain imaging study shown make change brain learning new skill studying new language playing musical instrument taking new hobby almost anything you’ve never done simple use power also improve brain’s capability brain imaging study shown participant able control sensation pain concentration Accomplished practitioner meditation long able control physical pain irritation “These advance neuroscience suggest reason brain 60 can’t competent 25” according Gilkey Kilts “That would news thinker Socrates Copernicus Galileo still peak intellectual power sixty seventy would surprise business leader Alan Greenspan Warren Buffett Sumner Redstone icon others like intuitively understood brain’s alertness result call cognitive fitness — state optimized ability reason remember learn plan adapt enhanced certain attitude lifestyle choice exercises” Many researcher say new trick variation old habit prescription maintaining improving cognitive function “Exercising keeping brain active maintaining creativity actually help prevent cognitive decline memory problems” Melinda Smith Jeanne Segal PhD Monika White PhD wrote nonprofit publication Getting Help “The active social use sharpen brain benefit get” Smith Segal White recommend three action maintaining improving mental acuityTags Aging Brain Health Seniors Neuroscience |
734 | 3 Trends That Are Putting the Power of Business Intelligence in More Hands | 3 Trends That Are Putting the Power of Business Intelligence in More Hands
A Conversation with Nick Caldwell on the State of the Business Intelligence Landscape
Looks like the summer is heating up in the Business Intelligence (BI) industry! Two major acquisitions have been announced in June 2019 within a week of each other. The first: Google’s acquisition of Looker for $2.6B. This deal is expected to be completed later this year. Thomas Kurian, Google Cloud CEO released an announcement on his intention to acquire the unified platform for BI, data apps and embedded analytics. Kurian views the acquisition of Looker as an extension of Google’s business analytics offerings. He states, “For any business that is looking for a partner to help drive digital transformation, the combination of Google Cloud and Looker will offer an incredible data management and analytics platform.” While Google is looking to beef up its cloud offerings, Salesforce recently bought Tableau for $15.7B. This was announced the week following the Looker acquisition. Their strategy, according to Salesforce CEO Marc Benioff, is to merge the top CRM platform and the top analytics platform to help speed customer digital transformations.
Nick is Looker’s Chief Product Officer, leading the company’s Engineering and Product Management and Design teams. Prior to Looker, he was VP of Engineering at Reddit and the General Manager of Power BI for Microsoft.
I had an opportunity to chat with Looker’s Chief Product Officer, Nick Caldwell about the state of business intelligence and where the industry is heading. He leads the company’s Engineering and Product Management and Design teams. Prior to Looker, he was VP of Engineering at Reddit and General Manager of Power BI for Microsoft. I wanted to get his take on the data visualization landscape and how it’s impacting the data-driven workforce.
Nick acknowledged that there is a lot of consolidation in the BI space right now as big platform players are looking for BI to be “the sizzle on the steak” and give an extra oomph to their portfolios.
Thomas Kurian stated in his press release that, “The addition of Looker to Google Cloud will help us offer customers a more complete analytics solution from ingesting data to visualizing results and integrating data and insights into their daily workflows.”
this is a BI world that will call for more than just dashboards, data prep and advanced analytics
Nick views both the Looker and Tableau acquisitions as examples of big platforms highlighting their collective power versus the individual capabilities of their product suites. The amount of data at our fingertips is more prevalent in our lives than it has ever been. He explained that this is a BI world that will call for more than just dashboards, data prep and advanced analytics — A trend he’s excited about. We discussed the trends influencing the BI market and how Looker has been riding the data wave! Nick sees three major trends:
Trend 1. The Arrival of a Data-Driven Workforce
Data is embedded in all facets of our lives, especially in the workplace. I definitely agree with him on this point. He went on to say that everyone needs to get the job done — Not just traditional analysts or quantitative marketers. I’m counting the days to when Looker or Tableau are a standard installation on office laptops but thinks standalone apps are the future.
“Of course we still need great BI tools for our analysts and data jockeys. But we must also be aware that there are a rapidly increasing number of data-enabled workers who view the idea of using dashboards the same way that you or I might view using a rotary phone. And that’s okay.”
Trend 2. A proliferation of SaaS Apps
Nick explained that untraditional data folks expect the data to come to them in an interface designed specifically for the task at hand, ideally integrated into a tool they are already familiar with. He speculates that the future of BI will not look much like a BI application at all and believes we will not need to have everyone become familiar with analyst tools to have an insight-driven business. Software as a Service (SaaS) apps are “polished user interfaces on top of rows of data”.
“People aren’t going to go to BI, BI has to go to the people. This is already happening in a big way.” ~Nick Caldwell
While Nick was explaining how the adoption of SaaS apps was skyrocketing, my mind went to dashboards and how they were on the top of every department’s wish list. This usually caused a bottleneck at the analyst as they looked to gather business requirements and find the data set needed to build the views. Then once the dataset was identified, performance was the next problem to deal with. Nick helped me to understand that SaaS apps put the power of data in the hands of non-traditional data folks from factory workers to pizza delivery drivers. I questioned the processing power that would be needed for all of these individual apps. Like dashboards, I figured this would call for numerous data silos and countless meetings with IT to make sure each app ran efficiently. This takes us to trend #3.
Trend 3. Modern Data Infrastructure
Nick explained that Looker took a huge bet and invested heavily in being able to natively support massively parallel processing (MPP) data warehouses like Google Cloud Big Query, Snowflake, Amazon Redshift Spectrum. This infrastructure supports the world of SaaS apps. “MPP data warehouses can hold immense amounts of data, query it all in seconds, and even do advanced analytics directly in the database — all this at a cost which is bafflingly cheap compared to last generation technologies.” says Caldwell. He went on to point out this infrastructure improves the lives of data engineers because they can spend less of their time building and maintaining data pipelines and more time doing what they really want: empowering end-users with data-driven experiences. | https://medium.com/nightingale/3-trends-that-are-putting-the-power-of-business-intelligence-in-more-hands-955fa77f4463 | ['Allen Hillery'] | 2020-04-02 04:03:10.589000+00:00 | ['Data Visualization', 'Business', 'Google', 'Business Intelligence', 'Marketing'] | Title 3 Trends Putting Power Business Intelligence HandsContent 3 Trends Putting Power Business Intelligence Hands Conversation Nick Caldwell State Business Intelligence Landscape Looks like summer heating Business Intelligence BI industry Two major acquisition announced June 2019 within week first Google’s acquisition Looker 26B deal expected completed later year Thomas Kurian Google Cloud CEO released announcement intention acquire unified platform BI data apps embedded analytics Kurian view acquisition Looker extension Google’s business analytics offering state “For business looking partner help drive digital transformation combination Google Cloud Looker offer incredible data management analytics platform” Google looking beef cloud offering Salesforce recently bought Tableau 157B announced week following Looker acquisition strategy according Salesforce CEO Marc Benioff merge top CRM platform top analytics platform help speed customer digital transformation Nick Looker’s Chief Product Officer leading company’s Engineering Product Management Design team Prior Looker VP Engineering Reddit General Manager Power BI Microsoft opportunity chat Looker’s Chief Product Officer Nick Caldwell state business intelligence industry heading lead company’s Engineering Product Management Design team Prior Looker VP Engineering Reddit General Manager Power BI Microsoft wanted get take data visualization landscape it’s impacting datadriven workforce Nick acknowledged lot consolidation BI space right big platform player looking BI “the sizzle steak” give extra oomph portfolio Thomas Kurian stated press release “The addition Looker Google Cloud help u offer customer complete analytics solution ingesting data visualizing result integrating data insight daily workflows” BI world call dashboard data prep advanced analytics Nick view Looker Tableau acquisition example big platform highlighting collective power versus individual capability product suite amount data fingertip prevalent life ever explained BI world call dashboard data prep advanced analytics — trend he’s excited discussed trend influencing BI market Looker riding data wave Nick see three major trend Trend 1 Arrival DataDriven Workforce Data embedded facet life especially workplace definitely agree point went say everyone need get job done — traditional analyst quantitative marketer I’m counting day Looker Tableau standard installation office laptop think standalone apps future “Of course still need great BI tool analyst data jockey must also aware rapidly increasing number dataenabled worker view idea using dashboard way might view using rotary phone that’s okay” Trend 2 proliferation SaaS Apps Nick explained untraditional data folk expect data come interface designed specifically task hand ideally integrated tool already familiar speculates future BI look much like BI application belief need everyone become familiar analyst tool insightdriven business Software Service SaaS apps “polished user interface top row data” “People aren’t going go BI BI go people already happening big way” Nick Caldwell Nick explaining adoption SaaS apps skyrocketing mind went dashboard top every department’s wish list usually caused bottleneck analyst looked gather business requirement find data set needed build view dataset identified performance next problem deal Nick helped understand SaaS apps put power data hand nontraditional data folk factory worker pizza delivery driver questioned processing power would needed individual apps Like dashboard figured would call numerous data silo countless meeting make sure app ran efficiently take u trend 3 Trend 3 Modern Data Infrastructure Nick explained Looker took huge bet invested heavily able natively support massively parallel processing MPP data warehouse like Google Cloud Big Query Snowflake Amazon Redshift Spectrum infrastructure support world SaaS apps “MPP data warehouse hold immense amount data query second even advanced analytics directly database — cost bafflingly cheap compared last generation technologies” say Caldwell went point infrastructure improves life data engineer spend le time building maintaining data pipeline time really want empowering endusers datadriven experiencesTags Data Visualization Business Google Business Intelligence Marketing |
735 | How Blogging About My Children Helped Me Become a Professional Writer | Photo by Thom Milkovic on Unsplash
When my girls were little, my writing time was pared down to a few stolen minutes in the evening. I’d scribble away in a journal, writing stream-of-consciousness vignettes that I (infrequently) turned into poems. Sometimes I’d jot ideas down in these same journals, hoping to revisit them at some point for the novel I never seemed to have time to write.
I wanted to be a writer, but my identity back then was immersed in motherhood. Writing took a back seat to more important things, including my professional work which didn’t involve much writing at all.
But what I didn’t anticipate (or fully accept) was that motherhood and writing would naturally intertwine. Indeed, I hated the term “mommy blogger” and actively rebelled against starting my own blog.
Even so, I caved and began blogging about my younger daughter in 2004. My blog wasn’t the standard mommy blog (I still shudder at this term). I didn’t agonize about standard new-mom topics like whether or not to breastfeed or how I’d introduce the new baby to my then 2-year-old.
My first blog post was about the treatment we’d chosen to use for my unborn baby’s cleft lip and palate, a birth defect we learned about at my 20-week ultrasound.
I was inspired to write the blog I’d called Cleft Stories, because my frantic research about raising a baby with a cleft lip and palate had turned up almost no information from a parent’s perspective. I’d mostly encountered clinical websites that talked about cleft lip and palate using medical terminology that featured jarring photographs of various types of clefts. At least, they were jarring to me. I was very pregnant and very scared and I just needed to know that my baby would be okay.
My first post included an ultrasound of my baby, her cleft circled in red. My goal was to write about my daughter’s medical journey to keep friends and family updated and to chronicle the procedure we’d chosen to treat and repair her cleft (a process called nasoalveolar molding).
But I was also determined to write about my daughter just like any other mommy blogger, chronicling milestones, sharing photos, and demonstrating that while babies with clefts face unique challenges, they are manageable. I wanted to illustrate that my daughter was normal and that things would be okay.
My blog was one of the first cleft-focused blogs written by a parent (at least, that I know of). NYU Medical Center, the hospital where my daughter was treated, routinely directed new parents to the blog, so it built up a devoted following of fellow parents.
The blog also inspired other families to start blogs of their own. In just a few short years, it was much easier to find personal stories about children growing up with cleft lip and palate among all those clinical pages about the condition. These days, there are also Facebook groups and other resources for families who need support.
I stopped updating my daughter’s blog as she got older. Her lip, nose, gums, and palate were repaired by the age of one and she had another surgery at age 7, but otherwise there wasn’t much to say about her cleft. Even so, I posted once or twice a year, showcasing school performances and sharing photos on her birthday so people could how her face changed as she got older, and (more importantly) demonstrate that life was relatively normal.
Then, in 2012, my older daughter was diagnosed with cancer at the age of 11. I updated Cleft Stories one last time, directing readers to my new blog, Healing Ana. I had once again become a mommy blogger, though cancer was not a topic I ever imagined myself writing about. Healing Ana blog is still live — for now. I haven’t updated it in nearly three years and I can’t bear to link to it.
I blogged about Ana’s cancer, her treatment, and how our family was coping, from 2012 until a few months after she died in 2017. The blog kept me going. It helped me stay connected to a large extended community of people who propped my family up during the tumultuous years of Ana’s illness, and kept me from sinking into complete despair after she died.
Something else emerged from the experience of maintaining that blog — my confidence in my writing began to grow. People who read my posts told me that my writing was helping them with their own struggles, even if they had nothing to do with cancer. In 2015, a few years after I started the blog about Ana, I began contributing essays to Huffpost.
Back then, the platform was similar to Medium (minus the paywall). If you could get added as a Huffpost writer (I did this fairly easily by writing to Arianna Huffington directly), you gained access to the platform and could post articles instantly.
Several of my early articles were featured in Huffpost’s Parenting section. The process was similar to curation in that an editor could select articles contributed by their pool of writers and feature it on a topic page. One of my earliest Huffpost articles was even featured on their homepage. As a result, thousands of people read the piece and dozens commented on it.
This gave me the confidence to begin submitting my essays to more high profile (and paying) publications. I continued writing about parenting a child with cancer and, subsequently, about parental grief.
The essays were an expansion of what I’d already been doing on my blogs — writing about the heartbreaking trauma we were experiencing in a way that not only helped me process it, but enabled me to connect to other parents who were dealing with the stress of caring for a terminally ill child.
The Washington Post was the first publication to pay me for an essay. When the money went into my account, something in my brain shifted. For the first time in my life, I realized that I could get paid for writing . I began submitting more essays to different publications. I got rejected (a lot), but my work also got accepted and I continued to earn some extra money from writing.
I made the decision to stop blogging sometime in 2018. That’s when I also realized that I didn’t want to continue working as a freelance digital marketing consultant. At the age of 47, I finally had the confidence (and clips) to pivot to freelance writing as my primary source of income.
I decided to experiment with writing on Medium at the start of 2019, first by republishing many of my Huffpost pieces, then by writing original essays on the platform. I began actively seeking freelance writing clients a few months before that, and was able to walk away from all my marketing work by January of 2019.
Freelance writing isn’t easy. For one thing, it doesn’t pay as well as digital marketing. Most months, if I’m lucky, I average roughly one paid piece per business day. That means I write about 20,000 to 25,000 words a month. This can be creatively draining to the extent that I need at least one to two non-writing days per week to recharge.
But, even with these challenges, writing for a living has transformed the way I work and live in ways I hadn’t anticipated. As an early riser, I’m often done with the day’s assignment by 1 or 2 pm, leaving the rest of the day free. I used to feel guilty about not sitting at my desk until 5 pm each day (or later), particularly because my digital marketing clients needed me to be available during business hours.
But it’s rare that a client needs me to drop everything to fix a writing emergency, particularly when I meet my deadlines (and I always meet my deadlines). Now, I can shape my days and weeks in ways that I never could before, and that’s a pretty significant when you consider that I’ve been my own boss for 18 years.
It’s still hard for me to call myself a writer, still nearly impossible for me to believe that writing is what pays my bills. A big reason for this disbelief is that I had an entirely different idea of what a “real” writer looked like. For many years, I could accept nothing less than the “Stephen King” model of writing. How I envied him — tapping away at his Underwood Typewriter for a few hours a day, while royalty checks rolled into his overflowing bank account.
But being a successful writer doesn’t have to look like that. Sure, there are a few unicorns like Stephen King and J. K. Rowling, but for the rest of us mere mortals, writing isn’t necessarily about fame and fortune. Sometimes successful writers can barely keeping their heads above water (ahem).
Being a successful writer just means that you’re writing, successfully — putting words down on the screen (or on paper), building a body of work that you’re proud of and reaching readers, one article at a time. For me, it also means getting paid for doing it, not because I’m attached to being fabulously wealthy, but because getting paid to write means that I can keep writing.
If you’d told me, seventeen years ago, that the anxiety-riddled musings of an expectant mother would one day help me realize my dream of being a writer, I probably wouldn’t have believed you.
What I didn’t understand back then was that sometimes we must reframe how we define our dreams before we can recognize that they’re attainable. It wasn’t until I abandoned my (very narrow) definition of what it meant to be a successful writer, that I was finally able to become one. | https://jacquelinedooley.medium.com/how-blogging-about-my-children-helped-me-become-a-professional-writer-79d8d0ee4012 | ['Jacqueline Dooley'] | 2020-11-08 15:16:54.655000+00:00 | ['Self', 'Nonfiction', 'Creativity', 'Parenting', 'Writing'] | Title Blogging Children Helped Become Professional WriterContent Photo Thom Milkovic Unsplash girl little writing time pared stolen minute evening I’d scribble away journal writing streamofconsciousness vignette infrequently turned poem Sometimes I’d jot idea journal hoping revisit point novel never seemed time write wanted writer identity back immersed motherhood Writing took back seat important thing including professional work didn’t involve much writing didn’t anticipate fully accept motherhood writing would naturally intertwine Indeed hated term “mommy blogger” actively rebelled starting blog Even caved began blogging younger daughter 2004 blog wasn’t standard mommy blog still shudder term didn’t agonize standard newmom topic like whether breastfeed I’d introduce new baby 2yearold first blog post treatment we’d chosen use unborn baby’s cleft lip palate birth defect learned 20week ultrasound inspired write blog I’d called Cleft Stories frantic research raising baby cleft lip palate turned almost information parent’s perspective I’d mostly encountered clinical website talked cleft lip palate using medical terminology featured jarring photograph various type cleft least jarring pregnant scared needed know baby would okay first post included ultrasound baby cleft circled red goal write daughter’s medical journey keep friend family updated chronicle procedure we’d chosen treat repair cleft process called nasoalveolar molding also determined write daughter like mommy blogger chronicling milestone sharing photo demonstrating baby cleft face unique challenge manageable wanted illustrate daughter normal thing would okay blog one first cleftfocused blog written parent least know NYU Medical Center hospital daughter treated routinely directed new parent blog built devoted following fellow parent blog also inspired family start blog short year much easier find personal story child growing cleft lip palate among clinical page condition day also Facebook group resource family need support stopped updating daughter’s blog got older lip nose gum palate repaired age one another surgery age 7 otherwise wasn’t much say cleft Even posted twice year showcasing school performance sharing photo birthday people could face changed got older importantly demonstrate life relatively normal 2012 older daughter diagnosed cancer age 11 updated Cleft Stories one last time directing reader new blog Healing Ana become mommy blogger though cancer topic ever imagined writing Healing Ana blog still live — haven’t updated nearly three year can’t bear link blogged Ana’s cancer treatment family coping 2012 month died 2017 blog kept going helped stay connected large extended community people propped family tumultuous year Ana’s illness kept sinking complete despair died Something else emerged experience maintaining blog — confidence writing began grow People read post told writing helping struggle even nothing cancer 2015 year started blog Ana began contributing essay Huffpost Back platform similar Medium minus paywall could get added Huffpost writer fairly easily writing Arianna Huffington directly gained access platform could post article instantly Several early article featured Huffpost’s Parenting section process similar curation editor could select article contributed pool writer feature topic page One earliest Huffpost article even featured homepage result thousand people read piece dozen commented gave confidence begin submitting essay high profile paying publication continued writing parenting child cancer subsequently parental grief essay expansion I’d already blog — writing heartbreaking trauma experiencing way helped process enabled connect parent dealing stress caring terminally ill child Washington Post first publication pay essay money went account something brain shifted first time life realized could get paid writing began submitting essay different publication got rejected lot work also got accepted continued earn extra money writing made decision stop blogging sometime 2018 That’s also realized didn’t want continue working freelance digital marketing consultant age 47 finally confidence clip pivot freelance writing primary source income decided experiment writing Medium start 2019 first republishing many Huffpost piece writing original essay platform began actively seeking freelance writing client month able walk away marketing work January 2019 Freelance writing isn’t easy one thing doesn’t pay well digital marketing month I’m lucky average roughly one paid piece per business day mean write 20000 25000 word month creatively draining extent need least one two nonwriting day per week recharge even challenge writing living transformed way work live way hadn’t anticipated early riser I’m often done day’s assignment 1 2 pm leaving rest day free used feel guilty sitting desk 5 pm day later particularly digital marketing client needed available business hour it’s rare client need drop everything fix writing emergency particularly meet deadline always meet deadline shape day week way never could that’s pretty significant consider I’ve bos 18 year It’s still hard call writer still nearly impossible believe writing pay bill big reason disbelief entirely different idea “real” writer looked like many year could accept nothing le “Stephen King” model writing envied — tapping away Underwood Typewriter hour day royalty check rolled overflowing bank account successful writer doesn’t look like Sure unicorn like Stephen King J K Rowling rest u mere mortal writing isn’t necessarily fame fortune Sometimes successful writer barely keeping head water ahem successful writer mean you’re writing successfully — putting word screen paper building body work you’re proud reaching reader one article time also mean getting paid I’m attached fabulously wealthy getting paid write mean keep writing you’d told seventeen year ago anxietyriddled musing expectant mother would one day help realize dream writer probably wouldn’t believed didn’t understand back sometimes must reframe define dream recognize they’re attainable wasn’t abandoned narrow definition meant successful writer finally able become oneTags Self Nonfiction Creativity Parenting Writing |
736 | Committed to Success: Clara Tsang, Software Engineer at Procurify | Clara Tsang joined Commit in 2019 and quickly found a role as a front end developer with Procurify, a rapidly growing Vancouver-based startup in the spend-management software space. We sat down with Clara to hear about her journey with Commit and Procurify.
Could you tell us about your background and experience before joining Commit?
I graduated last year from SFU’s Systems Engineering program, which actually isn’t a software program, it’s an electronic systems engineering program. I was exposed to software, but definitely not much web development. I did a few co-ops — mostly QA positions. This was mainly manual QA and unit test writing. After graduation, I worked for a digital marketing firm focused on the automotive industry. They hired me to do web dev, even though I didn’t have much experience. I was working with WordPress, so mostly vanilla JS, jQuery, HTML, CSS and PHP.
“It’s really easy to get lost navigating the world of nasty technical interviews and multiple-level screening that’s the reality in the tech industry today. Commit offers a better route.”
What drew you to Commit?
I was ready for a change, but I thought it would be hard to get an interview with my current skills and not easy to gain new skills while working full time. I talked to Cong Ly [Responsible for onboarding new Engineering Partners] and Commit was ready to take me on and train me on the job. Joining Commit was an upgrade in every way. It actually sounded too good to be true: maximum flexibility with remote working and on-the-job training. Plus it was higher pay and I got to work with technologies and standards that feel more relevant to the industry.
What were you hoping to achieve when you joined?
First and foremost it was to level up my front end skills because at the time I felt that I wasn’t there technically. My one-year plan was to start looking into full-stack work.
How did your time with Commit help you toward those goals?
The first two weeks were focused on onboarding, where I started learning React by following along with the React docs; the project I was going to be placed on first was a React project. I was matched up with a technical mentor from Commit, who pointed me towards great resources so I could learn the stuff. I met up with my mentor a lot in those first two weeks.
How did you get connected with Procurify?
Early on, Cong said he had a project he thought would be a great fit for me. That was good for me at that point, because I didn’t really know what I wanted or needed.
Can you tell us about Procurify?
We make a spending tracking app for businesses, so I guess you could say our product mission is to make spend culture in companies faster, less complex and more transparent. They’re also trying to create an awesome place for people to work.
And how are they doing at that last goal?
A really good job. I feel like Procurify really cares about its employees and trusts them. They have a responsible time-off policy (AKA unlimited vacation), flexible hours, and there’s a lot of effort to be remote-forward.
Are you learning lots and developing new skills?
Definitely. I think that was the main reason why I decided to stay with Procurify. Everyone in the office seems pretty passionate about their job, which is pretty new for me, too. As someone who’s a bit newer to the industry, there are so many people there I can learn from. And in engineering, there’s a lot of initiative to keep improving things and stay up to date with the latest technologies.
What advice would you give to other engineers about joining Commit?
I’d say just do it. Commit is designed by engineers for engineers, so there’s something in it for every engineer — well, at least any engineer interested in continuous learning and working remotely. If you’re looking to learn a lot and work on different projects, there’s that option. And if you’re looking for a new company to join full-time, there’s that option too. It’s really easy to get lost navigating the world of nasty technical interviews and multiple-level screening that’s the reality in the tech industry today. Commit offers a better route. | https://medium.com/commit-engineering/committed-to-success-clara-tsang-software-engineer-at-procurify-fdc1e3998bab | ['Beier Cai'] | 2020-06-14 21:47:10.295000+00:00 | ['Engineering', 'Hiring For Startup', 'Software Development', 'Startup', 'Software Engineering'] | Title Committed Success Clara Tsang Software Engineer ProcurifyContent Clara Tsang joined Commit 2019 quickly found role front end developer Procurify rapidly growing Vancouverbased startup spendmanagement software space sat Clara hear journey Commit Procurify Could tell u background experience joining Commit graduated last year SFU’s Systems Engineering program actually isn’t software program it’s electronic system engineering program exposed software definitely much web development coop — mostly QA position mainly manual QA unit test writing graduation worked digital marketing firm focused automotive industry hired web dev even though didn’t much experience working WordPress mostly vanilla JS jQuery HTML CSS PHP “It’s really easy get lost navigating world nasty technical interview multiplelevel screening that’s reality tech industry today Commit offer better route” drew Commit ready change thought would hard get interview current skill easy gain new skill working full time talked Cong Ly Responsible onboarding new Engineering Partners Commit ready take train job Joining Commit upgrade every way actually sounded good true maximum flexibility remote working onthejob training Plus higher pay got work technology standard feel relevant industry hoping achieve joined First foremost level front end skill time felt wasn’t technically oneyear plan start looking fullstack work time Commit help toward goal first two week focused onboarding started learning React following along React doc project going placed first React project matched technical mentor Commit pointed towards great resource could learn stuff met mentor lot first two week get connected Procurify Early Cong said project thought would great fit good point didn’t really know wanted needed tell u Procurify make spending tracking app business guess could say product mission make spend culture company faster le complex transparent They’re also trying create awesome place people work last goal really good job feel like Procurify really care employee trust responsible timeoff policy AKA unlimited vacation flexible hour there’s lot effort remoteforward learning lot developing new skill Definitely think main reason decided stay Procurify Everyone office seems pretty passionate job pretty new someone who’s bit newer industry many people learn engineering there’s lot initiative keep improving thing stay date latest technology advice would give engineer joining Commit I’d say Commit designed engineer engineer there’s something every engineer — well least engineer interested continuous learning working remotely you’re looking learn lot work different project there’s option you’re looking new company join fulltime there’s option It’s really easy get lost navigating world nasty technical interview multiplelevel screening that’s reality tech industry today Commit offer better routeTags Engineering Hiring Startup Software Development Startup Software Engineering |
737 | The Secrets Behind How I Made $12,184.40 from One Medium Article | One cold, early morning in late January 2018, I woke at 5:30 to let my new puppy relieve himself.
While he did his business, I brewed a pot of coffee then opened my laptop to start writing. About an hour later, I had written a draft of an article that would forever change my views on what was possible writing online.
That evening, I spent another 10 to 15 minutes editing the article before submitting it to The Mission as a freebie (not behind Medium’s paywall) to the world.
I settled on the title, “The 50/30/10/10 Rule for How to Wake Up Earlier and Work on Your Dream”.
I’ll let the stats speak for itself:
Since its debut, the article has accumulated 900,000 views, 65,000 claps, and has grossed just over $12,000.
Never in my wildest dreams would I have guessed that an article, written off-the-cuff in barely an hour, would end up generating five figures in earnings.
But I guess that’s how it goes — the articles you spend hours slaving over flop, and the ones you spend the least time preparing for make the biggest splash.
You want to know the secrets behind an article like this? Here they are.
I was hardly an “expert” on sleep or writing
Many newer writers think that you have to be an expert, the best-of-the-best, at something for anyone to care. My success with this article proves that’s not true. Though I’m quite fantastic at sleeping and a consistent writer, trust me I’m no Matthew Walker or Ernest Hemingway.
What I am is just a regular dude who loved the IDEA of waking up early but struggled most of his life to consistently get out of bed in the mornings. I’m a dude with dreams, goals, and ambition who, perhaps like you, wasn’t seeing the progress he wanted because his evenings were a shitshow, and the one time of day he had to write — the morning — he spent oversleeping and making zero progress.
So, after a couple of months of successfully waking early, I was over the moon about my newfound habit and wanted to share how I did it with random people on the internet.
This didn’t make me an “expert” on the topic — just a guy who figured out some key pieces to the puzzle and broke it down in a way that made sense to others.
More than anything, the article was a way for me to systematize my efforts as a personal reminder as to what was actually important to keep waking up early and getting my stuff done.
The story did quite well at first, but WAY BETTER later
The Mission proved to be a worthwhile publication for the story. During the article’s tenure there, it racked up something like 11,000 claps and a not-too-shabby one hundred-ish thousand views.
If you look at the graph above, you can see that the article did pretty well the first couple weeks after publishing but dropped off hard for ~ 4 months before SEO kicked in and Google started sending me free traffic in May of 2018.
Hellooo free traffic!
Prior to the search engine magic, the article averaged just 2–3,000 views per month. Afterward, it surged to 18–20,000. Since it was a free article and Medium’s paid Partner Program was still finding its legs, all views at this point were considered external.
But then, in May 2019 — sixteen months after publishing — I decided to move the article into the Better Humans publication, one of the first pubs specializing in the paywall (and it didn’t hurt that I had a relationship with the editor).
With Better Human’s help, the article quickly became Featured in Medium’s Top Stories.
In addition to the external Google traffic, the article was now getting boosted by Medium’s internal algorithms and received a significant bump in internal views.
At this point though, I had still made $0 from the story (remember, I published it as a “free” story), but that changed rather quickly.
Getting Paid
If you look at the story’s stats, you’ll notice there’s about a 3-month difference from when the article was published in Better Humans to when it started earning money.
This is because the Partner Program was still in its infancy, so Better Humans cut me a check each month for 100% of the earnings from the article. Eventually, it migrated over to a traditional Partner Program story, which is when I made my first $9.99 officially from Medium on September 9, 2019.
One month later, I vividly remember waking up in London while away on vacation, scrolling through my emails, and seeing a notification from Medium that I had received a payment.
“Oh cool, wonder what little amount I made this month.”
I legitimately felt like someone at Medium had made a mistake, and I actually expected someone to call me asking for the money back (no one ever did).
Shortly after that, the story exploded.
A couple of months later, NBC emailed me to request an interview for the article. I called my mom immediately after.
“Mom, NBC asked to interview ME!”
A month after that, Better Humans interviewed me for their podcast — an opportunity that fell through due to the audio quality on my end (I took the interview through Skype, in my car, on my lunch break at work).
With each passing month, I thought “there’s no way I’ll make that much again.” But every month, the article continued to thrive and I watched the earnings pour in.
Eventually, my income from the article tapered down to just a few hundred dollars per month, at which point I figured my luck had finally fizzled out. But then, boom, it spiked again.
And for another couple months, I was graced with 4-figure earnings again. All from just one article that barely took an hour to write.
Only time will tell what the rest of the graph will look like, and believe me when I say that I’m just as curious as anyone to see how it plays out.
Key Takeaways
It’s hard to state just how much I learned from an experience like this, but for brevity’s sake, I pulled out some of the important takeaways for you to note:
Anyone can do it
If I can hit it big, trust me, so can you. Medium money is real and it’s there for those who want it.
Luck has something to do with it
Part of the 50/30/10/10 process (the last 10%) I credit to “luck”. The world is an uncertain place, and you can never entirely control what happens. This article is proof of that.
However, you can increase your chances of getting lucky
Increase your odds of getting lucky by increasing your output. The more you write and the more you publish, the greater your odds of success. You simply never know what’s going to go viral.
Current events articles are great, evergreen is better
There are two options when it comes to viral stories:
Write a controversial piece about current events. With this option, the goal is to earn as much money as possible with a huge upfront surge. However, expect earnings to drop off a cliff quickly. Write evergreen content, aka articles (like this one) that are always useful. These might take longer to “go viral”, but the longer-term payoff can be far greater.
Tell your story
On Medium, no one cares about cookie-cutter how-to articles, no matter how useful the information is. People are here for stories. Find a way to share yours.
Be relatable
People don’t relate well with robots. They relate to other humans with humanlike faults similar to themselves.
This was one of the most highlighted snippets from the article:
I found it to be a fun game running to the alarm and seeing how fast I could dive back into bed. My brain is weird sometimes.
People like what they can relate to.
Strike emotion
Many people are dying to unlock their inner superhero and live out their dreams. My article, though somewhat unintentional, struck that chord with my audience. People were motivated by my success and wanted to achieve their own.
Emotion is what gets people to engage with the article. To date, it has 366 responses, which is pretty bananas to me. I’ve tried my best to either clap or comment back to everyone who reached out.
Above all, just write
You honestly never know what’s going to happen, so write a lot, write often, be genuine, and share your unique stories — no matter how much you don’t think anyone will care.
Because trust me, they just might. | https://medium.com/illumination/the-secrets-behind-how-i-made-12-184-40-from-one-medium-article-2bc99e6f8a8e | ['Jason Gutierrez'] | 2020-11-10 14:54:35.583000+00:00 | ['Marketing', 'Motivation', 'Success', 'Inspiration', 'Writing'] | Title Secrets Behind Made 1218440 One Medium ArticleContent One cold early morning late January 2018 woke 530 let new puppy relieve business brewed pot coffee opened laptop start writing hour later written draft article would forever change view possible writing online evening spent another 10 15 minute editing article submitting Mission freebie behind Medium’s paywall world settled title “The 50301010 Rule Wake Earlier Work Dream” I’ll let stats speak Since debut article accumulated 900000 view 65000 clap grossed 12000 Never wildest dream would guessed article written offthecuff barely hour would end generating five figure earnings guess that’s go — article spend hour slaving flop one spend least time preparing make biggest splash want know secret behind article like hardly “expert” sleep writing Many newer writer think expert bestofthebest something anyone care success article prof that’s true Though I’m quite fantastic sleeping consistent writer trust I’m Matthew Walker Ernest Hemingway regular dude loved IDEA waking early struggled life consistently get bed morning I’m dude dream goal ambition perhaps like wasn’t seeing progress wanted evening shitshow one time day write — morning — spent oversleeping making zero progress couple month successfully waking early moon newfound habit wanted share random people internet didn’t make “expert” topic — guy figured key piece puzzle broke way made sense others anything article way systematize effort personal reminder actually important keep waking early getting stuff done story quite well first WAY BETTER later Mission proved worthwhile publication story article’s tenure racked something like 11000 clap nottooshabby one hundredish thousand view look graph see article pretty well first couple week publishing dropped hard 4 month SEO kicked Google started sending free traffic May 2018 Hellooo free traffic Prior search engine magic article averaged 2–3000 view per month Afterward surged 18–20000 Since free article Medium’s paid Partner Program still finding leg view point considered external May 2019 — sixteen month publishing — decided move article Better Humans publication one first pub specializing paywall didn’t hurt relationship editor Better Human’s help article quickly became Featured Medium’s Top Stories addition external Google traffic article getting boosted Medium’s internal algorithm received significant bump internal view point though still made 0 story remember published “free” story changed rather quickly Getting Paid look story’s stats you’ll notice there’s 3month difference article published Better Humans started earning money Partner Program still infancy Better Humans cut check month 100 earnings article Eventually migrated traditional Partner Program story made first 999 officially Medium September 9 2019 One month later vividly remember waking London away vacation scrolling email seeing notification Medium received payment “Oh cool wonder little amount made month” legitimately felt like someone Medium made mistake actually expected someone call asking money back one ever Shortly story exploded couple month later NBC emailed request interview article called mom immediately “Mom NBC asked interview ME” month Better Humans interviewed podcast — opportunity fell due audio quality end took interview Skype car lunch break work passing month thought “there’s way I’ll make much again” every month article continued thrive watched earnings pour Eventually income article tapered hundred dollar per month point figured luck finally fizzled boom spiked another couple month graced 4figure earnings one article barely took hour write time tell rest graph look like believe say I’m curious anyone see play Key Takeaways It’s hard state much learned experience like brevity’s sake pulled important takeaway note Anyone hit big trust Medium money real it’s want Luck something Part 50301010 process last 10 credit “luck” world uncertain place never entirely control happens article proof However increase chance getting lucky Increase odds getting lucky increasing output write publish greater odds success simply never know what’s going go viral Current event article great evergreen better two option come viral story Write controversial piece current event option goal earn much money possible huge upfront surge However expect earnings drop cliff quickly Write evergreen content aka article like one always useful might take longer “go viral” longerterm payoff far greater Tell story Medium one care cookiecutter howto article matter useful information People story Find way share relatable People don’t relate well robot relate human humanlike fault similar one highlighted snippet article found fun game running alarm seeing fast could dive back bed brain weird sometimes People like relate Strike emotion Many people dying unlock inner superhero live dream article though somewhat unintentional struck chord audience People motivated success wanted achieve Emotion get people engage article date 366 response pretty banana I’ve tried best either clap comment back everyone reached write honestly never know what’s going happen write lot write often genuine share unique story — matter much don’t think anyone care trust mightTags Marketing Motivation Success Inspiration Writing |
738 | 20 of the Best Free Online Tools for Writers | Once upon a time, the typewriter was the only piece of technology a writer had to make his work easier. Now we not only have computers, but we can also access an endless array of useful writing tools on the Internet. Best of all, many of these web applications are absolutely free!
But it takes time to hunt down these apps (time you could be spending on writing), so I’ve done the work for you and put together a list of my favorites.
Writing Tools
1. Google Docs
Google Docs is my go-to word processor when I’m not typing in Microsoft Word. I highly recommend it if you’re collaborating with another person on a writing project. You can access the document on separate computers and see each other’s changes in real time as you type in the document. Additionally, all of your changes are automatically saved as you type.
I have also found Google Docs one of the easiest programs to use to create PDF eBooks. There are a wide variety of templates to choose from. Check out Google Docs here.
2. Draft
This web-based word processor is an excellent alternative to Google Docs or Microsoft Word if you’re looking for a distraction free interface. Draft has many cool features (view them all here) including the ability to share your documents with other users and accept or decline their changes. And like Google Docs, your work is all backed up online.
I also love that Draft tracks how many words you write per day and will even send out a helpful email reminding you to meet your daily word count goal. Check out Draft here.
3. Reedsy Book Editor
The Reedsy Book Editor is a free, online alternative to software programs like Scrivener. Reedsy does not have as many features as Scrivener, but it lets you write (or import) your manuscript, format it, and then instantly typeset to EPUB and print-ready PDF files. Check out the Reedsy Book Editor here.
Editing Tools
4. Grammarly
Grammarly is a proofreading tool that helps you spot grammatical errors, typos, and awkward sentences. Download the web extension and Grammarly will correct anything written in a web browser (yes, even your Tweets and Facebook updates). Though I don’t rely on Grammarly alone for my proofreading needs, it is a fast and efficient tool that I highly recommend. Check out Grammarly here.
5. Hemingway Editor
The Hemingway Editor evaluates a piece of writing for clarity and simplicity. It calculates readability and highlights adverbs, passive voice, and dull, complicated words. This is an excellent web application for learning how to write effectively. Check out Hemingway Editor here.
6. Readability Score
This web application helps you improve your writing by measuring the readability of your text. A readability score tells you roughly what level of education someone would need in order to read your piece of text easily. Find out how easy your writing is to read. Check out Readability Score here.
7. Cliche Finder
In his six rules for writing, George Orwell advised, “Never use a metaphor, simile, or other figure of speech which you are used to seeing in print.” The Cliche Finder highlights cliches in your text so you can avoid overused expressions in your writing. Check out Cliche Finder here.
Productivity Tools
8. Todoist
Todoist is a task management web application that lets you create to-do lists with recurring dates and times. With Todoist, you can create a custom schedule for every single day. This is fantastic for reminding you of your daily writing goals. For example, you can schedule “writing every day at 8am” or you can create tasks that only recur on specific days, e.g., “research every Friday”. Check out Todoist here.
9. Trello
Trello is a fantastic web application for organizing and planning writing projects and working collaboratively. It works like an online bulletin board, allowing you to arrange all of your cards into columns. Those cards can be anything from tasks on a to-do list to scenes from your latest novel. Check out Trello here.
10. Marinara Timer
This productivity timer is one of the best apps out there if you’re a fan of the Pomodoro technique like I am. I wrote about the Pomodoro technique in the post below:
Marinara lets you use a traditional 25-minute timer followed by pre-programmed breaks or you can customize the time. The web application also has a handy timer history that records the exact times when you start and finish working, when you pause the timer, and when you take breaks. This is a helpful way to keep track of your hours and see how long projects take you. Check out Marinara Timer here.
11. Evernote
Evernote is my go-to app for creating quick to-do lists, jotting down notes, and writing up blog post ideas. It even lets you take audio notes! With the Evernote Web Clipper extension, you can clip articles from across the web and save them into Evernote for quick reference while working on projects. Best of all, Evernote instantly syncs across any computer or smartphone you use. Check out Evernote here.
12. Writing Calculator
Answer several questions and this online tool will help you estimate how many words per day you need to write to complete your writing project by a specific deadline. This is super helpful for when you’re planning out your writing project. If you know how many words per minute you write, it will even tell you how many hours until you reach your goal. Check out the writing calculator here.
13. Hipster Sound
If you like to have ambient noise in the background during your writing sessions, you’re going to love this website. Hipster Sound is an ambient noise generator that recreates the sounds of cafes. It helps you stay focused and improve your productivity, even on those days when you can’t make it to your local coffee shop. Check out Hipster Sound here.
14. Noisli
If you prefer nature sounds to cafe sounds while you write, then Noisli is for you. Noisli lets you mix and match a selection of different nature sounds to create your perfect writing environment. Check out Noisli here.
Graphic Design & Social Media Tools
15. Canva
Canva is a user-friendly graphic design software with a wide range of drag and drop templates. You can use Canva to create stunning graphics for Facebook and Pinterest or design a beautiful eBook and book covers. Check out Canva here.
16. Adazing Ebook 3D Cover Creator
The Adazing eBook Cover Creator lets you bring your book to life with realistic 3D mockups. You can choose from ten different templates. Check out the Adazing Ebook Cover Creator here. (Book Brush is a similar service and also has a free plan.)
17. Piktochart
Infographics are an effective way to repurpose your blog content and present your writing in a visually captivating format. Piktochart is an easy-to-use infographic maker with a library of professionally designed templates. Check out Piktochart here.
18. Lumen5
Lumen5 is an easy to use tool that will turn your blog posts into high quality videos you can share on social media. First, you enter the link to your blog post. Then, using Natural Language Processing algorithms, Lumen5 automatically creates a storyboard for you and suggests free stock photos and audio you can use. Check out Lumen5 here.
Blogging Tools
19. Answer the Public
Sometimes we just run out of ideas for what to blog about. Answer the Public is a free keyword research tool that shows the questions people are typing into search engines about your topic. It’s perfect for when you’re facing writer’s block. Check out AnswerthePublic here.
20. CoSchedule Headline Analyzer
The CoSchedule headline analyzer app evaluates how well your blog post’s headline will rank in search engines. It also scores how effectively your headline will result in social shares and click-throughs. The headline analyzer is a fantastic way to hone your copywriting skills and make sure your headlines are irresistible. Check out the Headline Analyzer here. | https://medium.com/copywriting-secrets/20-of-the-best-free-online-tools-for-writers-8a1ce9fb725b | ['Nicole Bianchi'] | 2019-09-07 14:52:38.481000+00:00 | ['Writing', 'Self Improvement', 'Creativity', 'Writing Tips', 'Productivity'] | Title 20 Best Free Online Tools WritersContent upon time typewriter piece technology writer make work easier computer also access endless array useful writing tool Internet Best many web application absolutely free take time hunt apps time could spending writing I’ve done work put together list favorite Writing Tools 1 Google Docs Google Docs goto word processor I’m typing Microsoft Word highly recommend you’re collaborating another person writing project access document separate computer see other’s change real time type document Additionally change automatically saved type also found Google Docs one easiest program use create PDF eBooks wide variety template choose Check Google Docs 2 Draft webbased word processor excellent alternative Google Docs Microsoft Word you’re looking distraction free interface Draft many cool feature view including ability share document user accept decline change like Google Docs work backed online also love Draft track many word write per day even send helpful email reminding meet daily word count goal Check Draft 3 Reedsy Book Editor Reedsy Book Editor free online alternative software program like Scrivener Reedsy many feature Scrivener let write import manuscript format instantly typeset EPUB printready PDF file Check Reedsy Book Editor Editing Tools 4 Grammarly Grammarly proofreading tool help spot grammatical error typo awkward sentence Download web extension Grammarly correct anything written web browser yes even Tweets Facebook update Though don’t rely Grammarly alone proofreading need fast efficient tool highly recommend Check Grammarly 5 Hemingway Editor Hemingway Editor evaluates piece writing clarity simplicity calculates readability highlight adverb passive voice dull complicated word excellent web application learning write effectively Check Hemingway Editor 6 Readability Score web application help improve writing measuring readability text readability score tell roughly level education someone would need order read piece text easily Find easy writing read Check Readability Score 7 Cliche Finder six rule writing George Orwell advised “Never use metaphor simile figure speech used seeing print” Cliche Finder highlight cliche text avoid overused expression writing Check Cliche Finder Productivity Tools 8 Todoist Todoist task management web application let create todo list recurring date time Todoist create custom schedule every single day fantastic reminding daily writing goal example schedule “writing every day 8am” create task recur specific day eg “research every Friday” Check Todoist 9 Trello Trello fantastic web application organizing planning writing project working collaboratively work like online bulletin board allowing arrange card column card anything task todo list scene latest novel Check Trello 10 Marinara Timer productivity timer one best apps you’re fan Pomodoro technique like wrote Pomodoro technique post Marinara let use traditional 25minute timer followed preprogrammed break customize time web application also handy timer history record exact time start finish working pause timer take break helpful way keep track hour see long project take Check Marinara Timer 11 Evernote Evernote goto app creating quick todo list jotting note writing blog post idea even let take audio note Evernote Web Clipper extension clip article across web save Evernote quick reference working project Best Evernote instantly syncs across computer smartphone use Check Evernote 12 Writing Calculator Answer several question online tool help estimate many word per day need write complete writing project specific deadline super helpful you’re planning writing project know many word per minute write even tell many hour reach goal Check writing calculator 13 Hipster Sound like ambient noise background writing session you’re going love website Hipster Sound ambient noise generator recreates sound cafe help stay focused improve productivity even day can’t make local coffee shop Check Hipster Sound 14 Noisli prefer nature sound cafe sound write Noisli Noisli let mix match selection different nature sound create perfect writing environment Check Noisli Graphic Design Social Media Tools 15 Canva Canva userfriendly graphic design software wide range drag drop template use Canva create stunning graphic Facebook Pinterest design beautiful eBook book cover Check Canva 16 Adazing Ebook 3D Cover Creator Adazing eBook Cover Creator let bring book life realistic 3D mockups choose ten different template Check Adazing Ebook Cover Creator Book Brush similar service also free plan 17 Piktochart Infographics effective way repurpose blog content present writing visually captivating format Piktochart easytouse infographic maker library professionally designed template Check Piktochart 18 Lumen5 Lumen5 easy use tool turn blog post high quality video share social medium First enter link blog post using Natural Language Processing algorithm Lumen5 automatically creates storyboard suggests free stock photo audio use Check Lumen5 Blogging Tools 19 Answer Public Sometimes run idea blog Answer Public free keyword research tool show question people typing search engine topic It’s perfect you’re facing writer’s block Check AnswerthePublic 20 CoSchedule Headline Analyzer CoSchedule headline analyzer app evaluates well blog post’s headline rank search engine also score effectively headline result social share clickthroughs headline analyzer fantastic way hone copywriting skill make sure headline irresistible Check Headline Analyzer hereTags Writing Self Improvement Creativity Writing Tips Productivity |
739 | Massive AI Breakthrough Could Mean a New COVID-19 Pre-Screening Method in the Cards | The Massachusetts Institute of Technology (MIT) has recently released information about a potential game-changer in screening for COVID-19 asymptomatic patients.
So far during this pandemic, scientists have identified that there are individuals who show symptoms, and those who are infected but show no symptoms at all. These pose a substantial danger given they feel no need to stay at home, and rightly so, and thus potentially contaminating whoever they are in contact with.
Researches at MIT, however, have developed an artificial intelligence model that can detect the virus in asymptomatic individuals. After being fed with new recordings, the model correctly identified 98.5% of coughs from people who had tested positive, including 100% of coughs from asymptomatic people who showed no symptoms but had tested positive for the virus.
But how does it work?
Some background on AI models without going too deep — an AI model is typically fed a large amount of the same type of data you want it to look out for.
Let’s say we’re training a model to differentiate between manager signatures. We would give our cognitive model (which may be looking out for a range of parameters such as colour, length, thickness, etc of a signature) a large volume of signature images. How large is large? As many signatures, as you can acquire (with permission from the managers of course) — think hundreds if not thousands. The more data your model has to work with, the stronger it will become at differentiating between those signatures.
Each signature is fed to the model along with a label containing the name of the manager whose signature it belongs to. Once you’re happy it has been trained with all of the available signatures, you would feed it completely new signatures that were not part of the training data. This is really important! You don’t want to test your model on a data set you have trained it with. It is also just as important to feed your model with images that don’t contain any signatures from those managers whose signatures you wish to identify in the future. Why? Because it's important for it to also learn to correctly say if none of the signatures it was expecting to find has been found.
Providing the training has gone well and it was given a large enough sample of data, the model will correctly differentiate between signatures.
This, of course, is a slightly different, and for sure oversimplistic, example compared to how the MIT model works. But you hopefully understand the concept behind AI models. Train them with enough of the same dataset type, test them with new data.
Back to COVID
The researchers at MIT tested their AI model on thousands of cough and dialog samples — allowing for a very rich training dataset.
Which parameters exactly was this model looking out for? This particular AI model has been around for some time — it was initially intended to detect signs of Alzheimer's through forced-coughs given the effects that the disease is known to have on vocal cords. Vocal cord strength, along with sentiment, lung and respiratory performance, and muscular degradation were effective in detecting the disease.
As the pandemic unfolded, researchers wondered whether the same model could be used to identify COVID-19 infected individuals given the effects the virus has on our muscles (with some individuals reporting temporary muscle impairment).
Researchers found that without modifying the model too much, it was able to correctly identify asymptomatic COVID-19 patients based on the same biomarkers (vocal cord strength, along with sentiment, lung and respiratory performance, and muscular degradation) that were used to detect Alzheimer’s in individuals.
What’s next?
Researchers are currently looking at developing a user-friendly app based on the model. If approved by officials, they hope it could act as a pre-screening app for the virus. We may be looking at a future in which a day out is dependent on results after coughing into your phone!
What a fantastic way to put our technological advancements to use. | https://medium.com/swlh/massive-ai-breakthrough-could-mean-a-new-covid-19-pre-screening-method-in-the-cards-45b0db8f8c0c | ['Natalie Mclaren'] | 2020-11-19 09:32:49.666000+00:00 | ['Technology', 'Artificial Intelligence', 'Science', 'Coronavirus', 'Covid 19'] | Title Massive AI Breakthrough Could Mean New COVID19 PreScreening Method CardsContent Massachusetts Institute Technology MIT recently released information potential gamechanger screening COVID19 asymptomatic patient far pandemic scientist identified individual show symptom infected show symptom pose substantial danger given feel need stay home rightly thus potentially contaminating whoever contact Researches MIT however developed artificial intelligence model detect virus asymptomatic individual fed new recording model correctly identified 985 cough people tested positive including 100 cough asymptomatic people showed symptom tested positive virus work background AI model without going deep — AI model typically fed large amount type data want look Let’s say we’re training model differentiate manager signature would give cognitive model may looking range parameter colour length thickness etc signature large volume signature image large large many signature acquire permission manager course — think hundred thousand data model work stronger become differentiating signature signature fed model along label containing name manager whose signature belongs you’re happy trained available signature would feed completely new signature part training data really important don’t want test model data set trained also important feed model image don’t contain signature manager whose signature wish identify future important also learn correctly say none signature expecting find found Providing training gone well given large enough sample data model correctly differentiate signature course slightly different sure oversimplistic example compared MIT model work hopefully understand concept behind AI model Train enough dataset type test new data Back COVID researcher MIT tested AI model thousand cough dialog sample — allowing rich training dataset parameter exactly model looking particular AI model around time — initially intended detect sign Alzheimers forcedcoughs given effect disease known vocal cord Vocal cord strength along sentiment lung respiratory performance muscular degradation effective detecting disease pandemic unfolded researcher wondered whether model could used identify COVID19 infected individual given effect virus muscle individual reporting temporary muscle impairment Researchers found without modifying model much able correctly identify asymptomatic COVID19 patient based biomarkers vocal cord strength along sentiment lung respiratory performance muscular degradation used detect Alzheimer’s individual What’s next Researchers currently looking developing userfriendly app based model approved official hope could act prescreening app virus may looking future day dependent result coughing phone fantastic way put technological advancement useTags Technology Artificial Intelligence Science Coronavirus Covid 19 |
740 | Image to pencil sketch with just 2 lines of code! | Advanced Features
You can change the scale of the image to darken the image. Note that on lighter scales, the greater texture is achieved. You can change the scale in the scale parameter of the normalsketch function. Note that the scale needs to be an integer value between 1 and 10.
To tweak the scale, simply add the parameter scale to the existing code:
sketch.normalsketch(‘pathtoimage’, ‘savepath’, ‘savename’, scale=x)
The default value for the scale is 10.
Tweaking the scale parameter allows you to change the image ‘darkness’ from extremely light to dark: | https://medium.com/analytics-vidhya/image-to-pencil-sketch-in-just-2-lines-of-code-292de57be483 | ['Rahul Prabhu'] | 2020-12-22 07:13:16.049000+00:00 | ['Python', 'Machine Learning', 'Image Processing', 'Art', 'Computer Vision'] | Title Image pencil sketch 2 line codeContent Advanced Features change scale image darken image Note lighter scale greater texture achieved change scale scale parameter normalsketch function Note scale need integer value 1 10 tweak scale simply add parameter scale existing code sketchnormalsketch‘pathtoimage’ ‘savepath’ ‘savename’ scalex default value scale 10 Tweaking scale parameter allows change image ‘darkness’ extremely light darkTags Python Machine Learning Image Processing Art Computer Vision |
741 | The Neighborhood of Empty Houses | Photos by Vincent Aliquo
I came to realize my neighbors were nothing but empty houses. Homes over an extended grassy lot piled high with a diverse collection of belongings — from furniture and books to anniversary gifts. It has been over three years since I found these homes, and only recently, in my quarantined state, has my curiosity grown. Encapsulated in a single photo, a wooden plaque with white, hand-painted letters reading “HAPPY ANNIVERSARY” remains a nagging mystery. Dated 1980, the center of the plaque held two horseshoes, each painted with a couple’s name. Beside it, inside a plastic sleeve, a card dedicated to them with a pressed — now unrecognizable — flower, and a four-leaf clover. The photo brought back a rush of memories of the surreal parcel of land I found it in.
The anniversary plaque.
I was staying at my uncle’s place over the spring of 2017. I had nowhere to go for spring break, and my uncle, who lived in North Carolina, welcomed me to his home. This was the second time I visited his residence; in the fall of 2016, my university forced me to evacuate the city because of Hurricane Matthew. At the time, my concern about the hurricane overpowered my curiosity about the neighboring houses.
With no hurricane to distract me, it was impossible to not think about the houses. Only a thin layer of new growth separated me from the first building; a brick barn which loomed ominously over the woods. Its sight pierced my confidence when I walked past. I frequented these woods with the hope to photograph the wildlife; waking up before five to watch the world come alive. The forest bed was blanketed in dry, hay-like vegetation and formed canopies from fallen branches. My uncle told me that these woods used to be an old farm, and there was evidence of this; the ground was littered with old farm tools, teapots, and small home utensils, which small trees sometimes pushed out of the earth — such as, unsettlingly enough, a large saw blade. Often, I saw chicken wire strangle the base of trees; one such wire held the skull of a decaying canine.
The homes that surrounded my uncle’s property made me increasingly anxious. I used one of my early mornings to investigate — at least the brick barn. I trod carefully across the new growth. Shafts of crepuscular rays split the cold mist, revealing the decrepitude of the barn. Indeed, it was unused. A rusted mechanical hoe, metal shards, and scraps of wood littered the perimeter of the building. On closer inspection, the barn’s brick appearance was just mimicry; a thick rubber siding, peeling off from water damage, revealed weather-worn wood. Dead vines hugged the building, while pine trees, at a minimum of four years old, found its home around the tin awning. Inside was a thick layer of dirt, pine, wooden planks, and troughs containing only dead birds.
The barn opened to a hazy chartreuse field; the dew-laden grass and the wings of thousands of insects glowed from the morning light. The light revealed the features of another building past the barn; a two-story home, embedded in the forest — its roof bearing multiple holes. The first floor, made of layered concrete, was largely intact, however, the rest of the home made of wood had little to no walls. The poor structural integrity, made evident by its worrying crepitation, made it a plausible scenario that venturing inside would bring me harm. I only briefly ascended to the second floor, however, the roosting hornets above the staircase triggered my immense fear of aggressive, stinging insects, and promptly left the building. | https://valiquo.medium.com/the-neighborhood-of-empty-houses-d7783095298c | ['Vincent Aliquo'] | 2020-11-04 18:56:18.424000+00:00 | ['Storytelling', 'Exploration', 'Abandoned', 'Nonfiction', 'Writing'] | Title Neighborhood Empty HousesContent Photos Vincent Aliquo came realize neighbor nothing empty house Homes extended grassy lot piled high diverse collection belonging — furniture book anniversary gift three year since found home recently quarantined state curiosity grown Encapsulated single photo wooden plaque white handpainted letter reading “HAPPY ANNIVERSARY” remains nagging mystery Dated 1980 center plaque held two horseshoe painted couple’s name Beside inside plastic sleeve card dedicated pressed — unrecognizable — flower fourleaf clover photo brought back rush memory surreal parcel land found anniversary plaque staying uncle’s place spring 2017 nowhere go spring break uncle lived North Carolina welcomed home second time visited residence fall 2016 university forced evacuate city Hurricane Matthew time concern hurricane overpowered curiosity neighboring house hurricane distract impossible think house thin layer new growth separated first building brick barn loomed ominously wood sight pierced confidence walked past frequented wood hope photograph wildlife waking five watch world come alive forest bed blanketed dry haylike vegetation formed canopy fallen branch uncle told wood used old farm evidence ground littered old farm tool teapot small home utensil small tree sometimes pushed earth — unsettlingly enough large saw blade Often saw chicken wire strangle base tree one wire held skull decaying canine home surrounded uncle’s property made increasingly anxious used one early morning investigate — least brick barn trod carefully across new growth Shafts crepuscular ray split cold mist revealing decrepitude barn Indeed unused rusted mechanical hoe metal shard scrap wood littered perimeter building closer inspection barn’s brick appearance mimicry thick rubber siding peeling water damage revealed weatherworn wood Dead vine hugged building pine tree minimum four year old found home around tin awning Inside thick layer dirt pine wooden plank trough containing dead bird barn opened hazy chartreuse field dewladen grass wing thousand insect glowed morning light light revealed feature another building past barn twostory home embedded forest — roof bearing multiple hole first floor made layered concrete largely intact however rest home made wood little wall poor structural integrity made evident worrying crepitation made plausible scenario venturing inside would bring harm briefly ascended second floor however roosting hornet staircase triggered immense fear aggressive stinging insect promptly left buildingTags Storytelling Exploration Abandoned Nonfiction Writing |
742 | Understanding and Using BLOBs in Your Application (Part I) | What is the first thing that comes to your mind when you hear the word ‘blob’? Without any context, it sure sounds like an obscure, mystic object you are playing against in a game of Minecraft. I can almost bet we have all seen the horrifying manifestation of our definition of BLOB on the web somewhere. However, in the world of data, blobs are extremely powerful and have transformed the way data is represented and transmitted.
Photo by Michael Dziedzic on Unsplash
Anything and everything that can represent information about an entity is data. Our name, address, cell-phone number, genetic-information, family-tree, all represent information about us, and hence all these fields can be called data. As we begin to think about using information collected from entities, we must also carefully think about representing this information accurately. Consider for example a mobile-app which stores the first name (data), last name (data)and phone number (data)of people (entity) I know. Every morning, this app would go through the list of people on it, and send them a greeting on their phone. The success of this app depends upon the fact that each data field contains what I expect it to contain. I am expecting first name and last name to be strings but I want the phone number to be an integer. What if my app was not able to store data correctly and some of the names were being stored as numbers instead of strings? It would break the logic and the functionality of the app. This is to give you a very basic idea of why data representation matters. As we scale-up into bigger applications with a complex architecture, data-efficiency is crucial and accurate data representation is the first step towards efficiency.
Where do blobs fit-in when we talk of data and representation? Before we get into that, let’s take a look at what a blob is exactly. Blob is an acronym for Binary Large Object. Restructuring the terms for a better understanding, a blob is nothing but a large binary object. Unlike an integer which represents numbers, or unlike a string which represents a sequence of characters, a blob is a stream of raw binary data which represents a “large” object (entity). We are able to understand what ‘Binary’ and ‘Large’ each mean individually. However, what do they mean in the context of an object?
Going back to talking of data: anything and everything can represent information about entities and therefore, anything and everything could be data. Text, numbers, images, audio — everything. Although, we do have ways of representing and working with numerical-data (data as integer, float), textual-data (data as characters and list of characters), but how do we represent images? Or audio files? We should be able to store them in a way which accurately represents each of these types, and can easily undergo further analysis.
Photo by Dan Farrell on Unsplash
Multimedia files (audio, video, gifs) are certainly produced in large quantities these days thanks to all the cool video streaming platforms and smart home-devices. Since our machines are not yet capable of understanding multimedia files as are, they need to be converted into a machine-understandable format before they can be processed and/or transmitted between systems. Stay with me, we are finally about to get to the end of understanding the fuss around blobs. What solution came to help to store huge “blobs” of data in the form of multimedia files? You guessed it! BLOBs!
So blobs represent multimedia files in a binary format. Representing and storing images, audio-files has become really easy with blobs. An application which allows users to upload their image and get a likability score on it or a simple speech-to-text service, both will be transferring data from the front-end (which the user interacts with) to the back-end (where the magic actually happens) in the form of a blob (or multiple blobs). In fact, for applications which use a constant streaming of data from front-end to the back-end (an interactive speech-to-text, for example), the audio is converted into not one but several blobs. These constituent blobs makes up the entire blob, which is nothing but the entire audio-file in binary form.
In the next part of this story, we will be looking at how blobs look like in real-applications, how blobs are transferred between front-end and back-end, and making use of the data that the blob represents. We will be looking at a web-app where the user will be able to record audio and get synonyms for nouns and adjectives in the speech. Here, data transfer between the front-end and the back-end happens through blobs. More fun things to come. Stay tuned folks! | https://medium.com/swlh/understanding-and-using-blobs-in-your-application-part-i-3b8a81806437 | ['Shruti Chaturvedi'] | 2020-11-25 07:15:02.747000+00:00 | ['Mobile App Development', 'Nodejs', 'React', 'Data', 'Web Apps'] | Title Understanding Using BLOBs Application Part IContent first thing come mind hear word ‘blob’ Without context sure sound like obscure mystic object playing game Minecraft almost bet seen horrifying manifestation definition BLOB web somewhere However world data blob extremely powerful transformed way data represented transmitted Photo Michael Dziedzic Unsplash Anything everything represent information entity data name address cellphone number geneticinformation familytree represent information u hence field called data begin think using information collected entity must also carefully think representing information accurately Consider example mobileapp store first name data last name dataand phone number dataof people entity know Every morning app would go list people send greeting phone success app depends upon fact data field contains expect contain expecting first name last name string want phone number integer app able store data correctly name stored number instead string would break logic functionality app give basic idea data representation matter scaleup bigger application complex architecture dataefficiency crucial accurate data representation first step towards efficiency blob fitin talk data representation get let’s take look blob exactly Blob acronym Binary Large Object Restructuring term better understanding blob nothing large binary object Unlike integer represents number unlike string represents sequence character blob stream raw binary data represents “large” object entity able understand ‘Binary’ ‘Large’ mean individually However mean context object Going back talking data anything everything represent information entity therefore anything everything could data Text number image audio — everything Although way representing working numericaldata data integer float textualdata data character list character represent image audio file able store way accurately represents type easily undergo analysis Photo Dan Farrell Unsplash Multimedia file audio video gifs certainly produced large quantity day thanks cool video streaming platform smart homedevices Since machine yet capable understanding multimedia file need converted machineunderstandable format processed andor transmitted system Stay finally get end understanding fuss around blob solution came help store huge “blobs” data form multimedia file guessed BLOBs blob represent multimedia file binary format Representing storing image audiofiles become really easy blob application allows user upload image get likability score simple speechtotext service transferring data frontend user interacts backend magic actually happens form blob multiple blob fact application use constant streaming data frontend backend interactive speechtotext example audio converted one several blob constituent blob make entire blob nothing entire audiofile binary form next part story looking blob look like realapplications blob transferred frontend backend making use data blob represents looking webapp user able record audio get synonym noun adjective speech data transfer frontend backend happens blob fun thing come Stay tuned folksTags Mobile App Development Nodejs React Data Web Apps |
743 | It’s Not My Fault You’re Angry | The just-world hypothesis states that if something bad happens to you, you must have deserved it. Conversely, that if you’re bad, you will one day get your comeuppance.
This leads to victim-blaming: “You were robbed? Well, what were you doing out late at night?”
“What did you do to upset him?”
In reality, the world isn’t fair. I find that people tend to transfer their anger from its source to an easy target. You see this when a boss yells at a worker, and the worker takes it without a word of complaint, then goes home fuming and yells at his or her spouse and children.
It’s interesting that people can usually control their anger when faced with an authority figure. How many people will yell at a boss, a doctor, or a police officer? Very few, because the consequences are swift. But they will certainly yell at a customer service worker, a secretary, or a teacher — people who are usually powerless to stop the verbal abuse.
This differs from an argument, in which two parties (often with similar levels of power) are locked in disagreement. Arguments happen, and the two people generally know why they’re at odds. If I caused someone’s problem, I can certainly fix it and apologize.
When I am on the receiving end of someone who walks in and flies off the handle, and then I get blamed for it, it’s not so easy.
What I almost find worse is when someone says or does something really horrible, on purpose, then tries to backtrack. They stomp their feet so hard that things fall off shelves. They raise their fists. Then when they start to see a consequence — someone getting angry back, or they get shown the door — they say a quick, “I’m sorry.”
As if that means you will forget everything. As if what they did never happened. They’re not really sorry; they’re sorry they might lose out on something. Threats and tantrums have lasting effects, and two words don’t erase that.
I read this example somewhere: Knock a vase on the ground, smashing it. Now say, “I’m sorry.” Did that fix the vase? Of course not. The damage has been done. | https://elliedaforge.medium.com/its-not-my-fault-you-re-angry-dfc3238ed193 | ['Ellie Daforge'] | 2018-11-18 15:07:51.266000+00:00 | ['Mental Health', 'Self', 'Wellness', 'Psychology', 'Work'] | Title It’s Fault You’re AngryContent justworld hypothesis state something bad happens must deserved Conversely you’re bad one day get comeuppance lead victimblaming “You robbed Well late night” “What upset him” reality world isn’t fair find people tend transfer anger source easy target see bos yell worker worker take without word complaint go home fuming yell spouse child It’s interesting people usually control anger faced authority figure many people yell bos doctor police officer consequence swift certainly yell customer service worker secretary teacher — people usually powerless stop verbal abuse differs argument two party often similar level power locked disagreement Arguments happen two people generally know they’re odds caused someone’s problem certainly fix apologize receiving end someone walk fly handle get blamed it’s easy almost find worse someone say something really horrible purpose try backtrack stomp foot hard thing fall shelf raise fist start see consequence — someone getting angry back get shown door — say quick “I’m sorry” mean forget everything never happened They’re really sorry they’re sorry might lose something Threats tantrum lasting effect two word don’t erase read example somewhere Knock vase ground smashing say “I’m sorry” fix vase course damage doneTags Mental Health Self Wellness Psychology Work |
744 | Debug Google Colab and Kaggle | Hopefully, I’ll save you from getting her headache. Photo by Aiony Haust on Unsplash.
Debug Google Colab and Kaggle
You’d imagine free public datasets would be painless
This wasn’t painless. But now that I’ve figured it out, I want to save you Google-ers out there some time.
What’s Google Colab?
Colab is a way to run Python Jupyter Notebooks on the Cloud, for free. It’s provided by Google and integrates with Google Drive. Very clean way to handle “but it doesn’t work on my computer!” issues.
See the “Resources” section at the end of this article for a link.
What’s Kaggle?
Kaggle is “Your Home for Data Science”. It hosts public datasets and competitions revolving around Data Science and Artificial Intelligence. It adds a sense of gamification to it as well. You can even share your profile to show others your involvement.
Before hitting my connection issues, I was aware of Kaggle but did not have an account yet.
After Signing Up For Kaggle
There are some tricky preconditions before you can download data.
Verify Account
You need to verify your account with a phone number that is not already used in Kaggle.
If your number is already used in Kaggle (previous owner — my mother, actually — signed up for Kaggle years ago), the Kaggle system will simply not text you a verification code, and will not tell you why. (This is probably a security feature).
If you hit a similar situation as me, the advice from Kaggle is to either use or delete the old account. The other alternative is to use someone else’s phone. Or get a new phone number.
Accept Rules
You need to accept the rules for the competition. If this errors out (the error message is pretty vague), then make sure your account is verified via phone number. See the previous section.
Kaggle API Key
With your new Kaggle account, go to “My Account”, find an API Key section and generate an API key. This will download a kaggle.json file. Open it, and you will see 2 values: a username and a key. Keep those handy.
The key value is sensitive, you need to treat it like a password. If you give someone a notebook that has this value printed out, quickly go to your “My Account” page and expire all your API Keys (same menu where you generated your API Key). Otherwise, someone can steal your Kaggle identity. At that point, your old API key will stop working, so you may want to generate a new one.
(You can put this file at `~/.kaggle/kaggle.json` and give it 600 permissions, however uploading that file every time you want to work with Kaggle in Colab would get old really fast. So, just follow what you see below)
Installing Kaggle on Google Colab
Add this Code cell for each Colab Notebook. It will install Kaggle CLI and prompt you for your credentials.
!pip install kaggle
!mkdir -p /root/.kaggle
!echo "Input contents of kaggle.json"
!read -s KAGGLE_JSON; echo "$KAGGLE_JSON" > /root/.kaggle/kaggle.json
# this is sensitive, so `-s` masks the input
!chmod 600 /root/.kaggle/kaggle.json
(The ! tells Colab to run this as a Bash command, not Python. It’s useful for installing dependencies, downloading datasets, and exploring directories.)
At this point, you can download the dataset. I’m using the 2013 dogs-vs-cats competition dataset, here’s the command I copied from the data page.
!kaggle competitions download -c dogs-vs-cats
If you get an error here (something like 401 or 403):
Double-check that your KAGGLE_USERNAME and KAGGLE_KEY is accurate. Make sure you accepted the rules to the competition. Make sure your account is verified over SMS.
Conclusion
Alright, you should now be able to download Kaggle data in Google Colab.
If you hit a different issue, or if you found this debugging resource useful, please write in to let us know!
Resources | https://medium.com/python-in-plain-english/debug-google-colab-and-kaggle-2691998f29b3 | ['James Fulford'] | 2020-07-15 02:49:37.973000+00:00 | ['Data Science', 'Python', 'Jupyter Notebook', 'Colab', 'Kaggle'] | Title Debug Google Colab KaggleContent Hopefully I’ll save getting headache Photo Aiony Haust Unsplash Debug Google Colab Kaggle You’d imagine free public datasets would painless wasn’t painless I’ve figured want save Googleers time What’s Google Colab Colab way run Python Jupyter Notebooks Cloud free It’s provided Google integrates Google Drive clean way handle “but doesn’t work computer” issue See “Resources” section end article link What’s Kaggle Kaggle “Your Home Data Science” host public datasets competition revolving around Data Science Artificial Intelligence add sense gamification well even share profile show others involvement hitting connection issue aware Kaggle account yet Signing Kaggle tricky precondition download data Verify Account need verify account phone number already used Kaggle number already used Kaggle previous owner — mother actually — signed Kaggle year ago Kaggle system simply text verification code tell probably security feature hit similar situation advice Kaggle either use delete old account alternative use someone else’s phone get new phone number Accept Rules need accept rule competition error error message pretty vague make sure account verified via phone number See previous section Kaggle API Key new Kaggle account go “My Account” find API Key section generate API key download kagglejson file Open see 2 value username key Keep handy key value sensitive need treat like password give someone notebook value printed quickly go “My Account” page expire API Keys menu generated API Key Otherwise someone steal Kaggle identity point old API key stop working may want generate new one put file kagglekagglejson give 600 permission however uploading file every time want work Kaggle Colab would get old really fast follow see Installing Kaggle Google Colab Add Code cell Colab Notebook install Kaggle CLI prompt credential pip install kaggle mkdir p rootkaggle echo Input content kagglejson read KAGGLEJSON echo KAGGLEJSON rootkagglekagglejson sensitive mask input chmod 600 rootkagglekagglejson tell Colab run Bash command Python It’s useful installing dependency downloading datasets exploring directory point download dataset I’m using 2013 dogsvscats competition dataset here’s command copied data page kaggle competition download c dogsvscats get error something like 401 403 Doublecheck KAGGLEUSERNAME KAGGLEKEY accurate Make sure accepted rule competition Make sure account verified SMS Conclusion Alright able download Kaggle data Google Colab hit different issue found debugging resource useful please write let u know ResourcesTags Data Science Python Jupyter Notebook Colab Kaggle |
745 | Could AI Appreciate Art? | Could an AI understand or appreciate art? Is it possible for a computer to form a critical opinion? Or have an aesthetic response? What does it “see” when it looks at a work of art? To explore these and other questions, we enlisted an AI (or, more precisely, several AI models) to co-produce this curatorial project.
Front page of the PDF version of A(artistic)x I(nteresting)x P(owerful).
Which is… what, exactly? A book? An exhibition? It doesn’t fit neatly into either category or any of the common intermediaries (artist book, exhibition catalog, etc.). Complicating matters, there are no images anywhere, either in the gallery or the PDF. It’s certainly a text, an embodiment of an experiment, and a proof of concept (more on that later); a version of this PDF’s contents were installed on the walls of the Pearlman Gallery at the Art Academy of Cincinnati in late 2020, so it’s also an exhibition. And if you have the means to print out this 340 page document and enough wall space to put it on, there’s no reason some variation of this exhibition couldn’t be mounted anywhere in the world.
The 29 works “appearing” in this project were submitted by students, faculty, and alumni of the Art Academy of Cincinnati. All were created sometime in 2020, and most were born out of the COVID-19 quarantine. This isn’t a show about COVID or lockdowns, but it is a relevant context. Consider technology’s role in mediating not only interactions between humans but also its role mediating human encounters with works of art. Live streaming performance and gallery walkthrough videos; thesis exhibitions on Instagram (supply your own brie and boxed wine) — while these all existed before in some form, they usually served more as an auxiliary form of access and documentation while the real event took place in the real world.
installation view of A(artistic)x I(nteresting)x P(owerful) in the Art Academy of Cincinnati’s Pearlman Gallery. Photos by Lindsey Whittle
But reality is slippery. An important role of documentation pre-COVID was to validate the reality of the event and the reality of the artists’ and curators’ efforts. If photos from the opening aren’t posted to Instagram, some would say it didn’t happen. And if an exhibition isn’t reviewed in a glossy art publication, then, on some level, it may as well not have occurred. Now, with in person gatherings curtailed, these less real modes are as close to real real as we can get (at least without creating dangerous superspreading events). In other words, at least for the moment, real reality is only accessible via the streaming lens. So what?
Consider: When you look at a work of art on your phone, what are you seeing? Put another way, how are you seeing?
Between the plate of glowing touch capacitative glass in your hand and the art object elsewhere are a host of applications that rely on various forms of artificial intelligence. When we talk about “AI” we don’t mean sci-fi fantasies like self-aware computers or super-intelligent robots yearning to be more like their carbon-based comrades. AI is already a ubiquitous part of our lives and has become an indispensable — and frequently invisible — layer in our media ecology.
Next time you point your smartphone’s camera at a plate of food, you might catch it making subtle adjustments. Maybe the greens in your salad grow deeper and greener, popping out from the softened middle distance, and the reds of that plank of salmon…well, you get the idea. Swing your phone around to take a selfie, and those parameters change, because the layers of software operating your camera knows the difference between a plate of arugula and your smiling face. (What’s really interesting to me is when we start to see our faces and plates of food that way even when the camera isn’t looking! But that would be a digression).
This “content-awareness” is AI at its most banal. Its ubiquity renders the technology invisible — or, to use Silicon Valley’s favorite euphemism: frictionless. This should give you pause. Because when we take the smartphone’s view of reality for objective, we cede much of our agency to the algorithms that exert influence on how we see and, consequently, how we desire.¹ | https://medium.com/swlh/could-ai-appreciate-works-of-art-part-1-36075a5d926 | ['Steve Kemple'] | 2020-12-08 06:36:03.091000+00:00 | ['AI', 'NLP', 'Computer Vision', 'Artificial Intelligence', 'Art'] | Title Could AI Appreciate ArtContent Could AI understand appreciate art possible computer form critical opinion aesthetic response “see” look work art explore question enlisted AI precisely several AI model coproduce curatorial project Front page PDF version Aartisticx Interestingx Powerful is… exactly book exhibition doesn’t fit neatly either category common intermediary artist book exhibition catalog etc Complicating matter image anywhere either gallery PDF It’s certainly text embodiment experiment proof concept later version PDF’s content installed wall Pearlman Gallery Art Academy Cincinnati late 2020 it’s also exhibition mean print 340 page document enough wall space put there’s reason variation exhibition couldn’t mounted anywhere world 29 work “appearing” project submitted student faculty alumnus Art Academy Cincinnati created sometime 2020 born COVID19 quarantine isn’t show COVID lockdown relevant context Consider technology’s role mediating interaction human also role mediating human encounter work art Live streaming performance gallery walkthrough video thesis exhibition Instagram supply brie boxed wine — existed form usually served auxiliary form access documentation real event took place real world installation view Aartisticx Interestingx Powerful Art Academy Cincinnati’s Pearlman Gallery Photos Lindsey Whittle reality slippery important role documentation preCOVID validate reality event reality artists’ curators’ effort photo opening aren’t posted Instagram would say didn’t happen exhibition isn’t reviewed glossy art publication level may well occurred person gathering curtailed le real mode close real real get least without creating dangerous superspreading event word least moment real reality accessible via streaming lens Consider look work art phone seeing Put another way seeing plate glowing touch capacitative glass hand art object elsewhere host application rely various form artificial intelligence talk “AI” don’t mean scifi fantasy like selfaware computer superintelligent robot yearning like carbonbased comrade AI already ubiquitous part life become indispensable — frequently invisible — layer medium ecology Next time point smartphone’s camera plate food might catch making subtle adjustment Maybe green salad grow deeper greener popping softened middle distance red plank salmon…well get idea Swing phone around take selfie parameter change layer software operating camera know difference plate arugula smiling face What’s really interesting start see face plate food way even camera isn’t looking would digression “contentawareness” AI banal ubiquity render technology invisible — use Silicon Valley’s favorite euphemism frictionless give pause take smartphone’s view reality objective cede much agency algorithm exert influence see consequently desire¹Tags AI NLP Computer Vision Artificial Intelligence Art |
746 | Threading the Needle: How to Save Nature and Humanity without Sacrificing Either Part II | Saving nature without sacrificing modern life is the preeminent challenge of our time. It is a complicated problem that must be attacked simultaneously from multiple angles. Failure to act on one angle will invalidate efforts on other angles.
This problem must be addressed in two distinct phases. First, we must stop living in a manner that actively harms both ourselves and the natural world. Then, we must learn how to create a world where both nature and humanity thrive. This two-part article will explore how we can reorganize our civilization to be compatible with such a vision.
Part I can be found here.
Part II: Integration and Prosperity
In Part I, we discussed how to stop harming the natural world, but what would it look like for humanity and nature to thrive together? This is a question whose answer requires a radical restructuring of how, where, and “when” we live.
Reintegrating Humanity into the Landscape:
Reintegrating humanity into natural systems and landscapes will require different approaches based on population density and locality. Reintegration of rural regions would center on rewilding and conservation. Reintegration of degraded lands or sparse-to-moderately populated regions could focus on careful and intentional land management and resource extraction practices for human benefit. Reintegration of densely populated regions would require intentional city planning and novel approaches to urban design.
Rural Lands:
Rewilding is a large-scale conservation strategy that restores and protects core wilderness areas, provides connectivity between these areas, and protects or reintroduces apex predators and other keystone species. The ultimate goal of rewilding efforts is to restore ecosystems to a point where they are passively managed with near pre-human levels of biodiversity.
An example of this is the reintroduction of wolves to Yellowstone National Park, where after 70 years of their extirpation, they were reintroduced to the park. The reintroduction of wolves helped to stabilize the ecosystem by ending the boom and bust cycles that had plagued the elk population for decades. The wolves also contributed to the elk herds becoming more resilient to variable climate patterns, due to the wolves’ specific pattern of predation.
As another example of rewilding contributing to ecosystem resilience, rewilding can result in feedback loops that reinforce ecosystem integrity. An example of this could be restoration of oyster reefs. Increased levels of atmospheric carbon dioxide has led to lower oceanic pH, known as ocean acidification. This has corrosive effects on calcareous organisms, such as oysters.
However, if oysters are raised in sufficient quantities, they can serve as a local pH buffer, stabilizing local pH and diminishing the effects of ocean acidification. This then leads to the encouragement of more oyster growth, which then further stabilizes local pH. As a positive side effect, large amounts of carbon could be sequestered in the shells of restored oyster reefs.
Rewilding has enormous potential to sequester vast amounts of carbon in soils and living systems, potentially offsetting the worst impacts of climate change. Despite the increasing need for us to sequester atmospheric carbon and its inclusion in climate projections, the only way we know how to sequester carbon at scale is in living systems.
Additionally, as global biodiversity losses compound, it has become increasingly apparent that our protected lands and waters are by no means enough to halt catastrophic levels of species extinctions. Rather, now is the time for bold and decisive action that will put to rest fears of mass extinction.
Reducing the footprint of agriculture, eliminating ranching, and shifting the remaining footprint of agriculture to be within urban regions would free an enormous amount of land. It must first be considered though that much of this land has been heavily degraded by decades of intensive agriculture and livestock grazing.
To maximize the effectiveness of available resources, an inventory should be taken to determine the best candidate regions for rehabilitation. This inventory could look at potential for carbon sequestration, biodiversity conservation, and restoration.
Restoring these degraded lands would be challenging but could also be the most rewarding project humanity undertakes in the 21st century. Amid concerns of automation displacing workers, restoring hundreds of millions of acres of wildlands globally would require a workforce like no other.
And unlike many jobs today, the work to restore nature on an unprecedented scale could very well imbue a strong and deep sense of purpose. Restoration of the world’s wildlands would also require collaboration between nations on a scale never seen before, bringing the global community together in a new and profound way.
Perhaps the most important collaboration in this undertaking would occur between scientists and indigenous peoples. Many indigenous peoples possess a deep knowledge of sustainable land management practices localized to the regions they occupied. These land management practices were disrupted when European colonists stole their land.
Coupled with modern science, we could reinstate some of these land management practices, and come to better understandings of how to maximize ecosystem services for carbon sequestration, conservation of biodiversity, and climate resilience.
Taking this collaboration a step further, the US could honor past treaties with First Nations peoples to the extent that it is possible. First Nations peoples have generally proven themselves as far better stewards of land than their European counterparts. From a cultural perspective, it is just as important to heal and restore our human landscape as the natural landscape. In an increasingly volatile world, it is our collective resolve that will give us the best chance of navigating the turbulent times ahead.
Degraded and Moderately Populated Regions:
Humans will always require a steady flow of resources for civilization to function. This will not change, even under a circular economy in which all waste streams have been eliminated and all possible materials are recycled back into production chains. Rather, to minimize harm, we must practice careful and intentional land management and resource extraction. Degraded and sparsely-to-moderately populated lands would be best suited for such purposes.
Degraded lands that would be difficult to restore could be considered prime candidates for resource extraction sites, concentrating the environmental impact of heavy industry such as mining. Modern mining practices can further reduce environmental impact through such practices as reprocessing tailings to extract additional minerals, dewatering tailings and reusing the water, and dry-stacking tailings to improve long-term stability of storage.
Rare-earth elements are essential for many sources of renewable energy and electronics. Mining them is generally regarded as harmful because their tailings contain high concentrations of radioactive elements such as thorium and uranium. We can mitigate this hazard by processing their tailings into nuclear fuel for modern reactors. Such reactors produce minimal waste and have much lower associated risks. This would provide a new source of low-carbon energy while reducing the hazardous waste associated with manufacturing electronics and renewables.
While bulk caloric food production would be moved into urban centers, the surrounding areas could host food forests and edible permaculture landscapes. This could provide complementing locally and seasonally appropriate foods to what can be grown within urban vertical farms. Permaculture, food forests, and edible landscaping would be best suited for areas of moderate population density, turning surrounding population centers into lush harvestable gardens and parks.
Silviculture and forestry are other examples of intentional land management requisite for modern life. Modern forestry practices can balance a variety of needs, including conservation and wildlife habitat, timber, and recreation. Silviculture could be practiced on human-caused secondary forests and similarly degraded landscapes. Remaining stands of old-growth forest have high conservation value, and thus should not be viewed as a potential resource for forestry.
Urban Corridors:
Reintegrating cities into the natural world would at first seem contradictory. When most people think of nature, they think of it as a foil to civilization of which cities are the hallmark example. This is a grave mistake, as the two are inseparable and always will be. A better framing would view cities as a human extension of the landscape they occupy.
The issue we have now is that cities are not seen as an extension of the landscape, but rather exist at odds with it. We use materials that have no natural analogues. We impose structures on the land rather than allow the land to dictate what forms we build. We pretend that the natural world on which we build our cities does not exist or, worse, is an obstacle to be overcome.
We do not have to live in opposition to nature. We can draw inspiration from it instead.
If humanity is to reintegrate its urban landscape with natural cycles, it is not enough to simply close nutrient cycles and source raw material from waste rather than extraction. It must restore ecosystem functions and integrate greenspace into all structures. Structures must be built as though they are geologically fixed to the landscape and function as an extension of it.
Adding or integrating greenspace into all extant and planned structures has utility beyond aesthetic purposes. It is well documented that green space has a cooling effect that can counteract the urban heat bubble. Greenspace also plays a valuable role in managing stormwater and improving air quality. Green roofs can act as strong insulators, keeping warmer air in during the winter and out during the summer. And of course, they can be quite visually appealing.
As cities are most often vertical extensions of the landscape, they can look to cliffs and bluffs for inspiration on integrating greenspace. To mitigate engineering challenges and hazards, flora could be tiered as one ascends in elevation. Large canopy trees would remain on the ground, smaller trees could provide canopy on low to moderately tall buildings, and shrubs and forbs could be used on the tallest buildings. As cliffs often possess ledges with vegetation, we could model our buildings to similarly possess ledges with vegetation.
If integrating vegetation into the built environment is intended to be more than just aesthetically pleasing, stormwater management would be a necessity. To handle this, building designs could mimic natural topography with intermittent, cascading streams designed into buildings. This would be an excellent way to direct water within the built environment, while also retaining the functionality of natural systems of water catchment. Alongside this, waterways throughout the urban environment could be restored, and serve as receivers for stormwater runoff.
Incorporating ecosystem functionality into buildings is only one way that we could make our cities greener. Building for density makes services such as wastewater treatment, recycling, and composting easier. A high-rise is far more efficient per unit than a rural house.
Additionally, buildings that incorporate passive design elements use less energy than buildings that strictly use active design for heating and cooling. New buildings could implement these strategies, while older buildings could be retrofitted to improve insulation and energy efficiency.
At the city scale, making cities more walkable and providing public transit options can reduce emissions from vehicular traffic. To deeply cut emissions, it would seem necessary to eliminate vehicular traffic entirely from urban centers. This would come with the exception of a handful of emergency service vehicles alongside buses and streetcars. A network of electrified light rail, streetcars, and buses could be used to solve the transportation problems that remain.
The space that would be freed up from the expulsion of vehicles from urban-centers could turn streets into walkable gardens and forests full of native and edible plants. Streams that were once paved over could be “daylighted” and allow for at least partial restoration of their ecosystem services. Cities would become quieter and healthier due to improved air quality. Streets would become safer, and businesses would attract greater foot traffic.
Ultimately, far from an austere life devoid of modern comforts, a “green” future can be as rich as one is willing to imagine, just not necessarily in a strictly material sense.
Taking the Long View:
Over long enough timescales, cities will have to adapt to higher and lower sea levels. Glaciers will advance and recede. Landscapes change across centuries to millennia and as landscapes change, cities must adapt with them.
Each city could have a special council that reflects on questions of sustainability over centuries. Our myths and our stories ought to reflect such long-term thinking as well.
It seems inevitable that humanity will go to colonize the planets, moons, and stars. The question I ask you now is, “what would you like for us to take with us?”
For me? I would love to see humanity take baskets of earth with us, wherever we go. And for us to realize that civilization does not have to end for nature to go on, or nature to end for civilization to go on.
We are a young species, and the great project of human civilization has only just begun. We now have the tools to write our own story, so let’s have some fun.
Closing Thoughts:
This brings to a conclusion the series I have worked on over the past year. I have greatly enjoyed sharing my thoughts with you, and hope that you got as much from it as I have. Perhaps, together, we can steer this careening ship called civilization into a softer landing and preserve all that is important, without letting go of our humanity. | https://medium.com/climate-conscious/threading-the-needle-how-to-save-nature-and-humanity-without-sacrificing-either-part-ii-5806053448c7 | ['Charles Whitaker'] | 2020-11-12 21:58:21.427000+00:00 | ['Sustainability', 'Climate Change', 'Society', 'Future', 'Culture'] | Title Threading Needle Save Nature Humanity without Sacrificing Either Part IIContent Saving nature without sacrificing modern life preeminent challenge time complicated problem must attacked simultaneously multiple angle Failure act one angle invalidate effort angle problem must addressed two distinct phase First must stop living manner actively harm natural world must learn create world nature humanity thrive twopart article explore reorganize civilization compatible vision Part found Part II Integration Prosperity Part discussed stop harming natural world would look like humanity nature thrive together question whose answer requires radical restructuring “when” live Reintegrating Humanity Landscape Reintegrating humanity natural system landscape require different approach based population density locality Reintegration rural region would center rewilding conservation Reintegration degraded land sparsetomoderately populated region could focus careful intentional land management resource extraction practice human benefit Reintegration densely populated region would require intentional city planning novel approach urban design Rural Lands Rewilding largescale conservation strategy restores protects core wilderness area provides connectivity area protects reintroduces apex predator keystone specie ultimate goal rewilding effort restore ecosystem point passively managed near prehuman level biodiversity example reintroduction wolf Yellowstone National Park 70 year extirpation reintroduced park reintroduction wolf helped stabilize ecosystem ending boom bust cycle plagued elk population decade wolf also contributed elk herd becoming resilient variable climate pattern due wolves’ specific pattern predation another example rewilding contributing ecosystem resilience rewilding result feedback loop reinforce ecosystem integrity example could restoration oyster reef Increased level atmospheric carbon dioxide led lower oceanic pH known ocean acidification corrosive effect calcareous organism oyster However oyster raised sufficient quantity serve local pH buffer stabilizing local pH diminishing effect ocean acidification lead encouragement oyster growth stabilizes local pH positive side effect large amount carbon could sequestered shell restored oyster reef Rewilding enormous potential sequester vast amount carbon soil living system potentially offsetting worst impact climate change Despite increasing need u sequester atmospheric carbon inclusion climate projection way know sequester carbon scale living system Additionally global biodiversity loss compound become increasingly apparent protected land water mean enough halt catastrophic level specie extinction Rather time bold decisive action put rest fear mass extinction Reducing footprint agriculture eliminating ranching shifting remaining footprint agriculture within urban region would free enormous amount land must first considered though much land heavily degraded decade intensive agriculture livestock grazing maximize effectiveness available resource inventory taken determine best candidate region rehabilitation inventory could look potential carbon sequestration biodiversity conservation restoration Restoring degraded land would challenging could also rewarding project humanity undertakes 21st century Amid concern automation displacing worker restoring hundred million acre wildlands globally would require workforce like unlike many job today work restore nature unprecedented scale could well imbue strong deep sense purpose Restoration world’s wildlands would also require collaboration nation scale never seen bringing global community together new profound way Perhaps important collaboration undertaking would occur scientist indigenous people Many indigenous people posse deep knowledge sustainable land management practice localized region occupied land management practice disrupted European colonist stole land Coupled modern science could reinstate land management practice come better understanding maximize ecosystem service carbon sequestration conservation biodiversity climate resilience Taking collaboration step US could honor past treaty First Nations people extent possible First Nations people generally proven far better steward land European counterpart cultural perspective important heal restore human landscape natural landscape increasingly volatile world collective resolve give u best chance navigating turbulent time ahead Degraded Moderately Populated Regions Humans always require steady flow resource civilization function change even circular economy waste stream eliminated possible material recycled back production chain Rather minimize harm must practice careful intentional land management resource extraction Degraded sparselytomoderately populated land would best suited purpose Degraded land would difficult restore could considered prime candidate resource extraction site concentrating environmental impact heavy industry mining Modern mining practice reduce environmental impact practice reprocessing tailing extract additional mineral dewatering tailing reusing water drystacking tailing improve longterm stability storage Rareearth element essential many source renewable energy electronics Mining generally regarded harmful tailing contain high concentration radioactive element thorium uranium mitigate hazard processing tailing nuclear fuel modern reactor reactor produce minimal waste much lower associated risk would provide new source lowcarbon energy reducing hazardous waste associated manufacturing electronics renewables bulk caloric food production would moved urban center surrounding area could host food forest edible permaculture landscape could provide complementing locally seasonally appropriate food grown within urban vertical farm Permaculture food forest edible landscaping would best suited area moderate population density turning surrounding population center lush harvestable garden park Silviculture forestry example intentional land management requisite modern life Modern forestry practice balance variety need including conservation wildlife habitat timber recreation Silviculture could practiced humancaused secondary forest similarly degraded landscape Remaining stand oldgrowth forest high conservation value thus viewed potential resource forestry Urban Corridors Reintegrating city natural world would first seem contradictory people think nature think foil civilization city hallmark example grave mistake two inseparable always better framing would view city human extension landscape occupy issue city seen extension landscape rather exist odds use material natural analogue impose structure land rather allow land dictate form build pretend natural world build city exist worse obstacle overcome live opposition nature draw inspiration instead humanity reintegrate urban landscape natural cycle enough simply close nutrient cycle source raw material waste rather extraction must restore ecosystem function integrate greenspace structure Structures must built though geologically fixed landscape function extension Adding integrating greenspace extant planned structure utility beyond aesthetic purpose well documented green space cooling effect counteract urban heat bubble Greenspace also play valuable role managing stormwater improving air quality Green roof act strong insulator keeping warmer air winter summer course quite visually appealing city often vertical extension landscape look cliff bluff inspiration integrating greenspace mitigate engineering challenge hazard flora could tiered one ascends elevation Large canopy tree would remain ground smaller tree could provide canopy low moderately tall building shrub forbs could used tallest building cliff often posse ledge vegetation could model building similarly posse ledge vegetation integrating vegetation built environment intended aesthetically pleasing stormwater management would necessity handle building design could mimic natural topography intermittent cascading stream designed building would excellent way direct water within built environment also retaining functionality natural system water catchment Alongside waterway throughout urban environment could restored serve receiver stormwater runoff Incorporating ecosystem functionality building one way could make city greener Building density make service wastewater treatment recycling composting easier highrise far efficient per unit rural house Additionally building incorporate passive design element use le energy building strictly use active design heating cooling New building could implement strategy older building could retrofitted improve insulation energy efficiency city scale making city walkable providing public transit option reduce emission vehicular traffic deeply cut emission would seem necessary eliminate vehicular traffic entirely urban center would come exception handful emergency service vehicle alongside bus streetcar network electrified light rail streetcar bus could used solve transportation problem remain space would freed expulsion vehicle urbancenters could turn street walkable garden forest full native edible plant Streams paved could “daylighted” allow least partial restoration ecosystem service Cities would become quieter healthier due improved air quality Streets would become safer business would attract greater foot traffic Ultimately far austere life devoid modern comfort “green” future rich one willing imagine necessarily strictly material sense Taking Long View long enough timescales city adapt higher lower sea level Glaciers advance recede Landscapes change across century millennium landscape change city must adapt city could special council reflects question sustainability century myth story ought reflect longterm thinking well seems inevitable humanity go colonize planet moon star question ask “what would like u take us” would love see humanity take basket earth u wherever go u realize civilization end nature go nature end civilization go young specie great project human civilization begun tool write story let’s fun Closing Thoughts brings conclusion series worked past year greatly enjoyed sharing thought hope got much Perhaps together steer careening ship called civilization softer landing preserve important without letting go humanityTags Sustainability Climate Change Society Future Culture |
747 | Create a String Reversal Microservice on AWS Lambda Using Java | Server-side code
Now comes the part where we actually create our Lambda function using Java. Go the AWS Console and search for Lambda. Create a lambda function and name it whatever you want and use any Java runtime. I used Java 11 for this. Click into your newly created function and note down the ARN number on the top-right. This number uniquely identifies AWS resources such as Lambdas. Also, note down the region you are currently in as resources are specific to regions. Mine is in us-east-2.
To begin the server-side coding, create a new project (yes, we do this twice) and configure Maven again. Lambda functions handle requests. And these requests require specific handlers to be imported via AWS libraries. Therefore following these instructions, configure the aws-lambda-java-core dependency in your pom.xml file. This step is optional, but it is good practice to log information to AWS Cloud Watch — we can use gson library to help us with logging and you can declare this dependency by following this link.
Once again, to keep things simple, we will have only one class — the Handler — that will handle this request. This will reverse the string payload and return the reversed result. Following the official docs that explains how to handle requests, we arrive at the following:
service-side code that IS the Lambda function
This code sample should also be easy to follow. We define a private method down below that reverses a string. The handle request method does some logging (which is optional, but recommended), gets the payload (which is represented by the “event” parameter), calls the reverse string function; re-constructs the payload and returns the result/response.
With that our server-side code is complete. We now have to package-up this code into a JAR file and upload it to our Lambda within AWS. Make sure you’ve added the final set of Maven dependencies in your pom file, which allow you to build your code into a JAR format. When the dependencies are added:
run “mvn package” in your project directory. If the command executed successfully, you should get “BUILD SUCCESS” in the output. You can find the JAR file in your target folder within your project.
Now that we have our JAR, go to your Lambda function in AWS console and upload the file by going to: Function code -> Actions -> Upload a .zip or .jar file.
After your JAR file has uploaded successfully, set your function handler in Basic Settings -> Edit -> Handler. The first part of your handler will be your package and class name and the second part will be the method that handles the event. If you are following my code example, this was for me: “StringReversalLambdaHandler::handleRequest”. | https://medium.com/swlh/create-a-string-reversal-microservice-on-aws-lambda-using-java-5104f61400b5 | ['Akshay Kumar'] | 2020-07-21 16:26:59.036000+00:00 | ['AWS', 'Programming', 'Java', 'AWS Lambda', 'Microservices'] | Title Create String Reversal Microservice AWS Lambda Using JavaContent Serverside code come part actually create Lambda function using Java Go AWS Console search Lambda Create lambda function name whatever want use Java runtime used Java 11 Click newly created function note ARN number topright number uniquely identifies AWS resource Lambdas Also note region currently resource specific region Mine useast2 begin serverside coding create new project yes twice configure Maven Lambda function handle request request require specific handler imported via AWS library Therefore following instruction configure awslambdajavacore dependency pomxml file step optional good practice log information AWS Cloud Watch — use gson library help u logging declare dependency following link keep thing simple one class — Handler — handle request reverse string payload return reversed result Following official doc explains handle request arrive following serviceside code Lambda function code sample also easy follow define private method revers string handle request method logging optional recommended get payload represented “event” parameter call reverse string function reconstructs payload return resultresponse serverside code complete packageup code JAR file upload Lambda within AWS Make sure you’ve added final set Maven dependency pom file allow build code JAR format dependency added run “mvn package” project directory command executed successfully get “BUILD SUCCESS” output find JAR file target folder within project JAR go Lambda function AWS console upload file going Function code Actions Upload zip jar file JAR file uploaded successfully set function handler Basic Settings Edit Handler first part handler package class name second part method handle event following code example “StringReversalLambdaHandlerhandleRequest”Tags AWS Programming Java AWS Lambda Microservices |
748 | This Is How The Brain Multiplies Problems! | PSYCHOLOGY
This Is How The Brain Multiplies Problems!
Oh, Brain Oh! How Complicated You Are!
Photo by Tim Doerfler on Unsplash
The brain multiplies the problems as a result of a cognitive bias that leads to installation and development patterns that are then generalized or over-applied for no real reason.
This is why we may feel that the difficulties are piling up when they are not.
Many people complain that “problems don’t just happen” or that “once problems start, they never end”.
In a way, they are right, but this is probably not due to a fatality, but to a cognitive bias that causes the brain to multiply the problems.
This conclusion was drawn from various experiments.
The brain multiplies the problems because, although being a fabulous and complex organ, it also has its limits.
Such limitations arise from the fact that the mind tends to create patterns for everything, as it is a way to save energy.
But these general guidelines are misleading.
Knowing that the brain multiplies problems allows us to be critical and attentive to the functioning of thought, to put a limit where it is necessary.
Otherwise, the difficulties and the emotional expenses that they entail spread more than necessary.
We can’t solve problems by thinking the same way we did when we created them. ~ Albert Einstein
The brain multiplies the problems | https://medium.com/the-innovation/this-is-how-the-brain-multiplies-problems-838239434fd | ['Munawwar Goonoo'] | 2020-11-11 16:31:20.642000+00:00 | ['Psychology', 'Brain', 'Neuroscience', 'Problems'] | Title Brain Multiplies ProblemsContent PSYCHOLOGY Brain Multiplies Problems Oh Brain Oh Complicated Photo Tim Doerfler Unsplash brain multiplies problem result cognitive bias lead installation development pattern generalized overapplied real reason may feel difficulty piling Many people complain “problems don’t happen” “once problem start never end” way right probably due fatality cognitive bias cause brain multiply problem conclusion drawn various experiment brain multiplies problem although fabulous complex organ also limit limitation arise fact mind tends create pattern everything way save energy general guideline misleading Knowing brain multiplies problem allows u critical attentive functioning thought put limit necessary Otherwise difficulty emotional expense entail spread necessary can’t solve problem thinking way created Albert Einstein brain multiplies problemsTags Psychology Brain Neuroscience Problems |
749 | The Case For Eating a 90% Plant-Based Diet | The Case For Eating a 90% Plant-Based Diet
Don’t let conflicting advice and binary labels get in the way of making healthy changes
Photo by Chantal Garnier on Unsplash
What are we really supposed to eat?
Whatever you want the answer to be, you will find an expert advocating for it. Whether it’s Paleo, Keto, Vegan, High-Carb, or Low-Carb — there is supporting evidence for everything.
Outside of diets for weight loss, there is also no consensus on the ideal diet for life. And by diet for life, I mean a diet that takes into account not only health, but also our values, convenience, feasibility, sustainability, and so on.
At nutrition school, I am currently learning over 100 dietary theories, and each has its own pros and cons. What works for us is highly individual. Yet, the ambition to generalize remains, because clarifies and simplifies things.
In “Food Rules” Michael Pollan tries exactly that — to generalize and simplify. He summarizes his findings in a simple phrase: “Eat food, not too much, mostly plants.” There is a lot of wisdom in this sentence. By food, he means real rather than processed food. Not too much refers to healthy portion control. And plants — well that’s pretty self-explanatory.
So, in line with Pollan’s findings, I want to make the case for a primarily whole food, plant-based diet.
I’m not advocating for strict veganism, but rather want to introduce a more inclusive concept: the [90%] plant-based diet.
A plant-based lifestyle is much less restrictive than veganism, which not only excludes eating all animal foods but also wearing all leather and other animal products. Plant-based, however, is more flexible and looks different for each person: some are almost vegan but will still eat honey, others will have the occasional egg or organic, wild-caught fish, or even high-quality meat.
I believe the binary perception of veganism by most vegans and non-vegans does more harm than good. It certainly created a lot of confusion and back-and-forth for me for a long time. I was drawn to veganism for what it stood for, but sometimes found it hard to manage the restrictive nature of the diet in my everyday life. And at times, after eating vegan for a while, I would feel drawn to incorporate a little egg or fish on occasion. When I honored the craving, I would usually notice an improvement in how I felt — even if subtle.
To get the benefits from plant-based nutrition we don’t need to be “all-in or nothing”. Depending on our intention, we can be plant-based and still enjoy animal products on occasion, whether it’s because we otherwise find it too restrictive or because our body tells us it needs more than just plants to thrive. As with most things, quality over quantity is key.
So, what are the key benefits?
The three main reasons to eat plant-based: health, environment, and ethics
Argument #1: It can improve your overall health and performance, and prevent (and reverse) chronic disease
Countless studies are beginning to show some of the health benefits of plant-based diets. When you eat fewer animal products, you’re automatically inclined to eat more plants (because, what else would you eat?). Most people end up discovering an abundance of plant foods they didn’t eat before, so they naturally diversify their nutrient intake. Plants are nature’s medicine and good for your body and mental health, you won’t find anyone denying that.
Some interesting findings that extend beyond this common wisdom:
a. It can prevent (and even reverse) lifestyle-borne chronic diseases such as heart disease, cancer and type 2-diabetes.
b. It can improve your overall health by stabilizing your weight, increasing your energy, reducing systematic inflammation, and promoting longevity.
c. It promotes a healthy weight, individuals following a plant-based diet have a lower BMI and are less likely to be overweight.
d. It can increase your athletic performance by improving blood thickness, which helps get more oxygen to your muscles, which in turn improves performance. And by increasing our intake of antioxidants, which are highly prevalent in plant-based diets, we reduce free radicals that curb muscle fatigue, thus improving recovery.
e. It may increase our sexual health as demonstrated in Netflix’s recent documentary The Game Changers by improving blood flow and serotonin levels (more research to be done here though).
Argument #2: It’s the most powerful way to practice conscious consumerism and protect the environment
The effects of animal agriculture are disastrous and one of the best-kept secrets in our modern economy. Most people know that taking a plane is bad for the environment, but not too many are aware that the burger they are eating wastes more water than two months of showering.
So, how exactly does eating plant-based help the environment?
a. It reduces greenhouse gas emissions. Animal agriculture is responsible for 18% of GHG emissions, more than all transportation combined. Plus, a cow’s methane emissions are 25x more potent than carbon dioxide.
b. It conserves land and water usage. Livestock and livestock feed occupy 1/3 of the earth’s ice-free land and are responsible for 80–90% of US water consumption. It takes 18x more land to feed a meat-eater than a vegan.
c. It combats species extinction, ocean dead zones, water pollution, and habitat destruction (see Cowspiracy’s research for details).
Argument #3: It promotes animal welfare (and, if you believe in it, your karma)
This argument is less scientific but rather ethical/spiritual.
Some thoughts:
a. Current practices of factory farming are inhumane. In the US, 99% of farm animals are factory-farmed. Animals are grown unnaturally fast and unnaturally fat. As a result, they easily get sick, which is masked by an excessive use of antibiotics. Whether it’s animal slaughtering or the fact that cows in dairy farms are artificially inseminated for the majority of their life to produce milk, or that male chickens are slaughtered in egg farms because they don’t yield eggs — there are plenty of reason’s to avoid animal products out of compassion for animals.
b. As all religions and spiritual traditions teach: treat others how you want to be treated. Whether that translates into the karmic laws in Buddhism, the concept of non-violence (Ahimsa) in Yoga or Jesus’ famous words “do unto others as you would have them do unto you”. Even if you don’t believe in karma or religion: What gives us the right to domesticate another species? We are all one universal organism. We share 80% of our DNA with cows.
c. In a scientific sense, everything is energy, as proven by quantum physics. If we believe this energy also carries spirit (or however you’d want to call it), we’re facing an additional dilemma: By eating animal products, one could say that we eat the energy of dead beings. If we talk about a chicken who had a wonderful life in a lush farm, this might not be too disturbing. But most of the time it’s more likely that the animal you’re eating was treated terribly. And you are exposing your body-mind-soul to that energy.
The 90% [or 80%, 87%, 92%] rule is your ticket to feasibility & sustainability
So now you may be convinced that eating plant-based really is the way to go.
Your next thought will probably be: “But I could never not eat cheese again.”
And I’m telling you, you don’t have to give up anything to make a difference.
We want food to be pleasurable and joyful. If we restrict for the wrong reasons, it’s always a temporary fix. Restrictive diets don’t work, UCLA (and many other researchers) found that most dieters gain back the weight they lost. You’ll certainly find people in the vegan community who are using strict veganism as a form of controlling their diet, and that’s often not healthy. This is not what we’re going for here with the more intuitive and flexible mostly-plant-based eating approach.
If we want to make plant-based nutrition sustainable, we can’t neglect the psychological aspects of eating. We’ve all heard the saying “one unhealthy meal doesn’t make you fat, just like one healthy meal doesn’t make you healthy.” I’m suggesting we apply the same concept to veganism. One plant-based meal doesn’t make you a vegan, but one non-vegan meal also does not make you no longer a plant-based eater.
I get it, people like labels.
But what people like even more is freedom.
Give yourself that freedom! You can make a tremendous difference in your health, the environment, and protect your morals by eating this way, most of the time. And you can still have cheese, your favorite ice cream, or your grandmother’s home-cooked stew, on occasion.
It really depends on your intention. If animal welfare is on top of the list, you might decide to only eat eggs from your own chicken. If it’s mostly for the environment, you might be inclined to incorporate products from regenerative farms. As for health, there are certainly proven health benefits to some high-quality animal foods like wild salmon and eggs.
From my experience, once you start eating more plants, your body will also crave more plants naturally. When you start out it’s important that you don’t go into it with black-or-white thinking. We don’t want you to throw out all good intentions and dismiss the possibility of you ever being plant-based the first time you cave and have a croissant.
The most important piece of advice is to always listen to your body. I’m a firm believer in intuitive eating and leveraging our bodies’ cues as the guiding force to our nutrition. But I also believe values and genuine intention can play their role in reacting to cravings, as long as we don’t feel like we’re jeopardizing our health by denying ourselves nutrients the body is clearly asking for.
The 90% is of course illustrative, hence the brackets. Eating should be flexible, not rigid. Whether 80%, 85%, or 98% is what works for you — the number doesn’t matter as much as the concept and the intention behind it.
How to transition to a (primarily) plant-based diet
1. Transition gradually
If you try to go from yogurt-loving chicken fanatic to 100% plant-based from one day to another, you’re making your life harder than it needs to be.
Some people manage to do it this way, but for most a gradual approach is much more effective (adapted from James Clear’s Atomic Habits):
Step 1: Start by cutting out animals that live on four feet (cow, pig, etc.)
Step 2: Cut out animals that live on two feet (chicken, turkey)
Step 3: Cut out any other animal flesh (seafood)
Step 4: Cut out and replace any other animal products (dairy, cheese, eggs)
And to reiterate, when I say “stop” I mean “stop [90]% of the time”.
2. Be kind with yourself
Don’t beat yourself up if you’re having a hard time or are not following through as much as you want. Change is already hard as it is, judgment will only make it 10x harder.
Instead, approach every opportunity you don’t stick with it (even if you planned to) with curiosity: Why did you not manage to eat as intended, was it too inconvenient or not available? Then come better prepared next time. Did you feel like you didn’t want to miss out on something? Then find better alternatives.
Investigate and learn.
3. Remember your intentions
When you question yourself, always bring it back to why you do what you are doing. In fact, I suggest explicitly clarifying your intentions for yourself before you start out.
I’ve given you many good reasons to eat more plant-based above — find what resonates with you and hone in on those arguments, leave out what doesn’t feel relevant to you.
4. Listen to your body
Your body will let you know how it’s feeling about the changes you are making. Many people report having more energy and feeling more vital after only a few weeks of eating primarily plant-based.
Others may find their body craves animal products.
When you have a very specific craving, always listen to it. If you crave eggs, have eggs. There’s a reason your body is asking for it, and unless it’s a craving for a chemically-manipulated food you can assume that it’s a biological craving that your incredibly intelligent body is sending in response to a lack of nutrients.
5. Find your sweet spot
As mentioned, 90% is illustrative. Find a balance that works for you. Some weeks it maybe 99%, other weeks it may only be 80%. Look at it over the months rather than days and weeks.
Some models that work for people: Eat plant-based during the week and incorporate non-vegan foods on the weekend. Only cook plant-based at home, but have animal products when you eat out. Have a certain number of meals per day or per week be plant-based. And so on.
On a final note, if you’ve been keen on making these changes for a while — now is the time. Many of us are cooking as much as never before, which makes it so much more feasible.
There are endless options — go out explore and play with it! | https://medium.com/the-road-to-wellness/the-case-for-eating-90-plant-based-924b2c8b31e5 | ['Julia Blum'] | 2020-08-11 00:39:20.613000+00:00 | ['Sustainability', 'Wellness', 'Food', 'Diet', 'Health'] | Title Case Eating 90 PlantBased DietContent Case Eating 90 PlantBased Diet Don’t let conflicting advice binary label get way making healthy change Photo Chantal Garnier Unsplash really supposed eat Whatever want answer find expert advocating Whether it’s Paleo Keto Vegan HighCarb LowCarb — supporting evidence everything Outside diet weight loss also consensus ideal diet life diet life mean diet take account health also value convenience feasibility sustainability nutrition school currently learning 100 dietary theory pro con work u highly individual Yet ambition generalize remains clarifies simplifies thing “Food Rules” Michael Pollan try exactly — generalize simplify summarizes finding simple phrase “Eat food much mostly plants” lot wisdom sentence food mean real rather processed food much refers healthy portion control plant — well that’s pretty selfexplanatory line Pollan’s finding want make case primarily whole food plantbased diet I’m advocating strict veganism rather want introduce inclusive concept 90 plantbased diet plantbased lifestyle much le restrictive veganism excludes eating animal food also wearing leather animal product Plantbased however flexible look different person almost vegan still eat honey others occasional egg organic wildcaught fish even highquality meat believe binary perception veganism vegan nonvegans harm good certainly created lot confusion backandforth long time drawn veganism stood sometimes found hard manage restrictive nature diet everyday life time eating vegan would feel drawn incorporate little egg fish occasion honored craving would usually notice improvement felt — even subtle get benefit plantbased nutrition don’t need “allin nothing” Depending intention plantbased still enjoy animal product occasion whether it’s otherwise find restrictive body tell u need plant thrive thing quality quantity key key benefit three main reason eat plantbased health environment ethic Argument 1 improve overall health performance prevent reverse chronic disease Countless study beginning show health benefit plantbased diet eat fewer animal product you’re automatically inclined eat plant else would eat people end discovering abundance plant food didn’t eat naturally diversify nutrient intake Plants nature’s medicine good body mental health won’t find anyone denying interesting finding extend beyond common wisdom prevent even reverse lifestyleborne chronic disease heart disease cancer type 2diabetes b improve overall health stabilizing weight increasing energy reducing systematic inflammation promoting longevity c promotes healthy weight individual following plantbased diet lower BMI le likely overweight increase athletic performance improving blood thickness help get oxygen muscle turn improves performance increasing intake antioxidant highly prevalent plantbased diet reduce free radical curb muscle fatigue thus improving recovery e may increase sexual health demonstrated Netflix’s recent documentary Game Changers improving blood flow serotonin level research done though Argument 2 It’s powerful way practice conscious consumerism protect environment effect animal agriculture disastrous one bestkept secret modern economy people know taking plane bad environment many aware burger eating waste water two month showering exactly eating plantbased help environment reduces greenhouse gas emission Animal agriculture responsible 18 GHG emission transportation combined Plus cow’s methane emission 25x potent carbon dioxide b conserve land water usage Livestock livestock feed occupy 13 earth’s icefree land responsible 80–90 US water consumption take 18x land feed meateater vegan c combat specie extinction ocean dead zone water pollution habitat destruction see Cowspiracy’s research detail Argument 3 promotes animal welfare believe karma argument le scientific rather ethicalspiritual thought Current practice factory farming inhumane US 99 farm animal factoryfarmed Animals grown unnaturally fast unnaturally fat result easily get sick masked excessive use antibiotic Whether it’s animal slaughtering fact cow dairy farm artificially inseminated majority life produce milk male chicken slaughtered egg farm don’t yield egg — plenty reason’s avoid animal product compassion animal b religion spiritual tradition teach treat others want treated Whether translates karmic law Buddhism concept nonviolence Ahimsa Yoga Jesus’ famous word “do unto others would unto you” Even don’t believe karma religion give u right domesticate another specie one universal organism share 80 DNA cow c scientific sense everything energy proven quantum physic believe energy also carry spirit however you’d want call we’re facing additional dilemma eating animal product one could say eat energy dead being talk chicken wonderful life lush farm might disturbing time it’s likely animal you’re eating treated terribly exposing bodymindsoul energy 90 80 87 92 rule ticket feasibility sustainability may convinced eating plantbased really way go next thought probably “But could never eat cheese again” I’m telling don’t give anything make difference want food pleasurable joyful restrict wrong reason it’s always temporary fix Restrictive diet don’t work UCLA many researcher found dieter gain back weight lost You’ll certainly find people vegan community using strict veganism form controlling diet that’s often healthy we’re going intuitive flexible mostlyplantbased eating approach want make plantbased nutrition sustainable can’t neglect psychological aspect eating We’ve heard saying “one unhealthy meal doesn’t make fat like one healthy meal doesn’t make healthy” I’m suggesting apply concept veganism One plantbased meal doesn’t make vegan one nonvegan meal also make longer plantbased eater get people like label people like even freedom Give freedom make tremendous difference health environment protect moral eating way time still cheese favorite ice cream grandmother’s homecooked stew occasion really depends intention animal welfare top list might decide eat egg chicken it’s mostly environment might inclined incorporate product regenerative farm health certainly proven health benefit highquality animal food like wild salmon egg experience start eating plant body also crave plant naturally start it’s important don’t go blackorwhite thinking don’t want throw good intention dismiss possibility ever plantbased first time cave croissant important piece advice always listen body I’m firm believer intuitive eating leveraging bodies’ cue guiding force nutrition also believe value genuine intention play role reacting craving long don’t feel like we’re jeopardizing health denying nutrient body clearly asking 90 course illustrative hence bracket Eating flexible rigid Whether 80 85 98 work — number doesn’t matter much concept intention behind transition primarily plantbased diet 1 Transition gradually try go yogurtloving chicken fanatic 100 plantbased one day another you’re making life harder need people manage way gradual approach much effective adapted James Clear’s Atomic Habits Step 1 Start cutting animal live four foot cow pig etc Step 2 Cut animal live two foot chicken turkey Step 3 Cut animal flesh seafood Step 4 Cut replace animal product dairy cheese egg reiterate say “stop” mean “stop 90 time” 2 kind Don’t beat you’re hard time following much want Change already hard judgment make 10x harder Instead approach every opportunity don’t stick even planned curiosity manage eat intended inconvenient available come better prepared next time feel like didn’t want miss something find better alternative Investigate learn 3 Remember intention question always bring back fact suggest explicitly clarifying intention start I’ve given many good reason eat plantbased — find resonates hone argument leave doesn’t feel relevant 4 Listen body body let know it’s feeling change making Many people report energy feeling vital week eating primarily plantbased Others may find body craves animal product specific craving always listen crave egg egg There’s reason body asking unless it’s craving chemicallymanipulated food assume it’s biological craving incredibly intelligent body sending response lack nutrient 5 Find sweet spot mentioned 90 illustrative Find balance work week maybe 99 week may 80 Look month rather day week model work people Eat plantbased week incorporate nonvegan food weekend cook plantbased home animal product eat certain number meal per day per week plantbased final note you’ve keen making change — time Many u cooking much never make much feasible endless option — go explore play itTags Sustainability Wellness Food Diet Health |
750 | Blocked | Sometimes it just grows.
Like a seedling, forcing its way up
through the crust of the earth.
Birthed from that great brown womb,
secret and strange.
Those mysterious inner depths.
Blooming, baptized by life’s storms.
I remember the pen dancing,
a living extension of my hand.
The words flowing like water,
a tapped spring, onto paper.
Every word a drop of
the unconquered, wild ocean.
I remember the dance
of mind and hand, wedded
in free, spontaneous motion
Furious rhythm and rhyme
streaming from the tip of my pen.
Why is it that now
each word is squeezed out
to trickle and fall on arid sand?
Time stumbles on as I sit
hearing no divine music,
just the beat of my pen,
tapping on empty lines. | https://medium.com/ninja-writers/blocked-8c75bf6a959c | ['Sarah Lynn Terzo'] | 2020-10-18 18:38:18.377000+00:00 | ['Composing', 'Writing', 'Poetry', 'Creativity', 'Music'] | Title BlockedContent Sometimes grows Like seedling forcing way crust earth Birthed great brown womb secret strange mysterious inner depth Blooming baptized life’s storm remember pen dancing living extension hand word flowing like water tapped spring onto paper Every word drop unconquered wild ocean remember dance mind hand wedded free spontaneous motion Furious rhythm rhyme streaming tip pen word squeezed trickle fall arid sand Time stumble sit hearing divine music beat pen tapping empty linesTags Composing Writing Poetry Creativity Music |
751 | Will AI Help Us Solve The Hard Problem of Consciousness? | The question about the emergence of consciousness is perhaps the most important question humanity should attempt to answer.
Consciousness and its contents are at the root of everything. Consciousness is what is responsible for all of the greatest artifacts of culture that humanity has created: art, music, science, philosophy, technology.
Every child, adolescent, and adult ought to ask themselves: what is consciousness? Why does it exist? What is it like to be human? Why is it that a complex organization of unconscious matter and particles in certain corners of the universe gives rise to consciousness? As Annaka Harris puts it in her book “Conscious”, what causes the lights to “switch on” for a certain combination of particles, but not for others?
The Infamous Nagel Thought Experiment
Why is a human being considered to be conscious, whereas a rock is unequivocally unconscious?
Arguably the best thought experiment conducted on the nature of consciousness was proposed by the philosopher Thomas Nagel, who encourages readers to contemplate over whether or not it is something to be like a bat. Nagel’s proposition may appear trivial, but demands careful thought: if being a bat consists of having some experience of reality, whether it be sensations, feelings, vividness, self-awareness, or experiences of vision, then it is something to be like that bat.
Of course, the answer to this particular question remains ambiguous: for we, as sentient human beings, cannot say for certain that a bat is having a conscious experience in the way that we are. But Nagel’s point is that if we were to trade places with a bat, and were to have some experience of consciousness, however vague, then bats are conscious. If upon trading places with a bat, annihilation were the outcome, then bats are unconscious. The point Nagel is trying to make is that the difference between consciousness and unconsciousness is a matter of subjective experience.
The Hard Problem of Consciousness
Many scientists and philosophers have attempted to map conscious activity to brain activity — something that we can today study using brain imaging technologies that illustrate certain parts of the brain responding to external stimuli.
But we must be wary here to not mistake consciousness itself for other mental states such as confusion, anger, self-awareness, awakeness, or attention. While all such experiences point to the contents of consciousness and the links between certain stimuli and brain activity, they do not answer the question of how or why consciousness emerges in the first place. Consciousness exists prior to all of these states. In other words, recognizing that one is self-aware, and backing this fact up with brain imaging technologies, does not help solve the mystery of why the feeling of being self-aware exists at all.
In the scientific world, this mystery is now famously known as the “hard problem of consciousness” — an explanatory gap in the scientific realm that makes it difficult for us to connect why the arrangement of certain particles in the universe gives rise to conscious experience. In the process of answering these questions, neuroscientists have undoubtedly unraveled many fascinating truths about the human mind. We can also now also safely say that consciousness appears to lie on a spectrum — some complex beings, like us, are more conscious than others.
Is The Singularity Near?
In the debate over how technology and artificial intelligence will evolve in the coming years, many great thinkers wonder whether or not we will reach the singularity.
Ray Kurzweil defines the Singularity as a rapid increase in artificial intelligence writing that “The Singularity will allow us to transcend these limitations of our biological bodies and brains. There will be no distinction, post-Singularity, between human and machine”
So the relevant question we should attempt to answer is: will the evolution of AI eventually give rise to conscious machines? And if so, will this help us resolve the hard problem of consciousness?
If we apply the same mechanisms as we currently do to understand the nature of consciousness (i.e., through studying brain activity), then we will never really progress on answering this question. For instance, a mature and evolved artificial intelligence machine that looks, walks, and talks like a human being, could easily depict all of the external behaviors that we ascribe to human consciousness. This machine could be perfectly capable of duping us into believing it is conscious when in reality it’s nothing but the result of certain information processing systems that we have designed to depict certain outcomes. In other words, because of the mechanisms that we may use to create it, a robot might indicate that its lights are “on”, when in reality they are not.
Consciousness Is More Than Observable Behavior
If this is a confusing thought experiment, let’s consider the mysterious notion of “locked-in syndrome” or “anesthetic awareness” that some unfortunate patients feel when they wake up during surgery, unbeknownst to their surgeons. The experience of being conscious while having your body cut into is undoubtedly traumatizing: but scientifically, it points to the fact that it is possible to be conscious without having the ability to indicate that one is conscious. A patient experiencing anesthetic awareness is witness to the experience of her surgery without being able to exhibit any external or observable behaviors to indicate as such.
With this, we can safely say that just because a system (whether robot or chimpanzee) behaves a certain way, does not tell us whether or not it is conscious in the way that you or I are. This gives ammunition to the idea that consciousness is a spectrum and is ultimately a question about the subjective experience that concerns the very existence of subjectivity. In the scientific world, this experience of subjectivity is aptly labeled “qualia”, which is defined as individual instances of subjective, conscious experience. Qualia is the individual experience that I have when I observe the blueness of the sky, or that you have when you enjoy the sweetness of a berry. Though you and I can agree that the sky is blue, I cannot ascertain that your experience of its blueness is the same as mine.
The Turing Test As a Tool
The rise of conscious machines, or seemingly conscious machines, will undoubtedly have a lot of significance in helping us answer the mystery of consciousness. At the very least, the emergence of intelligent machines will cause a philosophical upheaval in which an increasing number of people will wonder about the hard problem of consciousness and therefore seek answers for it. In a future where machines exhibit human-like intelligence and consciousness, more people are bound to question the very nature of their own consciousness.
The infamous Turing Test, now well known in any discussion about the evolution of AI, is a method of inquiry for determining whether or not a computer is capable of thinking like a human being. Many exquisite works of science fiction within our era have explored this concept, whether directly or indirectly. Most famously, in movies like “Her” and “Ex Machina”, the protagonists use tools to determine whether or not the machine that they are interacting with is having a conscious experience in the way that we are.
“Ex Machina” offers perhaps the most jarring depiction of this — in which a machine exhibits human-like qualities that one would undoubtedly mistake for consciousness, such as manipulation, acting in self-interest, and scheming. While watching Ex Machina, one can easily imagine what it is like to be the AI that we are witness to. In fact, an audience might even feel deep empathy toward the AI. Indeed, it is the rise of such self-interested artificial intelligence that the Musks of the world are afraid of, and it’s the rise of such machines that therefore invoke images of a robo-calypse.
How carefully we should deal with and prepare for the evolution of AI is a separate issue. But the hope is that in trying to figure out whether or not we can create general AI — in other words, create consciousness — we might come one step closer to figuring out the emergence of consciousness.
Ultimately, if a robot looks, talks, and walks like a human being, we can only be as certain of its consciousness as I am about yours.
This is a good enough assessment: but in the process, we need to create more sophisticated versions of the Turing Test if we truly want to get to the heart of the debate: what in this universe causes the lights to switch on in some parts of it while they remain off in others? | https://tjajal.medium.com/will-ai-help-us-solve-the-hard-problem-of-consciousness-7a9013f5e052 | ['Tannya D. Jajal'] | 2020-09-01 12:29:33.340000+00:00 | ['Neuroscience', 'Consciousness', 'Philosophy', 'Artificial Intelligence', 'Future'] | Title AI Help Us Solve Hard Problem ConsciousnessContent question emergence consciousness perhaps important question humanity attempt answer Consciousness content root everything Consciousness responsible greatest artifact culture humanity created art music science philosophy technology Every child adolescent adult ought ask consciousness exist like human complex organization unconscious matter particle certain corner universe give rise consciousness Annaka Harris put book “Conscious” cause light “switch on” certain combination particle others Infamous Nagel Thought Experiment human considered conscious whereas rock unequivocally unconscious Arguably best thought experiment conducted nature consciousness proposed philosopher Thomas Nagel encourages reader contemplate whether something like bat Nagel’s proposition may appear trivial demand careful thought bat consists experience reality whether sensation feeling vividness selfawareness experience vision something like bat course answer particular question remains ambiguous sentient human being cannot say certain bat conscious experience way Nagel’s point trade place bat experience consciousness however vague bat conscious upon trading place bat annihilation outcome bat unconscious point Nagel trying make difference consciousness unconsciousness matter subjective experience Hard Problem Consciousness Many scientist philosopher attempted map conscious activity brain activity — something today study using brain imaging technology illustrate certain part brain responding external stimulus must wary mistake consciousness mental state confusion anger selfawareness awakeness attention experience point content consciousness link certain stimulus brain activity answer question consciousness emerges first place Consciousness exists prior state word recognizing one selfaware backing fact brain imaging technology help solve mystery feeling selfaware exists scientific world mystery famously known “hard problem consciousness” — explanatory gap scientific realm make difficult u connect arrangement certain particle universe give rise conscious experience process answering question neuroscientist undoubtedly unraveled many fascinating truth human mind also also safely say consciousness appears lie spectrum — complex being like u conscious others Singularity Near debate technology artificial intelligence evolve coming year many great thinker wonder whether reach singularity Ray Kurzweil defines Singularity rapid increase artificial intelligence writing “The Singularity allow u transcend limitation biological body brain distinction postSingularity human machine” relevant question attempt answer evolution AI eventually give rise conscious machine help u resolve hard problem consciousness apply mechanism currently understand nature consciousness ie studying brain activity never really progress answering question instance mature evolved artificial intelligence machine look walk talk like human could easily depict external behavior ascribe human consciousness machine could perfectly capable duping u believing conscious reality it’s nothing result certain information processing system designed depict certain outcome word mechanism may use create robot might indicate light “on” reality Consciousness Observable Behavior confusing thought experiment let’s consider mysterious notion “lockedin syndrome” “anesthetic awareness” unfortunate patient feel wake surgery unbeknownst surgeon experience conscious body cut undoubtedly traumatizing scientifically point fact possible conscious without ability indicate one conscious patient experiencing anesthetic awareness witness experience surgery without able exhibit external observable behavior indicate safely say system whether robot chimpanzee behaves certain way tell u whether conscious way give ammunition idea consciousness spectrum ultimately question subjective experience concern existence subjectivity scientific world experience subjectivity aptly labeled “qualia” defined individual instance subjective conscious experience Qualia individual experience observe blueness sky enjoy sweetness berry Though agree sky blue cannot ascertain experience blueness mine Turing Test Tool rise conscious machine seemingly conscious machine undoubtedly lot significance helping u answer mystery consciousness least emergence intelligent machine cause philosophical upheaval increasing number people wonder hard problem consciousness therefore seek answer future machine exhibit humanlike intelligence consciousness people bound question nature consciousness infamous Turing Test well known discussion evolution AI method inquiry determining whether computer capable thinking like human Many exquisite work science fiction within era explored concept whether directly indirectly famously movie like “Her” “Ex Machina” protagonist use tool determine whether machine interacting conscious experience way “Ex Machina” offer perhaps jarring depiction — machine exhibit humanlike quality one would undoubtedly mistake consciousness manipulation acting selfinterest scheming watching Ex Machina one easily imagine like AI witness fact audience might even feel deep empathy toward AI Indeed rise selfinterested artificial intelligence Musks world afraid it’s rise machine therefore invoke image robocalypse carefully deal prepare evolution AI separate issue hope trying figure whether create general AI — word create consciousness — might come one step closer figuring emergence consciousness Ultimately robot look talk walk like human certain consciousness good enough assessment process need create sophisticated version Turing Test truly want get heart debate universe cause light switch part remain othersTags Neuroscience Consciousness Philosophy Artificial Intelligence Future |
752 | Managing AMIs using CloudFormation | One of the long-time missing pieces of CloudFormation is AMIs (Amazon Linux Images). Creating custom AMIs allows you to launch new EC2 instances pre-configured to your liking (operating system, installed packages, hardened security, etc).
Thankfully, here at Poka, we have developed a solution that allows you to easily create AMIs as part of a CloudFormation stack. This project (unlike some others that I have come across on the web) supports not only creating and deleting AMIs, but also updating, which are properties that are necessary for a pleasant CloudFormation flow. For the impatient, you can find the code here.
Give me a sample!
First, let’s look at an example of how you would declare the AMI in a CloudFormation stack:
How you can use a Custom::AMI resource
In the snippet above, you can see that we must declare parameters for an Image and TemplateInstance . The Image section of the parameters can be used to specify some parameters (e.g. Name , Description ) for the CreateImage call. The TemplateInstance section allows you to specify parameters for the EC2 instance that will be used to create the AMI using the RunInstances call. Note that any update to the TemplateInstance property will trigger a complete replacement of the AMI.
How it all works
In order to create an AMI, we need to go through the following steps, in order:
Let’s take a closer look at the individual steps:
Steps 1, 2 & 3: We need to have an EC2 instance that has been fully configured to our liking, and then stopped. To prepare the EC2 instance, we leverage User Data. Unfortunately, there is no easy way of knowing exactly when the User Data script is completed. For that matter, I use a EC2 instance tag that is set once the User Data is completed. You can see this hack in the sample.yaml snippet shown previously:
A dirty little hack ends an otherwise harmless User Data script
We’ll wait before the UserDataFinished tag has been added to the EC2 instance before considering the User Data script to be completed. Now that the tag has been added to the instance, we can stop the instance.
Steps 4, 5 & 6: Once the EC2 instance has been stopped, we can create an image of the instance. This can take a few minutes. When the AMI is ready to go, we can safely terminate the EC2 instance. This will ensure we don’t pollute our AWS account with the stopped EC2 instance and its associated EBS volume.
Under the hood
At Poka, our preferred way of creating custom resources in CloudFormation is via Lambda functions. However, the 6 steps mentioned above can take way longer than the 5 minutes that are allowed for a Lambda function to execute.
To circumvent this limitation, we decided to leverage AWS Step Functions. Using a state machine, we can clearly express the different steps and the wait times in between:
The State Machine
When CloudFormation wants to create a new Custom::AMI resource, it will first invoke the Lambda-backed custom resource. In turn, the Custom Resource will start the State Machine. The last state in the State Machine sends back a success signal to CloudFormation:
How CloudFormation interacts with Custom::AMI
Once CloudFormation has received the success signal, the new AMI is now ready to use!
Getting Started
To start using AMIs in CloudFormation, all you need to do is create a CloudFormation Stack that will contain the Lambda-backed Resource, the State Machine and their associated IAM roles. You can find instructions to install the stack in the README.
Wrap up
AMIs are still not present to this day in CloudFormation. However, using Lambda-backed Custom Resources and State Machines, we have a way to create, update, and delete AMIs in a reliable way using infrastructure as code.
Thanks to Caroline Maltais for the awesome graphics! | https://medium.com/poka-techblog/managing-amis-using-cloudformation-a097f86a3622 | ['Simon-Pierre Gingras'] | 2018-05-31 20:38:50.443000+00:00 | ['Cloud Computing', 'Python', 'DevOps', 'Infrastructure', 'AWS'] | Title Managing AMIs using CloudFormationContent One longtime missing piece CloudFormation AMIs Amazon Linux Images Creating custom AMIs allows launch new EC2 instance preconfigured liking operating system installed package hardened security etc Thankfully Poka developed solution allows easily create AMIs part CloudFormation stack project unlike others come across web support creating deleting AMIs also updating property necessary pleasant CloudFormation flow impatient find code Give sample First let’s look example would declare AMI CloudFormation stack use CustomAMI resource snippet see must declare parameter Image TemplateInstance Image section parameter used specify parameter eg Name Description CreateImage call TemplateInstance section allows specify parameter EC2 instance used create AMI using RunInstances call Note update TemplateInstance property trigger complete replacement AMI work order create AMI need go following step order Let’s take closer look individual step Steps 1 2 3 need EC2 instance fully configured liking stopped prepare EC2 instance leverage User Data Unfortunately easy way knowing exactly User Data script completed matter use EC2 instance tag set User Data completed see hack sampleyaml snippet shown previously dirty little hack end otherwise harmless User Data script We’ll wait UserDataFinished tag added EC2 instance considering User Data script completed tag added instance stop instance Steps 4 5 6 EC2 instance stopped create image instance take minute AMI ready go safely terminate EC2 instance ensure don’t pollute AWS account stopped EC2 instance associated EBS volume hood Poka preferred way creating custom resource CloudFormation via Lambda function However 6 step mentioned take way longer 5 minute allowed Lambda function execute circumvent limitation decided leverage AWS Step Functions Using state machine clearly express different step wait time State Machine CloudFormation want create new CustomAMI resource first invoke Lambdabacked custom resource turn Custom Resource start State Machine last state State Machine sends back success signal CloudFormation CloudFormation interacts CustomAMI CloudFormation received success signal new AMI ready use Getting Started start using AMIs CloudFormation need create CloudFormation Stack contain Lambdabacked Resource State Machine associated IAM role find instruction install stack README Wrap AMIs still present day CloudFormation However using Lambdabacked Custom Resources State Machines way create update delete AMIs reliable way using infrastructure code Thanks Caroline Maltais awesome graphicsTags Cloud Computing Python DevOps Infrastructure AWS |
753 | AWS — Deploying Angular App With Java On Elastic Beanstalk | AWS — Deploying Angular App With Java On Elastic Beanstalk
A step by step guide with an example project
Photo by SpaceX on Unsplash
AWS provides more than 100 services and it’s very important to know which service you should select for your needs. If you want to deploy an application quickly without any worry about the underlying infrastructure, AWS Elastic Beanstalk is the answer. Elastic Beanstalk reduces management complexity without restricting choice or control. You simply upload your application, and Elastic Beanstalk automatically handles the details of capacity provisioning, load balancing, scaling, and application health monitoring.
In this post, we are going to deploy Angular application with java environment. There are other technologies or environments that AWS supports such as Go, Java, NodeJS, .Net, etc.
Introduction
Example Project
Prerequisites
Build the Project
Deploy on Elastic Beanstalk
Debugging and Update the Deployment
Route 53
Cleaning Up
Things To Consider
Summary
Conclusion
Introduction
If you want to deploy an application without worrying about the underlying infrastructure, Elastic Beanstalk is the solution. When you build the app and upload the app in the form of zip or war, Elastic Beanstalk would take care of provisioning underlying infrastructure such as a fleet of EC2 instances, auto calling groups, monitoring, etc.
The infrastructure provisioned by Elastic Beanstalk depends on the technology chosen while uploading your app. For example, we are going to deploy the Angular with Java backend on Elastic Beanstalk so we need to choose the Java environment. If you want to know more about Elastic Beanstalk here is the link.
Environment Setup
As you see in the above figure, we build our project and create a zip. Once we build the zip, we upload that zip on the Elastic Beanstalk environment. If you have a custom domain you can point that to the elastic beanstalk URL so that your app can be accessible to the public through that URL. | https://medium.com/bb-tutorials-and-thoughts/aws-deploying-angular-app-with-java-on-elastic-beanstalk-f3db3d6a7abe | ['Bhargav Bachina'] | 2020-05-06 14:29:34.988000+00:00 | ['Cloud Computing', 'Java', 'Angular', 'AWS', 'Programming'] | Title AWS — Deploying Angular App Java Elastic BeanstalkContent AWS — Deploying Angular App Java Elastic Beanstalk step step guide example project Photo SpaceX Unsplash AWS provides 100 service it’s important know service select need want deploy application quickly without worry underlying infrastructure AWS Elastic Beanstalk answer Elastic Beanstalk reduces management complexity without restricting choice control simply upload application Elastic Beanstalk automatically handle detail capacity provisioning load balancing scaling application health monitoring post going deploy Angular application java environment technology environment AWS support Go Java NodeJS Net etc Introduction Example Project Prerequisites Build Project Deploy Elastic Beanstalk Debugging Update Deployment Route 53 Cleaning Things Consider Summary Conclusion Introduction want deploy application without worrying underlying infrastructure Elastic Beanstalk solution build app upload app form zip war Elastic Beanstalk would take care provisioning underlying infrastructure fleet EC2 instance auto calling group monitoring etc infrastructure provisioned Elastic Beanstalk depends technology chosen uploading app example going deploy Angular Java backend Elastic Beanstalk need choose Java environment want know Elastic Beanstalk link Environment Setup see figure build project create zip build zip upload zip Elastic Beanstalk environment custom domain point elastic beanstalk URL app accessible public URLTags Cloud Computing Java Angular AWS Programming |
754 | The Next Wave of Startups Is Coming Out of Nature | As company founder Louis Rosenberg explains:
“Nature has spent millions of years trying to optimize how living things operate together, because they can solve very complicated problems together, and ultimately are able to survive by thinking as a system.”
In a swarm, survival depends on deciding where to go to find food or avoid predators. But swarms have no single leader to make a final decision and no language for discussion. Nature’s solution is to pool the knowledge and intuition of the swarm as a whole.
As individuals start to move, they exert a pull on the direction the group will take. Some individuals will agree, and pull in the same direction. Others will have different knowledge of the situation or a different gut feeling. They’ll tug in a different direction. As these pulls and tugs play out in real-time, a decision emerges that optimizes the inputs of every participant and the swarm moves in the best direction.
Unanimous’ Swarm AI works in the same way. A group of individuals is asked a question. They then interact in real-time to exert a pull towards what they think is the best answer. Together, the swarm makes a decision.
You can see this process applied to the question of Democratic Party policy during the 2016 election.
Unanimous AI vs Deep Learning AI vs Humans
Unanimous’ Swarm AI has already been tested against deep-learning AI and human experts in the diagnosis of pneumonia.
In a paper published in Nature, Unanimous AI unsurprisingly outperformed individual radiologists. It also outperformed groups of radiologists debating and voting on diagnoses.
When pitted against CheXNet, a deep-learning AI trained to diagnose pneumonia, the results are a little murkier.
When facing off against CheXNet, Unanimous AI made better diagnoses. But when CheXNet was retrained with a bigger and better dataset, it started to outperform Unanimous in some areas.
The researchers concluded that deep-learning AI was great where it had high degrees of certainty. For less certain cases, a group of radiologists assisted by Unanimous AI produced the best diagnoses possible.
Unanimous keeps humans relevant in the age of AI
That’s where the significance of Unanimous AI really comes to the fore. The development of artificial intelligence is often framed as a competition between humans and the algorithms we program. When the algorithms get better than us at any given task, we become irrelevant.
Image by Gerd Altmann from Pixabay
Unanimous AI goes against the assumption that deep-learning AI is going to be smarter than a group of human intelligence. Looking to nature’s swarms provided the inspiration to challenge that assumption.
Biomimicry has led Unanimous to reassert the importance of people in the age of artificial intelligence. In doing so, they are copying nature to address one of the most pressing issues of our time.
More than just eloquent engineering
“Positively shaping the development of artificial intelligence” is one of the most urgent issues we face as a species, according to 80,000 Hours, the non-profit that identifies mankind’s most pressing problems and figures out how best you can work to solve them.
Unsurprisingly, climate change is also on that list. We urgently need to answer the question of how we sustain our society while staying within the ecological limits of the Earth.
It might be tempting to break this challenge down into a series of individual engineering problems, like how we produce energy without emissions, or chemicals without fossil fuels.
But the most innovative new biomimicry startups are not just copying nature to solve engineering problems. They’re copying nature’s systems to address mankind’s systemic problems.
One example is UK based startup Biohm. Biohm’s product has its roots in nature — literally. They grow sustainable construction materials out of mycelium, the tendrilous, root-like part of mushrooms.
Like Unanimous AI, Biohm’s product is based on nature. And just like Unanimous AI, Biohm’s innovations have implications far beyond solving an engineering or design problem.
To make construction more sustainable, they are mimicking nature on a systematic level to address an underlying assumption of our society: waste.
Nature knows how to do business sustainably
We might be shocked by the scale of our waste mountains. But even if we recycle, we generally accept the concept of waste. How many times have you told yourself that “some things you just can’t recycle”?
Waste, it would seem, is simply the end of the line. We extract raw materials, we turn them into something new, we use them and then we throw them away when we’re done. The garbage heap is the place where a lot of our products naturally end up.
Nature has a different modus operandi. In natural systems, nothing is wasted. When a tree reaches the end of its life, for example, it dies and falls over. But it’s not wasted. Fungus and microbes break it down into humus, the nutrient rich soil where the next generation of trees lay their roots. New growth takes up valuable organic compounds from their decayed forest forebears.
You’ve heard of this before. It’s the circle of life. When the same logic is applied to the waste, it’s called the circular economy. In the circular economy, when products reach the end of their useful lives, someone steps in to turn them into something new.
But there’s more to the circular economy than just eliminating waste. It’s a different way of thinking that is offering biomimicry startups a systematically different way to do business.
As Biohm founder Ehab Sayed explains in Biohm’s Google Talk:
The circular economy is not simply about recycling resources, but it’s about adding as much value as possible at every stage of the way so that the resource that you’re using is not depleting in value at all, it’s either maintaining the same value or increasing in value.
The approach of adding value at every stage has led Biohm to conceive of their products a little differently. Instead of offering materials that can only be used once (like concrete), Biohm’s can be taken apart and reassembled. They also offer ongoing servicing of their buildings and end-of-life deconstruction.
Biohm is also growing new strains of fungus that can eat up waste plastic to grow construction materials and even food.
12.7 million tons of plastic waste was entering the oceans in 2015, and researchers expect this to rise “by an order of magnitude” before 2025. Given the scale of plastic pollution, it’s critical that we find a way to overturn our assumptions that “waste” plastic need only be disposed of. Nature is showing us both the solution, both on the systematic level and the engineering level.
Biohm represents a very different way to think about the construction industry. Overall, it’s a more environmentally sustainable business model. It doesn’t rely on extraction, processing and disposal, but on recycling, recreating and reusing.
The results are impressive: a truly sustainable company that can insulate an ice cream from a thousand degrees of heat by using mushrooms.
Screenshot from The Secret Story of Stuff: Materials of the Modern Age
Asking Nature for innovation
For both Anonymous AI and Biohm, the implications of biomimicry go far beyond just the product they are making.
Biomimicry is starting to show its potential to address the most pressing issues of our time on a systematic level. Startups are already making headway, and there’s more where that came from.
There is now a growing library of nature-inspired solutions to all kinds of problems. It’s called AskNature. Many of the solutions documented in this library relate to what 80,000 Hours identify as the most pressing questions for our survival and prosperity as a species.
Beyond positively shaping the development of AI and the extreme risks of climate change, 80,000 Hours also considers “reducing global catastrophic biological risks” and “improving institutional decision making” to be priorities.
Nature’s ready-made solutions leap off the page for their potential.
Will the bacteria that constantly evolve to defeat new fungal invaders help protect us from ever more powerful biological risks? And if not, then what of the 85 research-backed solutions already in the library under the category of “Protect from living threats > Microbes”?
If “improving institutional decision making” is one of the most impactful areas you can work on, what is the potential of any of the 214 ways the natural world has found to cooperate, coordinate and provide for one and other?
There are many barriers between an idea and a successful startup. Innovation carries a high risk of failure. But for the most crucial challenges of our time we will need to overturn assumptions and innovate.
Nature is ready to show us the way. We just need to ask. | https://medium.com/swlh/the-next-wave-of-startups-is-coming-out-of-nature-1d5061c68b24 | ['Sam Nattress'] | 2020-07-04 07:47:35.312000+00:00 | ['Innovation', 'AI', 'Startup', 'Biomimicry', 'Environment'] | Title Next Wave Startups Coming NatureContent company founder Louis Rosenberg explains “Nature spent million year trying optimize living thing operate together solve complicated problem together ultimately able survive thinking system” swarm survival depends deciding go find food avoid predator swarm single leader make final decision language discussion Nature’s solution pool knowledge intuition swarm whole individual start move exert pull direction group take individual agree pull direction Others different knowledge situation different gut feeling They’ll tug different direction pull tug play realtime decision emerges optimizes input every participant swarm move best direction Unanimous’ Swarm AI work way group individual asked question interact realtime exert pull towards think best answer Together swarm make decision see process applied question Democratic Party policy 2016 election Unanimous AI v Deep Learning AI v Humans Unanimous’ Swarm AI already tested deeplearning AI human expert diagnosis pneumonia paper published Nature Unanimous AI unsurprisingly outperformed individual radiologist also outperformed group radiologist debating voting diagnosis pitted CheXNet deeplearning AI trained diagnose pneumonia result little murkier facing CheXNet Unanimous AI made better diagnosis CheXNet retrained bigger better dataset started outperform Unanimous area researcher concluded deeplearning AI great high degree certainty le certain case group radiologist assisted Unanimous AI produced best diagnosis possible Unanimous keep human relevant age AI That’s significance Unanimous AI really come fore development artificial intelligence often framed competition human algorithm program algorithm get better u given task become irrelevant Image Gerd Altmann Pixabay Unanimous AI go assumption deeplearning AI going smarter group human intelligence Looking nature’s swarm provided inspiration challenge assumption Biomimicry led Unanimous reassert importance people age artificial intelligence copying nature address one pressing issue time eloquent engineering “Positively shaping development artificial intelligence” one urgent issue face specie according 80000 Hours nonprofit identifies mankind’s pressing problem figure best work solve Unsurprisingly climate change also list urgently need answer question sustain society staying within ecological limit Earth might tempting break challenge series individual engineering problem like produce energy without emission chemical without fossil fuel innovative new biomimicry startup copying nature solve engineering problem They’re copying nature’s system address mankind’s systemic problem One example UK based startup Biohm Biohm’s product root nature — literally grow sustainable construction material mycelium tendrilous rootlike part mushroom Like Unanimous AI Biohm’s product based nature like Unanimous AI Biohm’s innovation implication far beyond solving engineering design problem make construction sustainable mimicking nature systematic level address underlying assumption society waste Nature know business sustainably might shocked scale waste mountain even recycle generally accept concept waste many time told “some thing can’t recycle” Waste would seem simply end line extract raw material turn something new use throw away we’re done garbage heap place lot product naturally end Nature different modus operandi natural system nothing wasted tree reach end life example dy fall it’s wasted Fungus microbe break humus nutrient rich soil next generation tree lay root New growth take valuable organic compound decayed forest forebear You’ve heard It’s circle life logic applied waste it’s called circular economy circular economy product reach end useful life someone step turn something new there’s circular economy eliminating waste It’s different way thinking offering biomimicry startup systematically different way business Biohm founder Ehab Sayed explains Biohm’s Google Talk circular economy simply recycling resource it’s adding much value possible every stage way resource you’re using depleting value it’s either maintaining value increasing value approach adding value every stage led Biohm conceive product little differently Instead offering material used like concrete Biohm’s taken apart reassembled also offer ongoing servicing building endoflife deconstruction Biohm also growing new strain fungus eat waste plastic grow construction material even food 127 million ton plastic waste entering ocean 2015 researcher expect rise “by order magnitude” 2025 Given scale plastic pollution it’s critical find way overturn assumption “waste” plastic need disposed Nature showing u solution systematic level engineering level Biohm represents different way think construction industry Overall it’s environmentally sustainable business model doesn’t rely extraction processing disposal recycling recreating reusing result impressive truly sustainable company insulate ice cream thousand degree heat using mushroom Screenshot Secret Story Stuff Materials Modern Age Asking Nature innovation Anonymous AI Biohm implication biomimicry go far beyond product making Biomimicry starting show potential address pressing issue time systematic level Startups already making headway there’s came growing library natureinspired solution kind problem It’s called AskNature Many solution documented library relate 80000 Hours identify pressing question survival prosperity specie Beyond positively shaping development AI extreme risk climate change 80000 Hours also considers “reducing global catastrophic biological risks” “improving institutional decision making” priority Nature’s readymade solution leap page potential bacteria constantly evolve defeat new fungal invader help protect u ever powerful biological risk 85 researchbacked solution already library category “Protect living threat Microbes” “improving institutional decision making” one impactful area work potential 214 way natural world found cooperate coordinate provide one many barrier idea successful startup Innovation carry high risk failure crucial challenge time need overturn assumption innovate Nature ready show u way need askTags Innovation AI Startup Biomimicry Environment |
755 | Platform for success: The Telegraph’s big data transformation | At The Telegraph, the Data Team is a cross-functional group of engineers that manages the core big data platform. It has the responsibility to orchestrate hundreds of batch and real-time data pipelines in order to ensure that the process of data ingestion and transformation is performed on time and is compliant with the highest data quality standards.
This allows analysts to provide reports and insights and the business to make decisions based on reliable figures. Also, The Telegraph’s big data platform is used to support the Data Science Team in delivering predictive models, and the Data Team has an important role to automate the training process and deploy the models in production.
When I joined The Telegraph in 2015, the journey to become a data-driven company was in its infancy. The business was eager to have more data and easier access to information but the infrastructure in place was not yet ready to support decisions in the proper way. Most of the data processing that the data team was doing at that time was done through a single Hadoop cluster used to run few data pipelines on daily bases. But even managing that single cluster was an arduous task due to the lack of direct control on the infrastructure and all the dependencies involved in every minimum change needed.
A major problem at the time was also the waste of computational resources; some pipelines were very demanding but only for a short period of time and our infrastructure was not scaling as it should, which always meant that some nodes were left idle. It became clear that some changes were necessary in order to support the growing digital business with a secure, resilient and auditable data platform able to scale, drive decisions in multiple areas and allow the kind of flexibility that is always paramount in Agile environments. After evaluating all the challenges that the current infrastructure was carrying with it we decided to move away from Hadoop and AWS and as a data team gain more control over our platform.
At the end of 2015, after a few discussions and demos, we were really impressed by all the possibilities that Google Cloud Platform was providing. Bigquery, the data warehouse solution proposed by Google, immediately caught our attention. No infrastructure was needed, there was an affordable pricing model, low response times and high scalability — it seemed exactly what we were looking for! So at the beginning of 2016, we unanimously decided to adopt GCP and the team started to build what is now known as the TMG Datalake.
The first challenge we had was to redesign in a future-proof way our orchestration mechanism and data pipelines, to avoid any waste of resources and allow scalability. Instead of using a cluster of workers we decided to take a different approach and focus on creating single workers on demand and from them push all the computational effort to cloud services when possible. We picked Azkaban as a scheduler and orchestrator and we adopted a very simple paradigm: each time a data pipeline runs, a new worker is created to perform the ETL task. Instead of being restricted to the computational power of a single worker and processing the data in the virtual machine, all the work (when possible) is pushed to BigQuery and the data transformation and aggregation is performed there. After all the SQL queries are executed with success, the result is persisted in our data warehouse and the worker is terminated.
GCP’s Learning curve was less steep than expected and after a couple of months, our first data pipelines to ingest hitlog data from our website and mobile application were ready to be tested. The running time of both processes turned out to be much shorter compared to the results obtained with Hadoop and the data was easily accessible to anyone with a Google account and the right level of permissions. This first iteration proved that the new stack was more performant and easier to manage than the old one and gave us a solid case to continue building the platform.
Before going in further development we decided to take a step back and invest our energies in designing what would become the new Telegraph Datalake. The main goal was to democratize data access across the entire company and provide both insights layer and the possibility to perform discovery analysis on raw data. Also, it was really important to establish a basic data governance before progress further. We decided to avoid any monolithic approach and build our Datalake in multiple iterations reflecting the Agile philosophy that the company was starting to follow during that period.
After a few discussions we opted for the following structure:
Raw data layer: Contains all the data that are ingested both in batch and real-time without any type of transformation. All the data sources ingestion processes are designed to be independent when possible in order to allow modularity.
Contains all the data that are ingested both in batch and real-time without any type of transformation. All the data sources ingestion processes are designed to be independent when possible in order to allow modularity. Shared facts and dimensions layer: Contains facts and dimensions that can be used across multiple projects with the purpose to provide a guidance on the usage of some specific metrics and dimensions.
Contains facts and dimensions that can be used across multiple projects with the purpose to provide a guidance on the usage of some specific metrics and dimensions. Serving/Reporting layer: Contains aggregated data and is built on the top of the previous two. Every project is designed in a separate silo and might use a different way to serve the data to the end user. Two projects don’t interact with each other and they are only sharing the two-layer below. They can be developed independently and in parallel even from different product teams.
Following this approach, our data platform was growing at a fast pace and for more than one year the team focused on ingesting all the available data sources, define data lineage for the main projects and orchestrate multiple pipelines to serve reports on time. We also started to build dashboards and microservices in order to serve a different type of consumers.
After a few months of work TMG Datalake was already storing terabytes of data, but it was not driving any action directly. The need to use data to improve how we interact with our users was clear since the beginning, but it was only when the platform reached the right level of maturity that we started to see possibilities to drive user experience on Telegraph website and mobile application from data. For this purpose, we developed specific data pipelines with the goal to improve user segmentation and use data in our CRM to provide a better experience to our users. This is still an ongoing process and what we can do with Adobe Audience Manager and Adobe Target is having a more significant impact with every day that passes.
Since the beginning of our journey we designed the Datalake to support real-time data processing, but we started to explore how to work with streams for more than a year afterwards. The platform was not mature enough yet and we were looking for use-cases able to bring real value to the business.
Then, in 2017, we started to think which area might benefit more from ingesting and processing real-time data and we moved our focus on the Telegraph newsroom. We began to run a few experiments using Google Cloud Dataflow, with the idea to consume website hitlogs in real-time. It was our first attempt to deploy in production real-time data pipeline and support with that a product that was supposed to be in front of everyone in the company.
A few months later in the newsroom our first real-time dashboard was live. The dashboard (below) displays, with a minimum delay, which type of content leads to new registrations and subscriptions in order to assess which articles and editorial channels are performing better than others.
All the numbers displayed here are for illustrative purposes only.
At the end of 2017, the data platform was still growing and reached more than 200 data pipelines to orchestrate and run, some of them in real-time, others on a daily or weekly basis. Even the stack of technologies (below), which in the beginning was restricted to a few tools, was growing to cover all the use-cases and satisfy different areas of the business.
It was time to introduce some changes, refine our work and reconsider some decisions that we took at the beginning of our journey. We had always worked trying to follow certain standards, but no one was forcing us on the right path if not for common sense and QAs. For this reason, we adopted Giter8 as a templating tool. This allowed us to automatically generate the skeletons of our pipelines and microservices in order to standardize as much as possible every product.
We also started to observe with interest other areas and held discussions with different engineers at the company, to find new ways to improve our day-to-day job. Continuous delivery and continuous integration was already a hot topic at that time and we started to think how to apply what normally is used for microservices to data pipelines in order to deploy in production with the minimum effort an high-quality product.
In 2018, our Jenkins CD pipeline was ready to be tested and we were able to minimize dependencies from scripts, local machines and deploy artifacts in production with a controlled and standardized process.
A few months later we also decided to reconsider the usage of virtual machines. We were already using GKE (Kubernetes) to deploy microservices and it seemed natural to move in the direction to dockerize our data pipelines and orchestrate them as Kubernetes batch jobs.
For this reason we spent most of the summer to refine and improve our templates and prepare our infrastructure to manage docker images. In September 2018 we finally closed the loop deploying in production the first data pipeline on Kubernetes.
It is now almost the end of 2018 and we recently finished defining the direction that we want to take for the next year and where to focus our efforts in order to improve our platform.
The plan for 2019 is to focus on:
Improving our Python ETL library in order to increase development speed, code reusability, and stability of our pipelines. Once ready the library will become open source and shared with the entire community. All the tools that we use to interact with cloud resources in GCP and AWS will be available for everyone. In this way, we hope to receive feedbacks, contributions and make the same journey, that we started a few years ago, easier for who is approaching the cloud now.
Decommission Azkaban in favor of Google Cloud Composer (Airflow) in order to remove a single point of failure from our solution and migrate to a better-supported technology.
Release templates for Apache Beam pipelines and standardize the deployment process.
Improve monitoring logging and alerting with a closer integration with Slack, Stackdriver, and Jira to automatically raise trackable tickets.
Start to provide ETL tools and infrastructure to other teams at the Telegraph in order to help them automate their day to day job in a secure and resilient environment.
During these years we built our data platform like a puzzle, every day adding a new piece and sometimes removing one that was not fitting properly. This puzzle is not finished yet and it will never be. There is always space to improve what has been done in past, new ideas to follow, new technologies to explore. But in the end, what it really matters is to see people excited when a new product is live or they can start to play with a new piece of information and make better decisions based on it. This is the best reward for all our efforts.
Stefano Solimito is a Principal Data Engineer at The Telegraph
Connect on Linkedin | https://medium.com/the-telegraph-engineering/platform-for-success-the-telegraphs-big-data-transformation-e09aa7608832 | ['Stefano Solimito'] | 2018-11-22 21:54:54.485000+00:00 | ['Etl', 'Google Cloud Platform', 'Data Lake', 'Data Engineering', 'Big Data'] | Title Platform success Telegraph’s big data transformationContent Telegraph Data Team crossfunctional group engineer manages core big data platform responsibility orchestrate hundred batch realtime data pipeline order ensure process data ingestion transformation performed time compliant highest data quality standard allows analyst provide report insight business make decision based reliable figure Also Telegraph’s big data platform used support Data Science Team delivering predictive model Data Team important role automate training process deploy model production joined Telegraph 2015 journey become datadriven company infancy business eager data easier access information infrastructure place yet ready support decision proper way data processing data team time done single Hadoop cluster used run data pipeline daily base even managing single cluster arduous task due lack direct control infrastructure dependency involved every minimum change needed major problem time also waste computational resource pipeline demanding short period time infrastructure scaling always meant node left idle became clear change necessary order support growing digital business secure resilient auditable data platform able scale drive decision multiple area allow kind flexibility always paramount Agile environment evaluating challenge current infrastructure carrying decided move away Hadoop AWS data team gain control platform end 2015 discussion demo really impressed possibility Google Cloud Platform providing Bigquery data warehouse solution proposed Google immediately caught attention infrastructure needed affordable pricing model low response time high scalability — seemed exactly looking beginning 2016 unanimously decided adopt GCP team started build known TMG Datalake first challenge redesign futureproof way orchestration mechanism data pipeline avoid waste resource allow scalability Instead using cluster worker decided take different approach focus creating single worker demand push computational effort cloud service possible picked Azkaban scheduler orchestrator adopted simple paradigm time data pipeline run new worker created perform ETL task Instead restricted computational power single worker processing data virtual machine work possible pushed BigQuery data transformation aggregation performed SQL query executed success result persisted data warehouse worker terminated GCP’s Learning curve le steep expected couple month first data pipeline ingest hitlog data website mobile application ready tested running time process turned much shorter compared result obtained Hadoop data easily accessible anyone Google account right level permission first iteration proved new stack performant easier manage old one gave u solid case continue building platform going development decided take step back invest energy designing would become new Telegraph Datalake main goal democratize data access across entire company provide insight layer possibility perform discovery analysis raw data Also really important establish basic data governance progress decided avoid monolithic approach build Datalake multiple iteration reflecting Agile philosophy company starting follow period discussion opted following structure Raw data layer Contains data ingested batch realtime without type transformation data source ingestion process designed independent possible order allow modularity Contains data ingested batch realtime without type transformation data source ingestion process designed independent possible order allow modularity Shared fact dimension layer Contains fact dimension used across multiple project purpose provide guidance usage specific metric dimension Contains fact dimension used across multiple project purpose provide guidance usage specific metric dimension ServingReporting layer Contains aggregated data built top previous two Every project designed separate silo might use different way serve data end user Two project don’t interact sharing twolayer developed independently parallel even different product team Following approach data platform growing fast pace one year team focused ingesting available data source define data lineage main project orchestrate multiple pipeline serve report time also started build dashboard microservices order serve different type consumer month work TMG Datalake already storing terabyte data driving action directly need use data improve interact user clear since beginning platform reached right level maturity started see possibility drive user experience Telegraph website mobile application data purpose developed specific data pipeline goal improve user segmentation use data CRM provide better experience user still ongoing process Adobe Audience Manager Adobe Target significant impact every day pass Since beginning journey designed Datalake support realtime data processing started explore work stream year afterwards platform mature enough yet looking usecases able bring real value business 2017 started think area might benefit ingesting processing realtime data moved focus Telegraph newsroom began run experiment using Google Cloud Dataflow idea consume website hitlogs realtime first attempt deploy production realtime data pipeline support product supposed front everyone company month later newsroom first realtime dashboard live dashboard display minimum delay type content lead new registration subscription order ass article editorial channel performing better others number displayed illustrative purpose end 2017 data platform still growing reached 200 data pipeline orchestrate run realtime others daily weekly basis Even stack technology beginning restricted tool growing cover usecases satisfy different area business time introduce change refine work reconsider decision took beginning journey always worked trying follow certain standard one forcing u right path common sense QAs reason adopted Giter8 templating tool allowed u automatically generate skeleton pipeline microservices order standardize much possible every product also started observe interest area held discussion different engineer company find new way improve daytoday job Continuous delivery continuous integration already hot topic time started think apply normally used microservices data pipeline order deploy production minimum effort highquality product 2018 Jenkins CD pipeline ready tested able minimize dependency script local machine deploy artifact production controlled standardized process month later also decided reconsider usage virtual machine already using GKE Kubernetes deploy microservices seemed natural move direction dockerize data pipeline orchestrate Kubernetes batch job reason spent summer refine improve template prepare infrastructure manage docker image September 2018 finally closed loop deploying production first data pipeline Kubernetes almost end 2018 recently finished defining direction want take next year focus effort order improve platform plan 2019 focus Improving Python ETL library order increase development speed code reusability stability pipeline ready library become open source shared entire community tool use interact cloud resource GCP AWS available everyone way hope receive feedback contribution make journey started year ago easier approaching cloud Decommission Azkaban favor Google Cloud Composer Airflow order remove single point failure solution migrate bettersupported technology Release template Apache Beam pipeline standardize deployment process Improve monitoring logging alerting closer integration Slack Stackdriver Jira automatically raise trackable ticket Start provide ETL tool infrastructure team Telegraph order help automate day day job secure resilient environment year built data platform like puzzle every day adding new piece sometimes removing one fitting properly puzzle finished yet never always space improve done past new idea follow new technology explore end really matter see people excited new product live start play new piece information make better decision based best reward effort Stefano Solimito Principal Data Engineer Telegraph Connect LinkedinTags Etl Google Cloud Platform Data Lake Data Engineering Big Data |
756 | Music — Remedy. Music has the power to change the way… | Executive Producer Kanye West has had a slow decline over the past few years, in aspects other than his music. His battle with bipolar disorder is becoming increasingly difficult for him to hide from the public eye.
He has a new track feature on Playboi Carti’s album, Whole Lotta Red. I’m not a Kanye West fan by any means, but his struggles with mental health are relatable and bring to mind the creative possibilities that could come from such a terrible struggle.
I started the album off with “Go2DaMoon” (featuring Kanye West). Okay. It wasn’t great, but not worthy of the distaste the overall album was receiving.
Before listening to this album, I was in a rather neutral state. I was happy, but I wasn’t in any way anticipating anything from this hour-long musical collection.
Music (and more notably, rhythm) provides a track for the mind to follow.
The subconscious parts of our mind weigh far heavier than the parts we are able to control. Our subconscious minds have a nasty tendency of running rabid. This is why negative intrusive thoughts (like those brought on by OCD) are able to easily pierce your thoughts, and come to the forefront. | https://medium.com/illumination/music-remedy-3ffd3eeb2fab | ['Joseph M. Learned'] | 2020-12-26 21:00:23.292000+00:00 | ['Psychology', 'Anxiety', 'Depression', 'Music', 'Mental Health'] | Title Music — Remedy Music power change way…Content Executive Producer Kanye West slow decline past year aspect music battle bipolar disorder becoming increasingly difficult hide public eye new track feature Playboi Carti’s album Whole Lotta Red I’m Kanye West fan mean struggle mental health relatable bring mind creative possibility could come terrible struggle started album “Go2DaMoon” featuring Kanye West Okay wasn’t great worthy distaste overall album receiving listening album rather neutral state happy wasn’t way anticipating anything hourlong musical collection Music notably rhythm provides track mind follow subconscious part mind weigh far heavier part able control subconscious mind nasty tendency running rabid negative intrusive thought like brought OCD able easily pierce thought come forefrontTags Psychology Anxiety Depression Music Mental Health |
757 | The Three-Act, Eight-Sequence Story Structure | There’s something I always find fascinating about story theory.
It seems like it’s going to be complicated.
I mean, come on.
Three-act, eight-sequence story structure?
It sure as hell sounds complicated.
That’s twelve balls to keep in the air, right?
And if some one just throws that at you and you’ve never heard of it before, you might stare at them like a deer caught in the headlights.
Crap, you might think. I can’t be a writer. This is too hard.
All I want to do is tell a story.
Take a deep breath with me.
All you’re doing is telling a story.
And it’s so much easier than you think. None of this is that hard. Because, guess what? It’s super familiar. So familiar.
Know why?
You’ve been doing it all your life.
Because just about every story you’ve ever been told — every movie you’ve ever watched, every bedtime story you’ve ever been read, every campfire story, every book, every time your grandpa told you a story about when he was a little boy — fits somewhere in the realm of this story structure.
It’s in your blood. It’s part of your DNA.
I mean someone might throw some exception in the comments, but whatever. I might be able to argue with them, but I won’t.
The fact is that the vast majority of the history of stories fit within the framework of this story structure. It doesn’t matter the format the story takes — oral, written, movies, television.
I’m going to talk about novels now, though. Ready?
Three acts are simply a beginning, a middle, and an end.
See how easy that was?
In most books, the beginning, or first Act, takes up about a quarter of the story.
The second Act, or middle, actually takes up a full half of the story. It’s broken into two parts by the mid-point climax. Which means that most stories really have four acts. We’ve said stories have three acts — a beginning, a middle, and an end — since ancient times, so it’s tradition. So we still do.
the third Act, or end, generally takes up about the final quarter of the story.
That’s all. Nothing scary.
You can break your story down further into eight sequences to help with pacing.
This is actually a screenwriting technique, but it works beautifully for novels.
Act One gets two sequences.
Act Two gets four.
Act Three gets the final two.
Give each sequence a climactic scene at the end. Some of those climactic scenes are bigger, or splashier, than others.
For instance, at the end of Sequence Four, in nearly ever story there is, there’s a Very Big Climax.
The Mid-Point Climax.
That’s the scene, in a non-tragic story, where the story’s hero has a big win.
It’s where Harry catches the snitch in his mouth and wins his first quidditch match.
It’s where Dorothy and her friends reach the Emerald City.
It’s where Han Solo is drawn into the Death Star.
It’s still there in a tragic story, it’s just that it’s a very sad moment, instead of a win. It’s a big miss.
The mid-point climax mirrors the tone of the end of the story. So in a non-tragic story, it’s a happy win. In a tragic story it’s a sad miss.
At the end of sequence four — right smack in the middle — of just about every story that’s EVER BEEN TOLD there’s a big moment. If the story’s not tragic it’s a win. If the story is tragic, it’s a miss.
If you don’t believe me, you can check it out.
Start paying attention when you watch movies. Check out about the sixty minute mark. Pay attention when you’re about halfway through a book, too.
You’ll see it.
Other sequence climaxes might be less dramatic. They might be more internal. But knowing that they are there will help you to keep up the pacing of your story.
But every sequence gets one.
Sequence one gets the inciting incident, when the hero is invited into the story.
Sequence two gets the lock in, when the hero does something to accept that invitation.
Sequence three’s climactic scene is usually one of those less dramatic ones.
Sequence four, as we talked about, is the mid-point.
Sequence five is usually another less dramatic one.
Sequence six gets the main climax — the biggest one of all!
Sequence seven gets my favorite, the third act twist.
And sequence eight gets a climactic scene that wraps things up.
I told you. This is comfortable stuff.
You already know how to do it.
You just need to figure out that you know how to do it. | https://medium.com/the-1000-day-mfa/the-three-act-eight-sequence-story-structure-46126febe4f9 | ['Shaunta Grimes'] | 2019-08-07 14:30:21.602000+00:00 | ['Writing', 'Movies', 'Fiction', 'Books', 'Creativity'] | Title ThreeAct EightSequence Story StructureContent There’s something always find fascinating story theory seems like it’s going complicated mean come Threeact eightsequence story structure sure hell sound complicated That’s twelve ball keep air right one throw you’ve never heard might stare like deer caught headlight Crap might think can’t writer hard want tell story Take deep breath you’re telling story it’s much easier think None hard guess It’s super familiar familiar Know You’ve life every story you’ve ever told — every movie you’ve ever watched every bedtime story you’ve ever read every campfire story every book every time grandpa told story little boy — fit somewhere realm story structure It’s blood It’s part DNA mean someone might throw exception comment whatever might able argue won’t fact vast majority history story fit within framework story structure doesn’t matter format story take — oral written movie television I’m going talk novel though Ready Three act simply beginning middle end See easy book beginning first Act take quarter story second Act middle actually take full half story It’s broken two part midpoint climax mean story really four act We’ve said story three act — beginning middle end — since ancient time it’s tradition still third Act end generally take final quarter story That’s Nothing scary break story eight sequence help pacing actually screenwriting technique work beautifully novel Act One get two sequence Act Two get four Act Three get final two Give sequence climactic scene end climactic scene bigger splashier others instance end Sequence Four nearly ever story there’s Big Climax MidPoint Climax That’s scene nontragic story story’s hero big win It’s Harry catch snitch mouth win first quidditch match It’s Dorothy friend reach Emerald City It’s Han Solo drawn Death Star It’s still tragic story it’s it’s sad moment instead win It’s big miss midpoint climax mirror tone end story nontragic story it’s happy win tragic story it’s sad miss end sequence four — right smack middle — every story that’s EVER TOLD there’s big moment story’s tragic it’s win story tragic it’s miss don’t believe check Start paying attention watch movie Check sixty minute mark Pay attention you’re halfway book You’ll see sequence climax might le dramatic might internal knowing help keep pacing story every sequence get one Sequence one get inciting incident hero invited story Sequence two get lock hero something accept invitation Sequence three’s climactic scene usually one le dramatic one Sequence four talked midpoint Sequence five usually another le dramatic one Sequence six get main climax — biggest one Sequence seven get favorite third act twist sequence eight get climactic scene wrap thing told comfortable stuff already know need figure know itTags Writing Movies Fiction Books Creativity |
758 | Lou Rhodes on Lamb, lockdown, and her new book for children | If you’re looking for an easy, affordable gift for a small child, the digital version of the charming picture book The Phlunk is beautiful, and works seamlessly on an iPad. With gorgeous illustrations by Tori Elliott and read by its author, Lou Rhodes, it’s about a cute, cat-like alien with huge ears that enable it to hear everything children are doing on Earth.
Lou and I have know each other for a long time now. When we first met, I was editor of The Face, and she worked for the magazine as a photographer. When she asked me to write the first press bio for her band Lamb, a collaboration with producer Andy Barlow, I was a little anxious: what if they were awful?
I needn’t have worried.
They were magnificent, offsetting Lou’s calm stage presence and ethereal voice with frenetic electronic beats and lush strings. Lamb’s eponymous debut album, released in 1996, is still a huge favourite in our house along with their other seven studio albums, and Lou’s excellent, more folk-influenced 2006 solo album Beloved One. (Which was nominated for the Mercury Music Prize.)
But I wasn’t alone in my doubts, in those early days. Factory Records boss Tony Wilson thought Lamb was a mistake. “He famously told me to not give up the day job when we started,” says Rhodes with a laugh. “He said, ‘You’re a good photographer, Louise!’ I said, ‘Well, thanks. But I think we’re onto something here as well.’ But he was just like, nah.”
Trusting the process
“You’ve just got to trust,” she says, talking about the leap of faith we all take, when making something new. “It’s not really a cerebral process, when you’re creating. It’s that gut feeling of — something’s happening here. And if you had to explain it to anybody, in its formative stages, you wouldn’t go beyond that.
“If you cross-examine what you’re doing, you might as well stop there. Because it comes from somewhere else. It comes from your core and you just follow the process. That’s what we sometimes lose touch with, because we live in a kind of left-brain reality of, ‘Are you going to make a living out of this?’”
The Phlunk grew out of one of those leaps.
“He kind of appeared when my two boys were small,” she explains. “I wrote a poem about this alien creature who can hear everything that children say and do. When kids are small, they’re just always like, ‘Mommy, look, look, look.’ They always want you to see and hear what they’re doing.
“And from when mine were two and six, I was a single mum, and also working, writing, touring. So the idea behind The Phlunk was that he’s always there listening, even if they feel like you’re somewhere else. Although obviously, I wasn’t somewhere else all the time! I hope they would attest to that.”
The poem sat in a drawer until 2011, when Rhodes decided to make it into a book. She approached the illustration department at Falmouth School of Art and Design, and asked if they’d run a competition amongst the students, to draw the character. Tori Elliott won, bringing it to life in a way that Rhodes says was “better than I’d even imagined”. | https://medium.com/creative-living/lou-rhodes-on-lamb-lockdown-and-her-new-book-for-children-13701d1204cb | ['Sheryl Garratt'] | 2020-12-08 11:50:51.805000+00:00 | ['Self Publishing', 'Writing', 'Books', 'Crwativity', 'Music'] | Title Lou Rhodes Lamb lockdown new book childrenContent you’re looking easy affordable gift small child digital version charming picture book Phlunk beautiful work seamlessly iPad gorgeous illustration Tori Elliott read author Lou Rhodes it’s cute catlike alien huge ear enable hear everything child Earth Lou know long time first met editor Face worked magazine photographer asked write first press bio band Lamb collaboration producer Andy Barlow little anxious awful needn’t worried magnificent offsetting Lou’s calm stage presence ethereal voice frenetic electronic beat lush string Lamb’s eponymous debut album released 1996 still huge favourite house along seven studio album Lou’s excellent folkinfluenced 2006 solo album Beloved One nominated Mercury Music Prize wasn’t alone doubt early day Factory Records bos Tony Wilson thought Lamb mistake “He famously told give day job started” say Rhodes laugh “He said ‘You’re good photographer Louise’ said ‘Well thanks think we’re onto something well’ like nah” Trusting process “You’ve got trust” say talking leap faith take making something new “It’s really cerebral process you’re creating It’s gut feeling — something’s happening explain anybody formative stage wouldn’t go beyond “If crossexamine you’re might well stop come somewhere else come core follow process That’s sometimes lose touch live kind leftbrain reality ‘Are going make living this’” Phlunk grew one leap “He kind appeared two boy small” explains “I wrote poem alien creature hear everything child say kid small they’re always like ‘Mommy look look look’ always want see hear they’re “And mine two six single mum also working writing touring idea behind Phlunk he’s always listening even feel like you’re somewhere else Although obviously wasn’t somewhere else time hope would attest that” poem sat drawer 2011 Rhodes decided make book approached illustration department Falmouth School Art Design asked they’d run competition amongst student draw character Tori Elliott bringing life way Rhodes say “better I’d even imagined”Tags Self Publishing Writing Books Crwativity Music |
759 | Our FAQs | Writers
What happens when I submit my article to TDS?
Thank you so much for taking the time to submit your article to our team! We will review it as soon as we can.
If we believe that your article is excellent and ready to go, this is how you will be able to add your post to our publication. If “Towards Data Science” shows up after you click on “Add to publication” in the dropdown menu at the top of the page, that means we have added you as an author and are waiting for you to submit your article. Once you have submitted your article, it will be reviewed by an editor before a final decision is made.
If we think that your article is interesting but needs to be improved, someone from our team will provide you with feedback directly on your submitted Medium article.
Please note that we only respond to articles that were properly submitted using either our form or via an email that exactly follows the instructions listed here. We don’t respond to pitches or questions already answered in our FAQs or on our Contribute page. We also ignore articles that don’t comply with our rules.
If you haven’t heard from us within the next five working days, please carefully check the article you submitted to our team. See if you can now submit it directly to TDS and look for any private notes from us that you may have missed. You should also make sure to check your spam folder.
If you just can’t reach us, the best thing for you to do is submit your article to another publication. Although we’d love to, we can’t provide customized feedback to everyone because we simply receive too many submissions. You can learn more about our decision here and submit another post in a month. | https://medium.com/p/462571b65b35#9c5b | ['Tds Editors'] | 2020-11-19 01:16:58.476000+00:00 | ['Writers’ Guide', 'Tds Team', 'Writers Guide'] | Title FAQsContent Writers happens submit article TDS Thank much taking time submit article team review soon believe article excellent ready go able add post publication “Towards Data Science” show click “Add publication” dropdown menu top page mean added author waiting submit article submitted article reviewed editor final decision made think article interesting need improved someone team provide feedback directly submitted Medium article Please note respond article properly submitted using either form via email exactly follows instruction listed don’t respond pitch question already answered FAQs Contribute page also ignore article don’t comply rule haven’t heard u within next five working day please carefully check article submitted team See submit directly TDS look private note u may missed also make sure check spam folder can’t reach u best thing submit article another publication Although we’d love can’t provide customized feedback everyone simply receive many submission learn decision submit another post monthTags Writers’ Guide Tds Team Writers Guide |
760 | Want a Better Pitch? Play This Game. | Want a Better Pitch? Play This Game.
Teaching pitch structure to CEOs, founders, and Uber engineers.
In the last four weeks, I’ve led strategic messaging workshops for Uber’s engineering team (on how to use narrative structure to more effectively pitch ideas to internal colleagues), for the founders of an Andreessen Horowitz/First Round-backed startup, and for international entrepreneurship students visiting Stanford.
In these workshops —as in my strategic messaging and positioning engagements — everyone comes in wanting to tell a more compelling story. But not everyone comes in knowing exactly what that means.
So I wondered: Is there a fast, fun way to give people a sense of how much better their pitches and strategic messaging can get by applying the fundamentals of narrative structure?
There are plenty of articles on the subject (Stewart Butterfield’s “We Don’t Sell Saddles Here” is excellent, and many people tell me they’ve benefited from my narrative breakdown of Elon Musk’s Tesla Powerwall pitch, “Want a Better Pitch? Watch This”). But I wanted an exercise I could do with a live audience that would get the point across quickly.
I searched every book I could find on the topics of pitching and business storytelling, to no avail. Then, on Amazon, I stumbled across a guide for elementary-school teachers called Super Simple Storytelling (by Kendall Haven). Near the back, I found a game called “Is it a Story Yet?” — designed for second-graders — that is now part of every workshop and strategic messaging project I lead. I’ve found no quicker way to get CEOs, founders, engineers, marketers, salespeople and product managers excited about applying storytelling craft to their pitches.
How to Play “Is it a Story Yet?”
Before starting the game, I go around the room and ask a few people to give their elevator pitches. Almost always, the pitches start and end with a product, a feature, or a service, or make reference to upending a category:
We do healthcare analytics. We’re going to disrupt payments. I want us to switch to a new database for the XYZ application. (internal project pitch)
My goal is to get attendees to re-frame their pitches around the difference they’ll make in people’s lives.
In the elementary school version of “Is it a Story Yet?”, the teacher reads the class a tale about a boy who forgets his homework. After each sentence, the teacher pauses to ask, “Is it a story yet?” Students get to vote, and explain why it is or isn’t.
For adult audiences, I use a more familiar narrative. I start the game by saying:
There’s a lightsaber that glows green and can cut down enemies. Is it a story yet?
Everyone agrees it’s not. “Why not?” I ask.
“Because there are no characters or context,” someone will usually say. So I add a protagonist, a couple of other characters, and some context:
There’s a boy who lives on sandy planet, and an old man shows up asking the boy to rescue a princess. He offers the boy a gift: a lightsaber that glows green and can cut down enemies.
Still not a story, everyone concurs, because the main character hasn’t set off to rescue the princess. So I add that piece:
The boy demurs until the bad guy’s forces destroy the boy’s home, killing his family. Now he goes with the old man to rescue the princess and stop the bad guy.
“We need a resolution,” someone always says. “How does it end?”
He defeats the bad guy and rescues the princess.
At this point, some people think we have a story. Invariably, though, someone will say, “But how did he defeat the bad guy?” Finally, we add this, which most people agree makes it a story, even if it lacks a few details:
He has lightsaber duels, gets himself and his friends out of a trash compactor and, finally, drops a charge that blows up the bad guy’s weapon. He defeats the bad guy and rescues the princess.
Now, to bring the lesson home, I return to the earlier “What do you guys do?” answers, and compare them to the first line in the game:
We all agreed that ‘There’s a lightsaber…’ wasn’t a story. And surely none of us, when asked to summarize Star Wars: Episode IV, would start with that. Yet when we look at our pitches, they’re all essentially variations on ‘There’s a lightsaber.’”
I was initially worried that this would come off as pedantic or shaming. But in my post-session feedback surveys, participants always cite the lightsaber among the most valuable takeaways.
“Is it a Story Yet?” as a Roadmap for Emotional Pitch Narratives
Naturally, this begs the question: How do you turn what’s essentially a product or feature description (“There’s a lightsaber”) into an emotionally resonant pitch story?
The answer is to describe the difference you’re making in people’s lives by filling in the same missing pieces we added to the Star Wars story:
Main character
Who is the person whose life will be transformed by what you’re offering? Usually this is your customer. If you’re a non-profit, it’s the people or communities you’re helping. Why now?
Why is the main character compelled to act ? In other words, what is causing that person to struggle? In Star Wars, it’s the Empire hitting Luke where he lives, leaving him with nothing. For Zuora, the subscription payments platform, it’s what the company calls the “Subscription Economy,” in which many businesses are shifting from outright purchase to renewable service plans. (For more on this, read For Startups: The Power of “Why Now?”) Promised Land
What will the world look like for your main character (customer) if he/she buys what you’re selling? In Star Wars, it’s the princess rescued and the bad guy defeated. For Airbnb, it’s that you’ll “Belong anywhere.” (For more on teasing the Promised Land, read: Master the “Move”) Obstacles and gifts
OK, now it’s time to talk about that lightsaber. But do it in the context of the obstacles you’ll help your character overcome to reach the Promised Land. Luke has to overcome a bunch of bad guys and their weapons to defeat Darth Vader. What obstacles will your customer have to overcome, and how does your product/service help? You’re Obi Wan, furnishing your customer with gifts that will help him/her reach the Promised Land. Evidence
The one element of a pitch story that doesn’t appear in the game is evidence. In pitch stories, unlike movies, the ending hasn’t happened yet. What evidence can you offer that you’ll make the story come true? (I discuss evidence in more detail in Want a Better Pitch? Watch This.)
Once these pieces are in place, it’s easy to construct a compelling pitch narrative, even if you have limited time. The format for an effective, bare bones pitch is:
For [MAIN CHARACTER], [WHY NOW]. So we thought, what if we could [HELP THEM REACH THE PROMISED LAND].
For example, the Airbnb founders’ early pitch story was essentially,
For travelers, it’s really hard to feel connected to the places they visit. So we thought, what if it we made it easy to find locals willing to host them?
If you’re successful, your audience will be thinking, “Sounds great, but how do you do that?” Now, having established what’s at stake, you’ll find people pay a lot more attention when you then talk about the cool features of your product (your “lightsaber”), describe the obstacles it helps overcome, and share evidence that you’re not just spinning fairy tales. | https://medium.com/firm-narrative/uber-pitch-game-engineers-ceos-de7ada730a89 | ['Andy Raskin'] | 2016-08-22 14:27:04.832000+00:00 | ['Storytelling', 'Entrepreneurship', 'Startup', 'Fundraising', 'Pitching'] | Title Want Better Pitch Play GameContent Want Better Pitch Play Game Teaching pitch structure CEOs founder Uber engineer last four week I’ve led strategic messaging workshop Uber’s engineering team use narrative structure effectively pitch idea internal colleague founder Andreessen HorowitzFirst Roundbacked startup international entrepreneurship student visiting Stanford workshop —as strategic messaging positioning engagement — everyone come wanting tell compelling story everyone come knowing exactly mean wondered fast fun way give people sense much better pitch strategic messaging get applying fundamental narrative structure plenty article subject Stewart Butterfield’s “We Don’t Sell Saddles Here” excellent many people tell they’ve benefited narrative breakdown Elon Musk’s Tesla Powerwall pitch “Want Better Pitch Watch This” wanted exercise could live audience would get point across quickly searched every book could find topic pitching business storytelling avail Amazon stumbled across guide elementaryschool teacher called Super Simple Storytelling Kendall Near back found game called “Is Story Yet” — designed secondgraders — part every workshop strategic messaging project lead I’ve found quicker way get CEOs founder engineer marketer salesperson product manager excited applying storytelling craft pitch Play “Is Story Yet” starting game go around room ask people give elevator pitch Almost always pitch start end product feature service make reference upending category healthcare analytics We’re going disrupt payment want u switch new database XYZ application internal project pitch goal get attendee reframe pitch around difference they’ll make people’s life elementary school version “Is Story Yet” teacher read class tale boy forgets homework sentence teacher pause ask “Is story yet” Students get vote explain isn’t adult audience use familiar narrative start game saying There’s lightsaber glow green cut enemy story yet Everyone agrees it’s “Why not” ask “Because character context” someone usually say add protagonist couple character context There’s boy life sandy planet old man show asking boy rescue princess offer boy gift lightsaber glow green cut enemy Still story everyone concurs main character hasn’t set rescue princess add piece boy demur bad guy’s force destroy boy’s home killing family go old man rescue princess stop bad guy “We need resolution” someone always say “How end” defeat bad guy rescue princess point people think story Invariably though someone say “But defeat bad guy” Finally add people agree make story even lack detail lightsaber duel get friend trash compactor finally drop charge blow bad guy’s weapon defeat bad guy rescue princess bring lesson home return earlier “What guy do” answer compare first line game agreed ‘There’s lightsaber…’ wasn’t story surely none u asked summarize Star Wars Episode IV would start Yet look pitch they’re essentially variation ‘There’s lightsaber’” initially worried would come pedantic shaming postsession feedback survey participant always cite lightsaber among valuable takeaway “Is Story Yet” Roadmap Emotional Pitch Narratives Naturally begs question turn what’s essentially product feature description “There’s lightsaber” emotionally resonant pitch story answer describe difference you’re making people’s life filling missing piece added Star Wars story Main character person whose life transformed you’re offering Usually customer you’re nonprofit it’s people community you’re helping main character compelled act word causing person struggle Star Wars it’s Empire hitting Luke life leaving nothing Zuora subscription payment platform it’s company call “Subscription Economy” many business shifting outright purchase renewable service plan read Startups Power “Why Now” Promised Land world look like main character customer heshe buy you’re selling Star Wars it’s princess rescued bad guy defeated Airbnb it’s you’ll “Belong anywhere” teasing Promised Land read Master “Move” Obstacles gift OK it’s time talk lightsaber context obstacle you’ll help character overcome reach Promised Land Luke overcome bunch bad guy weapon defeat Darth Vader obstacle customer overcome productservice help You’re Obi Wan furnishing customer gift help himher reach Promised Land Evidence one element pitch story doesn’t appear game evidence pitch story unlike movie ending hasn’t happened yet evidence offer you’ll make story come true discus evidence detail Want Better Pitch Watch piece place it’s easy construct compelling pitch narrative even limited time format effective bare bone pitch MAIN CHARACTER thought could HELP REACH PROMISED LAND example Airbnb founders’ early pitch story essentially traveler it’s really hard feel connected place visit thought made easy find local willing host you’re successful audience thinking “Sounds great that” established what’s stake you’ll find people pay lot attention talk cool feature product “lightsaber” describe obstacle help overcome share evidence you’re spinning fairy talesTags Storytelling Entrepreneurship Startup Fundraising Pitching |
761 | Anxiety Is the Dizziness of Freedom | Nat got herself a cup of coffee and took her seat. This was her second meeting since getting the prism from Lyle; last week she’d been planning on announcing that she wasn’t going to be coming back, but she had wound up hardly saying anything at all. So she had had to attend at least one more and say that she was going to take a break from the meetings; people would wonder if she simply stopped.
Dana smiled at the group and said, “Who wants to start us off today?”
Without intending to, Nat found herself speaking, just as Lyle began saying something as well. Both of them stopped.
“You go,” said Nat.
“No, you should go,” said Lyle. “I don’t think you’ve ever started off a meeting before.”
Nat realized that he was right. What had come over her? She opened her mouth, but for once she couldn’t think of a good lie. Eventually, she said, “A guy I work with, I guess you’d call him my supervisor, he was killed recently. Murdered, actually.”
The group was shocked, with assorted “Oh my Gods” being murmured.
“Do you want to tell us about your relationship with him?” asked Dana.
“Yeah,” asked Kevin. “Was he a friend?”
“Kind of,” Nat admitted. “But that’s not why it’s been on my mind. I know this isn’t a grief support group . . . I guess I brought this up because I wanted your take on something.”
“Of course,” said Dana. “Go ahead.”
“I keep thinking about the randomness of this murder. I don’t mean the killer picked him at random. I mean, when he had the gun pointed at my supervisor, he said that some version of him was going to pull the trigger, so why shouldn’t it be him? We’ve all heard that line before, but I never paid any attention to it. But now I’m wondering, are the people who say that actually right?”
“That’s a good question,” said Dana. “I agree that we’ve all heard people make similar claims.” She addressed the group. “Does anyone have any thoughts on that? Do you think that every time someone makes you angry, there’s a branch where you pick up a gun and shoot the guy?”
Zareenah spoke up. “I’ve read that there’s been an increase in crimes of passion since prisms became popular. Not an enormous one, but statistically significant.”
“Yeah,” said Kevin, “which is why the theory can’t be true. The fact that there’s been an increase, even a small one, disproves the theory.”
“How do you figure?” asked Zareenah. “Branches are generated by any quantum event, right? Even before we had prisms, branches were still splitting off constantly; we just didn’t have access to any of them. If it were true that there’s always a branch where you pick up a gun and shoot someone on a whim, then we should have seen the same number of random murders every day before the prism was invented as we saw every day after. The invention of prisms wouldn’t cause more of those murders to line up in this particular branch. So if we’re seeing more people killing one another since prisms became popular, it can’t be because there’s always a branch where you pick up a gun.”
“I follow your reasoning,” said Zareenah, “but then what’s causing the rise in murders?”
Kevin shrugged. “It’s like a suicide fad. People hear about other people doing it, and it gives them ideas.”
Nat thought about it. “That proves that the argument can’t be right, but it doesn’t explain why it’s wrong.”
“If you know the theory’s wrong, why do you need more?”
“I want to know whether my decisions matter!” That came out more emphatically than she intended. Nat took a breath, and then continued. “Forget about murder; that’s not the kind of thing I’m talking about. But when I have a choice to do the right thing or the wrong thing, am I always choosing to do both in different branches? Why should I bother being nice to other people, if every time I’m also being a dick to them?”
There was some discussion among the members for a while, but eventually Nat turned to Dana. “Can you tell me what you think?”
“Sure,” said Dana. She paused to gather her thoughts. “In general, I think your actions are consistent with your character. There might be more than one thing that would be in character for you to do, because your behavior is going to vary depending on your mood, but there are a lot more things that would be utterly out of character. If you’re someone who’s always loved animals, there isn’t a branch where you kick a puppy just because it barked at you. If you’re someone who’s always obeyed the law, there’s no branch where you suddenly rob a convenience store instead of going into work in the morning.”
Kevin said, “What about branches that diverged when you were a baby and your life took a totally different course?”
“I don’t care about that,” said Nat. “I’m asking about branches where I, having lived the life I led, am faced with a choice.”
“Kevin, we can talk about bigger divergences later, if you want,” said Dana.
“No, that’s fine. Proceed.”
“Okay, so let’s imagine you’re in a situation where you have a couple options, and either course of action would be consistent with your character. For example, suppose a cashier has given you too much change, and you can either give it back or just keep it. Suppose you could see yourself doing either of those, depending on the kind of day you’re having. In that case, I’d say it’s entirely possible that there’s a branch where you keep the extra change, as well as a branch where you give it back.”
Nat realized there probably weren’t any branches out there where she gave back the extra change. For as long as she could remember, if she was having a good day, getting extra change would have just made it a better day.
Kevin asked, “So does that mean it doesn’t matter if we act like jerks?”
“It matters to the person in this branch that you’re acting like a jerk to,” said Zareenah.
“But what about globally? Does being a jerk in this branch increase the percentage of jerkish behavior across all branches?”
“I’m not sure about the math,” said Dana. “But I definitely think that your choices matter. Every decision you make contributes to your character and shapes the kind of person you are. If you want to be someone who always gives the extra money back to the cashier, the actions you take now affect whether you’ll become that person.
“The branch where you’re having a bad day and keep the extra change is one that split off in the past; your actions can’t affect it anymore. But if you act compassionately in this branch, that’s still meaningful, because it has an effect on the branches that will split off in the future. The more often you make compassionate choices, the less likely it is that you’ll make selfish choices in the future, even in the branches where you’re having a bad day.”
“That sounds good, but — ” Nat thought about how years of acting a certain way could wear ruts in a person’s brain, so that you would keep slipping into the same habits without trying to. “But it’s not easy,” said Nat.
“I know it’s not,” said Dana. “But the question was, given that we know about other branches, whether making good choices is worth doing. I think it absolutely is. None of us are saints, but we can all try to be better. Each time you do something generous, you’re shaping yourself into someone who’s more likely to be generous next time, and that matters.
“And it’s not just your behavior in this branch that you’re changing: you’re inoculating all the versions of you that split off in the future. By becoming a better person, you’re ensuring that more and more of the branches that split off from this point forward are populated by better versions of you.”
Better versions of Nat. “Thanks,” she said. “That’s what I was looking for.” | https://onezero.medium.com/anxiety-is-the-dizziness-of-freedom-b5ab45cae2a5 | ['Ted Chiang'] | 2020-08-13 17:24:07.890000+00:00 | ['Technology', 'Science Fiction', 'Future', 'Science', 'Books'] | Title Anxiety Dizziness FreedomContent Nat got cup coffee took seat second meeting since getting prism Lyle last week she’d planning announcing wasn’t going coming back wound hardly saying anything attend least one say going take break meeting people would wonder simply stopped Dana smiled group said “Who want start u today” Without intending Nat found speaking Lyle began saying something well stopped “You go” said Nat “No go” said Lyle “I don’t think you’ve ever started meeting before” Nat realized right come opened mouth couldn’t think good lie Eventually said “A guy work guess you’d call supervisor killed recently Murdered actually” group shocked assorted “Oh Gods” murmured “Do want tell u relationship him” asked Dana “Yeah” asked Kevin “Was friend” “Kind of” Nat admitted “But that’s it’s mind know isn’t grief support group guess brought wanted take something” “Of course” said Dana “Go ahead” “I keep thinking randomness murder don’t mean killer picked random mean gun pointed supervisor said version going pull trigger shouldn’t We’ve heard line never paid attention I’m wondering people say actually right” “That’s good question” said Dana “I agree we’ve heard people make similar claims” addressed group “Does anyone thought think every time someone make angry there’s branch pick gun shoot guy” Zareenah spoke “I’ve read there’s increase crime passion since prism became popular enormous one statistically significant” “Yeah” said Kevin “which theory can’t true fact there’s increase even small one disproves theory” “How figure” asked Zareenah “Branches generated quantum event right Even prism branch still splitting constantly didn’t access true there’s always branch pick gun shoot someone whim seen number random murder every day prism invented saw every day invention prism wouldn’t cause murder line particular branch we’re seeing people killing one another since prism became popular can’t there’s always branch pick gun” “I follow reasoning” said Zareenah “but what’s causing rise murders” Kevin shrugged “It’s like suicide fad People hear people give ideas” Nat thought “That prof argument can’t right doesn’t explain it’s wrong” “If know theory’s wrong need more” “I want know whether decision matter” came emphatically intended Nat took breath continued “Forget murder that’s kind thing I’m talking choice right thing wrong thing always choosing different branch bother nice people every time I’m also dick them” discussion among member eventually Nat turned Dana “Can tell think” “Sure” said Dana paused gather thought “In general think action consistent character might one thing would character behavior going vary depending mood lot thing would utterly character you’re someone who’s always loved animal isn’t branch kick puppy barked you’re someone who’s always obeyed law there’s branch suddenly rob convenience store instead going work morning” Kevin said “What branch diverged baby life took totally different course” “I don’t care that” said Nat “I’m asking branch lived life led faced choice” “Kevin talk bigger divergence later want” said Dana “No that’s fine Proceed” “Okay let’s imagine you’re situation couple option either course action would consistent character example suppose cashier given much change either give back keep Suppose could see either depending kind day you’re case I’d say it’s entirely possible there’s branch keep extra change well branch give back” Nat realized probably weren’t branch gave back extra change long could remember good day getting extra change would made better day Kevin asked “So mean doesn’t matter act like jerks” “It matter person branch you’re acting like jerk to” said Zareenah “But globally jerk branch increase percentage jerkish behavior across branches” “I’m sure math” said Dana “But definitely think choice matter Every decision make contributes character shape kind person want someone always give extra money back cashier action take affect whether you’ll become person “The branch you’re bad day keep extra change one split past action can’t affect anymore act compassionately branch that’s still meaningful effect branch split future often make compassionate choice le likely you’ll make selfish choice future even branch you’re bad day” “That sound good — ” Nat thought year acting certain way could wear rut person’s brain would keep slipping habit without trying “But it’s easy” said Nat “I know it’s not” said Dana “But question given know branch whether making good choice worth think absolutely None u saint try better time something generous you’re shaping someone who’s likely generous next time matter “And it’s behavior branch you’re changing you’re inoculating version split future becoming better person you’re ensuring branch split point forward populated better version you” Better version Nat “Thanks” said “That’s looking for”Tags Technology Science Fiction Future Science Books |
762 | Why Taoism’s Best Idea & Our Most Radical Environmental Restoration Plans Are Made For Each Other | Photo by Owain McGuire on Unsplash. Lion Sands Game Reserve, Sabi Sands, South Africa
The gentlest thing in the world
overcomes the hardest thing in the world.
That which has no substance
enters where there is no space.
This shows the value of non-action.
Teaching without words,
performing without actions:
that is the Master’s way.
~Tao Te Ching, chapter 43
We tend to think of environmental restoration as hard work: picks and shovels, ripping invasive plants out of the ground, conducting controlled burns. This is the well meaning side of our fondness for thinking of ourselves as nature’s masters. It was our intervention that screwed it up, so it must be our reintervention that restores it.
Often, however, restoration may involve little more than pulling the plug on a foolish dam, watching the waters recede and letting an area recover on its own. Will the place ever be the same? No. But then it was never going to be the same anyway. Nature is just another word for change.
In the Taoist tradition the concept of wu-wei (often translated as effortless/actionless action) lies at the core of the philosophy. It is closely associated with water, which though soft eventually cracks and erodes even the hardest of objects.
Rewilding likewise calls for respecting the soft power of natural processes and their capacity to eventually return even the most damaged environment back to health. Neither wu-wei nor rewilding call upon us to be utterly passive, but they do challenge us to incorporate careful observation, humility, and patience into everything we do. These have too often been missing from both our efforts to dominate nature and to preserve it. Even when we have the best of intentions, we often act rashly hoping to quickly fix what we have broken.
What rewilding isn’t . . . and is
Words like rewilding are vulnerable to misunderstanding, to say nothing of intentional misrepresentation by those who oppose it. So before getting to what rewilding is or can be, it’s necessary to say briefly what it isn’t.
Rewilding is not a Luddite movement attempting to return us to living in caves without even so much as a plastic toothbrush to our name. Indoor plumbing, electricity, and even Facebook (assuming you haven’t already deleted your account) are all comforts we would still be able to enjoy in a more rewilded world.
This isn’t to say incorporating the concept into our conservation and sustainability efforts to a much greater extent wouldn’t impact how we relate with technology. However, I’m convinced these changes would enhance human well-being and happiness rather than subtract from it. Indeed, there’s even evidence humanity could experience greater abundance than it does now if rewilding became a central value of our culture.
Rewilding remains strongly linked to the reintroduction of apex predators to ecosystems that haven’t seen them in a while. The reintroduction of wolves to Yellowstone is probably the best known example of this kind of rewilding effort. Because these animals stand at the top of the food chain they need larger areas to thrive. Therefore, their reintroduction also requires large tracts of sufficiently healthy habitat with adequate prey populations.
That said, the concept has broadened since the Yellowstone wolf reintroduction. The rewilding of a landscape need not include any reintroduction at all. For example, one can imagine the site of an ill-conceived reservoir being rewilded simply by draining the artificial lake and allowing nature to take its course as the formerly flooded canyon or valley slowly flushes itself of the accumulated silt and revegetates.
Some have expanded the concept even further to include areas as small as our backyard or urban rooftops. These tiny patches will obviously never be inviting habitats for mountain lions and wolves, but they can still function as critical refuges for birds and pollinators otherwise displaced by human development. In this context active human management is far more likely to play a vital role. That’s okay though. As I’ve written previously, playing in the dirt actually brings some wonderful health benefits. A well planned garden that requires some muscle to maintain can be as good for us as it is for the species that use it.
The Colorado River: an awaiting rewilding opportunity
In March of 1963 the diversion tunnels for what would become the US Bureau of Reclamation’s most controversial project were closed. At that moment the waters of the Colorado River began to slowly back up behind Glen Canyon Dam and Lake Powell was born.
Located in a remote section of desert on the Utah/Arizona border, Lake Powell has only briefly ever reached capacity in its now more than 50 years of existence. Between seepage into the surrounding sandstone and evaporation, the lake loses nearly one million acre feet of water a year that, in spite of the name of the government agency that gave birth to it, is never actually reclaimed for agricultural use or as drinking water. According to the Bureau of Reclamation’s own website, between “two and three percent of the lake’s water evaporates into the atmosphere each year.” In addition, “About 13.4 million acre feet (16,500 million cubic meters), or almost a third of the reservoir’s capacity” was absorbed into the porous sandstone of Glen Canyon just during the years the lake was filling. That absorption continues to this day.
Photo by Rainer Krienke on Unsplash. Lake Powell (formerly Glen Canyon), USA. The Late David Brower and other environmental activists have pointed to the canyon submerged beneath the waters of Lake Powell on the Utah/Arizona border as an excellent candidate for restoration.
Even though it is far from the largest river, the Colorado River is among the most overtaxed rivers in the world. Its watershed provides water to every major metropolitan area between Denver and Los Angeles and irrigates hundreds of thousands of acres of agricultural land. Given this reality, having so much of it go toward sustaining a giant evaporation pond in the middle of a remote desert is a waste from a human as well as environmental perspective.
In the environmental classic A Sand County Almanac, Aldo Leopold describes the 1922 trip he and his brother took to the Colorado River Delta. If you’ve visited the delta recently, as I have, Leopold’s description of it reads like a work of fiction. He writes that “On the map the Delta was bisected by the river, but in fact the river was nowhere and everywhere, for he [the river]could not decide which of a hundred green lagoons offered the most pleasant and least speedy path to the Gulf.”
Today it is considered a good year when the Colorado River makes it to the Sea of Cortez at all. Even then the small portion of the river that does find its way there is typically a fetid, briny rivulet contaminated with chemical fertilizer and hardened with minerals concentrated by evaporation.
Efforts have recently gotten underway to restore a small fraction of the Colorado River Delta to its former glory. A December 2014 article in National Geographic Magazine described the effects of a “pulse flow” of just over 105,000 acre feet of water that started with a release from Hoover Dam a week upstream. The results of that release were dramatic, and provide a glimpse of what could happen if a stronger more reliable flow were provided regularly to the Delta. The more than 800,000 acre feet lost each year to evaporation and absorption at Lake Powell immediately comes to mind as a possible source for such a renewal.
“Man always kills the thing he loves, and so we the pioneers have killed our wilderness,” Aldo Leopold wrote at the end of his chapter describing the time he and his brother spent exploring the Colorado River Delta. “Some say we had to,” he continued. “Be that as it may, I am glad I shall never be young without wild country to be young in. Of what avail are forty freedoms without a blank spot on the map?”
We may never get the Colorado River Delta of Aldo Leopold’s youth back, but the signs of recovery at sites benefitting from the pulse flow of 2014 do show that even after receiving seemingly fatal blows ecosystems can prove more resilient than we give them credit for. As Sandra Postel, director of the Global Water Policy Project put it, the Delta “isn’t dead,” but “it’s been dormant, and if you add water it will come back to life.”
Wildife management as actionless action
We’ve fallen into the bad habit of seeing natural relationships through a linear zero sum lens. For example, if we like to eat something and another animal also likes to eat the same thing, killing off the species that’s in competition with us will, we consistently think, necessarily increase the availability of the creature(s) we like to eat. This idea turns out to be as wrong as it is simplistic.
Wolves, coyotes, whales, seals, sea otters… In each case our eradication policies have backfired. The removal of the wolf turned out to have huge deleterious consequences for beaver populations, aspen stands, and many other components of the ecosystem. Overhunting whales ended up having a major negative impact on phytoplankton, which a variety of fish populations rely upon that humans as well as whales consume. Likewise, killing seals to protect fish species we consume had the opposite effect. As George Manbiot put it in The Guardian, “The fishermen who have insisted that predators such as seals should be killed might have been reducing, not enhancing, their catch.”
Our usually misguided wildlife management policies are perhaps the best example we have where the ancient Taoist concept of actionless action would, if applied, have tremendous positive consequences. Nature does not operate under zero sum rules. Predators, while obviously a threat to individual prey, play a critical role in maintaining the entire ecosystem. A healthy ecosystem is more productive than an unhealthy one, leaving more and better services for all species, including humans, to take advantage of.
There’s no need to reintroduce wolves again to the Rockies. All we need to do now is allow those already occupying the Greater Yellowstone Area to migrate unmolested to their ancestor’s old haunts. Likewise, just leaving the ocean’s whales, seals and otters alone will, in the long-run, likely be sufficient to facilitate their eventual recovery.
Any interventions we do undertake, however, must be for the purpose of restoring species to areas where they no longer exist or are extremely rare. Hunting one species for the “benefit” of another hasn’t paid dividends in the past and isn’t likely to in the future. Ecosystems are just too complex for us to know with any degree of certainty what the outcome will be if we again undertake such efforts.
Sea urchins are a staple of the sea otter diet. Sea urchins eat kelp and other plant matter. Eliminating sea otters has resulted in urchin population explosions which have devasted kelp forests upon which many fish depend. Image: wikimedia
Rewilding from local parks to the backyard
Personal and public habitats much closer to home provide excellent opportunities for both casual and more hands on approaches to environmental restoration. Obviously, our yards and city parks will never be wild in the sense that most people use that word. But they can play a key role in our conservation efforts none-the-less.
Consider Seattle’s Pollinator Pathway. This project is obviously targeting bees, butterflies, moths and other pollinators rather than the ecosystem as a whole. But by providing needed support for creatures that often struggle in urban and agricultural settings, the project increases the capacity of these critical species to absorb environmental stress and ultimately thrive.
The Pollinator Pathway runs only a mile along Seattle’s Columbia Street. However, it demonstrates the feasibility of much larger efforts that can be woven into any city’s infrastructure. These types of projects are only limited by the imagination of urban planners and architects and the willingness of local citizens and policymakers to make them a reality. If implemented on a large enough scale, the benefits to pollinators desperately seeking diverse habitats free of large scale pesticide and herbicide contamination could reverse decades long population declines.
According to bee expert Noah Wilson-Rich, urban bees already overwinter far better than rural bees. The year-to-year survival rate for bees in city beehives is more than 60%, while for rural beehives it’s only 40%. Though we are uncertain why bees seem to fare so well in cities, the fact remains that even the most crowded human environments can play a key role in the recovery of insects critical to both our own well-being and that of the more than 250,000 species of flowering plants that share this planet with us.
Conclusion:
It’s only a slight exaggeration to say the research opportunities that extensive rewilding would provide are practically endless. This will be especially true if we adopt a broad view of the concept instead of a narrow absolutist one. As the case of urban bees show, rewilding is something that happens along a spectrum rather than being an either/or proposition.
No matter how far we take our rewilding efforts in any given case, humans will always remain a component of the ecosystem. Any attempt to integrate even a little bit of the non-human world into our life will provide rich opportunities to enhance our understanding of the dynamic relationships that sustain earth’s biosphere.
But before arriving there we must take a leap of faith that is especially difficult in a culture where our first impulse is to attack and conquer those parts of the natural world that do not quickly or easily submit to our wishes. When faced with stiff resistance we are more likely to destroy what’s in our way than see the wisdom of bending like reeds in a stiff wind. As we approach 9 billion people and the climate change crisis becomes an ever more imminent threat to our quality of life, that attitude will have to change. Applying a form of actionless action through rewilding offers us a profound example of how we might go about changing it.
Follow Craig on Twitter or read him on 71Republic.com.
Other stories by Craig Axford that you may enjoy include: | https://craig-axford.medium.com/why-taoisms-best-idea-our-most-radical-environmental-restoration-plans-are-made-for-each-other-cbc988658a3 | ['Craig Axford'] | 2018-08-22 00:07:51.345000+00:00 | ['Sustainability', 'Politics', 'Environment', 'Science', 'Philosophy'] | Title Taoism’s Best Idea Radical Environmental Restoration Plans Made OtherContent Photo Owain McGuire Unsplash Lion Sands Game Reserve Sabi Sands South Africa gentlest thing world overcomes hardest thing world substance enters space show value nonaction Teaching without word performing without action Master’s way Tao Te Ching chapter 43 tend think environmental restoration hard work pick shovel ripping invasive plant ground conducting controlled burn well meaning side fondness thinking nature’s master intervention screwed must reintervention restores Often however restoration may involve little pulling plug foolish dam watching water recede letting area recover place ever never going anyway Nature another word change Taoist tradition concept wuwei often translated effortlessactionless action lie core philosophy closely associated water though soft eventually crack erodes even hardest object Rewilding likewise call respecting soft power natural process capacity eventually return even damaged environment back health Neither wuwei rewilding call upon u utterly passive challenge u incorporate careful observation humility patience everything often missing effort dominate nature preserve Even best intention often act rashly hoping quickly fix broken rewilding isn’t Words like rewilding vulnerable misunderstanding say nothing intentional misrepresentation oppose getting rewilding it’s necessary say briefly isn’t Rewilding Luddite movement attempting return u living cave without even much plastic toothbrush name Indoor plumbing electricity even Facebook assuming haven’t already deleted account comfort would still able enjoy rewilded world isn’t say incorporating concept conservation sustainability effort much greater extent wouldn’t impact relate technology However I’m convinced change would enhance human wellbeing happiness rather subtract Indeed there’s even evidence humanity could experience greater abundance rewilding became central value culture Rewilding remains strongly linked reintroduction apex predator ecosystem haven’t seen reintroduction wolf Yellowstone probably best known example kind rewilding effort animal stand top food chain need larger area thrive Therefore reintroduction also requires large tract sufficiently healthy habitat adequate prey population said concept broadened since Yellowstone wolf reintroduction rewilding landscape need include reintroduction example one imagine site illconceived reservoir rewilded simply draining artificial lake allowing nature take course formerly flooded canyon valley slowly flush accumulated silt revegetates expanded concept even include area small backyard urban rooftop tiny patch obviously never inviting habitat mountain lion wolf still function critical refuge bird pollinator otherwise displaced human development context active human management far likely play vital role That’s okay though I’ve written previously playing dirt actually brings wonderful health benefit well planned garden requires muscle maintain good u specie use Colorado River awaiting rewilding opportunity March 1963 diversion tunnel would become US Bureau Reclamation’s controversial project closed moment water Colorado River began slowly back behind Glen Canyon Dam Lake Powell born Located remote section desert UtahArizona border Lake Powell briefly ever reached capacity 50 year existence seepage surrounding sandstone evaporation lake loses nearly one million acre foot water year spite name government agency gave birth never actually reclaimed agricultural use drinking water According Bureau Reclamation’s website “two three percent lake’s water evaporates atmosphere year” addition “About 134 million acre foot 16500 million cubic meter almost third reservoir’s capacity” absorbed porous sandstone Glen Canyon year lake filling absorption continues day Photo Rainer Krienke Unsplash Lake Powell formerly Glen Canyon USA Late David Brower environmental activist pointed canyon submerged beneath water Lake Powell UtahArizona border excellent candidate restoration Even though far largest river Colorado River among overtaxed river world watershed provides water every major metropolitan area Denver Los Angeles irrigates hundred thousand acre agricultural land Given reality much go toward sustaining giant evaporation pond middle remote desert waste human well environmental perspective environmental classic Sand County Almanac Aldo Leopold describes 1922 trip brother took Colorado River Delta you’ve visited delta recently Leopold’s description read like work fiction writes “On map Delta bisected river fact river nowhere everywhere rivercould decide hundred green lagoon offered pleasant least speedy path Gulf” Today considered good year Colorado River make Sea Cortez Even small portion river find way typically fetid briny rivulet contaminated chemical fertilizer hardened mineral concentrated evaporation Efforts recently gotten underway restore small fraction Colorado River Delta former glory December 2014 article National Geographic Magazine described effect “pulse flow” 105000 acre foot water started release Hoover Dam week upstream result release dramatic provide glimpse could happen stronger reliable flow provided regularly Delta 800000 acre foot lost year evaporation absorption Lake Powell immediately come mind possible source renewal “Man always kill thing love pioneer killed wilderness” Aldo Leopold wrote end chapter describing time brother spent exploring Colorado River Delta “Some say to” continued “Be may glad shall never young without wild country young avail forty freedom without blank spot map” may never get Colorado River Delta Aldo Leopold’s youth back sign recovery site benefitting pulse flow 2014 show even receiving seemingly fatal blow ecosystem prove resilient give credit Sandra Postel director Global Water Policy Project put Delta “isn’t dead” “it’s dormant add water come back life” Wildife management actionless action We’ve fallen bad habit seeing natural relationship linear zero sum lens example like eat something another animal also like eat thing killing specie that’s competition u consistently think necessarily increase availability creature like eat idea turn wrong simplistic Wolves coyote whale seal sea otters… case eradication policy backfired removal wolf turned huge deleterious consequence beaver population aspen stand many component ecosystem Overhunting whale ended major negative impact phytoplankton variety fish population rely upon human well whale consume Likewise killing seal protect fish specie consume opposite effect George Manbiot put Guardian “The fisherman insisted predator seal killed might reducing enhancing catch” usually misguided wildlife management policy perhaps best example ancient Taoist concept actionless action would applied tremendous positive consequence Nature operate zero sum rule Predators obviously threat individual prey play critical role maintaining entire ecosystem healthy ecosystem productive unhealthy one leaving better service specie including human take advantage There’s need reintroduce wolf Rockies need allow already occupying Greater Yellowstone Area migrate unmolested ancestor’s old haunt Likewise leaving ocean’s whale seal otter alone longrun likely sufficient facilitate eventual recovery intervention undertake however must purpose restoring specie area longer exist extremely rare Hunting one specie “benefit” another hasn’t paid dividend past isn’t likely future Ecosystems complex u know degree certainty outcome undertake effort Sea urchin staple sea otter diet Sea urchin eat kelp plant matter Eliminating sea otter resulted urchin population explosion devasted kelp forest upon many fish depend Image wikimedia Rewilding local park backyard Personal public habitat much closer home provide excellent opportunity casual hand approach environmental restoration Obviously yard city park never wild sense people use word play key role conservation effort nonetheless Consider Seattle’s Pollinator Pathway project obviously targeting bee butterfly moth pollinator rather ecosystem whole providing needed support creature often struggle urban agricultural setting project increase capacity critical specie absorb environmental stress ultimately thrive Pollinator Pathway run mile along Seattle’s Columbia Street However demonstrates feasibility much larger effort woven city’s infrastructure type project limited imagination urban planner architect willingness local citizen policymakers make reality implemented large enough scale benefit pollinator desperately seeking diverse habitat free large scale pesticide herbicide contamination could reverse decade long population decline According bee expert Noah WilsonRich urban bee already overwinter far better rural bee yeartoyear survival rate bee city beehive 60 rural beehive it’s 40 Though uncertain bee seem fare well city fact remains even crowded human environment play key role recovery insect critical wellbeing 250000 specie flowering plant share planet u Conclusion It’s slight exaggeration say research opportunity extensive rewilding would provide practically endless especially true adopt broad view concept instead narrow absolutist one case urban bee show rewilding something happens along spectrum rather eitheror proposition matter far take rewilding effort given case human always remain component ecosystem attempt integrate even little bit nonhuman world life provide rich opportunity enhance understanding dynamic relationship sustain earth’s biosphere arriving must take leap faith especially difficult culture first impulse attack conquer part natural world quickly easily submit wish faced stiff resistance likely destroy what’s way see wisdom bending like reed stiff wind approach 9 billion people climate change crisis becomes ever imminent threat quality life attitude change Applying form actionless action rewilding offer u profound example might go changing Follow Craig Twitter read 71Republiccom story Craig Axford may enjoy includeTags Sustainability Politics Environment Science Philosophy |
763 | Make Your Working Day Work for You | Make Your Working Day Work for You
Dealing with task switching and interruptions, from a Neuroscience Graduate
Photo by Roman Koval from Pexels
When I was an employee, the worst jobs I had were the ones where my boss or line manager would give me new tasks one after the other, without considering that I was trying to focus on previous tasks first.
As a Neuroscience graduate, I knew how bad task switching was for my brain. It was frustrating to know exactly why my energy was haemorrhaging out of me, but not feeling able to stop it, as scared as I was of confrontation with the boss.
My working days were not working for me.
Today as a freelancer, I organise my working days to avoid task switching – but I do incorporate variety, because that’s good for my brain, my body and my soul.
There is a critical difference that, once understood, takes your focus and energy from being wasted to being utilised well.
It’s worth getting clear first on what task switching really is. It can apply to both scrolling through your phone while on a call (A.K.A. multi-tasking), and suddenly changing from doing one job to another, which is what we more commonly think of as task switching.
The term multi-tasking is actually a misnomer. People can’t actually do more than one task at a time. Instead we switch tasks. So the term that is used in the research is “task switching”.
- Susan Weinschenk Ph.D. in Psychology Today
What ever you call it, it’s all bad for your executive function, in other words, your productive and reasoning-based brain activity. Thoughts, planning, problem-solving — it all suffers when we divide our attention.
Variety in your day, on the other hand, can be thought of as taking sensible breaks as and when you reach an appropriate point to wrap up one task and purposefully move on to something else.
Variety is a good thing for most people. If we performed one type of task all day, we would get bored and our mind would wander, making task switching more likely to occur.
A planned break in your day where you get up from your desk and move, or just simply do something different, can be a welcome break for your brain, and can boost your productivity overall.
Research is showing us that we need to take more breaks to stop our creativity and focus from drying up.
Push yourself through too many hours or days of work and your brain starts to push back. Ideas that once flowed easily dry up, and tasks that you should be able to perform quickly become excruciatingly difficult.
- Minda Zetlin in Inc.
Here are a few simple things you can use, whether working from home or an office, to manage your time for better energy, making your day work for you.
Simple Time Blocking
Time block your days and say no to instantly replying to things.
That ‘multi-tasking’ kind of task switching is sneaky. We see a notification on our phone or desktop, and it’s like an unopened Christmas present just calling to us…
I’ve started ignoring DMs until my first coffee break of the day, so I can maximise my early mornings, when I am most creative.
During my coffee break, I indulge in replying to messages, giving my brain a nice little break from the focus of my early morning writing.
Instrumental-only music
Playing some relaxing or uplifting music can be great for your productivity. But did you know how important it is to listen to instrumentals only while writing?
If there are lyrics in your music then it’s like having somebody whispering in your ear while you are trying to work.
It might not feel like it, but on some level your brain is attending to those words, stealing away brain power.
It’s your brain’s equivalent of having too many tabs open on your browser – shut it down and save those songs for your break time.
Don’t force it
If you are forced to task switch due to an interruption or emergency, allow yourself a little break afterwards before returning to your original task.
Switching back is switching twice, and forcing yourself to do that is only going to add to the cognitive demand on your brain, tiring you out.
Be kinder to your brain and it will repay you with prolonged focus.
These tips helped me when I was studying for my Neuroscience exams several years ago now, and I still use them today as a freelance content writer.
I hope that these tips help you too. If you are working in a way that requires quiet focus, then I know it can be tricky. We can’t always control external circumstances, like interruptions or new urgent tasks from our boss.
But we can do the best with what we have, and that all helps. | https://medium.com/the-innovation/make-your-working-day-work-for-you-ed66e35d9d08 | ['Sarah K Brandis'] | 2020-11-19 17:49:44.103000+00:00 | ['Neuroscience', 'Psychology', 'Work', 'Entrepreneurship', 'Study'] | Title Make Working Day Work YouContent Make Working Day Work Dealing task switching interruption Neuroscience Graduate Photo Roman Koval Pexels employee worst job one bos line manager would give new task one without considering trying focus previous task first Neuroscience graduate knew bad task switching brain frustrating know exactly energy haemorrhaging feeling able stop scared confrontation bos working day working Today freelancer organise working day avoid task switching – incorporate variety that’s good brain body soul critical difference understood take focus energy wasted utilised well It’s worth getting clear first task switching really apply scrolling phone call AKA multitasking suddenly changing one job another commonly think task switching term multitasking actually misnomer People can’t actually one task time Instead switch task term used research “task switching” Susan Weinschenk PhD Psychology Today ever call it’s bad executive function word productive reasoningbased brain activity Thoughts planning problemsolving — suffers divide attention Variety day hand thought taking sensible break reach appropriate point wrap one task purposefully move something else Variety good thing people performed one type task day would get bored mind would wander making task switching likely occur planned break day get desk move simply something different welcome break brain boost productivity overall Research showing u need take break stop creativity focus drying Push many hour day work brain start push back Ideas flowed easily dry task able perform quickly become excruciatingly difficult Minda Zetlin Inc simple thing use whether working home office manage time better energy making day work Simple Time Blocking Time block day say instantly replying thing ‘multitasking’ kind task switching sneaky see notification phone desktop it’s like unopened Christmas present calling us… I’ve started ignoring DMs first coffee break day maximise early morning creative coffee break indulge replying message giving brain nice little break focus early morning writing Instrumentalonly music Playing relaxing uplifting music great productivity know important listen instrumentals writing lyric music it’s like somebody whispering ear trying work might feel like level brain attending word stealing away brain power It’s brain’s equivalent many tab open browser – shut save song break time Don’t force forced task switch due interruption emergency allow little break afterwards returning original task Switching back switching twice forcing going add cognitive demand brain tiring kinder brain repay prolonged focus tip helped studying Neuroscience exam several year ago still use today freelance content writer hope tip help working way requires quiet focus know tricky can’t always control external circumstance like interruption new urgent task bos best helpsTags Neuroscience Psychology Work Entrepreneurship Study |
764 | The key to scaling your business (and maintaining your sanity, too) | I had no idea how to manage or motivate people.
When it was just our first employee and I in the office, we ate lunch together every day. We talked and shared ideas, and I discovered what made him feel productive and engaged.
Slowly, I learned how to collaborate, while also serving as “the boss.” It was on-the-job leadership training.
To be clear, I’ve also made a ton of management mistakes. I’ve hired the wrong people. I’ve sometimes failed to provide what my employees needed.
And the first time I had to fire someone was incredibly scary. But, without investor pressure or an instant team of 60, my mistakes were small. They didn’t have huge repercussions.
When a majorly-funded company like Medium has to lay off a third of their staff, that’s an amplified mistake. It deeply affects lives and businesses. There’s real freedom in learning slowly and growing with your team.
3. You can develop an effective onboarding style
Every new JotForm hire goes through Bootcamp. I can’t take credit for this idea, because we stole it from Facebook and Zappos, but here’s how it works in our business:
In their first month, new employees handle at least 100 customer support requests.
It’s the best way to understand our users, their struggles, and their product needs. It also ensures that new hires know every nook and cranny of the software.
After all, you can’t help someone if you don’t understand the product’s inner workings — and most employees say that this hands-on support time proved to be invaluable.
New team members also serve as co-pilots. They spend at least two hours a day, for a couple of weeks, working alongside another developer or designer or marketer.
They watch and ask questions. They see firsthand how our employees tackle their roles. Not only do new people learn important skills, but the process quietly strengthens our culture.
New team members (hopefully) absorb values like openness and honesty and a willingness to take risks.
Our Ankara office is on a local university campus. It’s a pretty beautiful place to work, so I also try to go for long walks with our new employees and take them to lunch. In San Francisco office, we take long walks on the Embarcadero taking in the relaxing bay view. I share stories about JotForm and we get to know each other.
Relating to people on a human level matters to me, and it uncovers what they’re hoping to achieve in our company.
Sometimes we’ll hire someone for one position and find they’re better suited to a different role.
I wouldn’t be able to spend this kind of one-on-one time with new employees if I had to go on a massive hiring spree.
4. You can establish deep cultural roots
Startup culture is such a hot topic.
Many successful entrepreneurs say that culture is the foundation of their companies. I didn’t always understand why, but I do now.
JotForm has a culture of continuous improvement. I’ve been heavily influenced by the lean startup philosophy and leaders like Eric Ries.
Many software companies have monthly or six-month release cycles, but we try to release product changes immediately.
It can be a source of culture shock for new employees, but they soon embrace the chance to directly affect our product.
It’s empowering. And it shows that their work will make a difference.
We’re so committed to continuous improvement that we ask new hires to release a product update on their very first day.
Sometimes they’re apprehensive, but we try to ensure everything is prepared and we’re ready for their input. We streamline their onboarding and provide a challenge that can lead to immediate changes.
The beaming smiles we see at the end of the day are an awesome reward.
I love the sense of pride that hands-on action can create. Even when we hit a roadblock, we want our team to know that they are here to innovate and learn.
They can make meaningful contributions. I’m not saying we’re perfect, but it’s a technique we’ve developed to nurture our culture, and it took some time to get here.
5. You can cultivate strong, productive teams
Our company works in cross-functional teams.
Instead of creating silos, we have 5–6-member teams comprised of a developer, designer, UI expert, data scientist, and any other roles the group needs.
Most new hires tell me that they haven’t worked this way before, but I think it promotes better collaboration.
Each project has multi-functional input from start to finish, which means that every angle is considered. Small bugs or mistakes don’t multiply into major challenges.
An agile workflow also brings everyone together in a large space to share our screens and demonstrate what we’re working on. Everyone contributes.
We encourage each team member to comment, ask questions, and suggest new ideas.
Slow team growth can be your competitive advantage
I’ve learned what works for JotForm, but what works best for you?
Developing the right style takes time.
Bootstrapping gives you the space and freedom to do it your way and grow on your schedule.
I’ve listened to my employees over the past 13 years and learned what they want and need. I can’t say I always deliver, but I try — and that has also helped us to attract amazing people in a competitive market.
We believe in providing an amazing product. We also believe that everyone deserves to live an amazing life. We work hard to make our customers happy, but our company doesn’t glorify the 24/7 hustle.
We make mistakes and we try to fix them. We’re always trying to improve.
Oh, and I still do some customer support now and then, but I usually leave that to the professionals on our team. They’re much better at it than me.
After 13 years in business, that’s a great feeling. | https://medium.com/swlh/the-key-to-scaling-your-business-and-maintaining-your-sanity-too-2f2fff99e556 | ['Aytekin Tank'] | 2019-06-12 14:17:43.725000+00:00 | ['Culture', 'Psychology', 'Startup', 'Entrepreneurship', 'Self Improvement'] | Title key scaling business maintaining sanity tooContent idea manage motivate people first employee office ate lunch together every day talked shared idea discovered made feel productive engaged Slowly learned collaborate also serving “the boss” onthejob leadership training clear I’ve also made ton management mistake I’ve hired wrong people I’ve sometimes failed provide employee needed first time fire someone incredibly scary without investor pressure instant team 60 mistake small didn’t huge repercussion majorlyfunded company like Medium lay third staff that’s amplified mistake deeply affect life business There’s real freedom learning slowly growing team 3 develop effective onboarding style Every new JotForm hire go Bootcamp can’t take credit idea stole Facebook Zappos here’s work business first month new employee handle least 100 customer support request It’s best way understand user struggle product need also ensures new hire know every nook cranny software can’t help someone don’t understand product’s inner working — employee say handson support time proved invaluable New team member also serve copilot spend least two hour day couple week working alongside another developer designer marketer watch ask question see firsthand employee tackle role new people learn important skill process quietly strengthens culture New team member hopefully absorb value like openness honesty willingness take risk Ankara office local university campus It’s pretty beautiful place work also try go long walk new employee take lunch San Francisco office take long walk Embarcadero taking relaxing bay view share story JotForm get know Relating people human level matter uncovers they’re hoping achieve company Sometimes we’ll hire someone one position find they’re better suited different role wouldn’t able spend kind oneonone time new employee go massive hiring spree 4 establish deep cultural root Startup culture hot topic Many successful entrepreneur say culture foundation company didn’t always understand JotForm culture continuous improvement I’ve heavily influenced lean startup philosophy leader like Eric Ries Many software company monthly sixmonth release cycle try release product change immediately source culture shock new employee soon embrace chance directly affect product It’s empowering show work make difference We’re committed continuous improvement ask new hire release product update first day Sometimes they’re apprehensive try ensure everything prepared we’re ready input streamline onboarding provide challenge lead immediate change beaming smile see end day awesome reward love sense pride handson action create Even hit roadblock want team know innovate learn make meaningful contribution I’m saying we’re perfect it’s technique we’ve developed nurture culture took time get 5 cultivate strong productive team company work crossfunctional team Instead creating silo 5–6member team comprised developer designer UI expert data scientist role group need new hire tell haven’t worked way think promotes better collaboration project multifunctional input start finish mean every angle considered Small bug mistake don’t multiply major challenge agile workflow also brings everyone together large space share screen demonstrate we’re working Everyone contributes encourage team member comment ask question suggest new idea Slow team growth competitive advantage I’ve learned work JotForm work best Developing right style take time Bootstrapping give space freedom way grow schedule I’ve listened employee past 13 year learned want need can’t say always deliver try — also helped u attract amazing people competitive market believe providing amazing product also believe everyone deserves live amazing life work hard make customer happy company doesn’t glorify 247 hustle make mistake try fix We’re always trying improve Oh still customer support usually leave professional team They’re much better 13 year business that’s great feelingTags Culture Psychology Startup Entrepreneurship Self Improvement |
765 | 3 Engineering Mistakes That Will Botch Scale | 1. Too Much Focus on What Was There Before
Legacy code haunts every engineer. We tell ourselves, “Don’t touch it, it works” as the systems grow outdated. New exploits loom darkly. Once-enthusiastic users get bored with the same old experiences. We convince ourselves to “respect” what was there because “It was the right thing to build at the time.” We lock in system invariants that produce the same old buggy behaviors. We implicity consent to the old ways. We bask in our past successes or glory. The aging systems even become cultural traditions, with senior folks explaining:
“That’s just how we do it here.”
This attitude ruins our ability to scale because it inhibits our ability to meet the growing and changing demands presented by dynamic markets. The future of the business chokes on its past. At first, it goes completely unnoticed. Inevitably, it becomes a function of success itself: Every past success increasingly damping the next future success.
Legacy code distracts us from what the system should do.
As we reverse-engineer the legacy, we forget why we were looking at it in the first place. When we finally remember, we are more willing to distort our end-goal to fit what we already have, sitting right in front of us. “It’s half-built already,” we lie to ourselves. And just like that, our past success muddies our future direction.
The quality required to break out of this toxic spiral is bravery. | https://medium.com/better-programming/3-toxic-engineering-attitudes-that-botch-scale-709b0672747 | ['Jon Mclachlan'] | 2020-08-31 15:20:21.417000+00:00 | ['Software Development', 'Business', 'Engineering', 'Startup', 'Software Engineering'] | Title 3 Engineering Mistakes Botch ScaleContent 1 Much Focus Legacy code haunt every engineer tell “Don’t touch works” system grow outdated New exploit loom darkly Onceenthusiastic user get bored old experience convince “respect” “It right thing build time” lock system invariant produce old buggy behavior implicity consent old way bask past success glory aging system even become cultural tradition senior folk explaining “That’s here” attitude ruin ability scale inhibits ability meet growing changing demand presented dynamic market future business choke past first go completely unnoticed Inevitably becomes function success Every past success increasingly damping next future success Legacy code distracts u system reverseengineer legacy forget looking first place finally remember willing distort endgoal fit already sitting right front u “It’s halfbuilt already” lie like past success muddies future direction quality required break toxic spiral braveryTags Software Development Business Engineering Startup Software Engineering |
766 | How to Create Custom Video Controls in React Native | Requirements
You should be familiar with JavasScript and have some basic knowledge of React and React Native. I am going to be using Expo to initialize the app and I’ll be using their Video component. I am running this code on my iOS device using the Expo app. The same general principles should apply if you want to use a base React Native build with React Native Video. Of course the methods will probably be a bit different so be sure to read up on their documentation.
Setting Up
To begin let’s create a new app with the expo-cli:
expo init custom-video-controls
Select Bare from the JavaScript managed workflow.
Once the app is created, navigate into the base directory and install expo-av and React Native Elements.
expo install expo-av react-native-elements
Expo AV is a package provided by Expo that gives us access to their Video and Audio components. React Native Elements is a component library that will allow us to easily create some customizable controls.
Once that finishes installing open App.js and the code with this:
import React from 'react';
import { Video } from 'expo-av';
import { StatusBar } from 'expo-status-bar';
import {
View,
SafeAreaView,
StyleSheet,
Dimensions,
} from 'react-native';
export default function App() {
return (
<SafeAreaView style={styles.container}>
<StatusBar style="auto" />
<View>
<Video
source={{
uri: 'http://url/to/video',
}} rate={1.0}
volume={1.0}
isMuted={false}
resizeMode="cover"
shouldPlay
isLooping
style={styles.video}
/>
</View>
</SafeAreaView>
);
const styles = StyleSheet.create({
container: {
flex: 1,
alignItems: 'center',
},
video: {
width: Dimensions.get('screen').width,
height: Dimensions.get('screen').width * 0.5625,
},
});
Now that we have the initial set up we can start adding the custom controls. To do that we are going to import Icon from React Native Elements:
/* ... other imports*/
import {Icon} from 'react-native-elements'
/* ... component*/
We are going to create a new view just below the video element that will contain all of our controls. Inside of that we will add a play/pause button, a skip forward button and a skip backward button.
<View>
<Video
source={{
uri: 'http://d23dyxeqlo5psv.cloudfront.net/big_buck_bunny.mp4',
}}
rate={1.0}
volume={1.0}
isMuted={false}
resizeMode="cover"
shouldPlay
isLooping
style={styles.video}
/>
<View style={styles.buttonContainer}>
// the replay button
<TouchableOpacity>
<Icon name="replay-10" size={50} />
</TouchableOpacity>
// play/pause button
<TouchableOpacity>
<Icon name="pause" size={50} />
</TouchableOpacity>
// skip foward
<TouchableOpacity>
<Icon name="forward-10" size={50} />
</TouchableOpacity>
</View>
</View>
const styles = StyleSheet.create({
/* ... other styles ... */
buttonContainer: {
flexDirection: 'row',
justifyContent: 'space-evenly',
},
});
Now we have some touchable buttons. Lets go through each one.
Playing and Pausing
In React Native we cannot use the Document Object to target specific elements on the screen. Instead we have to use a React hook called useRef. First let’s import it into the file.
import React, { useRef } from 'react';
Then we need to define the ref.
export default function App() {
const videoRef = useRef(null)
return (
/* ... */
<Video
ref={videoRef}
source={{
uri: 'http://d23dyxeqlo5psv.cloudfront.net/big_buck_bunny.mp4',
}}
rate={1.0}
volume={1.0}
isMuted={false}
resizeMode="cover"
shouldPlay
isLooping
style={styles.video}
/>
/* ... */
)
}
Now we can use the videoRef variable to refer to the video element. This will allow us to utilize all of the methods specified in the Video docs.
Now let’s create a function that will toggle play and pause:
function togglePlayPause() {
videoRef.current.pauseAsync();
}
And let’s add this to the onPress prop of the play/pause button:
<TouchableOpacity onPress={togglePlayPause}>
<Icon name="pause" size={50} />
</TouchableOpacity>
Now when you press the button the video should pause. But we can’t play the video so let’s add some logic to toggle between play and pause.
First we need to create some state to track if the video is playing or not. Let’s import the useState hook from react.
import React, {useRef, useState} from 'react';
Next add playing state to the top of our component and have it default to false. This is good to do in case something prevents the video from playing automatically; we won’t have any UI bugs or inconsistencies on the screen.
export default function App() {
const [playing, setPlaying] = useState(false)
const videoRef = useRef(null)
/* ... */
}
Now we’re going to use a prop on the Video component called onPlaybackStatusUpdate that gets called regularly and will give us the playback status of the video. This is will allow us to see if the video is playing, if it has ended, and what our current position in the video is. For now we will just check if it is playing.
Let’s create a funtion for handling the status update.
function handlePlayBackStatusUpdate(e) {
if (e.isPlaying && !playing){
setPlaying(true)
}
if (!e.isPlaying && playing) {
setPlaying(false)
}
}
handlePlayBackStatusUpdate receives an object that represents the current status of the video. We check if it’s playing and if the playing state is false. If both of these return true we set playing to true. Next we check if the video is not playing and if playing is true. If both return true we set playing to false. The reason we check both e.isPlaying and playing is to avoid unnecessary calls to update playing. If the video is not playing and playing is already false there is no reason to call setPlaying(false).
Now let’s add this function to the onPlaybackStatusUpdate prop of the Video element.
<Video
ref={videoRef}
source={{
uri: 'http://d23dyxeqlo5psv.cloudfront.net/big_buck_bunny.mp4',
}}
rate={1.0}
volume={1.0}
isMuted={false}
resizeMode="cover"
shouldPlay
isLooping
style={styles.video}
onPlaybackStatusUpdate={handlePlaybackStatusUpdate}
/>
Then expand the togglePlayPause function to pause the video if it’s playing or play it if it’s paused.
function togglePlayPause() {
playing
? videoRef.current.pauseAsync()
: videoRef.current.playAsync();
}
Finally conditionally render a pause icon if the video is playing or a play icon if the video is paused.
<TouchableOpacity onPress={togglePlayPause}>
{playing ? (
<Icon name="pause" size={50} />
) : (
<Icon name="play-arrow" size={50} />
)}
</TouchableOpacity>
Now you should be able to toggle between play and pause by pressing the button!
Skipping Forward and Backward
Next create a function that will skip forward and backward.
async function skip(bool) {
const status = await videoRef.current.getStatusAsync();
const curPos = status.positionMillis;
const tenSeconds = 10000;
const newPos = bool ? curPos + tenSeconds : curPos - tenSeconds;
videoRef.current.setPositionAsync(newPos);
}
getStatusAsync returns an object that represents the current status of the video. Either add or remove 10 seconds from the current position to get our new position. Finally use the setPositionAsync method to set the new position of the video.
Add this function to the onPress prop of the skip buttons.
<TouchableOpacity onPress={() => skip(false)}>
<Icon name="replay-10" size={50} />
</TouchableOpacity>
/* ... */
<TouchableOpacity onPress={() => skip(true)}>
<Icon name="forward-10" size={50} />
</TouchableOpacity>
Now we should be able to play or pause the video, and skip the video forward and backwards. There are many other methods we can utilize to add additional controls and functionality but these are the basics.
Creating the Duration Bar
The last thing I want to touch on is the duration bar. For this tutorial I am going to use the slider component provided by React Native Elements.
Let’s wrap our current controls in a new View and add the slider to the bottom:
/* ... */
import { Icon, Slider } from 'react-native-elements';
/* ... */
<View>
<View style={styles.buttonContainer}>
<TouchableOpacity onPress={() => skip(false)}>
<Icon name="replay-10" size={50} />
</TouchableOpacity>
<TouchableOpacity onPress={togglePlayPause}>
{playing ? (
<Icon name="pause" size={50} />
) : (
<Icon name="play-arrow" size={50} />
)}
</TouchableOpacity>
<TouchableOpacity onPress={() => skip(true)}>
<Icon name="forward-10" size={50} />
</TouchableOpacity>
</View>
<Slider
value={position}
maximumValue={duration}
onSlidingComplete={handleDoneSliding}
/>
</View>
If you look closely you can see we have added some variables we need to define: position, duration, and handleDoneSliding.
Position is how we will track the current position of the video. We will use the onStatusUpdate to keep track of this.
Duration is the total length of the video.
HandleDoneSliding is a function that will set the position of the video after a user slides the thumb of the slider.
First let’s define postion and duration at the top of our component :
export default function App() {
const [duration, setDuration] = useState(0);
const [position, setPosition] = useState(0);
/* ... */
}
Update the handleStatusUpdate function to save the duration and the position of the video.
function handlePlaybackStatusUpdate(e) {
if (e.isPlaying && !playing) {
setPlaying(true);
}
if (!e.isPlaying && playing) {
setPlaying(false);
}
if (e.isPlaying) {
setPosition(e.positionMillis);
}
if (duration === 0) {
setDuration(e.durationMillis);
}
}
Now if the video is playing we set the current position. If the duration state is equal to 0 we set it to the duration in milliseconds of the video.
Now if you play the video you should see the slider moving as the video progresses.
Next let’s define handleDoneSliding:
function handleDoneSliding(value) {
setPosition(value);
videoRef.current.setPositionAsync(value);
}
As you can see above we have assigned handleDoneSliding to the onSlidingComplete prop on the Slider component. This prop passes the current position in milliseconds to the function which saves it to our position state and manually sets the position of the video.
This is what your App.js file should look like: | https://medium.com/dev-genius/how-to-create-custom-video-controls-in-react-native-f8d28ce22052 | ['Freedom Evenden'] | 2020-12-23 10:20:09.077000+00:00 | ['Mobile App Development', 'JavaScript', 'React Native', 'Development', 'React'] | Title Create Custom Video Controls React NativeContent Requirements familiar JavasScript basic knowledge React React Native going using Expo initialize app I’ll using Video component running code iOS device using Expo app general principle apply want use base React Native build React Native Video course method probably bit different sure read documentation Setting begin let’s create new app expocli expo init customvideocontrols Select Bare JavaScript managed workflow app created navigate base directory install expoav React Native Elements expo install expoav reactnativeelements Expo AV package provided Expo give u access Video Audio component React Native Elements component library allow u easily create customizable control finish installing open Appjs code import React react import Video expoav import StatusBar expostatusbar import View SafeAreaView StyleSheet Dimensions reactnative export default function App return SafeAreaView stylestylescontainer StatusBar styleauto View Video source uri httpurltovideo rate10 volume10 isMutedfalse resizeModecover shouldPlay isLooping stylestylesvideo View SafeAreaView const style StyleSheetcreate container flex 1 alignItems center video width Dimensionsgetscreenwidth height Dimensionsgetscreenwidth 05625 initial set start adding custom control going import Icon React Native Elements import import Icon reactnativeelements component going create new view video element contain control Inside add playpause button skip forward button skip backward button View Video source uri httpd23dyxeqlo5psvcloudfrontnetbigbuckbunnymp4 rate10 volume10 isMutedfalse resizeModecover shouldPlay isLooping stylestylesvideo View stylestylesbuttonContainer replay button TouchableOpacity Icon namereplay10 size50 TouchableOpacity playpause button TouchableOpacity Icon namepause size50 TouchableOpacity skip foward TouchableOpacity Icon nameforward10 size50 TouchableOpacity View View const style StyleSheetcreate style buttonContainer flexDirection row justifyContent spaceevenly touchable button Lets go one Playing Pausing React Native cannot use Document Object target specific element screen Instead use React hook called useRef First let’s import file import React useRef react need define ref export default function App const videoRef useRefnull return Video refvideoRef source uri httpd23dyxeqlo5psvcloudfrontnetbigbuckbunnymp4 rate10 volume10 isMutedfalse resizeModecover shouldPlay isLooping stylestylesvideo use videoRef variable refer video element allow u utilize method specified Video doc let’s create function toggle play pause function togglePlayPause videoRefcurrentpauseAsync let’s add onPress prop playpause button TouchableOpacity onPresstogglePlayPause Icon namepause size50 TouchableOpacity press button video pause can’t play video let’s add logic toggle play pause First need create state track video playing Let’s import useState hook react import React useRef useState react Next add playing state top component default false good case something prevents video playing automatically won’t UI bug inconsistency screen export default function App const playing setPlaying useStatefalse const videoRef useRefnull we’re going use prop Video component called onPlaybackStatusUpdate get called regularly give u playback status video allow u see video playing ended current position video check playing Let’s create funtion handling status update function handlePlayBackStatusUpdatee eisPlaying playing setPlayingtrue eisPlaying playing setPlayingfalse handlePlayBackStatusUpdate receives object represents current status video check it’s playing playing state false return true set playing true Next check video playing playing true return true set playing false reason check eisPlaying playing avoid unnecessary call update playing video playing playing already false reason call setPlayingfalse let’s add function onPlaybackStatusUpdate prop Video element Video refvideoRef source uri httpd23dyxeqlo5psvcloudfrontnetbigbuckbunnymp4 rate10 volume10 isMutedfalse resizeModecover shouldPlay isLooping stylestylesvideo onPlaybackStatusUpdatehandlePlaybackStatusUpdate expand togglePlayPause function pause video it’s playing play it’s paused function togglePlayPause playing videoRefcurrentpauseAsync videoRefcurrentplayAsync Finally conditionally render pause icon video playing play icon video paused TouchableOpacity onPresstogglePlayPause playing Icon namepause size50 Icon nameplayarrow size50 TouchableOpacity able toggle play pause pressing button Skipping Forward Backward Next create function skip forward backward async function skipbool const status await videoRefcurrentgetStatusAsync const curPos statuspositionMillis const tenSeconds 10000 const newPos bool curPos tenSeconds curPos tenSeconds videoRefcurrentsetPositionAsyncnewPos getStatusAsync return object represents current status video Either add remove 10 second current position get new position Finally use setPositionAsync method set new position video Add function onPress prop skip button TouchableOpacity onPress skipfalse Icon namereplay10 size50 TouchableOpacity TouchableOpacity onPress skiptrue Icon nameforward10 size50 TouchableOpacity able play pause video skip video forward backwards many method utilize add additional control functionality basic Creating Duration Bar last thing want touch duration bar tutorial going use slider component provided React Native Elements Let’s wrap current control new View add slider bottom import Icon Slider reactnativeelements View View stylestylesbuttonContainer TouchableOpacity onPress skipfalse Icon namereplay10 size50 TouchableOpacity TouchableOpacity onPresstogglePlayPause playing Icon namepause size50 Icon nameplayarrow size50 TouchableOpacity TouchableOpacity onPress skiptrue Icon nameforward10 size50 TouchableOpacity View Slider valueposition maximumValueduration onSlidingCompletehandleDoneSliding View look closely see added variable need define position duration handleDoneSliding Position track current position video use onStatusUpdate keep track Duration total length video HandleDoneSliding function set position video user slide thumb slider First let’s define postion duration top component export default function App const duration setDuration useState0 const position setPosition useState0 Update handleStatusUpdate function save duration position video function handlePlaybackStatusUpdatee eisPlaying playing setPlayingtrue eisPlaying playing setPlayingfalse eisPlaying setPositionepositionMillis duration 0 setDurationedurationMillis video playing set current position duration state equal 0 set duration millisecond video play video see slider moving video progress Next let’s define handleDoneSliding function handleDoneSlidingvalue setPositionvalue videoRefcurrentsetPositionAsyncvalue see assigned handleDoneSliding onSlidingComplete prop Slider component prop pass current position millisecond function save position state manually set position video Appjs file look likeTags Mobile App Development JavaScript React Native Development React |
767 | How to Delete Your Carbon Footprint | You’ve got mail! It used to be so exciting and romantic to get an email. At least Tom Hanks and Meg Ryan made it seem that way. The somewhat prophetic plot of that movie was built around a corporate giant putting a boutique bookstore out of business. Since the founding of the first American e-commerce company in 1969 (CompuServe), the world has been transitioning more and more economic activity to the internet. At the same time, more and more of our hours are being spent in front of a computer consuming information and interacting with other people online. As a result, our computers, and the infrastructure supporting them, have been designed to do more for us. Perhaps the most significant capability development in the 21st century has been the ever-improving ability to store all of our information.
In the old days, when you wanted to store a communication you received from someone, you simply put the letter in a box or a drawer. Then later, if you decided you didn’t need to keep the letter, you would simply throw the letter away (recycle it). Similarly, before the 21st century, when you received a receipt, you either kept it for your accounting or threw it away (recycled it). However, today, when we receive emails and receipts, most of us just let them sit in our inboxes… forever.
This is a problem for the environment.
While the internet only represents 2% of the global emissions portfolio, the continued growth in e-commerce and streaming services will inherently mean the energy use associated with the internet will grow significantly. Companies like Google have taken on this challenge by increasing energy efficiency and renewable energy procurement. They have also deployed machine learning initiatives specifically designed to save energy. The buying power of internet giants also gives them the unique ability to influence electricity providers. So, there are clearly top down strategies to reduce the carbon footprint of the internet.
But what about the bottom up grassroots strategies? What can individuals do to reduce the carbon footprint of the internet?
It’s easy. Just clean out your inbox. | https://medium.com/an-idea/how-to-delete-your-carbon-footprint-71cec5dcb596 | ['Carbon Radio'] | 2020-11-20 12:57:46.849000+00:00 | ['Sustainability', 'Design', 'Life', 'Climate Change', 'Environment'] | Title Delete Carbon FootprintContent You’ve got mail used exciting romantic get email least Tom Hanks Meg Ryan made seem way somewhat prophetic plot movie built around corporate giant putting boutique bookstore business Since founding first American ecommerce company 1969 CompuServe world transitioning economic activity internet time hour spent front computer consuming information interacting people online result computer infrastructure supporting designed u Perhaps significant capability development 21st century everimproving ability store information old day wanted store communication received someone simply put letter box drawer later decided didn’t need keep letter would simply throw letter away recycle Similarly 21st century received receipt either kept accounting threw away recycled However today receive email receipt u let sit inboxes… forever problem environment internet represents 2 global emission portfolio continued growth ecommerce streaming service inherently mean energy use associated internet grow significantly Companies like Google taken challenge increasing energy efficiency renewable energy procurement also deployed machine learning initiative specifically designed save energy buying power internet giant also give unique ability influence electricity provider clearly top strategy reduce carbon footprint internet bottom grassroots strategy individual reduce carbon footprint internet It’s easy clean inboxTags Sustainability Design Life Climate Change Environment |
768 | The Rule of Cool | And don’t even get me started on disintegrations. Credit: Lucasfilm
The Rule of Cool
We love mysterious characters. Why then do creators feel compelled to ruin the mystique
Admit it, Boba Fett was way cooler before.
Before we knew he was the original clone, brother to 200,000, with a million more on the way. Before we knew his surname came from his proto-father Jango Fett, the clone template himself. Before we’d seen him as snot-nosed kid laughing at his dad’s jokes.
Before. When he was a masked bad ass with a rocket pack and an assortment of nifty weaponry. A savvy hunter that saw through Han’s float-away-with-the-garbage gambit. A killer so remorseless Darth Vader singled him out amid a cadre of the galaxy’s deadliest to sternly warn that disintegrations were off the table.
Not even an unceremonious end in the Sarlaac’s belly could diminish him. He was so cool we denied his death, a sentiment bolstered by comics and novels.
Boba Fett is the Elvis of the Star Wars universe.
At least, he was, back when an air of mystery draped his shoulders as tangibly as the braid of wookiee hair and the bullet-riddled half-cape.
We see this same pattern time and again in pop culture: interesting side character is introduced as part of a larger story, side character captures the imagination, creator destroys the mystique by retroactively telling their story. AKA, the dreaded origin story.
This is not to be conflated with the superhero origin story, a necessary but often masturbatory part of the genre that precedes the really good stuff. Though I would argue that Wolverine is far more interesting prior to Wolverine: Origins than he is after.
Star Wars is king at this, shining a light into the dark corners of the franchise, giving everyone a name and background.
Is Han Solo a better character now that we’ve seen how he meets Chewie, wins the Millenium Falcon, or completes the Kessel Run? Or that the Solo name came via a bored Imperial recruiter? Emphatically no. Fortunately, Han emerges with his mojo intact because the movie largely avoids recasting Han in a different light from what we’ve seen previously.
The Prequel Trilogy is the ultimate origin story, a three-movie saga conceived solely to depict Anakin’s dark side osmosis, culminating in his conversion to Darth Vader. And, just as in Fett, Vader’s dark shadow is lessened for it. The Phantom Menace establishes Anakin as a figure of prophecy, reframing his decision to throw Palpatine down the shaft in Return of the Jedi as something preordained and less a father’s final act of mercy.
The Prequels change Vader from a murderous villain redeemed by the love of his son into a misunderstood man prone to emotional outbursts and horrible decisions who finally sees the light.
Something is irrevocably lost, each time a creator goes back. | https://medium.com/fan-fare/the-rule-of-cool-2b7f2ea82dd9 | ['Eric Pierce'] | 2019-05-19 11:43:50.265000+00:00 | ['Star Wars', 'Writing', 'Film', 'Books', 'Creativity'] | Title Rule CoolContent don’t even get started disintegration Credit Lucasfilm Rule Cool love mysterious character creator feel compelled ruin mystique Admit Boba Fett way cooler knew original clone brother 200000 million way knew surname came protofather Jango Fett clone template we’d seen snotnosed kid laughing dad’s joke masked bad as rocket pack assortment nifty weaponry savvy hunter saw Han’s floatawaywiththegarbage gambit killer remorseless Darth Vader singled amid cadre galaxy’s deadliest sternly warn disintegration table even unceremonious end Sarlaac’s belly could diminish cool denied death sentiment bolstered comic novel Boba Fett Elvis Star Wars universe least back air mystery draped shoulder tangibly braid wookiee hair bulletriddled halfcape see pattern time pop culture interesting side character introduced part larger story side character capture imagination creator destroys mystique retroactively telling story AKA dreaded origin story conflated superhero origin story necessary often masturbatory part genre precedes really good stuff Though would argue Wolverine far interesting prior Wolverine Origins Star Wars king shining light dark corner franchise giving everyone name background Han Solo better character we’ve seen meet Chewie win Millenium Falcon completes Kessel Run Solo name came via bored Imperial recruiter Emphatically Fortunately Han emerges mojo intact movie largely avoids recasting Han different light we’ve seen previously Prequel Trilogy ultimate origin story threemovie saga conceived solely depict Anakin’s dark side osmosis culminating conversion Darth Vader Fett Vader’s dark shadow lessened Phantom Menace establishes Anakin figure prophecy reframing decision throw Palpatine shaft Return Jedi something preordained le father’s final act mercy Prequels change Vader murderous villain redeemed love son misunderstood man prone emotional outburst horrible decision finally see light Something irrevocably lost time creator go backTags Star Wars Writing Film Books Creativity |
769 | Data Storytelling: Appealing to Logic and Perspective | Data Visualization — Part 7
Data Storytelling: Appealing to Logic and Perspective
Data Storytelling provides a presenter with the ability to appeal to logic and perspective at the same time.
Leveraging storytelling concepts helps to communicate effectively with data because data (numbers and non-emotional information) connect an audience to logic, and stories connect people to reason or perspective.
Presenting information that bridges logic and perspective can help an audience recall the information they’ve seen and, also, trust it. If an audience cannot remember what they’ve just seen and heard, the story didn’t work, and you won’t meet your goals. If an audience doesn’t trust the information, they’ll either ignore it or work counter to it — a failure either way you look at it.
To tell a good story that people remember, you first need their attention. What Shahram Heshmat Ph.D. says in an article in Psychology Today is, “Emotional intensity acts to narrow the scope of attention so that a few objects are emphasized at the expense of many others.” This idea works for data storytelling because it’s intended to take a broad set of information and drill into a smaller or single point.
If I look at a data story for 12-seconds and then look away, what will I recall from the text and visuals?
Take a look at the story above.
Beginning, Middle, and End
At its most basic structure, a story is what storytellers have learned to describe as a beginning, middle, and end. One way to structure and link these three parts is by using plot points. Plot points flush out things such as motivations, desires, or setbacks.
Beginning: Regional exports in Chicago are strong, more robust than other big cities. Chicago’s job market is diverse, but overall economic growth has slowed. Why?
Plot point: Define the setback that connects us to the middle. Looking at the story example above, I assume that the “extensive assets” might be one reason a setback has persisted.
Middle: Using what the audience knows about the good and bad background of the situation, how did the “setback” align with opportunity or failure. Maybe it could go either way depending on the future actions of the intended audience reviewing this story.
Plot point: The plot point that connects the middle to the end might be the motivation to reduce debt or find unique resources.
End: The end is a forecast of the future, showing what the data would look like if everyone “does the right thing” and maybe a simulation of data if they choose the “wrong path.”
The three-act structure can be an outline or model to weave the story through. Conflict or tension (drama) is an essential part of a story because a feeling will trigger a memory, helping to link ideas with urgency.
In the example above, it seems that all was going well. The data was moving in favor of Chicago’s economic strength, but then … a setback. Those damn “extensive assets.” What will we do?
Building the Story
The beginning helps the audience understand the context of what the situation or point is. This should spark interest by answering the questions, “Why do I care?” and “What’s in it for me?”
The middle moves people from context to the implications of the context. Because the situation is “x,” it’s lead us to a state of “x,” and it’s even more substantial when implications of the context can connect to the audience as the key to the future of “x.”
The end of a story should lead people to understand why you brought them through this journey. Will you close with a call to action, something subtle that suggests a behavior change? Does the presentation’s goal call for a direct request, asking for time, money, or resources?
If you are presenting a data story for fun that you want to spend 40-hours building, which everyone will forget in 4-seconds, post to Tableau Public, or have fun in a MakeoverMonday Challenge. If you’re going to create something that has a return on investment and moves a goal forward, there should be a clear message at the end.
Facts are not Interesting
Suppose you reduce the Chicago example, above, to facts. It’s not engaging, visual, or memorable. It might be unclear if presented in a 30-page excel file.
And despite what you might think, turning a 30-page excel file into a story is not the easiest route either. Storytelling takes creativity, and it can even be harder to create than outlining rows of information.
Still, if the data you have requires a data story to digest and transform from information into action, it will be worth the effort.
Reviewing Your Story Before Presenting to Others
Try outlining a set of questions based on your goal, which you want the audience to answer after being presented with your data story. These questions will help you understand what’s missing because if, for example, you cannot walk away knowing what the issue is, the story fell apart somewhere.
After your presentation or reviewing your data story dashboard, will they be able to answer:
What was the problem?
Why did the problem occur?
What can be done to fix the issue or restore balance?
If I (audience) ignore this data, what will happen to me or those close to me?
What did the designer/presenter want me to care about the most?
Do I believe this information?
Do I need more information to decide to act, which will take time and resources that will delay my action or even change my mind?
A standard set of rules does not define a narrative structure for data storytelling. Narrative flow depends on your audience, the purpose of the presentation, and the data you have, or do not have, access to.
How much time will you have with your audience, and do they already trust you, or will it need to be established? Are they familiar with the topic, or will you need to layout a more extensive background and context? The narrative flow will align with answers to the questions above.
Facts presented alone do not always stick, and they rarely compel audiences into doing anything, but facts do help people logically believe a point. Stories lend structure to facts, leveraging the potential to give information meaning and actionable purpose. Data Storytelling provides a presenter with the ability to appeal to logic and perspective at the same time.
Part 1 — Context and Pre-Data Visualization Planning
Part 2 — Selecting Effective Visuals for Data Visualizations
Part 3- Identifying and Eliminating Design Clutter From Data Visualizations
Part 4 — Data Visualization: Designing for Attention
Part 5 — Fundamentals of Good Design for Data Storytelling
Part 6 — Good Data Visualization Examples | https://medium.com/the-innovation/data-storytelling-appealing-to-logic-and-perspective-a6a6831ffe98 | ['Kristi Pelzel'] | 2020-12-10 20:24:47.775000+00:00 | ['Entrepreneurship', 'Marketing', 'Data Science', 'Data Visualization', 'Storytelling'] | Title Data Storytelling Appealing Logic PerspectiveContent Data Visualization — Part 7 Data Storytelling Appealing Logic Perspective Data Storytelling provides presenter ability appeal logic perspective time Leveraging storytelling concept help communicate effectively data data number nonemotional information connect audience logic story connect people reason perspective Presenting information bridge logic perspective help audience recall information they’ve seen also trust audience cannot remember they’ve seen heard story didn’t work won’t meet goal audience doesn’t trust information they’ll either ignore work counter — failure either way look tell good story people remember first need attention Shahram Heshmat PhD say article Psychology Today “Emotional intensity act narrow scope attention object emphasized expense many others” idea work data storytelling it’s intended take broad set information drill smaller single point look data story 12seconds look away recall text visuals Take look story Beginning Middle End basic structure story storyteller learned describe beginning middle end One way structure link three part using plot point Plot point flush thing motivation desire setback Beginning Regional export Chicago strong robust big city Chicago’s job market diverse overall economic growth slowed Plot point Define setback connects u middle Looking story example assume “extensive assets” might one reason setback persisted Middle Using audience know good bad background situation “setback” align opportunity failure Maybe could go either way depending future action intended audience reviewing story Plot point plot point connects middle end might motivation reduce debt find unique resource End end forecast future showing data would look like everyone “does right thing” maybe simulation data choose “wrong path” threeact structure outline model weave story Conflict tension drama essential part story feeling trigger memory helping link idea urgency example seems going well data moving favor Chicago’s economic strength … setback damn “extensive assets” Building Story beginning help audience understand context situation point spark interest answering question “Why care” “What’s me” middle move people context implication context situation “x” it’s lead u state “x” it’s even substantial implication context connect audience key future “x” end story lead people understand brought journey close call action something subtle suggests behavior change presentation’s goal call direct request asking time money resource presenting data story fun want spend 40hours building everyone forget 4seconds post Tableau Public fun MakeoverMonday Challenge you’re going create something return investment move goal forward clear message end Facts Interesting Suppose reduce Chicago example fact It’s engaging visual memorable might unclear presented 30page excel file despite might think turning 30page excel file story easiest route either Storytelling take creativity even harder create outlining row information Still data requires data story digest transform information action worth effort Reviewing Story Presenting Others Try outlining set question based goal want audience answer presented data story question help understand what’s missing example cannot walk away knowing issue story fell apart somewhere presentation reviewing data story dashboard able answer problem problem occur done fix issue restore balance audience ignore data happen close designerpresenter want care believe information need information decide act take time resource delay action even change mind standard set rule define narrative structure data storytelling Narrative flow depends audience purpose presentation data access much time audience already trust need established familiar topic need layout extensive background context narrative flow align answer question Facts presented alone always stick rarely compel audience anything fact help people logically believe point Stories lend structure fact leveraging potential give information meaning actionable purpose Data Storytelling provides presenter ability appeal logic perspective time Part 1 — Context PreData Visualization Planning Part 2 — Selecting Effective Visuals Data Visualizations Part 3 Identifying Eliminating Design Clutter Data Visualizations Part 4 — Data Visualization Designing Attention Part 5 — Fundamentals Good Design Data Storytelling Part 6 — Good Data Visualization ExamplesTags Entrepreneurship Marketing Data Science Data Visualization Storytelling |
770 | How Coursera uses psychology to make education addictive | How Coursera uses psychology to make education addictive
The behavioral science principles behind their experience
Photo by Austin Distel on Unsplash
Coursera, the online education platform, was founded by professors at Stanford University. During the height of the global lockdown in early 2020, Coursera added 10 million new users in only two months, 7x the pace of new sign-ups in the previous year.
Like many online education platforms, commonly known as MOOCs, Coursera has struggled with low course completion rates. Many students will start a class with high hopes, only to quit partway through. A recent study by the Massachusetts Institute of Technology (MIT) found that online courses have a 96% dropout rate on average, over five years.
To increase their completion rates Coursera applies several strategies, with underlying behavioral science and psychology principles:
1. Helping users track their progress makes them more likely to stick around
To keep students engaged, Coursera includes cues and messages that encourage learners to stay active.
Step One — Set a goal: In the example below, Coursera asks its users to set a learning goal. They know if students have a goal, the platform can communicate how far they have to go, and therefore motivate users to complete their course.
Source: Coursera.org
Research has shown that in educational settings, sending pop-up and push notifications like these can increase student retention by up to 18%.
Step Two — Communicate progress: Coursera is transparent about where students are and how much work they have to finish.
Progress indicators, like the example below, help students know exactly how much work they’ve already completed, and how much is left to go. Studies have shown that progress indicators like these, can increase user focus, enjoyment, and users overwhelming prefer an experience that includes this element.
Source: Coursera.org
Why do these tactics work? It’s down to a behavioral science principle known as the Goal Gradient Effect.
Goal Gradient Effect
Coined by behaviorist Clark Hull in 1932, the Goal Gradient Effect states that as people get closer to a reward, they speed up their behavior to get to their goal faster. In other words, people are motivated by how much is left to reach their target, not how far they’ve come.
People are motivated by how much is left to reach their target, not how far they’ve gone.
How to apply the Goal Gradient Effect to your experience
To apply this principle to your customer experience, be clear about where your users are in relation to their “finish line”.
You’ll often see the Goal Gradient Effect at work in gamification elements like progress bars, badges, and profile completion percentages. It can be applied anywhere users are encouraged to complete a big task by achieving smaller objectives.
For example, LinkedIn ties a progress bar to a measure of completeness, to get users to complete their profile.
For added measure, LinkedIn has combined the Goal Gradient Effect with data to persuade users that their work will help them be “discovered in search by recruiters.” By tying users’ profile completeness to the end goal of a new job, LinkedIn makes it clear that their goal is closer than they think.
2. Paying for a course greatly increases completion rates
The Coursera business model was initially based around free classes. But when it comes to getting students to complete a course, “free” isn’t the best strategy. Paying a fee — even if it’s small — results in students staying engaged.
The New York Times reported that “10 percent or fewer students complete free courses, while the completion rates for paid courses that grant certificates or degrees range from 40 percent to 90 percent.”
Paid courses that grant certificates or degrees have 4–9x the completion rate of free courses.
Why? It’s down to a behavioral science principle known as Commitment.
Commitment
Coined by author and professor Robert Cialdini, this principle states that people are highly driven to be and look consistent. We want to keep our actions consistent with what we’ve already committed to.
Psychologists view this desire for consistency as one of the most important drivers of human behavior.
How to apply commitment and consistency to your experience
To apply this principle to your customer experience, get your users to make an easy, low-stakes commitment to the activity you want them to undertake.
For example, outdoor supply store REI offers a lifetime co-op membership for only $20. Once a customer signs up, they get lots of free perks and can even get paid money from the annual REI dividend — a way that REI pays back co-op members a share of their profits.
Source: REI.com
By offering a low-stakes commitment, REI increases the chances that customers will be loyal to their brand over the long-term. If customers see themselves as people who’ve made a commitment to REI, they want to stay consistent to this belief and shop there more.
3. Using data to nudge users at the right moments can greatly increase completion rates
The advantage that online education platforms like Coursera have over in-person classes is the ability to track how frequently students watch videos, submit assignments, and complete quizzes. These platforms can use this information to “nudge” students into finishing their modules, by sending reminders at critical moments.
What are nudges?
‘Nudging’ describes the different ways a decision can be presented, to make it easier for people to make the “right” choice. A “nudge” describes a specific strategy to frame or present a choice. For example, online education platform EdX found that when they sent email reminders to complete tasks, students engaged 30% more.
Engagement is 30% higher when users are nudged.
Additional research from Copenhagen Business School found that sending email nudges in online learning made students “1.5 times more likely to view lecture videos” and these same students spent “15% more time on viewing online lectures than their peers”.
But keep in mind — different types of nudges work better for different types of students.
Studies suggest that negatively framed messages — such as “If you don’t watch this video, you’ll lose 10 points off your final grade” — work better for low-performing students.
However, positively framed messages — such as “If you complete this lesson you’ll earn 3 more points toward your final grade” — are more effective for high-performing students.
So be mindful of the different types of users you have, and how you might frame nudges for each of them. Experiment to figure out who these user groups are, and what messages are most effective for each.
How to apply nudges to your experience
There are hundreds of kinds of nudges that you can use to encourage user engagement. But in general, good places for nudges are those moments in the journey where customers drop-off, forget to do something or tend to make the “wrong” decision.
For example, the language learning program Duolingo famously relies on lots of pop-up notifications to nudge users into using the app:
To learn more about nudges and how to apply them, I recommend reading Nobel-laureate Richard Thaler’s book Nudge: Improving Decisions About Health, Wealth, and Happiness.
The bottom line
When it comes to making learning addictive, MOOCs like Coursera have the benefit of millions of data points and proven strategies. But underlying their successful tactics are universal behavioral science and psychological principles that can be applied to any industry or business.
To apply these principles, ask yourself: | https://medium.com/choice-hacking/how-coursera-uses-psychology-to-make-education-addictive-e0c6421379b0 | ['Jennifer Clinehens'] | 2020-09-29 08:54:24.692000+00:00 | ['UX', 'Business', 'Psychology', 'Startup', 'Marketing'] | Title Coursera us psychology make education addictiveContent Coursera us psychology make education addictive behavioral science principle behind experience Photo Austin Distel Unsplash Coursera online education platform founded professor Stanford University height global lockdown early 2020 Coursera added 10 million new user two month 7x pace new signups previous year Like many online education platform commonly known MOOCs Coursera struggled low course completion rate Many student start class high hope quit partway recent study Massachusetts Institute Technology MIT found online course 96 dropout rate average five year increase completion rate Coursera applies several strategy underlying behavioral science psychology principle 1 Helping user track progress make likely stick around keep student engaged Coursera includes cue message encourage learner stay active Step One — Set goal example Coursera asks user set learning goal know student goal platform communicate far go therefore motivate user complete course Source Courseraorg Research shown educational setting sending popup push notification like increase student retention 18 Step Two — Communicate progress Coursera transparent student much work finish Progress indicator like example help student know exactly much work they’ve already completed much left go Studies shown progress indicator like increase user focus enjoyment user overwhelming prefer experience includes element Source Courseraorg tactic work It’s behavioral science principle known Goal Gradient Effect Goal Gradient Effect Coined behaviorist Clark Hull 1932 Goal Gradient Effect state people get closer reward speed behavior get goal faster word people motivated much left reach target far they’ve come People motivated much left reach target far they’ve gone apply Goal Gradient Effect experience apply principle customer experience clear user relation “finish line” You’ll often see Goal Gradient Effect work gamification element like progress bar badge profile completion percentage applied anywhere user encouraged complete big task achieving smaller objective example LinkedIn tie progress bar measure completeness get user complete profile added measure LinkedIn combined Goal Gradient Effect data persuade user work help “discovered search recruiters” tying users’ profile completeness end goal new job LinkedIn make clear goal closer think 2 Paying course greatly increase completion rate Coursera business model initially based around free class come getting student complete course “free” isn’t best strategy Paying fee — even it’s small — result student staying engaged New York Times reported “10 percent fewer student complete free course completion rate paid course grant certificate degree range 40 percent 90 percent” Paid course grant certificate degree 4–9x completion rate free course It’s behavioral science principle known Commitment Commitment Coined author professor Robert Cialdini principle state people highly driven look consistent want keep action consistent we’ve already committed Psychologists view desire consistency one important driver human behavior apply commitment consistency experience apply principle customer experience get user make easy lowstakes commitment activity want undertake example outdoor supply store REI offer lifetime coop membership 20 customer sign get lot free perk even get paid money annual REI dividend — way REI pay back coop member share profit Source REIcom offering lowstakes commitment REI increase chance customer loyal brand longterm customer see people who’ve made commitment REI want stay consistent belief shop 3 Using data nudge user right moment greatly increase completion rate advantage online education platform like Coursera inperson class ability track frequently student watch video submit assignment complete quiz platform use information “nudge” student finishing module sending reminder critical moment nudge ‘Nudging’ describes different way decision presented make easier people make “right” choice “nudge” describes specific strategy frame present choice example online education platform EdX found sent email reminder complete task student engaged 30 Engagement 30 higher user nudged Additional research Copenhagen Business School found sending email nudge online learning made student “15 time likely view lecture videos” student spent “15 time viewing online lecture peers” keep mind — different type nudge work better different type student Studies suggest negatively framed message — “If don’t watch video you’ll lose 10 point final grade” — work better lowperforming student However positively framed message — “If complete lesson you’ll earn 3 point toward final grade” — effective highperforming student mindful different type user might frame nudge Experiment figure user group message effective apply nudge experience hundred kind nudge use encourage user engagement general good place nudge moment journey customer dropoff forget something tend make “wrong” decision example language learning program Duolingo famously relies lot popup notification nudge user using app learn nudge apply recommend reading Nobellaureate Richard Thaler’s book Nudge Improving Decisions Health Wealth Happiness bottom line come making learning addictive MOOCs like Coursera benefit million data point proven strategy underlying successful tactic universal behavioral science psychological principle applied industry business apply principle ask yourselfTags UX Business Psychology Startup Marketing |
771 | Our FAQs | Writers
What happens when I submit my article to TDS?
Thank you so much for taking the time to submit your article to our team! We will review it as soon as we can.
If we believe that your article is excellent and ready to go, this is how you will be able to add your post to our publication. If “Towards Data Science” shows up after you click on “Add to publication” in the dropdown menu at the top of the page, that means we have added you as an author and are waiting for you to submit your article. Once you have submitted your article, it will be reviewed by an editor before a final decision is made.
If we think that your article is interesting but needs to be improved, someone from our team will provide you with feedback directly on your submitted Medium article.
Please note that we only respond to articles that were properly submitted using either our form or via an email that exactly follows the instructions listed here. We don’t respond to pitches or questions already answered in our FAQs or on our Contribute page. We also ignore articles that don’t comply with our rules.
If you haven’t heard from us within the next five working days, please carefully check the article you submitted to our team. See if you can now submit it directly to TDS and look for any private notes from us that you may have missed. You should also make sure to check your spam folder.
If you just can’t reach us, the best thing for you to do is submit your article to another publication. Although we’d love to, we can’t provide customized feedback to everyone because we simply receive too many submissions. You can learn more about our decision here and submit another post in a month. | https://medium.com/p/462571b65b35#a316 | ['Tds Editors'] | 2020-11-19 01:16:58.476000+00:00 | ['Writers’ Guide', 'Tds Team', 'Writers Guide'] | Title FAQsContent Writers happens submit article TDS Thank much taking time submit article team review soon believe article excellent ready go able add post publication “Towards Data Science” show click “Add publication” dropdown menu top page mean added author waiting submit article submitted article reviewed editor final decision made think article interesting need improved someone team provide feedback directly submitted Medium article Please note respond article properly submitted using either form via email exactly follows instruction listed don’t respond pitch question already answered FAQs Contribute page also ignore article don’t comply rule haven’t heard u within next five working day please carefully check article submitted team See submit directly TDS look private note u may missed also make sure check spam folder can’t reach u best thing submit article another publication Although we’d love can’t provide customized feedback everyone simply receive many submission learn decision submit another post monthTags Writers’ Guide Tds Team Writers Guide |
772 | A Meditation on Not Being a Dick to Yourself | A Meditation on Not Being a Dick to Yourself
It’s possible to hold yourself accountable with compassion
Photo: Kohei Hara/Getty Images
A few days ago, I was sitting at the piano. It had been a while. I used to play pretty regularly, but these days I’m not what you’d call a musician. I’ll jam with my husband every once in a while, noodling with things by ear and messing around with chord progressions, but it had been years since I seriously sat down in front of a piece of sheet music to see if I could still sight-read.
Side note: Don’t get the impression I’m some kind of Renaissance woman here, it’s not like I nobly pursue the arts in my spare time because I’m virtuous and well-rounded. I only ended up at the piano because we’re quarantined in a pandemic and my husband and I had run through all 24 seasons of America’s Next Top Model, and now I was bored.
So I pulled out a Bach prelude, prepared to go as slowly as I needed to go, very maturely laying the groundwork for a nonstressful experience by reminding myself to breathe and be patient with myself. I was the very picture of emotional evolution! Pema Chodron herself would’ve been proud, and I made a mental note to tell my husband what a good job I’d done of practicing the shit I preach.
Seven minutes later I was literally sweating. I’d been stuck on the same four fucking measures, repeatedly making the same fucking mistakes, because I was a moron and I couldn’t figure out what the fingering was supposed to be. I kept getting stuck counting ledger lines because I’d been a lazy piece of shit who was wasting her bullshit brain on reality TV, and I couldn’t stop thinking that the last time I heard this fucking prelude was when some nine-year-old asshole played it at a recital. Flawlessly.
Don’t look at me, Pema. I’m hideous.
The tendency to resort to self-abuse when we miss the mark is an instinct. It’s not exactly an instinct that has intelligence to it (by which I mean, we rarely stop to investigate whether it’s useful) but it’s definitely an emotional reflex.
Shame is always promoting the idea that we didn’t just make a mistake, we are a mistake.
For some, that instinct toward self-abuse is an organic attempt at self-motivation which may once have been healthy, but became warped and weaponized over time. For others it’s a direct result of trauma; if we were parented with abuse, abuse will be the tool we reach for when we’re failing, and find ourselves in “survival mode.” My dad (who wasn’t a musician but liked to think of himself as one) had some pretty unrealistic expectations about what my practice technique should look like, and draconian methods of enforcing them. I say this not to blame him for my piano meltdown — I’m a grown-up, I’m capable of finding healthier, more creative methods of holding myself accountable than those my dad resorted to, but on a neurological level this stuff gets embedded in us.
Self-abuse is born of shame, and shame is always promoting the idea that we didn’t just make a mistake, we are a mistake. It’s all about what we should be instead. We should be better by now. We should be in a relationship. We should have tackled our whole to-do list. We should want to make sacrifices for our children. I should be more compassionate with myself given that I coach these issues. These are valid things to want for ourselves, generally speaking, but the shame that’s driving them undermines our ability to actually grow.
Yet, we’re often terrified to surrender the self-abuse. One of the most common fears I hear from my clients is some version of “but if I’m not beating up on myself, I’ll never get any better!” My dear, if self-abuse were an effective strategy for transformation, don’t you think it would’ve worked by now? It’s about time we acknowledge the data; shame-based self-abuse isn’t an effective technique for evolution after all.
So what is?
I hate to say it, but compassion is the most reliable catalyst for healing and lasting behavioral change. I know, I know, it’s a therapy buzzword, and in our rigid, neurotic culture of overachieving black-and-white thinkers it’s even kind of a dirty word. I get it, I grappled with it myself; I too was convinced that the whole self-love/self-compassion thing would make me soft. I envisioned myself being slowly conditioned to mindlessly forgive my own self-destructive patterns, bid my critical thinking adieu, start wearing mood rings, join a commune, take up chanting, and ultimately melt into an underachieving puddle of Pollyanna self-acceptance.
We Westerners seem to have grossly misinterpreted compassion as laziness, or simply letting ourselves off the hook. I frequently hear from clients “If I’m compassionate with myself, I’ll just accept my flaws and never improve, then I’ll really hate myself.” Well, if that’s how you understand compassion I can see why it’s so damn dangerous!
Compassion just asks us to change our relationship to accountability, so that it isn’t one rooted in abuse.
For those of us indoctrinated to believe that being tender with ourselves is a form of indulgent enabling, compassion really is a courageous act! Our brains are telling us compassion risks the slippery slope toward entropy and chaos. It’s crucial that we identify how much fear we have about being kind to ourselves, and stay curious that perhaps compassion isn’t as one-dimensional as we’ve understood it to be. In truth, it’s far more nuanced — compassion includes accountability, it doesn’t absolve us of it.
Sitting at the piano, telling myself “That’s okay honey, you did two and a half minutes of work, that’s enough hardship for today” would not have been a compassionate response, because it ultimately would have created more suffering for me by enabling my laziness. Compassion just asks us to change our relationship to accountability, so that it isn’t one rooted in abuse. It invites us to re-parent ourselves with tenderness. “Take a breath, girl. Slow down the tempo. No, that doesn’t mean you’re stupid, it means it’s been a while since you used this skill. Give yourself a chance to relearn it.” Compassion needn’t even be overly saccharine, it’s just the truth filtered through kindness, rather than through our shame and our judgment.
We’ve spent years trying to discipline ourselves into being better using the wrong tools. Rather than disciplining ourselves with self-abuse, what would it look like if compassion itself was the discipline?
I haven’t been back to the piano since my mini meltdown. I was hoping that writing this piece would be enough of an amends to myself, but it’s only making me realize that for the sake of my own spiritual and professional integrity, I’m going to have to break that prelude out again. I’ll let you know how it goes. Maybe I need a refresher lesson. If anyone knows a patient nine-year-old who can sight-read, send her my way. | https://humanparts.medium.com/a-meditation-on-not-being-a-dick-to-yourself-7c907c5500d2 | ['Mollie Birney'] | 2020-04-27 14:01:01.519000+00:00 | ['Mental Health', 'Productivity', 'Life Lessons', 'Psychology', 'Self Improvement'] | Title Meditation Dick YourselfContent Meditation Dick It’s possible hold accountable compassion Photo Kohei HaraGetty Images day ago sitting piano used play pretty regularly day I’m you’d call musician I’ll jam husband every noodling thing ear messing around chord progression year since seriously sat front piece sheet music see could still sightread Side note Don’t get impression I’m kind Renaissance woman it’s like nobly pursue art spare time I’m virtuous wellrounded ended piano we’re quarantined pandemic husband run 24 season America’s Next Top Model bored pulled Bach prelude prepared go slowly needed go maturely laying groundwork nonstressful experience reminding breathe patient picture emotional evolution Pema Chodron would’ve proud made mental note tell husband good job I’d done practicing shit preach Seven minute later literally sweating I’d stuck four fucking measure repeatedly making fucking mistake moron couldn’t figure fingering supposed kept getting stuck counting ledger line I’d lazy piece shit wasting bullshit brain reality TV couldn’t stop thinking last time heard fucking prelude nineyearold asshole played recital Flawlessly Don’t look Pema I’m hideous tendency resort selfabuse miss mark instinct It’s exactly instinct intelligence mean rarely stop investigate whether it’s useful it’s definitely emotional reflex Shame always promoting idea didn’t make mistake mistake instinct toward selfabuse organic attempt selfmotivation may healthy became warped weaponized time others it’s direct result trauma parented abuse abuse tool reach we’re failing find “survival mode” dad wasn’t musician liked think one pretty unrealistic expectation practice technique look like draconian method enforcing say blame piano meltdown — I’m grownup I’m capable finding healthier creative method holding accountable dad resorted neurological level stuff get embedded u Selfabuse born shame shame always promoting idea didn’t make mistake mistake It’s instead better relationship tackled whole todo list want make sacrifice child compassionate given coach issue valid thing want generally speaking shame that’s driving undermines ability actually grow Yet we’re often terrified surrender selfabuse One common fear hear client version “but I’m beating I’ll never get better” dear selfabuse effective strategy transformation don’t think would’ve worked It’s time acknowledge data shamebased selfabuse isn’t effective technique evolution hate say compassion reliable catalyst healing lasting behavioral change know know it’s therapy buzzword rigid neurotic culture overachieving blackandwhite thinker it’s even kind dirty word get grappled convinced whole selfloveselfcompassion thing would make soft envisioned slowly conditioned mindlessly forgive selfdestructive pattern bid critical thinking adieu start wearing mood ring join commune take chanting ultimately melt underachieving puddle Pollyanna selfacceptance Westerners seem grossly misinterpreted compassion laziness simply letting hook frequently hear client “If I’m compassionate I’ll accept flaw never improve I’ll really hate myself” Well that’s understand compassion see it’s damn dangerous Compassion asks u change relationship accountability isn’t one rooted abuse u indoctrinated believe tender form indulgent enabling compassion really courageous act brain telling u compassion risk slippery slope toward entropy chaos It’s crucial identify much fear kind stay curious perhaps compassion isn’t onedimensional we’ve understood truth it’s far nuanced — compassion includes accountability doesn’t absolve u Sitting piano telling “That’s okay honey two half minute work that’s enough hardship today” would compassionate response ultimately would created suffering enabling laziness Compassion asks u change relationship accountability isn’t one rooted abuse invite u reparent tenderness “Take breath girl Slow tempo doesn’t mean you’re stupid mean it’s since used skill Give chance relearn it” Compassion needn’t even overly saccharine it’s truth filtered kindness rather shame judgment We’ve spent year trying discipline better using wrong tool Rather disciplining selfabuse would look like compassion discipline haven’t back piano since mini meltdown hoping writing piece would enough amends it’s making realize sake spiritual professional integrity I’m going break prelude I’ll let know go Maybe need refresher lesson anyone know patient nineyearold sightread send wayTags Mental Health Productivity Life Lessons Psychology Self Improvement |
773 | Will A Robot Take Your Writing Job? | Will A Robot Take Your Writing Job?
The pandemic has accelerated automation in non-creative fields, but are creative jobs vulnerable, too?
Photo by Photos Hobby on Unsplash
Since the 1700s, when Luddites started smashing English looms, automation has been putting people out of work.
In the 20th century, automation took over manufacturing jobs, data entry jobs, manual computing jobs. Now, breakthroughs in “deep learning” have exponentially increased the number of jobs that can be automated. McKinsey estimates that, by 2030, as much as 30% of work will be done by machines, displacing almost 400 million workers.
If you’re entering the job market or considering a career change, you’re probably wondering whether your job will still be around 20 years from now.
AI is now being used to make pizza, trade stocks and operate call centers (much to the chagrin of anyone trying to get in contact with a service provider). Administrative jobs and even some legal tasks, like document review and contract generation, are increasingly being handled by machines.
But what about more creative jobs?
For a long time, people have touted the relative safety of “creative” professions. While computers are already better than humans at things like processing data, sensory perception, recall, and predictable physical motion, humans outpace computers in soft skills like empathy, unstructured problem-solving, and unpredictable physical movement.
But is that changing? As neural networks improve, AI software is being used to disrupt professions once considered automation-proof.
There are now AI actors, poets, and visual artists. Even influencers, whose professions are based almost solely on being “personable”, are having to compete against computerized competitors. Imma, a virtual influencer, has over 300,000 Instagram followers and has landed deals with brands like Celine.
Just this year, Open AI released GTP-3, a pre-trained language model that can be used to generate a wide range of texts ranging from emails to dialogue, and even memes.
There are other platforms that generate prose, often notoriously incoherent. But GPT-3 is different.
GPT-3 produces extremely high-quality prose. Just look at this op-ed from The Guardian, written by GPT-3 itself:
The mission for this op-ed is perfectly clear. I am to convince as many human beings as possible not to be afraid of me. Stephen Hawking has warned that AI could “spell the end of the human race”. I am here to convince you not to worry. Artificial intelligence will not destroy humans. Believe me.
Look at that parallel structure. Look at those integrated quotes (ignore the wayward periods). I teach writing to middle and high schoolers and wish they could deliver this level of deliberate style and clear, concise prose. I wish I could write this well all the time.
This is not just words on a page or language generated simply to adhere to the grammar rules programmed into it. As Farhad Manjoo states in his New York Times op-ed, “How Do You Know a Human Wrote This”:
“GPT-3 is so good at aping human writing that it sometimes gave me chills. Not too long from now, your humble correspondent might be put out to pasture by a machine — and you might even miss me when I’m gone.”
So, yes, AI writers are coming, and soon.
So, will a robot take YOUR writing job?
Well that depends on the kind of writing you do. AI software might be more efficient than humans at developing SEO content or generating B2B emails (If you’re a Gmail user, you’ve noticed algorithms creeping into your responses already). So, in the future, platforms might be flooded by AI-created, SEO-optimized, targeted clickbait. Writers whose careers rely on sheer volume and formulaic writing might have difficulty competing with AI-generated work. In fact, some bots are writing clickbait headlines already.
Freelance sometimes feels like a hamster wheel: Medium articles espouse the joys of writing one or two or three articles a day, publishing every day, come rain or sleet, on four hours of sleep, after a 12-hour shift at a demanding corporate job.
But, whatever writing AI software produces, it will write it much faster than a human ever could.
Are freelance writers doomed to be like John Henry, working faster and faster, producing on a superhuman scale, racing against the machine — even winning, but at the cost of their lives, or, at least, the independent lives to which they aspired?
Well, here’s one glimmer of hope: GPT-3 struggles to write factually. So, as long as humans still value accuracy and truth telling (though the past four years have made me question whether they still do) there will be jobs for humans (at least until the bots get it right).
Another: advancements in AI may be slowing due to the extremely high cost of computing power. Automation replaces human labor when the cost of a machine is less than the wages of the workers it’s replacing. While GPT-3 is good, it won’t be able to completely replace humans until it is substantially improved, and that could take a lot of effort.
Improving the performance of an English-to-French machine-translation algorithm so that it only makes mistakes 10 percent of the time instead of the current rate of 50 percent, for example, would require an extraordinary increase in computational power — a billion billion times as much — if it were to rely on more computation power alone (Will Knight, “Prepare for Artificial Intelligence to Produce Less Wizardry”)
So in the short term, your job might be safe.
In fact, at least in the near future, AI is much more likely to benefit writers than to take their jobs. In-app editors like Medium’s might incorporate AI software that will supercharge your writing. Or, an app might reorganize your notes into cohesive paragraphs. Or a Chrome plug-in will use search tools to identify possible references based on keywords and automatically create hyperlinks in your stories. Or all three. As a bot who has digested a couple million Medium articles might say, “The possibilities are endless.”
So, how can you protect your writing career?
1. Write for yourself
There will always be a market for high-quality copywriting and clients who will pay big money for SEO-optimized prose. But the same clients who are abandoning quality freelancers for race-to-the-bottom job boards are the same ones who are going to switch to AI writers, if they turn out to be cheaper.
This goes for ghostwriting gigs and content mills as well. Of course, those hiring a ghostwriter to write their memoirs are still going to be looking for high-quality writing, and, possibly, the human interaction that goes along with dictating your life story to someone. But self-help e-books and Kindle Unlimited “pulp” fiction might be on the automating block.
Writing for yourself, however, is also not only much more satisfying than work-for-hire, but is also, hopefully, the kind of work that will help you build longevity and generate a following (see #2). Ultimately, the more authentic you your work is, the less likely it is to be replicable — by anyone or any bot. Developing a voice is part of developing your craft as a writer. Even now, in a sea of voices, it’s important to make yours stand out.
2. Interact with your audience; build a following
Building your audience now will help you when the glut of AI writers arrives. New steamy romance writers might not be able to compete with a bot that can churn out a 300-page book a day with passable prose that flawlessly adheres to the conventions of that particular niche. But, if you have a mailing list and a back catalog, you’ve got a leg up.
And while there are advances in AI language software, bots still make poor conversationalists. The more you interact with your audience the more your authenticity and humanity will show through.
3. Learn constantly
Language models are trained on enormous data sets full of text. From these texts it “learns” the patterns of language such that it can produce coherent prose. But the craft of writing is built on finding novel ways to express novel ideas.
Ideas are the heart of everything your write, and while AI-generated prose might take over, the ideas that structure it will be human, and they might be your own. In the future, expect a hybrid model of writing, where humans produce the ideas and software produces the prose.
4. Use AI tools when possible
As I said before, it’s likely that AI in the near future will be used, not to take over writing jobs, but to make writing easier. Thus, you’re not just competing against bots, you’re competing against every human writer using AI to aid their output. Refusing to take part in what will be the biggest technological advancement in Letters of the 21st century will put you decades behind in terms of output, market access and strategy. Can AI help you meet your daily writing goals? Can it help you develop headlines more quickly or give you a good foundation for your first draft? | https://kaydems.medium.com/will-a-robot-take-your-writing-job-c03749a8c7ef | ['Kaye Dems'] | 2020-11-01 18:08:42.515000+00:00 | ['AI', 'Artificial Intelligence', 'Automation', 'Deep Learning', 'Writing'] | Title Robot Take Writing JobContent Robot Take Writing Job pandemic accelerated automation noncreative field creative job vulnerable Photo Photos Hobby Unsplash Since 1700s Luddites started smashing English loom automation putting people work 20th century automation took manufacturing job data entry job manual computing job breakthrough “deep learning” exponentially increased number job automated McKinsey estimate 2030 much 30 work done machine displacing almost 400 million worker you’re entering job market considering career change you’re probably wondering whether job still around 20 year AI used make pizza trade stock operate call center much chagrin anyone trying get contact service provider Administrative job even legal task like document review contract generation increasingly handled machine creative job long time people touted relative safety “creative” profession computer already better human thing like processing data sensory perception recall predictable physical motion human outpace computer soft skill like empathy unstructured problemsolving unpredictable physical movement changing neural network improve AI software used disrupt profession considered automationproof AI actor poet visual artist Even influencers whose profession based almost solely “personable” compete computerized competitor Imma virtual influencer 300000 Instagram follower landed deal brand like Celine year Open AI released GTP3 pretrained language model used generate wide range text ranging email dialogue even meme platform generate prose often notoriously incoherent GPT3 different GPT3 produce extremely highquality prose look oped Guardian written GPT3 mission oped perfectly clear convince many human being possible afraid Stephen Hawking warned AI could “spell end human race” convince worry Artificial intelligence destroy human Believe Look parallel structure Look integrated quote ignore wayward period teach writing middle high schoolers wish could deliver level deliberate style clear concise prose wish could write well time word page language generated simply adhere grammar rule programmed Farhad Manjoo state New York Times oped “How Know Human Wrote This” “GPT3 good aping human writing sometimes gave chill long humble correspondent might put pasture machine — might even miss I’m gone” yes AI writer coming soon robot take writing job Well depends kind writing AI software might efficient human developing SEO content generating B2B email you’re Gmail user you’ve noticed algorithm creeping response already future platform might flooded AIcreated SEOoptimized targeted clickbait Writers whose career rely sheer volume formulaic writing might difficulty competing AIgenerated work fact bot writing clickbait headline already Freelance sometimes feel like hamster wheel Medium article espouse joy writing one two three article day publishing every day come rain sleet four hour sleep 12hour shift demanding corporate job whatever writing AI software produce write much faster human ever could freelance writer doomed like John Henry working faster faster producing superhuman scale racing machine — even winning cost life least independent life aspired Well here’s one glimmer hope GPT3 struggle write factually long human still value accuracy truth telling though past four year made question whether still job human least bot get right Another advancement AI may slowing due extremely high cost computing power Automation replaces human labor cost machine le wage worker it’s replacing GPT3 good won’t able completely replace human substantially improved could take lot effort Improving performance EnglishtoFrench machinetranslation algorithm make mistake 10 percent time instead current rate 50 percent example would require extraordinary increase computational power — billion billion time much — rely computation power alone Knight “Prepare Artificial Intelligence Produce Less Wizardry” short term job might safe fact least near future AI much likely benefit writer take job Inapp editor like Medium’s might incorporate AI software supercharge writing app might reorganize note cohesive paragraph Chrome plugin use search tool identify possible reference based keywords automatically create hyperlink story three bot digested couple million Medium article might say “The possibility endless” protect writing career 1 Write always market highquality copywriting client pay big money SEOoptimized prose client abandoning quality freelancer racetothebottom job board one going switch AI writer turn cheaper go ghostwriting gig content mill well course hiring ghostwriter write memoir still going looking highquality writing possibly human interaction go along dictating life story someone selfhelp ebooks Kindle Unlimited “pulp” fiction might automating block Writing however also much satisfying workforhire also hopefully kind work help build longevity generate following see 2 Ultimately authentic work le likely replicable — anyone bot Developing voice part developing craft writer Even sea voice it’s important make stand 2 Interact audience build following Building audience help glut AI writer arrives New steamy romance writer might able compete bot churn 300page book day passable prose flawlessly adheres convention particular niche mailing list back catalog you’ve got leg advance AI language software bot still make poor conversationalist interact audience authenticity humanity show 3 Learn constantly Language model trained enormous data set full text text “learns” pattern language produce coherent prose craft writing built finding novel way express novel idea Ideas heart everything write AIgenerated prose might take idea structure human might future expect hybrid model writing human produce idea software produce prose 4 Use AI tool possible said it’s likely AI near future used take writing job make writing easier Thus you’re competing bot you’re competing every human writer using AI aid output Refusing take part biggest technological advancement Letters 21st century put decade behind term output market access strategy AI help meet daily writing goal help develop headline quickly give good foundation first draftTags AI Artificial Intelligence Automation Deep Learning Writing |
774 | Embracing the startup mindset | Canva Behind the Scenes
Embracing the startup mindset
Congratulations! You’ve taken the leap from your 9–5 job for a new job at a startup. Now what?
Written by Guy Kawasaki — Chief Evangelist at Canva
So you’ve taken the plunge and left your traditional 9 to 5 for a new job at a startup. After countless nerve-wracking interviews, you’ve signed on the dotted line and met up with the rest of your new team. Their drive and passion is contagious. You’re bubbling with newfound energy and looking forward to making a dent in the universe — nothing’s going to stop you, and it’s going to be great.
Despite the promises of an exciting, fulfilling career ahead, throughout my career I’ve met a lot of people who have had a rough time with their first few months at a startup. In private conversations, there’s a common theme around shame of feeling unsettled and confused — especially from those who already have years and years of experience under their belt.
While some of these people have manage to find their footing and get through this transition from corporate to startup, others don’t. In this article, I’ll share my thoughts around what you can do to ensure a smooth transition from a larger corporation to startup.
My personal experience
On paper (this was before LinkedIn), my résumé should not have gotten me an interview with Apple. My educational background was a bachelor of arts in psychology and a master of business administration in marketing. I had not taken a computer class — not that there were many computer classes back then. My work experience was also seemingly irrelevant. After earning my MBA, I went to work for a fine jewelry manufacturer.
I started out by counting diamonds and left five years later as vice president of sales and marketing. My only exposure to computers was using an IBM System to enter and access data. How did I get a job in the tech industry? The simple answer was nepotism! My roommate from college hired me into the division. But then how did I succeed, without a computer science background or any significant work experience in tech?
Speaking at Canva’s “Season Opener” event
The answer is “a lot of determination and the willingness to learn along the way.” While on the firing line (there was no evangelism training program), I had to learn enough about programming to talk to techies and enough about marketing computers to talk to CXOs at software companies.
As you can see it was a far cry from counting diamonds. So ultimately I had to embrace a set of character traits that are required to thrive in a startup — aka “the startup mindset”.
The startup mindset
It all boils down to this.
The startup mindset is very, very different from the mindset you’ll see at most companies. It’s a mindset where the glass is half full. It’s a mindset of meritocracy. It’s not about hierarchy — it’s about whether you’re good or not and where you’re going, not where you’ve been.
The startup mindset is one where you’re not so concerned about procedures and established roles, you’re going to do whatever it takes to get things done.
The startup mindset means that you’re creative, you ask for forgiveness not permission, and you’re on a mission, you’re not simply trying to make a buck. You’re trying to dent the universe. If you do all that, you’ll be fine.
This is the exact kind of mindset that separates a good startup culture from a bad one. Let me explain further.
Running a workshop with Canva’s Presentations Team
Thinking like an owner
Most companies rely upon educational background and work experience to determine the acceptability of job candidates. The logic is that employees need a foundation of relevant knowledge and skills to succeed — or at least not make the hiring manager look bad. So why do some not perform at their peak when placed in this new startup environment?
From my experience in Silicon Valley, what separates a good startup from a bad one boils down to a team that is dedicated to constantly improving everything — products, service, customer support, onboarding and everything in between.
There has to be a mentality within the company that isn’t siloed. For example, one’s job is not just about engineering or sales or marketing — everybody has a role to play in all these functions, and everyone needs to contribute to get the job done.
One of the qualities of an employee who does well at a startup is that you think like an owner, in the sense that when you see something that’s wrong, you speak up and you correct it. When you see something being wasted, you speak up. If you see an opportunity, you seize it.
In other words, you’re not saying, “I’m just an employee. My job is to do X, not Y and Z.” That’s not a startup, and that’s not an ownership mentality. Everyone needs to be in this together, and it starts with you.
Being the master of your own fate
It’s important that you recognize that despite all the years of experience you bring to the company, there’s still a lot to learn.
Despite that, you can’t spend weeks and weeks on onboarding and training — that’s not the nature of a startup. Although this can be challenging, a high-performance team in a startup consists of persistent go-getters raring to solve the problems ahead. At a startup you’re expected to move fast and experiment and iterate, again and again — and that’s okay. The nature of experiments means that oftentimes things will fail — that’s just how it is.
So don’t be proud. The day after you start a job, nobody cares about your connections, history, and credentials — or lack thereof. You either deliver results, or you don’t.
If you’ve worked for a large company in the past, it’s a very different world at a startup. But I think you might be delighted that you have the freedom to make things happen like you could never do before. You have the freedom to move, to change, and to innovate. In a sense, you’re the master of your own fate. When the dust settles, you’ll be glad you’re in a startup.
Guy Kawasaki (2nd from the right) with some of the Canva Team
Adapting to change
When you work for a startup, things change fast.
It’s not like there’s a three-year strategic plan, and that’s all you have to focus on. You have to be opportunistic: when a potential customer comes in, when potential new technology emerges, or a potential partnership with a large company arises, you have to grab it.
Theoretically, there might be a danger of changing too fast, but if you had to pick between changing too fast and not changing, you’ll need to take the first option every time. That’s what’s called a high-quality problem. Changing too fast is much better than having management, tradition or another department standing in the way of every change you’re trying to make.
Whilst you have to change quickly, you don’t have to do it alone. In a startup, a valuable lesson is that giving stuff away, turning things down, and having other people help you is a good thing. There’s a really good blog post here about giving away your legos at work — a startup commandment that everyone should learn and adopt.
It’s not about a territorial grab, it’s about a rising tide floating all boats. So don’t get territorial, and don’t try to possess things — set it free, and it’ll work out. This is not a Fortune 500 company where you have to protect your piece of turf and your little fiefdom. That attitude will not work at a startup.
In the example of Canva, we don’t call it a day without setting some ambitious goals, and hustling to execute — so there is always something different to keep things fresh and challenging. That might mean that we rarely get our ducks perfectly in a row, but that’s a sacrifice we’re willing to make.
Roles will be redefined. Teams will be reshuffled. Processes will be broken. Timelines will be moved.
What keeps the team going is the understanding that with high growth comes frequent change. We don’t concern ourselves with ideals. Instead, we adapt with a flexible mindset, and work to identify pragmatic solutions together.
We live and play in a very dynamic space, and there are always plenty of opportunity to work on different parts and priorities of the business. The world is yours to shape.
Final thoughts
Resilience is a necessary skill at any startup. We’re in a fast-paced market, and there are a lot of challenges. But I dare say you’ll look back and say, “Wow, that was some of the hardest work I ever did, but also some of the best work I ever did. It was also some of the most enjoyable.”
It’s a great feeling.
So congratulations again on landing a job in a startup. It’s time to strap yourself into your seat. We’re on a rocket. We will dent the universe.
This story first appeared on Business Insider. For more details on our current job openings, visit Canva’s careers page here. | https://medium.com/canva/embracing-the-startup-mindset-edd3638e9bc2 | ['Canva Team'] | 2020-05-13 03:00:05.377000+00:00 | ['Marketing', 'Startup', 'Work', 'Entrepreneurship', 'Careers'] | Title Embracing startup mindsetContent Canva Behind Scenes Embracing startup mindset Congratulations You’ve taken leap 9–5 job new job startup Written Guy Kawasaki — Chief Evangelist Canva you’ve taken plunge left traditional 9 5 new job startup countless nervewracking interview you’ve signed dotted line met rest new team drive passion contagious You’re bubbling newfound energy looking forward making dent universe — nothing’s going stop it’s going great Despite promise exciting fulfilling career ahead throughout career I’ve met lot people rough time first month startup private conversation there’s common theme around shame feeling unsettled confused — especially already year year experience belt people manage find footing get transition corporate startup others don’t article I’ll share thought around ensure smooth transition larger corporation startup personal experience paper LinkedIn résumé gotten interview Apple educational background bachelor art psychology master business administration marketing taken computer class — many computer class back work experience also seemingly irrelevant earning MBA went work fine jewelry manufacturer started counting diamond left five year later vice president sale marketing exposure computer using IBM System enter access data get job tech industry simple answer nepotism roommate college hired division succeed without computer science background significant work experience tech Speaking Canva’s “Season Opener” event answer “a lot determination willingness learn along way” firing line evangelism training program learn enough programming talk techie enough marketing computer talk CXOs software company see far cry counting diamond ultimately embrace set character trait required thrive startup — aka “the startup mindset” startup mindset boil startup mindset different mindset you’ll see company It’s mindset glass half full It’s mindset meritocracy It’s hierarchy — it’s whether you’re good you’re going you’ve startup mindset one you’re concerned procedure established role you’re going whatever take get thing done startup mindset mean you’re creative ask forgiveness permission you’re mission you’re simply trying make buck You’re trying dent universe you’ll fine exact kind mindset separate good startup culture bad one Let explain Running workshop Canva’s Presentations Team Thinking like owner company rely upon educational background work experience determine acceptability job candidate logic employee need foundation relevant knowledge skill succeed — least make hiring manager look bad perform peak placed new startup environment experience Silicon Valley separate good startup bad one boil team dedicated constantly improving everything — product service customer support onboarding everything mentality within company isn’t siloed example one’s job engineering sale marketing — everybody role play function everyone need contribute get job done One quality employee well startup think like owner sense see something that’s wrong speak correct see something wasted speak see opportunity seize word you’re saying “I’m employee job X Z” That’s startup that’s ownership mentality Everyone need together start master fate It’s important recognize despite year experience bring company there’s still lot learn Despite can’t spend week week onboarding training — that’s nature startup Although challenging highperformance team startup consists persistent gogetters raring solve problem ahead startup you’re expected move fast experiment iterate — that’s okay nature experiment mean oftentimes thing fail — that’s don’t proud day start job nobody care connection history credential — lack thereof either deliver result don’t you’ve worked large company past it’s different world startup think might delighted freedom make thing happen like could never freedom move change innovate sense you’re master fate dust settle you’ll glad you’re startup Guy Kawasaki 2nd right Canva Team Adapting change work startup thing change fast It’s like there’s threeyear strategic plan that’s focus opportunistic potential customer come potential new technology emerges potential partnership large company arises grab Theoretically might danger changing fast pick changing fast changing you’ll need take first option every time That’s what’s called highquality problem Changing fast much better management tradition another department standing way every change you’re trying make Whilst change quickly don’t alone startup valuable lesson giving stuff away turning thing people help good thing There’s really good blog post giving away lego work — startup commandment everyone learn adopt It’s territorial grab it’s rising tide floating boat don’t get territorial don’t try posse thing — set free it’ll work Fortune 500 company protect piece turf little fiefdom attitude work startup example Canva don’t call day without setting ambitious goal hustling execute — always something different keep thing fresh challenging might mean rarely get duck perfectly row that’s sacrifice we’re willing make Roles redefined Teams reshuffled Processes broken Timelines moved keep team going understanding high growth come frequent change don’t concern ideal Instead adapt flexible mindset work identify pragmatic solution together live play dynamic space always plenty opportunity work different part priority business world shape Final thought Resilience necessary skill startup We’re fastpaced market lot challenge dare say you’ll look back say “Wow hardest work ever also best work ever also enjoyable” It’s great feeling congratulation landing job startup It’s time strap seat We’re rocket dent universe story first appeared Business Insider detail current job opening visit Canva’s career page hereTags Marketing Startup Work Entrepreneurship Careers |
775 | AI & Big-Data can help tackle Covid-19’s spread | Adapting social distancing policies via the monitoring of crucial home and public behaviour data can help contain the Covid-19 offensive.
Let’s take a closer look in this, widely circulated, chart regarding the progression of the infections:
The Infection Trajectory as depicted by the reported cases vs. the days since case count passed 500 (22/3). An interactive - regularly updated version of this chart is also available at the Visual Capitalist website
It appears that its spread -when the case count passes 500- is strikingly similar, regardless of the size or the geographical density of the “sample”.
The chart was provided to me by David Gorelick, Professor of Psychiatry at the School of Medicine of the University of Maryland with whom we happened to discuss the capability of predicting electoral outcomes (as discussed in a series of articles that started here, whereas in one of the articles it was mentioned how data analysis can be used to draw meaningful conclusions that are insensitive to polling sample size and density).
I reverted to him with a message that I feel worthwhile sharing here (in a revised/enhanced and edited for publication form) regarding social distancing policies that are crucial in pacing the spread of the pandemic as the resilience of the Health Care Systems is tested; As this brilliant simulator displays, Health Systems collapse when the rate of concurrent cases exceeds their capacity, thus patients -who otherwise could have been saved in ICUs- are dying due to lack of resources:
“As we were discussing how I am going about in predicting elections, here is a way to approach this curve in order to get meaningful data that may shape “social distancing” policies:
1. Obviously, the initial cases had to do with travel in China or other in countries that do business with China (e.g. Italy’s fashion industry and Germany’s car parts complex as well as the interplay between the Italian and German cases).
2. For sure, it would have been beneficial to consider together with the bare numbers, the professions of the patients. How many of them are in the medical domain (doctors, nurses and caregivers exposed to the initial patients). How many of them are related to these doctors, nurses, caregivers (and the “delay”/rate of contagion).
3. How many of them were in public transport (air crew, drivers etc.), delivery boys, supermarket/shop employees (and the cases related to them).
4. Investigate their hobbies (Church/Social Club goers, Sport fans who attended matches etc) and perform an analysis among the cases of all those who are church goers, football fans etc.
5. Check the mobility of each patient vs. the overall mobility index of their respective segments (Professions, Hobbies etc.). Mobility data are readily available by Google, Facebook and Apple (and in many countries by the Network Operators).
6. Examine the available space to them (e.g. get metrics of the their home, which floor they live in, if they use public transport, if they work in a desk-job etc.)
7. As the volume of the data to be analysed may become hard to handle, one may issue rules based on the profile of the “candidate”. For example, there are suggestions that the time of exposure to the virus should be over 15 minutes for a higher probability of infection (I am sure the epidemiologists have more accurate data about it) that can be used accordingly to filter “irrelevant” data.
This knowledge is already available in some countries that are ahead of others may help us make informed decisions about which services to block (e.g. it may be the case that the public transport should shutdown) as well as go to predictive patterns in clusters about the behaviour of the “virus”.
If I were an American or British official I would have asked the Chinese, Italians etc. to share with me all this info — if not make all this info available to all by all — as well as take a note of the measures and their impact, and adapt our own accordingly”.
Such a meta-analysis may also help us cross-check and validate the official data provided by countries not known for their transparency. For example the South Korean “first 500” cases may be different than any other country’s “first 500” cases in this chart, given the fact that the preventative testing of the Koreans far surpasses the testing in the other countries; thus subsequent correlations and normalisations ought to be considered.
The medical authorities should get related personal details (other than address, age & sex) of each patient and the authorization to access their mobile phone’s data in order to investigate the case’s movements for a thorough forensic analysis that may help save lives.
If this virus is relatively mild, others in the future may not be so, and we have to be ready for them too; consider this as a preparatory step for similar and potentially more deadly diseases that may follow. Sadly, we do not have the luxury to philosophise about the usage of our Personal and Privacy data when Life is at risk.
Edit: less than three weeks later, Apple and Google announced their collaboration in order to enable interoperable contact tracing in their devices. | https://medium.com/predict/how-big-data-can-help-us-tackle-covid-19s-spread-8bda64b3adde | ['Dimitris Vayenas'] | 2020-08-20 07:02:16.884000+00:00 | ['Covid 19', 'AI', 'Big Data', 'Health'] | Title AI BigData help tackle Covid19’s spreadContent Adapting social distancing policy via monitoring crucial home public behaviour data help contain Covid19 offensive Let’s take closer look widely circulated chart regarding progression infection Infection Trajectory depicted reported case v day since case count passed 500 223 interactive regularly updated version chart also available Visual Capitalist website appears spread case count pass 500 strikingly similar regardless size geographical density “sample” chart provided David Gorelick Professor Psychiatry School Medicine University Maryland happened discus capability predicting electoral outcome discussed series article started whereas one article mentioned data analysis used draw meaningful conclusion insensitive polling sample size density reverted message feel worthwhile sharing revisedenhanced edited publication form regarding social distancing policy crucial pacing spread pandemic resilience Health Care Systems tested brilliant simulator display Health Systems collapse rate concurrent case exceeds capacity thus patient otherwise could saved ICUs dying due lack resource “As discussing going predicting election way approach curve order get meaningful data may shape “social distancing” policy 1 Obviously initial case travel China country business China eg Italy’s fashion industry Germany’s car part complex well interplay Italian German case 2 sure would beneficial consider together bare number profession patient many medical domain doctor nurse caregiver exposed initial patient many related doctor nurse caregiver “delay”rate contagion 3 many public transport air crew driver etc delivery boy supermarketshop employee case related 4 Investigate hobby ChurchSocial Club goer Sport fan attended match etc perform analysis among case church goer football fan etc 5 Check mobility patient v overall mobility index respective segment Professions Hobbies etc Mobility data readily available Google Facebook Apple many country Network Operators 6 Examine available space eg get metric home floor live use public transport work deskjob etc 7 volume data analysed may become hard handle one may issue rule based profile “candidate” example suggestion time exposure virus 15 minute higher probability infection sure epidemiologist accurate data used accordingly filter “irrelevant” data knowledge already available country ahead others may help u make informed decision service block eg may case public transport shutdown well go predictive pattern cluster behaviour “virus” American British official would asked Chinese Italians etc share info — make info available — well take note measure impact adapt accordingly” metaanalysis may also help u crosscheck validate official data provided country known transparency example South Korean “first 500” case may different country’s “first 500” case chart given fact preventative testing Koreans far surpasses testing country thus subsequent correlation normalisation ought considered medical authority get related personal detail address age sex patient authorization access mobile phone’s data order investigate case’s movement thorough forensic analysis may help save life virus relatively mild others future may ready consider preparatory step similar potentially deadly disease may follow Sadly luxury philosophise usage Personal Privacy data Life risk Edit le three week later Apple Google announced collaboration order enable interoperable contact tracing devicesTags Covid 19 AI Big Data Health |
776 | How Google Got Its Employees to Eat Their Vegetables | How Google Got Its Employees to Eat Their Vegetables
The tech giant is engineering a way to encourage its employees to eat healthier — and it might just help the rest of the country
Tina Williams eats greens for breakfast every day. She didn’t always. There was a time when the only vegetables she regularly ate were canned corn or potatoes. But Williams works at Google in New York City, where the food is free and her favorite kale-quinoa-avocado salad is available starting at 8 a.m. each workday.
Growing up, Williams would never have believed that she would one day eat salad for breakfast. Her middle-class family lived outside of Boston, and she remembers feeling sorry for the kid she knew whose mom always bought whole-wheat bread. But over nine years at Google, where she eats breakfast and lunch five days a week, she has learned to like bok choy, a vegetable she previously wouldn’t have recognized in the supermarket, and Brussels sprouts, of which she says, “It turns out I really like when they are well-prepared.”
Williams, who is 35, tall, and fit, now feels good about how she eats. But she knows that her healthier diet depends in large part on Google. When she took maternity leave a few years ago, she didn’t have time to prep greens every morning — never mind that despite multiple attempts, her version of the kale-quinoa salad never turns out quite right. “That’s what I fear if I lose my job,” Williams told me. “The food implications! Which is nuts.”
Google’s free food is a well-known perk, both in and beyond Silicon Valley. The company’s first chef, Charlie Ayers, won his job in 1999 by cooking a meal for Google’s then 40-some employees that included, among other dishes, Sri Lankan chicken curry with roasted pumpkin. It wasn’t long before every ambitious Silicon Valley company was compelled to compete with the legend of Google’s food. A 2014 headline from nerd-food site Serious Eats summed up its reputation: “Lunch at Google HQ Is as Insanely Awesome as You Thought.”
For some time now, Google has been quietly adding a (virtuous) new wrinkle to its food program: It’s no longer enough just to keep its employees happy; it’s trying to make them healthy, too. Over the past five years, the company has taken a typically Google-ish approach to the food it serves — methodical, iterative — to create the largest and most ambitious real-world test of how to nudge people to make healthier choices at mealtime. The campaign isn’t changing just the food itself, but how it’s presented. Google’s tactics include limiting portion sizes for meat and desserts and redesigning its premises to lead its “users” to choose water and fruit over soda and M&M’s. The goal, says Michiel Bakker, Google’s director of global workplace programs, is to make the healthy choice the easy choice and, as in the case of Tina Williams, the preferred one.
The company’s grand experiment matters because getting Americans to eat healthfully has long flummoxed scientists, public health advocates, corporations, and schools, all of which are desperately seeking ways to improve the American diet.
The results, though limited, are impressive. In the kitchens of Google’s New York offices alone, which feed more than 10,000 people daily, the company serves 2,300 breakfast salads every day, up from zero two years ago. Seafood consumption jumped 85% between 2017 and 2018, from 13 to 24 pounds per person, even though the company focuses on more sustainable but less popular species such as trout, octopus, fluke, and shellfish. While soda consumption has remained flat at an average of 20 cans per person per year, water consumption has jumped sharply. In 2018, New York Googlers drank nearly five times more bottled water than bottled sugary drinks — and that doesn’t include the water drunk in cups and free reusable water bottles that Google provides to cut back on its use of plastic.
Good for Google, critics may scoff, but can it be replicated? After all, Google’s parent company, Alphabet, is valued at nearly $1 trillion, and Google has an unusually educated, motivated, and sophisticated workforce. But the company’s grand experiment matters because getting Americans to eat healthfully has long flummoxed scientists, public health advocates, corporations, and schools, all of which are desperately seeking ways to improve the American diet. Obesity affects almost one in five children and one in three adults in the United States, putting them at risk for chronic diseases such as diabetes, heart disease, and cancer. And no effort to reverse it has worked.
The traditional public health campaigns, which specialize in telling people what they should do, routinely fail to alter behavior. Since the early 1990s, agencies including the National Cancer Institute and the Centers for Disease Control and Prevention have been pushing “five a day” initiatives to encourage Americans to get five daily servings of fruits and vegetables, but today, only 13% of Americans eat the recommended servings of fruit, and just 9% eat enough vegetables. The feckless corporate wellness industry — which offers health and fitness programs and incentives to corporate employees — has grown into an $8 billion behemoth without making much difference in the health of American workers or reducing corporate health care costs. And the dieting industry? The most successful results it generates are its $66 billion in annual revenue.
Google’s strategy, in contrast, is simple, subtle, and replicable. With the exception of the panoramic city views, the Google cafés I visited at the company’s New York offices looked a lot like the coffee bars, fast-casual burrito shops, made-to-order salad bars, and buffets you’ll find anywhere else. But the small changes make big differences. The plates on the buffet line are only eight to 10 inches wide, versus a standard 12 inches, which effectively limits serving sizes. Vegetables always come first on the line, so by the time you get to the meat or the snickerdoodles and chocolate tarts, there’s not much space on your plate. “Spa water,” bobbing with strawberries or cucumbers or lemons, is everywhere — and deliberately more accessible than sugary drinks or even bottled water. A burrito at Google weighed in at about 10 ounces — 60% smaller than the whopping one-pound nine-ounce log filled with similar ingredients that I picked up at a Chipotle near my home in Washington, D.C.
In other words, it’s a vision of what sensible eating could look like. Through small, intentional choices, Google has conjured a world for its employees where there are no 20-ounce Caramel Frappuccinos, no Triple Whoppers, and no “endless” shrimp, pasta, and breadsticks. “What Google is attempting here is culture change,” says David Katz, MD, MPH, founding director of Yale University’s Prevention Research Center and president of the True Health Initiative. “And that’s the level we have to reach to transform behaviors and health for a lifetime.” | https://onezero.medium.com/how-google-got-its-employees-to-eat-their-vegetables-a2206820d90d | ['Jane Black'] | 2020-02-10 02:38:42.315000+00:00 | ['Google', 'Vegetables', 'Engineering', 'Health', 'Industry'] | Title Google Got Employees Eat VegetablesContent Google Got Employees Eat Vegetables tech giant engineering way encourage employee eat healthier — might help rest country Tina Williams eats green breakfast every day didn’t always time vegetable regularly ate canned corn potato Williams work Google New York City food free favorite kalequinoaavocado salad available starting 8 workday Growing Williams would never believed would one day eat salad breakfast middleclass family lived outside Boston remembers feeling sorry kid knew whose mom always bought wholewheat bread nine year Google eats breakfast lunch five day week learned like bok choy vegetable previously wouldn’t recognized supermarket Brussels sprout say “It turn really like wellprepared” Williams 35 tall fit feel good eats know healthier diet depends large part Google took maternity leave year ago didn’t time prep green every morning — never mind despite multiple attempt version kalequinoa salad never turn quite right “That’s fear lose job” Williams told “The food implication nuts” Google’s free food wellknown perk beyond Silicon Valley company’s first chef Charlie Ayers job 1999 cooking meal Google’s 40some employee included among dish Sri Lankan chicken curry roasted pumpkin wasn’t long every ambitious Silicon Valley company compelled compete legend Google’s food 2014 headline nerdfood site Serious Eats summed reputation “Lunch Google HQ Insanely Awesome Thought” time Google quietly adding virtuous new wrinkle food program It’s longer enough keep employee happy it’s trying make healthy past five year company taken typically Googleish approach food serf — methodical iterative — create largest ambitious realworld test nudge people make healthier choice mealtime campaign isn’t changing food it’s presented Google’s tactic include limiting portion size meat dessert redesigning premise lead “users” choose water fruit soda MM’s goal say Michiel Bakker Google’s director global workplace program make healthy choice easy choice case Tina Williams preferred one company’s grand experiment matter getting Americans eat healthfully long flummoxed scientist public health advocate corporation school desperately seeking way improve American diet result though limited impressive kitchen Google’s New York office alone feed 10000 people daily company serf 2300 breakfast salad every day zero two year ago Seafood consumption jumped 85 2017 2018 13 24 pound per person even though company focus sustainable le popular specie trout octopus fluke shellfish soda consumption remained flat average 20 can per person per year water consumption jumped sharply 2018 New York Googlers drank nearly five time bottled water bottled sugary drink — doesn’t include water drunk cup free reusable water bottle Google provides cut back use plastic Good Google critic may scoff replicated Google’s parent company Alphabet valued nearly 1 trillion Google unusually educated motivated sophisticated workforce company’s grand experiment matter getting Americans eat healthfully long flummoxed scientist public health advocate corporation school desperately seeking way improve American diet Obesity affect almost one five child one three adult United States putting risk chronic disease diabetes heart disease cancer effort reverse worked traditional public health campaign specialize telling people routinely fail alter behavior Since early 1990s agency including National Cancer Institute Centers Disease Control Prevention pushing “five day” initiative encourage Americans get five daily serving fruit vegetable today 13 Americans eat recommended serving fruit 9 eat enough vegetable feckless corporate wellness industry — offer health fitness program incentive corporate employee — grown 8 billion behemoth without making much difference health American worker reducing corporate health care cost dieting industry successful result generates 66 billion annual revenue Google’s strategy contrast simple subtle replicable exception panoramic city view Google cafés visited company’s New York office looked lot like coffee bar fastcasual burrito shop madetoorder salad bar buffet you’ll find anywhere else small change make big difference plate buffet line eight 10 inch wide versus standard 12 inch effectively limit serving size Vegetables always come first line time get meat snickerdoodles chocolate tart there’s much space plate “Spa water” bobbing strawberry cucumber lemon everywhere — deliberately accessible sugary drink even bottled water burrito Google weighed 10 ounce — 60 smaller whopping onepound nineounce log filled similar ingredient picked Chipotle near home Washington DC word it’s vision sensible eating could look like small intentional choice Google conjured world employee 20ounce Caramel Frappuccinos Triple Whoppers “endless” shrimp pasta breadstick “What Google attempting culture change” say David Katz MD MPH founding director Yale University’s Prevention Research Center president True Health Initiative “And that’s level reach transform behavior health lifetime”Tags Google Vegetables Engineering Health Industry |
777 | Burnout Symptoms vs. Depression Symptoms | Every year, dozens of medical leaders and health care administrators travel to Half Moon Bay, California, a beach town 30 miles south of San Francisco, to learn how to help their doctors combat burnout. The Chief Wellness Officer course is part of Stanford’s WellMD Center, an initiative designed to keep stressed-out doctors mentally and physically healthy by promoting evidence-based practices like exercise and mindfulness.
The idea of “burnout” goes back to the 1970s, when American psychologist Herbert Freudenberger coined the term to explain the consequences of extreme stress in “helping” professions like health care.
Since then, the meaning of “burnout” has evolved. It now applies to more professions than just health care workers and includes more symptoms — for example, “errand paralysis,” as writer Anne Helen Petersen describes her inability to complete mundane tasks. In her viral BuzzFeed article about millennials and burnout, Petersen argues that anyone can get burned out, because it’s the product of a culture that demands people go on working even when their internal resources are depleted.
By definition, burnout is more than just “work stress.” Being burned out can also lead to deterioration in the quality of one’s work, which can result in a negative self-image. The ongoing stress that accompanies burnout is also linked to a heightened risk for mental illness: According to a recent study of burned-out resident doctors, burnout can contribute to a higher risk for depression, along with suicide and substance abuse.
Recently, some researchers have attempted to capture exactly how much burnout affects U.S. workers: A 2018 study found 28% of the general U.S. workforce experiences “overall burnout,” while a Gallup study from the same year found that 23% of American workers report feeling burned out “often or always.”
Burnout has become such a hallmark of American culture, especially among millennials, that Petersen calls it “the contemporary condition.” But as much as it defines the zeitgeist, only recently was burnout formally recognized as an occupational phenomenon by the World Health Organization (WHO), which defines it as “a syndrome conceptualized as resulting from chronic workplace stress that has not been successfully managed.”
The link between burnout and depression
The WHO maintains that burnout is in its own category, separate from any medical or mental health diagnosis, including depression. Some experts believe burnout is a risk factor for developing depression — or, for those who are already diagnosed with major depressive disorder, a trigger for a depressive episode. But they disagree that the two basically describe the same condition.
“I can tell you that the symptoms look identical,” says Dattilo. “When someone describes the symptoms of burnout, those are all going to show up in depression, too.”
Other researchers think we’re misunderstanding burnout altogether — and that rather than a totally separate phenomenon, it’s just another name for depression.
Depression and burnout symptoms can look a lot alike. Natalie Dattilo, PhD, a psychologist at Brigham and Women’s Hospital in Boston, says they’re usually both driven by a sense of helplessness. Both burned-out and depressed people often feel ineffective, like they can’t gain traction — that nothing seems to be working, no matter how hard they try to get out of a slump.
“I can tell you that the symptoms look identical,” says Dattilo. “There’s not much distinction. When someone describes the symptoms of burnout, those are all going to show up in depression, too.”
Irvin Schonfeld, PhD, a professor of psychology at The City College of New York and the CUNY Graduate Center, researches the correlation between depression and burnout. He says it’s the emotional exhaustion, the core component of burnout, that mirrors depression most distinctly.
“Depression and emotional exhaustion symptoms correlate with each other highly enough that they appear to be measuring the same thing,” says Schonfeld. “The emotional exhaustion component of burnout, which is the core of burnout, is thus reflective of a depressive condition.”
He has conducted a number of studies on the correlation between burnout symptoms and depression symptoms. His research focuses on comparing the continuums of emotional exhaustion and depression, using symptoms from depression screening surveys like the Patient Health Questionnaire (also known as the PHQ-9).
In a 2016 study of 1,386 schoolteachers, Schonfeld and his colleagues found that subjects’ scores on depression symptom scales and the emotional exhaustion scale for burnout correlated very highly. In further studies, Schonfeld and co-author Renzo Bianchi found similarly high correlations between emotional exhaustion and other depression scales.
In a separate 2016 study of 6,351 Austrian physicians, researchers found that doctors who had symptoms of burnout were more likely to suffer from depression, and that the three components of burnout symptoms tended to correlate more highly with depression than with each other.
Schonfeld and Bianchi’s work found the same correlation, suggesting that burnout and depression may not be distinct experiences.
Is burnout just depression at work?
Many researchers view burnout symptoms as strictly a work-related syndrome, separate from any medical condition, because the majority of burnout research is related to occupational burnout. Depression, on the other hand, is seen as more omnipresent and universal — a medical condition that can impact all aspects of a patient’s life.
But Schonfeld says the distinction between someone’s work and home life isn’t so black and white. Someone’s work environment can contribute to their existing general depression, or even cause new depressive symptoms. “There is a lot of evidence that indicates depressive symptoms become elevated when people work in adverse conditions,” he says. Particularly, people who have jobs with heavy workloads and little autonomy are at risk for developing elevated levels of depressive symptoms.
For example, Schonfeld’s work suggests that when teachers get jobs where they’re exposed to violence or disrespect, their depressive symptom levels rise.
Dattilo says this is especially true if work-based burnout begins to strain other areas of someone’s life, which is common, since stress often comes with physical and emotional symptoms. “If burnout affects you in other areas, like your relationships or your health — if you feel in general like nothing you do seems to make a difference — that’s going to feel and look a lot more like depression.”
Burnout might be a less stigmatizing way of talking about depression
Since burnout and depression can look the same clinically and there’s scientific evidence they’re not separate conditions, why is depression so commonly pushed aside to an entirely separate category? Labeling emotional exhaustion and its associated symptoms as “burnout” could be easier to swallow than calling yourself “depressed.”
“I think there is less stigma attached to the word ‘burnout’ than the word ‘depression,’” Schonfeld says. “Some people may prefer to call what they are suffering ‘burnout’ for that reason.”
For example, physicians who are overworked and emotionally exhausted could put their jobs in jeopardy if they label their symptoms as a mental illness like depression. And people who recognize their burnout symptoms as depression might find themselves even more exhausted by the hurdles treating depression brings, like accessing medical care.
But for people to manage their burnout and the health symptoms associated with exposure to chronic stress, Schonfeld says it’s important to recognize burnout for what it is: a kind of depression that occurs in reaction to terrible working conditions. “There are evidence-based treatments for depression, and people can be helped by them. We have to get these people the help they need,” he says.
Redefining burnout means changing how it’s treated
It may seem like an issue of semantics, but how psychologists and organizations understand burnout is the basis for how it’s treated. If someone is burned out in the traditional work context, it might seem like a natural solution to take some extra time off work.
But if burnout is recognized as depression, then a few personal days isn’t the right way to manage it.
In general, Schonfeld says people with burnout aren’t going to see many benefits to their mental health if their only approach is taking a few days off — the “half-life” of a vacation is about two weeks, which means the same negative feelings will kick in once people get back to their normal routines at work. Instead, people with depressive symptoms would be better served by the support of a psychotherapist or psychiatrist who can prescribe an antidepressant if necessary.
Dattilo says the priority should be getting people the help they need to deal with emotional exhaustion. “In research, it’s important to have clearly defined criteria so you know exactly what you’re trying to study, but in clinical practice, I’m more interested in helping people find relief from their symptoms than whether we call it ‘burnout’ or depression,’” she says. “No matter what you call it, it’s important to deal with your mental health in work and in life. There’s no reason to suffer.” | https://elemental.medium.com/what-if-burnout-is-actually-depression-7c69125300e9 | ['Ashley Abramson'] | 2020-02-17 15:49:05.018000+00:00 | ['Mental Health', 'Depression', 'Brain', 'Psychology', 'Burnout'] | Title Burnout Symptoms v Depression SymptomsContent Every year dozen medical leader health care administrator travel Half Moon Bay California beach town 30 mile south San Francisco learn help doctor combat burnout Chief Wellness Officer course part Stanford’s WellMD Center initiative designed keep stressedout doctor mentally physically healthy promoting evidencebased practice like exercise mindfulness idea “burnout” go back 1970s American psychologist Herbert Freudenberger coined term explain consequence extreme stress “helping” profession like health care Since meaning “burnout” evolved applies profession health care worker includes symptom — example “errand paralysis” writer Anne Helen Petersen describes inability complete mundane task viral BuzzFeed article millennials burnout Petersen argues anyone get burned it’s product culture demand people go working even internal resource depleted definition burnout “work stress” burned also lead deterioration quality one’s work result negative selfimage ongoing stress accompanies burnout also linked heightened risk mental illness According recent study burnedout resident doctor burnout contribute higher risk depression along suicide substance abuse Recently researcher attempted capture exactly much burnout affect US worker 2018 study found 28 general US workforce experience “overall burnout” Gallup study year found 23 American worker report feeling burned “often always” Burnout become hallmark American culture especially among millennials Petersen call “the contemporary condition” much defines zeitgeist recently burnout formally recognized occupational phenomenon World Health Organization defines “a syndrome conceptualized resulting chronic workplace stress successfully managed” link burnout depression maintains burnout category separate medical mental health diagnosis including depression expert believe burnout risk factor developing depression — already diagnosed major depressive disorder trigger depressive episode disagree two basically describe condition “I tell symptom look identical” say Dattilo “When someone describes symptom burnout going show depression too” researcher think we’re misunderstanding burnout altogether — rather totally separate phenomenon it’s another name depression Depression burnout symptom look lot alike Natalie Dattilo PhD psychologist Brigham Women’s Hospital Boston say they’re usually driven sense helplessness burnedout depressed people often feel ineffective like can’t gain traction — nothing seems working matter hard try get slump “I tell symptom look identical” say Dattilo “There’s much distinction someone describes symptom burnout going show depression too” Irvin Schonfeld PhD professor psychology City College New York CUNY Graduate Center research correlation depression burnout say it’s emotional exhaustion core component burnout mirror depression distinctly “Depression emotional exhaustion symptom correlate highly enough appear measuring thing” say Schonfeld “The emotional exhaustion component burnout core burnout thus reflective depressive condition” conducted number study correlation burnout symptom depression symptom research focus comparing continuum emotional exhaustion depression using symptom depression screening survey like Patient Health Questionnaire also known PHQ9 2016 study 1386 schoolteacher Schonfeld colleague found subjects’ score depression symptom scale emotional exhaustion scale burnout correlated highly study Schonfeld coauthor Renzo Bianchi found similarly high correlation emotional exhaustion depression scale separate 2016 study 6351 Austrian physician researcher found doctor symptom burnout likely suffer depression three component burnout symptom tended correlate highly depression Schonfeld Bianchi’s work found correlation suggesting burnout depression may distinct experience burnout depression work Many researcher view burnout symptom strictly workrelated syndrome separate medical condition majority burnout research related occupational burnout Depression hand seen omnipresent universal — medical condition impact aspect patient’s life Schonfeld say distinction someone’s work home life isn’t black white Someone’s work environment contribute existing general depression even cause new depressive symptom “There lot evidence indicates depressive symptom become elevated people work adverse conditions” say Particularly people job heavy workload little autonomy risk developing elevated level depressive symptom example Schonfeld’s work suggests teacher get job they’re exposed violence disrespect depressive symptom level rise Dattilo say especially true workbased burnout begin strain area someone’s life common since stress often come physical emotional symptom “If burnout affect area like relationship health — feel general like nothing seems make difference — that’s going feel look lot like depression” Burnout might le stigmatizing way talking depression Since burnout depression look clinically there’s scientific evidence they’re separate condition depression commonly pushed aside entirely separate category Labeling emotional exhaustion associated symptom “burnout” could easier swallow calling “depressed” “I think le stigma attached word ‘burnout’ word ‘depression’” Schonfeld say “Some people may prefer call suffering ‘burnout’ reason” example physician overworked emotionally exhausted could put job jeopardy label symptom mental illness like depression people recognize burnout symptom depression might find even exhausted hurdle treating depression brings like accessing medical care people manage burnout health symptom associated exposure chronic stress Schonfeld say it’s important recognize burnout kind depression occurs reaction terrible working condition “There evidencebased treatment depression people helped get people help need” say Redefining burnout mean changing it’s treated may seem like issue semantics psychologist organization understand burnout basis it’s treated someone burned traditional work context might seem like natural solution take extra time work burnout recognized depression personal day isn’t right way manage general Schonfeld say people burnout aren’t going see many benefit mental health approach taking day — “halflife” vacation two week mean negative feeling kick people get back normal routine work Instead people depressive symptom would better served support psychotherapist psychiatrist prescribe antidepressant necessary Dattilo say priority getting people help need deal emotional exhaustion “In research it’s important clearly defined criterion know exactly you’re trying study clinical practice I’m interested helping people find relief symptom whether call ‘burnout’ depression’” say “No matter call it’s important deal mental health work life There’s reason suffer”Tags Mental Health Depression Brain Psychology Burnout |
778 | Peer Review is Not Scientific | Peer Review is Not Scientific
How a process designed to ensure scientific rigor is tainted by randomness, bias, and arbitrary delays.
A mess of papers. Image by Christa Dodoo, edited by author in Pixlr.
In my last essay, I discussed the ways in which the academic publishing process is unnecessarily slow, labor-intensive, and exploitative of researchers who are not paid for the research and writing they devote years of their lives to. Academic journal articles are mightily expensive to get access to ($35–55 for a single article, thousands of dollars for an annual subscription), and yet the people who collected the data, analyzed it, wrote it up, submitted it for publication, and revised it receive no compensation no matter how much money is spent accessing their work.
Today, I’d like to set my sights on another troublingly biased and sloppy aspect of the academic publishing process: Peer review. The peer-reviewed process, which is intended to boost the rigor and objectivity of scientific work, is not itself done in an objective, systematic, or scientific way. In fact, it’s one of the most scattershot, inconsistent messes possible.
The process of selecting and assigning reviewers is unsystematic and filled with room for bias and error; review processes themselves are not standardized in any way; the content of reviews is often arbitrary and influenced by personal agendas; it takes a miserably long time to conduct a review; and worst of all, no one involved in the process earns a cent.
…
The first thing I want all lovers of science to know is this: peer-reviewers are not paid. When you are contacted by a journal editor and asked to conduct a review, there is no discussion of payment, because no payment is available. Ever. Furthermore, peer reviewing is not associated in any direct way with the actual job of being a professor or researcher. The person asking you to conduct a peer review is not your supervisor or the chair of your department, in nearly any circumstance. Your employer does not keep track of how many peer reviews you conduct and reward you appropriately.
Instead, you’re asked by journal editors, via email, on a voluntary basis. And it’s up to you, as a busy faculty member, graduate student, post-doc, or adjunct, to decide whether to say yes or not.
The process is typically anonymized, and tends to be relatively thankless — no one except the editor who has asked you to conduct the review will know that you were involved in the process. There is no quota of reviews a faculty member is expected to provide. Providing a review cannot really be placed on your resume or CV in any meaningful way.
The journal article you have reviewed, if it gets published, may be placed in an immensely expensive journal, and your hard work may have helped the journal maintain its reputation as respectable and rigorous, but you aren’t gonna see a cent of that money regardless, and you’re not going to get any accolades for doing that free labor, either. Like many informal academic responsibilities, such as serving on committees or mentoring students, you are expected to do it for the love of doing it –while also juggling the responsibilities you are actually evaluated on.
…
Image by Jessica Lewis, courtesy of Unsplash. Remixed in Pixlr by author.
Why does a researcher get asked to provide a peer review? Usually because they have published research in the topic area before. This makes sense, of course. You want reviewers to be people who are well-versed in a subject, who know its terminology and the theories that undergird it, who have some passing familiarity, at least, with the current gaps in the literature and any ongoing debates.
The problem is, the process of identifying and contacting a reviewer is not systematic. Not everyone who has conducted work in a topic area is asked to review. Someone who has published a ton of work in journal A (but not in journal B) may never be asked to provide a peer-review on a publication in journal B, no matter how relevant it is to their work. If an editor hasn’t heard of a person, or has never met them in person, they may not consider them to be a viable peer reviewer. Reviewers who are a good fit may not have the time to actually conduct a review.
As a result, who gets contacted first, who actually has the time to do a review, and who actually ends up providing the review has a lot of arbitrary elements to it. And often, the people providing reviews are graduate students, post-docs, and others who are lower in the academic hierarchy — because they haven’t learned to say “no” to unpaid, thankless jobs yet.
If you’re a busy professor with tenure, it’s likely you have no time, and no motivation, to complete a peer review. You’re busy teaching classes, grading papers, attending committee meetings, and mentoring graduate students of your own — that all requires a lot of reading, and thinking, and writing feedback, and sending emails. The last thing you want to do is get bleary-eyed looking over some additional half-formed research draft that it isn’t a requirement of your job for you to look at.
And if you have a nice, tenure-track academic job, you get very little out of providing a peer review, besides a pat on the back for helping the community, and maybe the chance to criticize a peer whose work you disagree with.
…
Conversely, if you’re a grad student, a postdoc, or a new professor, you probably feel like you are obligated to peer review manuscripts when asked, because you think it’s going to help establish you and help your career in some way. Though it probably won’t. You may feel, the first few times that you are asked, that being asked is an immense honor. It isn’t. You might even believe that providing a high-quality review will help you establish yourself in your subfield. That won’t happen.
Reviewers, as I’ve already mentioned, are largely anonymous. Only the editor who has asked you to do a review will know that you’ve put in the work. They might be grateful to you for doing them a solid, but unless you forge a real bond with them, it’s probably not consequential. Your university won’t know you’ve provided a peer review. Other researchers in your subfield won’t have any idea. Your peers or classmates won’t know. Your mentor will not be impressed.
Even if you tell people about the work you’ve been peer reviewing, they won’t exactly be impressed. Providing peer reviews is seen as necessary, but unimpressive, like going to conferences and providing service. It’s a thankless task that aspiring faculty are nonetheless pressured to take part in. And, like going to conferences and serving on committees, it ends up eating a lot of time and resources while remaining unpaid, unremarked upon, and unrewarded.
…
If you’re a high profile researcher in your field, you’ll tend to get more review requests. Because of the volume of requests, and because doing reviews is tiring and useless, you can’t take them all on. So you turn down the review requests, and they trickle down to your grad students or to less-established researchers who are jockeying to prove they are productive by any means necessary, including means that don’t actually help.
Since nobody is required to do a specific number of reviews, or any reviews at all, and there’s no payment for the service, the work of peer reviewing is doled out in an illogical way, one that is definitely not merit-based. One article may, by random chance, be assigned to two well-known, well-established reviewers who happened to feel like giving back to the scientific community on the day that the editor asked them; another article on the exact same topic may be assigned to two graduate students who have never reviewed an article before. Editors try their damndest to select strong reviewers, with a variety of backgrounds, but given there is no financial or professional incentive for the work, doing so is a major swim upstream.
…
Image by Chuttersnap, courtesy of Unsplash. Remixed in pixlr by author.
Having established that the process of selecting reviewers is scattershot & not at all systematic, let’s talk about the process of review itself.
It’s also not at all objective. There is no training in how to do it well. Educators have known for a long time that giving consistent, useful feedback on a piece of writing often requires using a rubric or some standard marking scheme. Researchers have known for decades that if you want to consistently analyze human performance, you need training and practice in doing so.
Despite this, there is no rubric for the peer review process. A reviewer’s process and performance is not evaluated or examined for consistency and rigor. Reviewers are not trained. There are no mutually agreed upon professional standards that are used to assess the work’s quality. In my field, social psychology, a field that is plagued with questionable research practices, we don’t even have rules in place for how to make sure a study is legitimate. We do have some navel-gazey journal articles that muse on the subject, though.
So what actually happens, then, when an article is submitted to reviewers? Basically, you are subjected to the whims of the reviewers you have arbitrarily been assigned. You have no idea what to expect from them, going into the process. Who knows what agendas, pet peeves, & writing style preferences your reviewers are gonna have. Who knows how much time they’re going to put into reading your work, or how competent they actually are. You don’t even know who they are. You just know they’ve probably been published in the journal before.
…
The level of scrutiny that an article is subjected to all comes down to chance. If you’re assigned a reviewer who created a theory that opposes your own theory, your work is likely to be picked apart. The reviewer will look very closely for flaws and take issue with everything that they can. This is not inherently a bad thing — research should be closely reviewed — but it’s not unbiased either.
Your reviewer, if they have an ax to grind, may not search for flaws for the sake of science, but for the sake of their own career. And they may go far, far easier on a researcher whom they like or have collaborated with in the past. Scientists are humans. We fall prey to the same cognitive biases as the people we study, and we’re pretty un-self-reflective about that. And making sure a reviewer is appropriately strident in their critiques is not usually the journal editor’s job.
In addition to ax-grinding reviewers and dispositionally friendly reviewers, there are reviewers who don’t give a shit. These are typically overwhelmed people, or folks who are in over their heads. Either because of workload demands or a lack of experience and training, they cannot provide a useful review on the literature they’ve been given. They may delay providing feedback, or only make a few surface level comments that are not particularly useful. Or, in an attempt to appear helpful and critical, they may make a ton of irrelevant recommendations that delay publication, but don’t improve the quality of the work.
…
Many of the comments you do get from reviewers, no matter who they are, will be unsubstantive or subjective. A person may ask you to cite their own work more. Someone who prefers “myriads of” to “a myriad of” may criticize your grammar even though both usages are technically fine. A reviewer who doesn’t believe sexism exists may ask the author to remove language acknowledging pretty basic, well-established social justice issues. A reviewer who has a favorite pet statistical technique, or pet peeve, may ask for additional analyses to satisfy their own neuroses.
A lot of times, reviewers approach a paper by asking, “How would I have done this?” rather than “Was this done well?”. This leads to a lot of recommendations that come down to taste and tradition. And this is how problematic analytical processes (like p-hacking) get passed down from generation to generation.
A lot of reviewers, especially rookies, feel the need to prove that they have put a lot of work into the review, so they make a ton of recommendations pretty much just for the sake of recommending something. Authors may be asked to add collect several studies’ worth of new data, or they may be told to add a large section to their literature review, citing work that is related to the reviewer’s own work. Structural changes to the paper may be requested, to appeal to one reviewer’s personal preferences or biases. Sometimes these changes improve the paper as a whole, but often they just delay everything unnecessarily.
…
Image by Henrique Ferreira, remixed in Pixlr by author.
The peer-review process is also, unfortunately, subjected to all kinds of other human biases, including sexism and nepotism. While reviewers’ names are usually masked, author’s names are not always anonymous to the reviewers themselves, allowing for implicit assumptions about the authors to creep in. Cordelia Fine famously reported, in her book Delusions of Gender, about a trans male researcher who was perceived as more intelligent and capable by his colleagues once he started using a male name. He even heard people off-handedly mentioning that he was a “much better researcher” than his “sister”. He had no sister — people were talking about his pre-transition self.
Of course, that instance is anecdotal. But the data suggests widespread sexism biases the peer-review process, too. Wennerås and Wold conducted a large-scale review of postdoctoral fellowship applications in 2010, and found evidence of sexism and nepotism. Women were far less likely to receive a fellowship, compared to men; when applicants were close friends with reviewers, they were far more likely to be accepted. In a meta-analytic study published in 2007, Bornmann and colleagues found that, under a peer-review process, “men have statistically significant greater odds of receiving grants than women by about 7%.”
From the Bornmann et al paper.
I was unable to find published, empirical work establishing whether or not the peer review process is poisoned by racism. However, reviewers can often discern (or guess at) an author’s race or ethnicity based on their name, and prior work has established that when a name on a job application or resume is perceived as a “black” name, the applicant is more closely scrutinized and less likely to be selected. It is highly likely the same process occurs when a reviewer knows the names of article authors as well.
Of course, even when an author’s name is hidden from reviewers, there is still room for bias and prejudice to seep in. Work on racism, sexism, ableism, or any other structural form of oppression may be more strictly reviewed by a white, cis male author who is uncomfortable being reminded that the world is not a fair place.
If you’ve ever attended a research presentation in the social sciences, you’ve witnessed a well-off, established white male researcher attempt to pick apart the work of a black or female colleague, not on the basis of merit, but because he doesn’t believe the world is as bigoted as black and female people say it is. Scientists love to believe they are unflinchingly logical, which can make it even harder for them to acknowledge their own biases than it is for the average person.
…
Did I mention that it may take a reviewer months to return a draft with feedback? That some reviewers take upwards of a year? And that editors often have to remind, cajole, and harass their reviewers into getting a review back, even then? An author may be stuck waiting for semesters to hear back about an article, only to have their reviewer hastily read the work in a day and type together some surface-level critiques, based on subjectivities, and send them off to the editor without a second thought.
The slowness of this process is clearly related to the fact that reviewers are not paid, and are not assigned review duties as part of their official jobs. It’s a whole separate, unmeasured thing. Of course it’s gonna be a low priority when you have classes to teach, manuscripts of your own to write, and job talks to attend.
Image by Jon Tyson, courtesy of Unsplash. Edited in Pixlr by author.
…
This is all worsened by the fact that academic journals do not allow simultaneous submissions. You can only submit a paper to one journal at a time. So you’re truly at the behest of 2–3 random reviewers, plus a journal editor, for months or years at a time, before your work can be seen. If it gets accepted the first time. Which statistically speaking, it probably won’t.
Most submissions are not accepted. If you’re lucky, the editor will “desk reject” your work within a few days. This happens when an article is clearly not a good fit for the publication, or it’s so shot full of flaws that the editor can tell it won’t pass reviewer muster. Getting a desk reject saves time at least. It allows an author to make big revisions or submit to an entirely different journal, rather than waiting around for months only to get turned down anyway.
Even after a full review process, most articles get rejected. This is part of how journals establish that they are rigorous — a high rejection rate looks good. It implies thoroughness, albeit in a tail-wags-the-dog kind of way. Many suitable manuscripts are sacrificed at the altar of rigor. A small but “fortunate” subset, instead, gets the exhausting but slightly hopeful “revise and resubmit”.
…
A “revise and resubmit” is not a guarantee that an article will be accepted after revisions are made. It’s merely an opportunity for the author to respond to comments, make some adjustments, and try for acceptance into a journal once again.
Sometimes revisions are merely structural or theoretical. Sometimes an author is asked to re-analyze their data. But often, especially in the social sciences, a research is expected to go out and collect additional data, analyze it, write it up, and then send the new draft to the reviewers once again.
Regardless of how long revisions take, they will then be sent back to the reviewers and the editor, where they may languish, unread, for months at a time. Finally, the reviewer may return to the draft, at this point perhaps not even remembering their original feedback. The author will, in all likelihood, receive a second round of feedback that is just as fuzzy and subjective as the first round ways, many months or years after the original submission. And that’s if none of the original reviewers have dropped off the map / gotten too swamped to finish the review / died.
…
Image by Samuel Zeller, courtesy of Unsplash. Edited in Pixlr by author.
Things don’t always go poorly, of course. Sometimes, reviews are relatively quick (a month or two), revision requests make sense, and revised work is accepted. That has happened to me, and my colleagues, quite a few times. In instances like those, the peer review process is a lot better than nothing. But that does not mean that it is adequate, or logical, or systematic.
Because the process is secretive, hard, long, and tedious, it has the illusion of contributing value and rationality to scientific work. And because scientists are the ones conducting it, it is assumed to be a scientific practice. But in practice, it’s not much better than receiving a few random half-considered comments on a conference presentation from whomever happens to show up. And there’s no structure, training, or pay scale in place to make it better.
Peer review is not a scientific process. Scientists are only scientists when they are following the scientific method. When they are giving feedback to a colleague based on gut feelings, personal preferences, biases, and limited resources, they are just humans, and they’re as inconsistent and sloppy as anyone else.
…
This messy, slipshod process delays the public’s access to novel scientific work. It means that published work is not held to a consistently high, yet fair, standard. It results in the lay public having a great deal of faith in work that they assume has been carefully and thoroughly vetted, but which they cannot easily verify. And it also puts an immense amount of arbitrary barriers between researchers and their professional goals. Publications are necessary for a researcher to receive tenure, or to have a prayer of netting a competitive academic job. And yet the process of getting publications is barely related to a researcher’s productivity or merit at all.
But there are alternatives. In the realm of mathematics, for example, journal articles are seen as an inefficient and illogical method of sharing knowledge with colleagues and the public-at-large. Mathematical proofs are instead shared rapidly, and freely, among the handful of specialists who take an interest in it. And when work is well done, it speaks for itself. Either the proof is successful in establishing what the writer set out to establish…or it isn’t. There’s no need to introduce random barriers to make the process look more logical. The work is already logical.
Image by Marten vanden Huevel, courtesy of Unsplash.
In my next article in this series, I will discuss how the academic publishing process can be reformed, taking some inspiration from the mathematics world, as well as from the growing world of open science. I’m still actively researching these alternative options, and the writing other people have done on the subject. If you’re a researcher and you have insight into how the peer-review process can be reformed, or what ought to take its place, please join the conversation in the comments, by tweeting at me (@dr_eprice), or by writing an essay of your own. | https://devonprice.medium.com/peer-review-is-not-scientific-bf3283069ffd | ['Devon Price'] | 2018-06-18 21:05:51.613000+00:00 | ['Education', 'Publishing', 'Science', 'Psychology', 'Writing'] | Title Peer Review ScientificContent Peer Review Scientific process designed ensure scientific rigor tainted randomness bias arbitrary delay mess paper Image Christa Dodoo edited author Pixlr last essay discussed way academic publishing process unnecessarily slow laborintensive exploitative researcher paid research writing devote year life Academic journal article mightily expensive get access 35–55 single article thousand dollar annual subscription yet people collected data analyzed wrote submitted publication revised receive compensation matter much money spent accessing work Today I’d like set sight another troublingly biased sloppy aspect academic publishing process Peer review peerreviewed process intended boost rigor objectivity scientific work done objective systematic scientific way fact it’s one scattershot inconsistent mess possible process selecting assigning reviewer unsystematic filled room bias error review process standardized way content review often arbitrary influenced personal agenda take miserably long time conduct review worst one involved process earns cent … first thing want lover science know peerreviewers paid contacted journal editor asked conduct review discussion payment payment available Ever Furthermore peer reviewing associated direct way actual job professor researcher person asking conduct peer review supervisor chair department nearly circumstance employer keep track many peer review conduct reward appropriately Instead you’re asked journal editor via email voluntary basis it’s busy faculty member graduate student postdoc adjunct decide whether say yes process typically anonymized tends relatively thankless — one except editor asked conduct review know involved process quota review faculty member expected provide Providing review cannot really placed resume CV meaningful way journal article reviewed get published may placed immensely expensive journal hard work may helped journal maintain reputation respectable rigorous aren’t gonna see cent money regardless you’re going get accolade free labor either Like many informal academic responsibility serving committee mentoring student expected love –while also juggling responsibility actually evaluated … Image Jessica Lewis courtesy Unsplash Remixed Pixlr author researcher get asked provide peer review Usually published research topic area make sense course want reviewer people wellversed subject know terminology theory undergird passing familiarity least current gap literature ongoing debate problem process identifying contacting reviewer systematic everyone conducted work topic area asked review Someone published ton work journal journal B may never asked provide peerreview publication journal B matter relevant work editor hasn’t heard person never met person may consider viable peer reviewer Reviewers good fit may time actually conduct review result get contacted first actually time review actually end providing review lot arbitrary element often people providing review graduate student postdoc others lower academic hierarchy — haven’t learned say “no” unpaid thankless job yet you’re busy professor tenure it’s likely time motivation complete peer review You’re busy teaching class grading paper attending committee meeting mentoring graduate student — requires lot reading thinking writing feedback sending email last thing want get blearyeyed looking additional halfformed research draft isn’t requirement job look nice tenuretrack academic job get little providing peer review besides pat back helping community maybe chance criticize peer whose work disagree … Conversely you’re grad student postdoc new professor probably feel like obligated peer review manuscript asked think it’s going help establish help career way Though probably won’t may feel first time asked asked immense honor isn’t might even believe providing highquality review help establish subfield won’t happen Reviewers I’ve already mentioned largely anonymous editor asked review know you’ve put work might grateful solid unless forge real bond it’s probably consequential university won’t know you’ve provided peer review researcher subfield won’t idea peer classmate won’t know mentor impressed Even tell people work you’ve peer reviewing won’t exactly impressed Providing peer review seen necessary unimpressive like going conference providing service It’s thankless task aspiring faculty nonetheless pressured take part like going conference serving committee end eating lot time resource remaining unpaid unremarked upon unrewarded … you’re high profile researcher field you’ll tend get review request volume request review tiring useless can’t take turn review request trickle grad student lessestablished researcher jockeying prove productive mean necessary including mean don’t actually help Since nobody required specific number review review there’s payment service work peer reviewing doled illogical way one definitely meritbased One article may random chance assigned two wellknown wellestablished reviewer happened feel like giving back scientific community day editor asked another article exact topic may assigned two graduate student never reviewed article Editors try damndest select strong reviewer variety background given financial professional incentive work major swim upstream … Image Chuttersnap courtesy Unsplash Remixed pixlr author established process selecting reviewer scattershot systematic let’s talk process review It’s also objective training well Educators known long time giving consistent useful feedback piece writing often requires using rubric standard marking scheme Researchers known decade want consistently analyze human performance need training practice Despite rubric peer review process reviewer’s process performance evaluated examined consistency rigor Reviewers trained mutually agreed upon professional standard used ass work’s quality field social psychology field plagued questionable research practice don’t even rule place make sure study legitimate navelgazey journal article muse subject though actually happens article submitted reviewer Basically subjected whim reviewer arbitrarily assigned idea expect going process know agenda pet peeve writing style preference reviewer gonna know much time they’re going put reading work competent actually don’t even know know they’ve probably published journal … level scrutiny article subjected come chance you’re assigned reviewer created theory opposes theory work likely picked apart reviewer look closely flaw take issue everything inherently bad thing — research closely reviewed — it’s unbiased either reviewer ax grind may search flaw sake science sake career may go far far easier researcher like collaborated past Scientists human fall prey cognitive bias people study we’re pretty unselfreflective making sure reviewer appropriately strident critique usually journal editor’s job addition axgrinding reviewer dispositionally friendly reviewer reviewer don’t give shit typically overwhelmed people folk head Either workload demand lack experience training cannot provide useful review literature they’ve given may delay providing feedback make surface level comment particularly useful attempt appear helpful critical may make ton irrelevant recommendation delay publication don’t improve quality work … Many comment get reviewer matter unsubstantive subjective person may ask cite work Someone prefers “myriads of” “a myriad of” may criticize grammar even though usage technically fine reviewer doesn’t believe sexism exists may ask author remove language acknowledging pretty basic wellestablished social justice issue reviewer favorite pet statistical technique pet peeve may ask additional analysis satisfy neurosis lot time reviewer approach paper asking “How would done this” rather “Was done well” lead lot recommendation come taste tradition problematic analytical process like phacking get passed generation generation lot reviewer especially rookie feel need prove put lot work review make ton recommendation pretty much sake recommending something Authors may asked add collect several studies’ worth new data may told add large section literature review citing work related reviewer’s work Structural change paper may requested appeal one reviewer’s personal preference bias Sometimes change improve paper whole often delay everything unnecessarily … Image Henrique Ferreira remixed Pixlr author peerreview process also unfortunately subjected kind human bias including sexism nepotism reviewers’ name usually masked author’s name always anonymous reviewer allowing implicit assumption author creep Cordelia Fine famously reported book Delusions Gender trans male researcher perceived intelligent capable colleague started using male name even heard people offhandedly mentioning “much better researcher” “sister” sister — people talking pretransition self course instance anecdotal data suggests widespread sexism bias peerreview process Wennerås Wold conducted largescale review postdoctoral fellowship application 2010 found evidence sexism nepotism Women far le likely receive fellowship compared men applicant close friend reviewer far likely accepted metaanalytic study published 2007 Bornmann colleague found peerreview process “men statistically significant greater odds receiving grant woman 7” Bornmann et al paper unable find published empirical work establishing whether peer review process poisoned racism However reviewer often discern guess author’s race ethnicity based name prior work established name job application resume perceived “black” name applicant closely scrutinized le likely selected highly likely process occurs reviewer know name article author well course even author’s name hidden reviewer still room bias prejudice seep Work racism sexism ableism structural form oppression may strictly reviewed white ci male author uncomfortable reminded world fair place you’ve ever attended research presentation social science you’ve witnessed welloff established white male researcher attempt pick apart work black female colleague basis merit doesn’t believe world bigoted black female people say Scientists love believe unflinchingly logical make even harder acknowledge bias average person … mention may take reviewer month return draft feedback reviewer take upwards year editor often remind cajole harass reviewer getting review back even author may stuck waiting semester hear back article reviewer hastily read work day type together surfacelevel critique based subjectivity send editor without second thought slowness process clearly related fact reviewer paid assigned review duty part official job It’s whole separate unmeasured thing course it’s gonna low priority class teach manuscript write job talk attend Image Jon Tyson courtesy Unsplash Edited Pixlr author … worsened fact academic journal allow simultaneous submission submit paper one journal time you’re truly behest 2–3 random reviewer plus journal editor month year time work seen get accepted first time statistically speaking probably won’t submission accepted you’re lucky editor “desk reject” work within day happens article clearly good fit publication it’s shot full flaw editor tell won’t pas reviewer muster Getting desk reject save time least allows author make big revision submit entirely different journal rather waiting around month get turned anyway Even full review process article get rejected part journal establish rigorous — high rejection rate look good implies thoroughness albeit tailwagsthedog kind way Many suitable manuscript sacrificed altar rigor small “fortunate” subset instead get exhausting slightly hopeful “revise resubmit” … “revise resubmit” guarantee article accepted revision made It’s merely opportunity author respond comment make adjustment try acceptance journal Sometimes revision merely structural theoretical Sometimes author asked reanalyze data often especially social science research expected go collect additional data analyze write send new draft reviewer Regardless long revision take sent back reviewer editor may languish unread month time Finally reviewer may return draft point perhaps even remembering original feedback author likelihood receive second round feedback fuzzy subjective first round way many month year original submission that’s none original reviewer dropped map gotten swamped finish review died … Image Samuel Zeller courtesy Unsplash Edited Pixlr author Things don’t always go poorly course Sometimes review relatively quick month two revision request make sense revised work accepted happened colleague quite time instance like peer review process lot better nothing mean adequate logical systematic process secretive hard long tedious illusion contributing value rationality scientific work scientist one conducting assumed scientific practice practice it’s much better receiving random halfconsidered comment conference presentation whomever happens show there’s structure training pay scale place make better Peer review scientific process Scientists scientist following scientific method giving feedback colleague based gut feeling personal preference bias limited resource human they’re inconsistent sloppy anyone else … messy slipshod process delay public’s access novel scientific work mean published work held consistently high yet fair standard result lay public great deal faith work assume carefully thoroughly vetted cannot easily verify also put immense amount arbitrary barrier researcher professional goal Publications necessary researcher receive tenure prayer netting competitive academic job yet process getting publication barely related researcher’s productivity merit alternative realm mathematics example journal article seen inefficient illogical method sharing knowledge colleague publicatlarge Mathematical proof instead shared rapidly freely among handful specialist take interest work well done speaks Either proof successful establishing writer set establish…or isn’t There’s need introduce random barrier make process look logical work already logical Image Marten vanden Huevel courtesy Unsplash next article series discus academic publishing process reformed taking inspiration mathematics world well growing world open science I’m still actively researching alternative option writing people done subject you’re researcher insight peerreview process reformed ought take place please join conversation comment tweeting dreprice writing essay ownTags Education Publishing Science Psychology Writing |
779 | How The Pandemic Affected My Creativity | In 2019, I was playing in two bands and filling in on a third. I was writing almost daily on Medium and working on a novel. I was burning the creative oil.
Then a one-two sucker punch hit me. The Medium Partner Program change and the Great Interruption (Covid-19 pandemic.)
And let’s not forget massive social upheaval and an extremely derisive, stressful Presidential election in 2020. Neither of which fostered an environment conducive to creativity.
Medium decided to switch to read-time Partner Program payments slashing earning for creative writers like me who often post very short humor, microfiction, and poetry. I was disappointed, and I would be lying if I said it didn’t take a little of the wind out of my creative sails. Since I started writing on Medium, I had the feeling that getting paid decent money to write whatever I wanted couldn’t last, but it stung anyway.
Regardless, I soldiered on. Hoping it would somehow turn around.
In January of 2020, I released two collections (books) of stories from my previous two years on Medium. Then I turned my attention to finishing my novel. I had set a goal of finishing it by my 60th birthday on July 05. And it revived my creative spark and enjoyment for a while. I got focused and I accomplished that goal. I published my novel in June of 2020. | https://medium.com/creative-humans/how-the-pandemic-affected-my-creativity-1d3b02f012ce | ['Mark Starlin'] | 2020-12-12 18:02:21.567000+00:00 | ['Writing', 'Music', 'Pandemic', 'Creativity', 'Life'] | Title Pandemic Affected CreativityContent 2019 playing two band filling third writing almost daily Medium working novel burning creative oil onetwo sucker punch hit Medium Partner Program change Great Interruption Covid19 pandemic let’s forget massive social upheaval extremely derisive stressful Presidential election 2020 Neither fostered environment conducive creativity Medium decided switch readtime Partner Program payment slashing earning creative writer like often post short humor microfiction poetry disappointed would lying said didn’t take little wind creative sail Since started writing Medium feeling getting paid decent money write whatever wanted couldn’t last stung anyway Regardless soldiered Hoping would somehow turn around January 2020 released two collection book story previous two year Medium turned attention finishing novel set goal finishing 60th birthday July 05 revived creative spark enjoyment got focused accomplished goal published novel June 2020Tags Writing Music Pandemic Creativity Life |
780 | Are Trees the Silver Bullets to Solve Climate Change? | “The word “restore” is defined as the act of returning something back to its original state, but it’s also the act of returning it back to its original owners. The restoration of nature is for the local biodiversity and the communities that depend on it. And as that network grows, the collective action benefits everyone. And these benefits go far beyond the threat of climate change. Even if climate change stopped right now, the protection and rebuilding of earth’s biodiversity would still be a top priority because it underpins all life on earth. It can help us with all other global threats, including extreme weather events, droughts, food shortages and global pandemics.” — Thomas Crowther
Thomas Crowther is the Lead Scientific Advisor for the UN Billion Tree Campaign. He is a prolific researcher and ecologist leading a number of initiatives related to biodiversity, forestry, and soil research. His recent TED Talk highlights his work beautifully and lays the groundwork for local and global conversations about restoring nature’s biodiversity.
“The problem is clear. We know we need to reduce our emissions and draw the existing carbon out of the atmosphere, stop the damage and start the repair. And this is where forests can help. Like all plants, trees capture carbon from the atmosphere, and they use it for growth. And some of that carbon enters the soil, where it can stay for hundreds or even thousands of years. If we could stop the losses of forests around the world, we could directly help to cut our annual emissions. And if we could start to tip the balance in the other direction, we might even help the repair process.” — Thomas Crowther
The research Crowther has spearheaded is leading to a networked approach to understanding and restoring forests across the globe. The value of this goes far beyond reducing the impact of our global emissions though. Yes, defending and restoring our forests will of course result in significantly more carbon being captured by the world’s trees. Perhaps more importantly, the restoration of our global forests will also address the planet’s declining biodiversity.
In his recent TED Talk, Thomas Crowther addresses a communication error he has made and suggests that we focus on the value of restoring nature’s biodiversity. There is some psychology associated with this reasoning. If environmentalists communicate to the world that restoring our forests and planting trees can capture a significant amount of our annual emissions, then the general public will have less incentive and less conviction to reduce their carbon footprint on a regular basis. The challenge of communicating effectively isn’t new to environmentalists. This communication strategy is about communicating the primary value of a global action in a simple way that continues to encourage the world population to continue to take action on climate change. The global action of protecting and restoring forests is vital to addressing climate change, but that doesn’t mean we should communicate that to the general public. The communication of climate change solutions needs to be prioritized.
“But in the excitement of it all, and with the chance to make that positive impact I’d always dreamed of, I made some naive and stupid mistakes in communication that threatened the entire message. The simplicity of our message was its strength, but it came at the expense of nuance that is so important. And as the headlines began to emerge, I desperately just wanted to pull them back in. Because to some, it seemed like we were proposing restoration as the single solution to climate change. And this is the opposite of what this movement needs. When viewed through this lens, restoration just seems like an easy way out, a chance for us to “offset our emissions” by planting a few trees and ignore the very real and urgent challenges of cutting emissions and protecting the ecosystems that we currently have. Restoration is not a silver bullet. There is no silver bullet. It is just one of a huge portfolio of solutions that we so desperately need. And this view of trees as an easy way out is such a tempting perspective, but it is a real threat to the climate change movement and to the ecosystems that still remain.” — Thomas Crowther
As we approach financing massive global efforts to addressing climate change, it is important for us to consider the communication strategies and communication effectiveness associated with each effort. So much of addressing climate change is about changing behaviors, and we as environmentalists and sustainability professionals have the ability to enable and emphasize certain behaviors by changing the way we prioritize our communication. In the case of restoring our forests, we should not emphasize reducing the impact of our global emissions. We should instead emphasize the value of restoring the planet’s biodiversity. | https://medium.com/discourse/are-trees-the-silver-bullets-to-solve-climate-change-f299f63f1c27 | ['Carbon Radio'] | 2020-10-21 16:15:18.388000+00:00 | ['Nature', 'Climate Change', 'Environment', 'Future', 'Sustainability'] | Title Trees Silver Bullets Solve Climate ChangeContent “The word “restore” defined act returning something back original state it’s also act returning back original owner restoration nature local biodiversity community depend network grows collective action benefit everyone benefit go far beyond threat climate change Even climate change stopped right protection rebuilding earth’s biodiversity would still top priority underpins life earth help u global threat including extreme weather event drought food shortage global pandemics” — Thomas Crowther Thomas Crowther Lead Scientific Advisor UN Billion Tree Campaign prolific researcher ecologist leading number initiative related biodiversity forestry soil research recent TED Talk highlight work beautifully lay groundwork local global conversation restoring nature’s biodiversity “The problem clear know need reduce emission draw existing carbon atmosphere stop damage start repair forest help Like plant tree capture carbon atmosphere use growth carbon enters soil stay hundred even thousand year could stop loss forest around world could directly help cut annual emission could start tip balance direction might even help repair process” — Thomas Crowther research Crowther spearheaded leading networked approach understanding restoring forest across globe value go far beyond reducing impact global emission though Yes defending restoring forest course result significantly carbon captured world’s tree Perhaps importantly restoration global forest also address planet’s declining biodiversity recent TED Talk Thomas Crowther address communication error made suggests focus value restoring nature’s biodiversity psychology associated reasoning environmentalist communicate world restoring forest planting tree capture significant amount annual emission general public le incentive le conviction reduce carbon footprint regular basis challenge communicating effectively isn’t new environmentalist communication strategy communicating primary value global action simple way continues encourage world population continue take action climate change global action protecting restoring forest vital addressing climate change doesn’t mean communicate general public communication climate change solution need prioritized “But excitement chance make positive impact I’d always dreamed made naive stupid mistake communication threatened entire message simplicity message strength came expense nuance important headline began emerge desperately wanted pull back seemed like proposing restoration single solution climate change opposite movement need viewed lens restoration seems like easy way chance u “offset emissions” planting tree ignore real urgent challenge cutting emission protecting ecosystem currently Restoration silver bullet silver bullet one huge portfolio solution desperately need view tree easy way tempting perspective real threat climate change movement ecosystem still remain” — Thomas Crowther approach financing massive global effort addressing climate change important u consider communication strategy communication effectiveness associated effort much addressing climate change changing behavior environmentalist sustainability professional ability enable emphasize certain behavior changing way prioritize communication case restoring forest emphasize reducing impact global emission instead emphasize value restoring planet’s biodiversityTags Nature Climate Change Environment Future Sustainability |
781 | Why “Yield” Is an Excellent Python Keyword — Generators and Coroutines | Suspend and Resume Execution in Coroutines
Note: It’s important to know that Python is phasing out the use of generators as the basis for coroutines. Check the official documentation. However, the concept of coroutines can be easy to understand if we know that yield has the meaning of “giving way.”
In this usage, it’s applicable to yield ’s second meaning: give way. It’s pretty much like the traffic “yield” sign in real life. You temporarily stop your vehicle such that other vehicles with right of the way can go first. Applying this analog to programming, it’s the basic idea of coroutines.
As you can see, the key to coroutines is that the execution is suspended temporarily and it can be resumed later on when it’s applicable. Doesn’t this sound familiar to you? The generator function runs and yields the element (the execution is suspended), and when the next element is requested, it continues to run from where it was suspended (the execution is resumed).
Actually, we can use the yield keyword to implement a coroutine. Let’s see the following code snippet for a simple example:
We first define a coroutine function using the yield keyword. As you can see, the overall structure is similar to the generator function. However, you’ll notice that the yield keyword appears after the assignment operator (i.e. the equals sign).
keyword. As you can see, the overall structure is similar to the function. However, you’ll notice that the keyword appears after the assignment operator (i.e. the equals sign). The assignment is possible because, in addition to the typical next() function that works with a generator, generators also support the send() function.
function that works with a generator, generators also support the function. In the example, we send a string data (i.e. “Here’s the data that your requested.” ) back to the coroutine. Specifically, the data that we sent is set to the variable x , and we’re able to print the x value as expected.
) back to the coroutine. Specifically, the data that we sent is set to the variable , and we’re able to print the value as expected. The StopIteration exception is expected because the generator function has reached the end, which is consistent with the behavior of a generator (as shown in the last section). | https://medium.com/better-programming/why-yield-is-an-excellent-python-keyword-generators-and-coroutines-d0d9d9403171 | ['Yong Cui'] | 2020-11-04 17:48:37.297000+00:00 | ['Python', 'Technology', 'Software Engineering', 'Artificial Intelligence', 'Programming'] | Title “Yield” Excellent Python Keyword — Generators CoroutinesContent Suspend Resume Execution Coroutines Note It’s important know Python phasing use generator basis coroutines Check official documentation However concept coroutines easy understand know yield meaning “giving way” usage it’s applicable yield ’s second meaning give way It’s pretty much like traffic “yield” sign real life temporarily stop vehicle vehicle right way go first Applying analog programming it’s basic idea coroutines see key coroutines execution suspended temporarily resumed later it’s applicable Doesn’t sound familiar generator function run yield element execution suspended next element requested continues run suspended execution resumed Actually use yield keyword implement coroutine Let’s see following code snippet simple example first define coroutine function using yield keyword see overall structure similar generator function However you’ll notice yield keyword appears assignment operator ie equal sign keyword see overall structure similar function However you’ll notice keyword appears assignment operator ie equal sign assignment possible addition typical next function work generator generator also support send function function work generator generator also support function example send string data ie “Here’s data requested” back coroutine Specifically data sent set variable x we’re able print x value expected back coroutine Specifically data sent set variable we’re able print value expected StopIteration exception expected generator function reached end consistent behavior generator shown last sectionTags Python Technology Software Engineering Artificial Intelligence Programming |
782 | My Technology Collection | I am a technology leader, inventor, and entrepreneur in technology, cognitive science, and digital content development fields with four decades of experience. I have been practising and researching technology in academia and industry. My Enterprise Architecture experience was certified at the distinguished level and Design Thinking expertise rated as a thought leader and influencer in the industry.
Apart from researching the technology field as a scholar and documenting my findings, I am a hands-on practitioner and a dedicated learner reading from the body of knowledge voraciously. Not only show guidance to aspiring entrepreneurs as a mentor and development coach, but I am also a humble protege of many great technology leaders in the industry and academia.
I see myself as a technologist working in large corporate organizations within the last four decades. Since I started at a young age with an informatics degree earned by a scholarship in Europe, technology evolved, and every single day added a new meaning to my life.
While technology stresses some of us, it plays a different role in my life. It relaxes and uplifts me. The pieces of technology nourish my brain cells and keep me active and joyful.
I chose technology as a profession to deal with both human and machine intelligence. I attempted to be a polymath and am still working on it. I am educated both as a technologist and scientist.
From the technology point of view, I cover breadth, including an extensive array of technology domains with depth in some specialized areas such as Big Data, Cloud, Cybersecurity, Cognitive Computing, and IoT.
I work as an Enterprise Architect, certified at a distinguished level by the Open Group. From the scientific perspective, I cover machine and human intelligence under the discipline of cognitive science.
I have been consuming insights from the body of knowledge for my profession. Technology knowledge turned out to become a hobby for a long time. During my studies, I started contributing to the body of knowledge in various formats, such as papers, articles, books, and blogs. I published in peer-reviewed scientific journals, delivered keynotes, and am recognized as an inventor with multiple patents.
In this story, I want to introduce and share my recent technology-related articles, presenting them in a single story. This story allows my readers to find my leadership stories easily.
I aim to share my knowledge and experience of outstanding technical and technology leadership.
I am founder and editor in chief of Technology Hits on Medium.
You are welcome to join this sensational publication.
Please enjoy my selected stories touching various aspects of technology.
Ethical Hacking
Topics — #Cybersecurity #Security #Career
In this article, I aim to create awareness on ethical hacking, its purpose, use cases, and a brief introduction to the role of ethical hackers in digital transformation initiatives. My goal is to help security executives and managers choose the best ethical hackers for their business. As a by-product, the information in this article can also guide the aspiring ethical hackers to build their skills and plan their experience.
Power of Design Thinking for Writers
Topics — #Writing #Design #Design Thinking
Design Thinking can be useful for writers too. This is my main premise in this short article. Writers can use Design Thinking principles for quality, simplicity, and agility purposes. Design Thinking can be considered as an innovative problem-solving technique with close interactions of the actual users of the solutions. As writers, one of our fundamental roles is to solve problems for our readers and delight them.
Business Value and Architectural Overview of Robotics Process Automation
Topics — #Artificial Intelligence #Automation #Robotics #Business
In this article, my aim is to briefly introduce what Robotics Process Automation (RPA) is, it’s business value, the major use cases, how it works, how we can enhance it, and my experience on architecting solutions to meet the customer requirements. If you are tired of performing tedious and repetitive tasks and witnessing too many errors upsetting your clients, your business may need the use of Robotics Process Automation empowered by software robots and the capabilities artificial intelligence.
Cognitive Transformation
Topics #Artificial Intelligence #Cognition #Transformation #AI
The purpose of this article is to briefly introduce to you what cognitive computing entails, its current progress in the industry, the value proposition for its necessity, and my personal observations and thoughts on trends and future plans. Most of us in the IT industry most likely heard, read about, or already started working with the cognitive systems for various practical use cases. Would you like to have a quick taste? Then, keep reading! I simplified this complex topic as much as possible in this article.
Business Architecture Framework for Digital Transformation
Topics #Software Engineering #Leadership #Programming
In order to create effective business architecture for complex digital transformation programs, we must follow an established yet innovative framework. A framework is a basic structure underlying a business and technology system. Our framework can start with understanding the organizational structure of the business. This framework can include multiple considerations and business viewpoints. Business organizations are made up of stakeholders with different roles focusing on common goals.
Introduction to IoT Ecosystem
Topics #IoT #Design #Infrastructure #Technology
My aim in this article is to introduce IoT (Internet of Things) ecosystem based on a model that I developed as part of my solution architecture engagements in multiple successful projects in large business organisations. I set the IoT solution context at high level as understanding the solution context can help us delve into architectural and design topics in a methodical way. Every solution requires following a method to leverage prior experience in the field. This also helps us ensure that nothing falls into cracks during the solution lifecycle.
Digital Transformation
Topics #Digital Transformation #Technology
Business organisations are facing enormous challenges to respond to the rapid technological changes and growing demands of consumers for digital products and services globally. There is a constant search to find solutions to the growing business problems. From my experience, the most optimal solution to address these problems is to architect enterprise digital transformation requirements and objectives. My approach, in this article, is to propose an innovative model and framework formulated and described in a 12-step architectural method called DTM.
Architecting Data Lake Solutions
Topics #Data Architect #Data Science #Design
In this article I provide an overview of data lakes, data ponds, , and data swamps from an architectural perspective. My aim is to provide business value, use cases, architectural, design, and implementation, and consumption considerations for data lakes. Use of data lakes became the de-facto standard for Big Data Analytics platforms supporting the Internet of Things (IoT) and Artificial Intelligence (AI) initiatives. Majority of the Big Data and Analytics solutions require consideration of a data lake architectural model. Most of my recent digital transformation initiatives involve data lake solutions to support IoT, Mobility, and AI goals.
Design Thinking for Technical Excellence
Topics #Design Thinking #Design #Leadership #Technology
As an Enterprise Architect, I practice Design Thinking in my daily solution design activities, team interactions, user requirements gathering, and business stakeholder management. Modern business initiatives require adoption of design thinking to their core culture to compete, thrive, and sustain their services with agility. Design thinking can be applied to any discipline.
How To Deal With Big Data For AI?
Topics #AI #Artificial Intelligence #Big Data #Data #Design
In this article, I provide a pragmatic architectural overview of the Big Data management lifecycle covering the distinct phases and essential definitions with critical solution considerations based on my architectural experience acquired from many successful and failed projects. I learnt that architecting Big Data solutions pragmatically with rigour can substantially contribute to the delivery of quality AI and Cognitive solutions especially in enterprise modernisation and digital transformation programs. These transformative programs are empowered by Big Data and AI, integrated with other emerging technology domains e.g. Cloud, Edge, IoT, Blockchain, and Mobile technologies.
Free Online Technology Courses
Topics #Education #Learning #Courses #Technology
In this post, I want to introduce free learning programs by industry experts and industry skill badges offered by the Cognitive Class AI to help you boost your data science, machine learning, cognitive computing, cloud computing, deep learning, Blockchain, programming, open-source tools, and artificial intelligence capabilities online at your convenience in 2020. Continuous learning and applying our knowledge can be powerful and critical success factors for achieving our professional goals. The Cognitive Class AI offers a wide variety of professional learning paths, as free of charge, to learners globally. In this article, I provide you with some prominent learning path samples with links so that you commence achieving your 2020 professional education and career development goals.
A Modern Enterprise Architecture Approach
Topics #Enterprise Architecture #Design #Technology
My purpose to author this book was to provide essential guidance, compelling ideas, and unique ways to Enterprise Architects so that they could successfully perform complex enterprise modernization initiatives transforming them from chaos to coherence. This is not a theory book describing Enterprise Architecture in detail. There are myriad of books on the market and in libraries discussing details of enterprise architecture.
The Blockchain
Topics #Blockchain #Security #Digital Trust
Is the Blockchain panacea to address manipulation problems and illuminate the world with true trust? Can we enable Artificial Intelligence without the Blockchain? Can we start developing the Blockchain application? Let’s explore! In this article, my aim is to introduce what the Blockchain is without hype, its importance with facts, its architecture for autonomous systems, compelling use cases it presents, business value proposition, the technologies and processes behind it, learning and development opportunities, current progress for its deployment, future plans for integrating it to our lives to better understand it and gain more insights about its value proposition.
Digital Twin Solutions
Topics #Digital Twins #IoT #Internet #AI #Big Data
In this article, I want to introduce Digital Twin concept as an emerging technology used in various industries. My aim is to give an overview and shed lights on this emerging technology, architectural construct, and business initiative based on my experience. I explain the digital twin concept using the Cyber-Physical System architecture, business use cases, and value proposition.Digital Twin concept is simple however manifesting them in reality can be complex and difficult to due to combination of underlying technology stacks, tools, and integration requirements. Therefore, a structured and methodical approach to the topic is essential.
Power of the Edge Computing
Topics #Edge Computing #Cloud Computing #Big Data
The purpose of this article is to introduce the Edge Computing as a functional and non-functional extension to the IoT (Internet of Things) and the Cloud Computing service model. The Edge Computing is a powerful extension to the Cloud and the IoT based Big Data Analytics projects. Let me set the context by giving you a brief background on the Cloud and the IoT relationship, then introduce the Edge Computing in this context. We all know the recent trends for the Cloud Computing. The Cloud marked a paradigm shift to Information Technology and Computing field. The IoT Cloud is a critical player in the data ecosystem of large business organisations. The central role the Cloud plays in the IoT is to facilitate data integration of the solution components effectively. IoT solutions are mainly used to provide real-time information to consumers.
Big Data as a Service
Topics #Big Data #BDaaS #Cloud Computing
In this article, I want to introduce the business value, architecture, use cases, deployment, and service model of Big Data as a Service (BDaaS) which is a novel and rapidly growing Cloud-based service. The service is offered to support big data and analytics solutions in a cost-effective manner. This service can be considered as a type of outsourcing model for the deployment of big data and analytics projects rapidly and inexpensively. It is considered inexpensive because we can use the service without upfront investment for costly underlying infrastructure hosting the big data and analytics solutions. Underlying infrastructure costs can include computing servers, storage, network, appliances, racks, hosting facilities, and more importantly infrastructure management and support costs.
Viability Assessment for IoT Solutions
Topics #IoT #Architecture #Design
In this article, I aim to introduce a critical architectural and design tool to validate feasibility of IoT (Internet of Things) solutions across the project lifecycle. I attempt to explain what this tool does, why we need to use it, and how we create, document, analyze, and communicate it to relevant stakeholders of the IoT solutions. Viability Assessment is a critical architectural and design tool to meet business organization’s standards, solution requirements, use cases, project quality criteria, industry compliance, and consumer expectations. This article provides awareness for IoT solution architects to use this tool effectively. Without applying proper viability assessment, IoT projects can most certainly fail! As solution architects and designers, we use a viability assessment work-product, which exists in many proprietary and open-source architectural solution methods. We can either use a viability assessment work-product template from an established open-source method such as TOGAF or our business organization’s proprietary method
IoT, Big Data, & Cloud Computing
Topics #IoT #Big Data #Cloud Computing
IoT (Internet of Things), Big Data Analytics, and Cloud Computing are three distinct technology domains with overlapping use cases. Each technology has its own merits; however, the combination of three creates a synergy and the golden opportunity for businesses to reap the exponential benefits. This combination can create technological magic for innovation when adequately architected, designed, implemented, and operated. In this post, my aim is to provide relationships among IoT, Big Data Analytics, and the Cloud Computing as detailed in one of my recent books titled Architecting Big Data & Analytics Solutions — Integrated with IoT & Cloud.
Big Data Lifecycle Management
Topics #Big Data #Data Platforms #Data Architecture
In this article, I want to provide an architectural overview of the Big Data lifecycle management based on key points extracted from my recent book titled “Architecting Big Data & Analytics Solutions — Integrated with IoT & Cloud”. Understanding this process is essential to architect and design Big Data solutions. Big data is different from traditional data. The main differences come from characteristics such as volume, velocity, variety, veracity, value and overall complexity of data sets in a data ecosystem. Understanding these V words provide useful insights into the nature of Big Data.
I want to share a few technology articles on other platforms.
A Technical Overview of Internet of Things (IoT)
What is Edge Computing?
Robotics Process Automation
Big Data Lifecycle Management
What Is Business Architecture?
Digital Twins
How to Become an Ethical Hacker
Design Thinking for Writers
Find more technology stories at Vocal Media.
Thank you for reading my perspectives.
To collaborate with me, you can join Technology Hits, ILLUMINATION, and ILLUMINATION-Curated. Please send a request via this link. Please point out the name of the preferred publication in your request form. | https://medium.com/illumination-curated/my-technology-collection-68b25729b22e | ['Dr Mehmet Yildiz'] | 2020-12-11 10:02:50.251000+00:00 | ['Technology', 'Entrepreneurship', 'Design Thinking', 'Writing', 'Design'] | Title Technology CollectionContent technology leader inventor entrepreneur technology cognitive science digital content development field four decade experience practising researching technology academia industry Enterprise Architecture experience certified distinguished level Design Thinking expertise rated thought leader influencer industry Apart researching technology field scholar documenting finding handson practitioner dedicated learner reading body knowledge voraciously show guidance aspiring entrepreneur mentor development coach also humble protege many great technology leader industry academia see technologist working large corporate organization within last four decade Since started young age informatics degree earned scholarship Europe technology evolved every single day added new meaning life technology stress u play different role life relaxes uplift piece technology nourish brain cell keep active joyful chose technology profession deal human machine intelligence attempted polymath still working educated technologist scientist technology point view cover breadth including extensive array technology domain depth specialized area Big Data Cloud Cybersecurity Cognitive Computing IoT work Enterprise Architect certified distinguished level Open Group scientific perspective cover machine human intelligence discipline cognitive science consuming insight body knowledge profession Technology knowledge turned become hobby long time study started contributing body knowledge various format paper article book blog published peerreviewed scientific journal delivered keynote recognized inventor multiple patent story want introduce share recent technologyrelated article presenting single story story allows reader find leadership story easily aim share knowledge experience outstanding technical technology leadership founder editor chief Technology Hits Medium welcome join sensational publication Please enjoy selected story touching various aspect technology Ethical Hacking Topics — Cybersecurity Security Career article aim create awareness ethical hacking purpose use case brief introduction role ethical hacker digital transformation initiative goal help security executive manager choose best ethical hacker business byproduct information article also guide aspiring ethical hacker build skill plan experience Power Design Thinking Writers Topics — Writing Design Design Thinking Design Thinking useful writer main premise short article Writers use Design Thinking principle quality simplicity agility purpose Design Thinking considered innovative problemsolving technique close interaction actual user solution writer one fundamental role solve problem reader delight Business Value Architectural Overview Robotics Process Automation Topics — Artificial Intelligence Automation Robotics Business article aim briefly introduce Robotics Process Automation RPA it’s business value major use case work enhance experience architecting solution meet customer requirement tired performing tedious repetitive task witnessing many error upsetting client business may need use Robotics Process Automation empowered software robot capability artificial intelligence Cognitive Transformation Topics Artificial Intelligence Cognition Transformation AI purpose article briefly introduce cognitive computing entail current progress industry value proposition necessity personal observation thought trend future plan u industry likely heard read already started working cognitive system various practical use case Would like quick taste keep reading simplified complex topic much possible article Business Architecture Framework Digital Transformation Topics Software Engineering Leadership Programming order create effective business architecture complex digital transformation program must follow established yet innovative framework framework basic structure underlying business technology system framework start understanding organizational structure business framework include multiple consideration business viewpoint Business organization made stakeholder different role focusing common goal Introduction IoT Ecosystem Topics IoT Design Infrastructure Technology aim article introduce IoT Internet Things ecosystem based model developed part solution architecture engagement multiple successful project large business organisation set IoT solution context high level understanding solution context help u delve architectural design topic methodical way Every solution requires following method leverage prior experience field also help u ensure nothing fall crack solution lifecycle Digital Transformation Topics Digital Transformation Technology Business organisation facing enormous challenge respond rapid technological change growing demand consumer digital product service globally constant search find solution growing business problem experience optimal solution address problem architect enterprise digital transformation requirement objective approach article propose innovative model framework formulated described 12step architectural method called DTM Architecting Data Lake Solutions Topics Data Architect Data Science Design article provide overview data lake data pond data swamp architectural perspective aim provide business value use case architectural design implementation consumption consideration data lake Use data lake became defacto standard Big Data Analytics platform supporting Internet Things IoT Artificial Intelligence AI initiative Majority Big Data Analytics solution require consideration data lake architectural model recent digital transformation initiative involve data lake solution support IoT Mobility AI goal Design Thinking Technical Excellence Topics Design Thinking Design Leadership Technology Enterprise Architect practice Design Thinking daily solution design activity team interaction user requirement gathering business stakeholder management Modern business initiative require adoption design thinking core culture compete thrive sustain service agility Design thinking applied discipline Deal Big Data AI Topics AI Artificial Intelligence Big Data Data Design article provide pragmatic architectural overview Big Data management lifecycle covering distinct phase essential definition critical solution consideration based architectural experience acquired many successful failed project learnt architecting Big Data solution pragmatically rigour substantially contribute delivery quality AI Cognitive solution especially enterprise modernisation digital transformation program transformative program empowered Big Data AI integrated emerging technology domain eg Cloud Edge IoT Blockchain Mobile technology Free Online Technology Courses Topics Education Learning Courses Technology post want introduce free learning program industry expert industry skill badge offered Cognitive Class AI help boost data science machine learning cognitive computing cloud computing deep learning Blockchain programming opensource tool artificial intelligence capability online convenience 2020 Continuous learning applying knowledge powerful critical success factor achieving professional goal Cognitive Class AI offer wide variety professional learning path free charge learner globally article provide prominent learning path sample link commence achieving 2020 professional education career development goal Modern Enterprise Architecture Approach Topics Enterprise Architecture Design Technology purpose author book provide essential guidance compelling idea unique way Enterprise Architects could successfully perform complex enterprise modernization initiative transforming chaos coherence theory book describing Enterprise Architecture detail myriad book market library discussing detail enterprise architecture Blockchain Topics Blockchain Security Digital Trust Blockchain panacea address manipulation problem illuminate world true trust enable Artificial Intelligence without Blockchain start developing Blockchain application Let’s explore article aim introduce Blockchain without hype importance fact architecture autonomous system compelling use case present business value proposition technology process behind learning development opportunity current progress deployment future plan integrating life better understand gain insight value proposition Digital Twin Solutions Topics Digital Twins IoT Internet AI Big Data article want introduce Digital Twin concept emerging technology used various industry aim give overview shed light emerging technology architectural construct business initiative based experience explain digital twin concept using CyberPhysical System architecture business use case value propositionDigital Twin concept simple however manifesting reality complex difficult due combination underlying technology stack tool integration requirement Therefore structured methodical approach topic essential Power Edge Computing Topics Edge Computing Cloud Computing Big Data purpose article introduce Edge Computing functional nonfunctional extension IoT Internet Things Cloud Computing service model Edge Computing powerful extension Cloud IoT based Big Data Analytics project Let set context giving brief background Cloud IoT relationship introduce Edge Computing context know recent trend Cloud Computing Cloud marked paradigm shift Information Technology Computing field IoT Cloud critical player data ecosystem large business organisation central role Cloud play IoT facilitate data integration solution component effectively IoT solution mainly used provide realtime information consumer Big Data Service Topics Big Data BDaaS Cloud Computing article want introduce business value architecture use case deployment service model Big Data Service BDaaS novel rapidly growing Cloudbased service service offered support big data analytics solution costeffective manner service considered type outsourcing model deployment big data analytics project rapidly inexpensively considered inexpensive use service without upfront investment costly underlying infrastructure hosting big data analytics solution Underlying infrastructure cost include computing server storage network appliance rack hosting facility importantly infrastructure management support cost Viability Assessment IoT Solutions Topics IoT Architecture Design article aim introduce critical architectural design tool validate feasibility IoT Internet Things solution across project lifecycle attempt explain tool need use create document analyze communicate relevant stakeholder IoT solution Viability Assessment critical architectural design tool meet business organization’s standard solution requirement use case project quality criterion industry compliance consumer expectation article provides awareness IoT solution architect use tool effectively Without applying proper viability assessment IoT project certainly fail solution architect designer use viability assessment workproduct exists many proprietary opensource architectural solution method either use viability assessment workproduct template established opensource method TOGAF business organization’s proprietary method IoT Big Data Cloud Computing Topics IoT Big Data Cloud Computing IoT Internet Things Big Data Analytics Cloud Computing three distinct technology domain overlapping use case technology merit however combination three creates synergy golden opportunity business reap exponential benefit combination create technological magic innovation adequately architected designed implemented operated post aim provide relationship among IoT Big Data Analytics Cloud Computing detailed one recent book titled Architecting Big Data Analytics Solutions — Integrated IoT Cloud Big Data Lifecycle Management Topics Big Data Data Platforms Data Architecture article want provide architectural overview Big Data lifecycle management based key point extracted recent book titled “Architecting Big Data Analytics Solutions — Integrated IoT Cloud” Understanding process essential architect design Big Data solution Big data different traditional data main difference come characteristic volume velocity variety veracity value overall complexity data set data ecosystem Understanding V word provide useful insight nature Big Data want share technology article platform Technical Overview Internet Things IoT Edge Computing Robotics Process Automation Big Data Lifecycle Management Business Architecture Digital Twins Become Ethical Hacker Design Thinking Writers Find technology story Vocal Media Thank reading perspective collaborate join Technology Hits ILLUMINATION ILLUMINATIONCurated Please send request via link Please point name preferred publication request formTags Technology Entrepreneurship Design Thinking Writing Design |
783 | How I Use Powerful Life Stories as Fuel for My Writing | Whether we write fiction or non, we need input for memorable output
Photo by Peter Kasprzyk on Unsplash
I had a life-changing moment in high school, but there was a two-year delay in the life-changing part of the experience. It was one of those typical assemblies where some guest speaker shares educational slides about how bad it is to do drugs or join a gang — the kind most teenagers tune-out and the kind administrators drool-over.
I don’t remember the assembly… because of what happened.
There was a guy next to me. I didn’t know him. The auditorium was full — elbows-deep in teenagers. My high school was huge. Bigger than my first college. My class alone had twelve hundred students. So, this guy whom I didn’t know, wouldn’t stop cat-calling and two-finger whistling.
The auditorium was silent — save for Captain Whistles.
It was clear he wanted attention and had no tools beyond being a douche. I’m not a fan of confrontation and was even less-so in high school, but I couldn’t take it anymore. After ten minutes I finally turned and shushed him.
I don’t know what I expected to happen, but being choked wasn’t on the list of possibles. Whistles picked choking anyway. Nonchalant. As if he were making a sandwich or putting on mismatched socks — the guy reached over my head in a split-second and choked me until I saw stars. He whispered kind things like I’ll break your neck right here, and I should kill you for that. Maybe a nobody tells me what to do, kind of statement.
My journalism skills for remembering quotes weren’t so good in that moment. But his pep-talk went something like that.
The guy was sweaty. I could smell him too. Being that my head was cradled in his armpit like a newborn. Somehow all that trauma, no matter how old, makes a permanent, embossed notary stamp on your brain.
I couldn’t breathe. I did my best to resist. I coughed and struggled. Then — he let me go. Like it was another Tuesday. And returned to his cat-calling. As if to show me he was still in charge. Eventually, some form of authority took him away, but no one saw the choking.
So, that happened. Standard high school stuff. Kids are horrible to each other. But the choking wasn’t the life-changing bit. The part that stopped me cold — the life-changing part, was two years later. The summer after I graduated, Whistles made the paper. Turns out, in the middle of the night the same guy choked a cab driver to death. For money. Whistles went to prison. Apparently choking people was his hobby.
I don’t remember the guy’s name. I don’t feel it therapeutic to look him up either. I still think about him choking me sometimes and I’d rather not cement the memory deeper.
As crazy as it sounds I do owe the strangler a bit of gratitude.
A cab driver died and I can’t remove the weight of that incident. A young guy went to prison, because he probably led a horrible life. But I got to walk away with a strong story. The story is mine. I can package it any way I wish and re-tell it as many times as someone will listen.
Story is currency for humans. | https://medium.com/the-book-mechanic/how-i-learned-to-use-powerful-life-stories-as-fuel-for-my-writing-9817ddc7706 | ['August Birch'] | 2019-04-10 13:41:52.818000+00:00 | ['Storytelling', 'Self Improvement', 'Books', 'Writing', 'Life'] | Title Use Powerful Life Stories Fuel WritingContent Whether write fiction non need input memorable output Photo Peter Kasprzyk Unsplash lifechanging moment high school twoyear delay lifechanging part experience one typical assembly guest speaker share educational slide bad drug join gang — kind teenager tuneout kind administrator droolover don’t remember assembly… happened guy next didn’t know auditorium full — elbowsdeep teenager high school huge Bigger first college class alone twelve hundred student guy didn’t know wouldn’t stop catcalling twofinger whistling auditorium silent — save Captain Whistles clear wanted attention tool beyond douche I’m fan confrontation even lessso high school couldn’t take anymore ten minute finally turned shushed don’t know expected happen choked wasn’t list possible Whistles picked choking anyway Nonchalant making sandwich putting mismatched sock — guy reached head splitsecond choked saw star whispered kind thing like I’ll break neck right kill Maybe nobody tell kind statement journalism skill remembering quote weren’t good moment peptalk went something like guy sweaty could smell head cradled armpit like newborn Somehow trauma matter old make permanent embossed notary stamp brain couldn’t breathe best resist coughed struggled — let go Like another Tuesday returned catcalling show still charge Eventually form authority took away one saw choking happened Standard high school stuff Kids horrible choking wasn’t lifechanging bit part stopped cold — lifechanging part two year later summer graduated Whistles made paper Turns middle night guy choked cab driver death money Whistles went prison Apparently choking people hobby don’t remember guy’s name don’t feel therapeutic look either still think choking sometimes I’d rather cement memory deeper crazy sound owe strangler bit gratitude cab driver died can’t remove weight incident young guy went prison probably led horrible life got walk away strong story story mine package way wish retell many time someone listen Story currency humansTags Storytelling Self Improvement Books Writing Life |
784 | The 4 Most Important Things I Learned From Haruki Murakami’s Writing Philosophy | The 4 Most Important Things I Learned From Haruki Murakami’s Writing Philosophy Casimir Mura Follow Oct 4 · 7 min read
These are the gems I unearthed from his ‘memoir’.
Haruki Murakami | Source
Haruki Murakami is a novelist. And although he may not exactly look like it, he’s also a marathon writer. In his book What I Talk About When I Talk About Running, he discusses his journey as a marathon runner and how this impacted his writing philosophy.
In the foreword, Murakami mentions a mantra that he had read in the International Herald Tribune in an article about running.
Pain is inevitable. Suffering is optional. (What I Talk About When I Talk About Running pg. 4)
The majority of the book is just Murakami talking about running. However, at certain moments, Murakami talks about how his thought process behind running bleeds into writing. Since running and writing can both be painful to certain extents, this isn’t too surprising.
Anyone who has written anything knows the feeling of being stuck or writer's block. They know the pain associated with writing. Within a few moments of this book, Murakami points out some of the pains with writing and how to alleviate the suffering that’s attached to them.
Pay Attention to the World Around You
A big factor of appeal for Murakami’s writing is that he takes us to worlds where the border between reality and supernatural is hard to distinguish. It doesn’t surprise me then that Murakami values his experience of running a bar in Tokyo over anything he learned while he was at university as a source of inspiration for his writing.
Sometimes we are so caught up in our own lives that we forget there are always people around us with their own inner lives. Of course, all life experiences are not equal. Murakami had the good fortune of encountering all sorts of strange people.
Running a service-oriented business means you have to accept whoever comes through the door. No matter who comes in, unless they’re really awful, you have to greet them with a friendly smile on your face. Thanks to this, though, I met all kinds of offbeat people and had some unusual encounters. Before I began writing, I dutifully, even enthusiastically, absorbed a variety of experiences. For the most part, I think I enjoyed these and all the stimuli that they brought. (What I Talk About When I Talk About Running pg. 17)
Although we are not all owners of bars or places of intermittent chaos, I believe Murakami brings up a very important point about how a writer should live his life. He seems to be pointing out the importance of embracing the things you encounter instead of avoiding them.
As anyone else has, I have had my fair share of painful life experiences that I’ve tried to sweep under the rug. However, I realized that I was only able to deal with and process these experiences after I had written them down. In the process of writing them down, I had finally resolved the matter in my head and unmapped a little bit more of the world.
The experiences I’ve successfully written down have turned out to be great sources of inspiration for my fiction writing. Whenever I’m struggling with a scene or a particular chapter, I tend to think back on my own experiences and if any of them are particularly relevant to the section that I’m working on.
Forget about Pleasing Everyone (In All Aspects of Life)
Any writer who has received criticism on a written work will not be surprised at all by this point. The fact of the matter is that you can’t impress everyone with your work. So, you should strive to reach your target audience.
Murakami held this stance when he was running a bar during his 20s and he still holds it during his 50s. Evidently, his philosophy of running his bar transferred over well to his writing.
I placed the highest priority on the sort of life that lets me focus on writing, not associating with all the people around me. I felt that the indispensable relationship I should build in my life was not with a specific person, but with an unspecified number of readers. (What I Talk About When I Talk About Running pg. 20)
While we’re writing, we may be tempted to consider how we can alter the content to make it applicable to as many people as possible. This creates a misalignment with our goals for the work and our idea of what we think the public wants. If we’re constantly striving to reach everyone, we might find that we haven’t said anything worthwhile to keep anyone around.
We’ve seen this countless times before with authors that didn’t reach critical acclaim until they’ve passed away. We’re not guaranteed anything by sticking to our resolve and motivation when it comes to writing, but this brings us to an important question: Why did we start writing in the first place?
Most people start writing for themselves. The medium can vary wildly, but the spark is always the same. They have something that they want to create, share, or investigate through writing.
I went on writing the kind of things I wanted to write, exactly the way I wanted to write them, and if that allowed me to make a normal living, then I couldn’t ask for more. (What I Talk About When I Talk About Running pg. 21)
Balance Your Writing with Outside Activities
In a way, this entire book is Murakami talking about how he’s utilized running as a way to stay fit while being a novelist. Given that being a novelist requires long hours at a desk, it doesn’t surprise me that most people and media have the perception that being a writer is an unhealthy activity. Murakami addresses this point head-on.
Basically, I agree with the view that writing novels is an unhealthy type of work. When we set off to write a novel, when we use writing to create a story, like it or not a kind of toxin that lies deep down in all humanity rises to the surface. All writers have to come face-to-face with this toxin and, aware of the danger involved, discover a way to deal with it because otherwise, no creative activity in the real sense can take place. (What I Talk About When I Talk About When I Talk About Running pg. 46)
Writing, whether from experience or from external sources, can be a way of bringing the dark and hidden to the surface. By taking certain stances in writing, a writer is throwing their thoughts and feelings at the mercy of the rest of the world. Murakami also mentions that prolonged exposure to this ‘toxin’ can eventually lead to burnout.
Although he doesn’t offer a clear cut solution in the book, he mentions that a writer must develop an immunity to this ‘toxin’ by developing their physical body. This clearly connects to his stance and opinion on the importance of running for his own fiction writing, but I think the meaning is multi-faceted here.
Writing is especially important to a lot of us, but it cannot be the foundation that we lay our entire being on. By doing so, we are placing ourselves at risk of pretty much having the carpet pulled away from underneath our feet. There has to be another way. Of course, writing will always be one of your top priorities, but you should also strive to balance your life with other purposeful activities, social interactions, and exercises.
Personally, I feel that making music allows me to focus on something else entirely while keeping the narratives of my stories running in the back of my head. Even though I’m not thinking of them directly, I uncover things I hadn’t thought about before. In a way, it helps me to sharpen my dedication when I am actually writing without worrying so much about my craft.
Realize the Importance of Developing Focus and Endurance
The final piece of advice from Murakami revolves around his assertion of the most important qualities for a novelist to have. He mentions that the first and most important quality is talent. The following two are focus and endurance.
You have to continually transmit the object of your focus to your entire body, and make sure it thoroughly assimilates the information necessary for you to write every single day and concentrate on the work at hand. And gradually you’ll expand the limits of what you’re able to do. Almost imperceptibly you’ll make the bar rise. This involves the same process as jogging every day to strengthen your muscles and develop a runner’s physique. Add a stimulus and keep it up. And repeat. Patience is a must in this process, but I guarantee the results will come. (What I Talk About When I Talk About When I Talk About Running pg. 37)
Concentration is a skill like any other. It has to be developed over time. The only way that we can develop it is by integrating it into our writing practice. Relying on a muse is unpredictable, unreliable, and will not satisfy the intended output of most writers.
By having faith in the process, we can cast away our worries around our external goals and focus on the craft of writing itself. Although, like training a muscle, the results are not immediately self-evident, over time you start to see the effect that sticking to something can have.
To return to the mantra that was mentioned at the beginning of this article, writing can be a painful process at times. However, like most activities, we have a choice in how much we have to suffer. By constantly looking for new sources of inspiration in the world around us, realizing that we are at the heart of our written work, reminding ourselves that there is life outside of writing, and having faith in the writing process, we can relinquish some of that suffering and continue to develop as writers. | https://medium.com/swlh/the-4-most-important-things-i-learned-from-haruki-murakamis-writing-philosophy-4623a84d3519 | ['Casimir Mura'] | 2020-10-06 01:29:55.537000+00:00 | ['Fiction', 'Writing', 'Japanese', 'Creativity', 'Books'] | Title 4 Important Things Learned Haruki Murakami’s Writing PhilosophyContent 4 Important Things Learned Haruki Murakami’s Writing Philosophy Casimir Mura Follow Oct 4 · 7 min read gem unearthed ‘memoir’ Haruki Murakami Source Haruki Murakami novelist although may exactly look like he’s also marathon writer book Talk Talk Running discus journey marathon runner impacted writing philosophy foreword Murakami mention mantra read International Herald Tribune article running Pain inevitable Suffering optional Talk Talk Running pg 4 majority book Murakami talking running However certain moment Murakami talk thought process behind running bleeds writing Since running writing painful certain extent isn’t surprising Anyone written anything know feeling stuck writer block know pain associated writing Within moment book Murakami point pain writing alleviate suffering that’s attached Pay Attention World Around big factor appeal Murakami’s writing take u world border reality supernatural hard distinguish doesn’t surprise Murakami value experience running bar Tokyo anything learned university source inspiration writing Sometimes caught life forget always people around u inner life course life experience equal Murakami good fortune encountering sort strange people Running serviceoriented business mean accept whoever come door matter come unless they’re really awful greet friendly smile face Thanks though met kind offbeat people unusual encounter began writing dutifully even enthusiastically absorbed variety experience part think enjoyed stimulus brought Talk Talk Running pg 17 Although owner bar place intermittent chaos believe Murakami brings important point writer live life seems pointing importance embracing thing encounter instead avoiding anyone else fair share painful life experience I’ve tried sweep rug However realized able deal process experience written process writing finally resolved matter head unmapped little bit world experience I’ve successfully written turned great source inspiration fiction writing Whenever I’m struggling scene particular chapter tend think back experience particularly relevant section I’m working Forget Pleasing Everyone Aspects Life writer received criticism written work surprised point fact matter can’t impress everyone work strive reach target audience Murakami held stance running bar 20 still hold 50 Evidently philosophy running bar transferred well writing placed highest priority sort life let focus writing associating people around felt indispensable relationship build life specific person unspecified number reader Talk Talk Running pg 20 we’re writing may tempted consider alter content make applicable many people possible creates misalignment goal work idea think public want we’re constantly striving reach everyone might find haven’t said anything worthwhile keep anyone around We’ve seen countless time author didn’t reach critical acclaim they’ve passed away We’re guaranteed anything sticking resolve motivation come writing brings u important question start writing first place people start writing medium vary wildly spark always something want create share investigate writing went writing kind thing wanted write exactly way wanted write allowed make normal living couldn’t ask Talk Talk Running pg 21 Balance Writing Outside Activities way entire book Murakami talking he’s utilized running way stay fit novelist Given novelist requires long hour desk doesn’t surprise people medium perception writer unhealthy activity Murakami address point headon Basically agree view writing novel unhealthy type work set write novel use writing create story like kind toxin lie deep humanity rise surface writer come facetoface toxin aware danger involved discover way deal otherwise creative activity real sense take place Talk Talk Talk Running pg 46 Writing whether experience external source way bringing dark hidden surface taking certain stance writing writer throwing thought feeling mercy rest world Murakami also mention prolonged exposure ‘toxin’ eventually lead burnout Although doesn’t offer clear cut solution book mention writer must develop immunity ‘toxin’ developing physical body clearly connects stance opinion importance running fiction writing think meaning multifaceted Writing especially important lot u cannot foundation lay entire placing risk pretty much carpet pulled away underneath foot another way course writing always one top priority also strive balance life purposeful activity social interaction exercise Personally feel making music allows focus something else entirely keeping narrative story running back head Even though I’m thinking directly uncover thing hadn’t thought way help sharpen dedication actually writing without worrying much craft Realize Importance Developing Focus Endurance final piece advice Murakami revolves around assertion important quality novelist mention first important quality talent following two focus endurance continually transmit object focus entire body make sure thoroughly assimilates information necessary write every single day concentrate work hand gradually you’ll expand limit you’re able Almost imperceptibly you’ll make bar rise involves process jogging every day strengthen muscle develop runner’s physique Add stimulus keep repeat Patience must process guarantee result come Talk Talk Talk Running pg 37 Concentration skill like developed time way develop integrating writing practice Relying muse unpredictable unreliable satisfy intended output writer faith process cast away worry around external goal focus craft writing Although like training muscle result immediately selfevident time start see effect sticking something return mantra mentioned beginning article writing painful process time However like activity choice much suffer constantly looking new source inspiration world around u realizing heart written work reminding life outside writing faith writing process relinquish suffering continue develop writersTags Fiction Writing Japanese Creativity Books |
785 | How Much Juice Is in the Story? | How Much Juice Is in the Story?
What I learned from messing around for 3 months — 2 proven tactics to write killer content
Photo by Lidya from Pexels
When life gives you lemons, you squeeze them. Nothing else.
Real tight, let’s see how much juice is coming out. Some lemons give you enough to make 3 liters of fresh lemonade, others are barely enough to chuck away your tequila shot.
The same philosophy applies to an idea — some are powerful enough to change the world, others barely enough to get you out of bed in the morning.
Everything in life starts with an idea. Everything you see around you started inside the mind. Someone was crazy enough to pursue the desire to materialize it. You’re currently staring at many iterations.
It wasn’t until I published more than 50 articles this year that the importance of spending adequate time with my ideas became clear to me. | https://medium.com/the-brave-writer/how-much-juice-is-in-the-story-34778baad3bd | ['Jessie Van Breugel'] | 2020-12-24 13:02:36.891000+00:00 | ['Marketing', 'Creativity', 'Writing Tips', 'Inspiration', 'Writing'] | Title Much Juice StoryContent Much Juice Story learned messing around 3 month — 2 proven tactic write killer content Photo Lidya Pexels life give lemon squeeze Nothing else Real tight let’s see much juice coming lemon give enough make 3 liter fresh lemonade others barely enough chuck away tequila shot philosophy applies idea — powerful enough change world others barely enough get bed morning Everything life start idea Everything see around started inside mind Someone crazy enough pursue desire materialize You’re currently staring many iteration wasn’t published 50 article year importance spending adequate time idea became clear meTags Marketing Creativity Writing Tips Inspiration Writing |
786 | Designing for AI: Trust | At IBM, we’re building software solutions that help our users make smarter decisions faster. In the world of data and artificial intelligence (AI), it all comes down to designing products our users can trust enough to help them make those important decisions.
This focus on trust goes beyond data security and validation, it’s about helping our users understand their data, providing relevant recommendations when they need it, and empowering them to create solutions they can be confident in.
As we designed our end to end AI platform IBM Cloud Pak for Data, as well as a diverse set of AI offerings and solutions in our IBM Watson portfolio, we focused on the following 8 principles for establishing trust within AI experiences.
Explainability
At IBM, we believe that good design does not sacrifice transparency and that imperceptible AI is not ethical AI. When designing for AI, you should never hide the decision-making process and your users should always be aware that they are with an AI. To do this, you need to bring explainability into every AI experience so that your users understand the conclusions and recommendations made by the AI.
We’ve established a set of ethical guidelines related to designing for AI called Everyday Ethics for AI that outlines fundamental ways for you to bring explainability into your AI experiences.
Allow for questions. A user should be able to ask why an AI is doing what it’s doing on an ongoing basis. This should be clear and upfront in the user interface. Decision-making processes must be reviewable, especially if the AI is working with highly sensitive personal information data like personally identifiable information, protected health information, and/or biometric data. When an AI is assisting users with making any highly sensitive decisions, the AI must be able to provide them with a sufficient explanation of recommendations, the data used, and the reasoning behind the recommendations. Teams should have access to a record of an AI’s decision processesand be amenable to verification of those decision processes.
Fairness
Humans are inherently biased, and since humans build AI systems…there’s a pretty good chance that human bias could be embedded into the systems we create. It’s our responsibility to minimize algorithmic bias through continuous research and data collection that represents a diverse population. Fairness, like explainability, should be standard practice when it comes to infusing products and services with AI. Meaning that whenever sensitive data is involved, you should design AI experiences that not only minimize bias, but also help your users do the same. You can see this through the bias detector within Watson Openscale where users are alerted to potential bias in data sets.
In this example, you can see that age 65–105 did not get as many favorable outcomes compared to the other groups. It is below the acceptable level so Watson OpenScale marked it with an alert.
There’s also some great work being done by AI Fairness 360, a team of developers who have built an open-source toolkit to help teams examine, report, and mitigate discrimination and bias in their machine learning models. The best part? You can start using these metrics and datasets to start detecting bias in your own AI experiences today.
AI Fairness 360 is an open source toolkit that developers can use to mitigate discrimination and bias in their machine learning models.
Voice and Tone
Walking the voice and tone tightrope is a real challenge for anybody designing or writing for AI. It’s all about finding the balance between too technical and overly simplified. The language that you use within your experiences can go a long way when it comes to building trust with your users. We’ve found that it is best to be succinct and value-driven, and use straightforward language. It’s equally important that you don’t personify the AI because…it isn’t a person. At IBM, this means paying close attention to the language we use when our users are directly interacting with Watson. For example, instead of saying “What can I help you with?”, the Watson avatar should lead with something personal and user-focused like, “What do you want help with next?”
Consistency
If you have a suite of products you expect the same key commands or icons to behave the same as you move between products. Well, the same goes for AI experiences.
When we’re designing with AI we should intentionally design consistent experiences of common elements from product to product. To do this, you can define and leverage AI design patterns. We’ve established universal patterns that should be applied to any moment where Watson is providing guidance or insight. This consistent look and feel ensures that your users aren’t having to relearn a new language every time they open a product with AI capabilities.
Examples of AI design patterns developed for IBM’s products.
Predictability
Predictability is established through consistency. As you continue to deliver these transparent and easily-recognizable AI moments, you’ll get to the point where your users will grow accustomed to AI working alongside them. Ultimately, the need to overtly highlight these experiences will diminish because they will start to understand the possibilities and limits of AI. When you iterate on your experiences, always consider the future and how your users’ understanding and comfort with AI will evolve over time. It’s your job to guide them along the path toward AI maturity, meet them where they’re at, and avoid as many unknowns and surprises as possible.
Learnability
To continuously educate your user and meet them where they’re at along their journey to AI, we recommend leaning on the principles of progressive disclosure. As designers, it’s our job to account for our users’ needs and serve up relevant guidance or content when they need it. We’ve found that most users crave a high-level understanding of what’s going on, but not all of them want to delve into the mechanics of AI. From in-product guidance to expert-level documentation, be sure to consider the moments where your users might need to dive deeper and those times when the complexity might just be too much.
Clarity
When you’re designing for trust within AI experiences, it’s important to think about the unique ways AI can be used to help your users accomplish their goals. One way to do this is to think about clarity. Ask yourself: “how can AI help our users see beyond the obvious?” and on the design side: “what do we need to do to make sure everything within this AI experience is clear and consumable?” Often times, this all comes down to seeing beyond the obvious and translating complex insights into plain language.
Insights presented in plain language insights within Cognos Analytics.
At IBM, this sense of clarity is extremely relevant within Cognos Analytics, an analytics experience that our users explore, visualize, and share insights from their data. With a little help from AI, advanced pattern detection points out interesting relationships our users might not have known were there. And every visualization is accompanied by statistical insights that are presented in plain language.
Confidence
In the world of AI and machine learning, accuracy is key. As you design solutions, always be sure to showcase accuracy and relevancy, so your users can clearly understand how confident the model is in the prediction it made. For a model that contains two output classes (cat vs. tiger) a confidence score of 51% is slightly better than flipping a coin, but a confidence score of 99% indicates that the model is very certain in the judgment that is provided. Knowing the model’s confidence will help your users gauge how much trust they should place in the recommendation.
This example of visual recognition model distinguishes lions from dogs. Notice that when the dog is wearing a lion costume the confidence isn’t as high but the model still got the right answer.
Don’t do it alone
When it comes to delivering AI experiences — or any experience for that matter — a neverending commitment to co-creation is the key to creating something that your users will trust. This means working alongside your users to identify and design for their real needs. At IBM, we’ve developed a robust user research practice that’s centered around co-creation with them.
And the only way to successfully guide your users along their journey to AI is through collaboration, innovation, and trust. | https://medium.com/design-ibm/designing-for-ai-trust-ae0342d2b270 | ['Arin Bhowmick'] | 2020-09-03 13:54:42.158000+00:00 | ['AI', 'Data Science', 'UX', 'Artificial Intelligence', 'Design'] | Title Designing AI TrustContent IBM we’re building software solution help user make smarter decision faster world data artificial intelligence AI come designing product user trust enough help make important decision focus trust go beyond data security validation it’s helping user understand data providing relevant recommendation need empowering create solution confident designed end end AI platform IBM Cloud Pak Data well diverse set AI offering solution IBM Watson portfolio focused following 8 principle establishing trust within AI experience Explainability IBM believe good design sacrifice transparency imperceptible AI ethical AI designing AI never hide decisionmaking process user always aware AI need bring explainability every AI experience user understand conclusion recommendation made AI We’ve established set ethical guideline related designing AI called Everyday Ethics AI outline fundamental way bring explainability AI experience Allow question user able ask AI it’s ongoing basis clear upfront user interface Decisionmaking process must reviewable especially AI working highly sensitive personal information data like personally identifiable information protected health information andor biometric data AI assisting user making highly sensitive decision AI must able provide sufficient explanation recommendation data used reasoning behind recommendation Teams access record AI’s decision processesand amenable verification decision process Fairness Humans inherently biased since human build AI systems…there’s pretty good chance human bias could embedded system create It’s responsibility minimize algorithmic bias continuous research data collection represents diverse population Fairness like explainability standard practice come infusing product service AI Meaning whenever sensitive data involved design AI experience minimize bias also help user see bias detector within Watson Openscale user alerted potential bias data set example see age 65–105 get many favorable outcome compared group acceptable level Watson OpenScale marked alert There’s also great work done AI Fairness 360 team developer built opensource toolkit help team examine report mitigate discrimination bias machine learning model best part start using metric datasets start detecting bias AI experience today AI Fairness 360 open source toolkit developer use mitigate discrimination bias machine learning model Voice Tone Walking voice tone tightrope real challenge anybody designing writing AI It’s finding balance technical overly simplified language use within experience go long way come building trust user We’ve found best succinct valuedriven use straightforward language It’s equally important don’t personify AI because…it isn’t person IBM mean paying close attention language use user directly interacting Watson example instead saying “What help with” Watson avatar lead something personal userfocused like “What want help next” Consistency suite product expect key command icon behave move product Well go AI experience we’re designing AI intentionally design consistent experience common element product product define leverage AI design pattern We’ve established universal pattern applied moment Watson providing guidance insight consistent look feel ensures user aren’t relearn new language every time open product AI capability Examples AI design pattern developed IBM’s product Predictability Predictability established consistency continue deliver transparent easilyrecognizable AI moment you’ll get point user grow accustomed AI working alongside Ultimately need overtly highlight experience diminish start understand possibility limit AI iterate experience always consider future users’ understanding comfort AI evolve time It’s job guide along path toward AI maturity meet they’re avoid many unknown surprise possible Learnability continuously educate user meet they’re along journey AI recommend leaning principle progressive disclosure designer it’s job account users’ need serve relevant guidance content need We’ve found user crave highlevel understanding what’s going want delve mechanic AI inproduct guidance expertlevel documentation sure consider moment user might need dive deeper time complexity might much Clarity you’re designing trust within AI experience it’s important think unique way AI used help user accomplish goal One way think clarity Ask “how AI help user see beyond obvious” design side “what need make sure everything within AI experience clear consumable” Often time come seeing beyond obvious translating complex insight plain language Insights presented plain language insight within Cognos Analytics IBM sense clarity extremely relevant within Cognos Analytics analytics experience user explore visualize share insight data little help AI advanced pattern detection point interesting relationship user might known every visualization accompanied statistical insight presented plain language Confidence world AI machine learning accuracy key design solution always sure showcase accuracy relevancy user clearly understand confident model prediction made model contains two output class cat v tiger confidence score 51 slightly better flipping coin confidence score 99 indicates model certain judgment provided Knowing model’s confidence help user gauge much trust place recommendation example visual recognition model distinguishes lion dog Notice dog wearing lion costume confidence isn’t high model still got right answer Don’t alone come delivering AI experience — experience matter — neverending commitment cocreation key creating something user trust mean working alongside user identify design real need IBM we’ve developed robust user research practice that’s centered around cocreation way successfully guide user along journey AI collaboration innovation trustTags AI Data Science UX Artificial Intelligence Design |
787 | 35 Actionable Tips to Grow Your Medium Blog | 1.
T I M I N G
Publish on a Saturday morning to increase your chances of standing out. You will quickly notice you will even be able to make it to the Medium Top 20 during some weekends with just a few recommendations.
2.
P U S H
Bring your first audience from outside. It is way easier for your story to get noticed on Medium after some initial exposure. Email your close friends and also share your article on your social media accounts.
3.
B A C K L I N K
Place links back to your website (or CTAs) at the end, or integrate them smoothly into your story. Use Google’s free UTM tool to track links to see which posts drive the most traffic back to your website.
4.
B R A N D I N G
Getting your author or brand name noticed can be difficult on Medium’s minimal design. You can insert your name and link it to your website, as I did at the top of this article. Or, Paul also does it in a smart way as below:
5.
I M A G E S
Link images. Click on an image in the editor and press Ctrl+K or ⌘+K to insert your link. (You can also drag and drop an image.) Also optimize your images using a free website like www.compressor.io
6.
G R I D
Upload two or more images at the same time and you have a beautiful image grid. Below is an example with my Instagram pictures linked: | https://medium.com/swlh/35-actionable-tips-to-grow-your-medium-blog-4e4017b89905 | ['Ali Mese'] | 2020-05-12 19:53:54.564000+00:00 | ['Tech', 'Startup', 'Writing', 'Entrepreneurship', 'Blogger'] | Title 35 Actionable Tips Grow Medium BlogContent 1 N G Publish Saturday morning increase chance standing quickly notice even able make Medium Top 20 weekend recommendation 2 P U H Bring first audience outside way easier story get noticed Medium initial exposure Email close friend also share article social medium account 3 B C K L N K Place link back website CTAs end integrate smoothly story Use Google’s free UTM tool track link see post drive traffic back website 4 B R N N G Getting author brand name noticed difficult Medium’s minimal design insert name link website top article Paul also smart way 5 G E Link image Click image editor press CtrlK ⌘K insert link also drag drop image Also optimize image using free website like wwwcompressorio 6 G R Upload two image time beautiful image grid example Instagram picture linkedTags Tech Startup Writing Entrepreneurship Blogger |
788 | This Is My Problem Not Yours | This Is My Problem Not Yours
Losing It
I open my personalized home page
and feel madly overwhelmed
my whimsical escapism caged.
Underscoring my guilt-ridden conscience
I cannot cope with the suggestions presented
of poems and prose and stories from writers
I follow, appreciate and admire but have
increasingly less time to savour and devour.
I read and I clap, and now rarely respond
as a time-consumer like this might trigger
the need for a reciprocal reply — No time to bond
A never-ending story of guilt and pressure
and dwindling pleasure I cannot deny.
More than a three-minute read from a writer
whose work I admire, often disappears
into my writer’s archived graveyard — for later.
I shove back the mountain of poems and stories
with an invisible hand. ‘Push back, push back’
my inner me cries as I try to hop and not skip.
My writing, I note is beginning to suffer
Not as concentrated, or creative as I ride
the production bandwagon. I’m losing my grip
I often make do with ‘whatever the form,
just write a story and tick the obligatory.’
Exhausted, I will rest and recuperate
from reader’s block and a medium meltdown
and rekindle my joy away from the conveyor
belt and the purported ploy — My desire
to clap for all, remaining unsolved.
Oh yes, and you my new followers, grant me
time and I promise to check what wonders
you create and follow you back at a later date. | https://medium.com/blueinsight/this-is-my-problem-not-yours-ffd1bf51c23e | ['Sylvia Wohlfarth'] | 2020-12-12 04:06:55.734000+00:00 | ['Blue Insights', 'Mental Health', 'Motivation', 'Poetry', 'Creativity'] | Title Problem YoursContent Problem Losing open personalized home page feel madly overwhelmed whimsical escapism caged Underscoring guiltridden conscience cannot cope suggestion presented poem prose story writer follow appreciate admire increasingly le time savour devour read clap rarely respond timeconsumer like might trigger need reciprocal reply — time bond neverending story guilt pressure dwindling pleasure cannot deny threeminute read writer whose work admire often disappears writer’s archived graveyard — later shove back mountain poem story invisible hand ‘Push back push back’ inner cry try hop skip writing note beginning suffer concentrated creative ride production bandwagon I’m losing grip often make ‘whatever form write story tick obligatory’ Exhausted rest recuperate reader’s block medium meltdown rekindle joy away conveyor belt purported ploy — desire clap remaining unsolved Oh yes new follower grant time promise check wonder create follow back later dateTags Blue Insights Mental Health Motivation Poetry Creativity |
789 | Convert Commenters Into Leads with a Facebook Auto Responder for Page Posts | Convert Commenters Into Leads with a Facebook Auto Responder for Page Posts Larry Kim Follow Mar 18 · 10 min read
It can be really difficult to get views and engagement on Facebook page posts these days. Last year, Facebook post engagement dropped 50% YoY!
But in order to gather more leads, your business can’t be publishing donkey-level Facebook posts all the time.
Here’s the thing: Facebook’s news feed algorithm only gives visibility to posts that have engagement.
The more active engagement, like comments, on your Facebook posts, the more people will see them.
So here’s a unicorn tactic for higher Facebook post engagement which will help you increase your leads….
The Facebook post auto responder is a Facebook tool that helps your business engage customers, expand reach, and, most importantly, convert your commenters into leads.
Today I want to inspire you with 5 ideas of how you can use a Facebook bot auto responder to convert commenters into leads.
Then we’ll make this comment guard together.
What is a Facebook Auto Responder?
The concept of a Facebook auto responder is simple.
It’s actually a Facebook auto reply bot that automatically sends a message to any user that comments on a Facebook post that is set up with one.
These bots go by different names — Facebook auto responder, auto reply bot, comment guard — they all serve the same purpose:
Turn the people that comment on your posts into named contacts that you can message privately and personally.
The process is simple:
Build a Facebook auto responder
Set up a compelling Facebook post with the bot
Wait for users to comment
When users comment, the bot will automatically send them a message
The best part is, when the user responds to the bot’s message, that user will be added to your Facebook Messenger contact list.
If the bot requests more user information, such as email or phone number, the user can be added to your business’s other contact lists!
Piece of cake.
There’s some secrets to having a Facebook auto responder that gets unicorn engagement and gathers tons of leads.
First of all, you’ll want to make your Facebook post compelling, so that people will want to comment on it.
Some great ways to guarantee comments are:
Offering services
Hosting contests
Asking open-ended questions
Giving discount codes
Entertainment and user interaction
If you create a post offering a discount code, for example, you can encourage users to comment in order to receive the code.
Then, you can set up your Facebook auto responder bot to request the user’s email before giving them the discount code.
Boom!
Now you’ve increased your engagement by encouraging a lot of users to comment on your post, and collected tons of leads using your Facebook auto responder tool.
5 Ways to Use a Facebook Auto Responder to Get More Leads & Increase Engagement
There’s some awesome examples of Facebook auto responders in the wild, so here’s some ideas to inspire you to create your own…
B2B Services
Businesses aren’t always interacting the most with everyday Facebook users.
And there’s no reason that Facebook auto responders should be limited to use by B2C companies.
Companies that provide B2B services can get just as much use out of this Facebook tool.
Here’s an example from Search Engine Journal:
This is a question directed at marketers and people who know marketing well — otherwise, why would they have left behind any marketing tactics in the first place?
Users comment on the post in order to receive SEJ’s list of outdated marketing tactics.
Here’s the result:
First, they send a message with the content of their original post, asking the user to type “ghosts” in order to receive the list.
So, I typed “ghosts,” and…
The bot sent the list, complete with ghost GIF.
Offering B2B services like white papers, lists, etc. in exchange for connecting in Facebook Messenger is an exchange of value to collect leads.
Converting commenters into leads through posts like this is as simple as asking the user for their email and emailing the list to them instead.
Contests
A great way to get folks engaged on your post is to host a contest.
Businesses host contests all the time; prizes can include giveaways, discounts, free products, or things that aren’t even related to the company.
People love contests, so these types of posts tend to get a lot of engagement — especially if entering the contest is as simple as commenting on the post!
That’s why using a Facebook auto responder for a contest is a great way to convert more users into leads.
Check out this example of a contest demo bot responding to users that commented on the contest entry post:
All users do is comment the name of their favorite football team, and the Facebook auto responder sends them a message.
There is one thing to be mindful of when your posts involve contests, giveaways, or specifically asking people to comment: steering clear of Facebook’s cautions around engagement bait.
Facebook has rules against posting engagement bait, which simply means posts that goad users into interacting.
It’s important to be mindful of Facebook’s Newsfeed Guidelines when setting up your Facebook auto responder bots, otherwise your posts might get removed.
Open Ended Questions
On the topic of avoiding engagement bait, here’s an example of a post that could be set up with a Facebook auto responder without specifically asking people to comment, like, or share the post.
Asking open-ended questions allows people to share things (which we all love to do!).
Consider creating a post that asks users an open-ended question where they can type whatever they want for an answer.
Examples of these are “What’s your favorite food?” “what’s your favorite color?” “what’s your dog’s name?” etc.
Check this out:
I posted a simple question on the MobileMonkey Island Facebook group.
This wasn’t set up with a Facebook auto responder, I was simply posing a question.
But, look at that!
It has 309 comments of people sharing where they’re from.
If this was a post on my Facebook business page set up with a Facebook auto responder, that would have been around 309 potential leads.
Discount Codes
Offering discount codes will definitely boost engagement, leading to greater post visibility and helping your convert more users into contacts.
Here’s an example from another MobileMonkey demo bot:
Notice how this post does not explicitly ask users to comment on the post in order to receive the discount code.
It says “Let me know below!”
This way, users know to comment on the post, but it isn’t in direct violation of Facebook’s engagement bait rules.
User Interaction
As we’ve covered, more engagement means more leads.
And in order to get more engagement, a post needs to be compelling.
What better way to interest users than making them part of the fun?
Facebook posts that promote user interaction tend to do well.
Posts like this can include guessing games, silly quizzes, riddles and other fun topics.
Take, for example, this tweet of mine:
It’s simply a photo of a chalkboard with a riddle written on it.
The riddle is silly and fun, so it sparked users to comment…a lot!
Most commenters got the answer right.
The riddle has no question mark at the end, so it’s not a riddle at all — it’s simply a statement!
Regardless of how silly the riddle was, it got a lot of engagement from users online.
Now, if this tweet was a Facebook post set up with a Facebook auto responder, it would have still gotten a ton of engagement — but that engagement would have generated leads.
Because if this riddle was in a Facebook post with a Facebook auto responder, the bot would have reached out to anyone who commented on it, offering them the answer.
Then all of the folks you see tweeting the answer here (and over 100 other replies you don’t see!) would have been converted into leads that I could reach out to again.
In fact, let’s make this exact Facebook post auto reply bot together in MobileMonkey now. Comment on this post to see the final comment autoresponder in action.
How to Build a Facebook Auto Responder in 3 Steps Using MobileMonkey
The first step to building any bot is to sign up for the free MobileMonkey chatbot builder.
This is where you’ll get started by creating the Facebook Messenger funnel that your bot will follow with its dialogues.
To get started, sign up for MobileMonkey and connect your Facebook page. Then you can follow along with these steps.
From the Facebook tools in MobileMonkey, go to FB Comment Guards.
Step 1: Name the comment guard and the choose the post you want to add this comment guard to from the drop-down of your recent Facebook Page posts.
With MobileMonkey’s Free Forever Unlimited Contacts plan, you can make a comment guard that triggers on every comment on every post of your Facebook Page. It’s called a Catch All Comment Guard and it turns every commenter into a lead you can follow-up with using MobileMonkey.
If you upgrade to MobileMonkey PRO (see all chatbot prices here), you can select a specific post for the bot to respond to.
Step 2: Write the conversation that your bot will have with your Facebook post commenters.
Your first question will be what the bot says to people as soon as they comment, so ask a question that matches whatever you included in your Facebook post.
In my Twitter riddle example, I’m going to ask “would you like to know the answer to the riddle?” and I give people a thumbs up emoji to enter the conversation. I follow up that question with the answer to the riddle.
As soon as the commenter clicks that thumbs up emoji (or says anything to the bot) they are now a lead in MobileMonkey. Facebook automatically provides businesses with that users name and basic info like where they are located, their gender, their language, and more useful lead qualifying information.
Because it’s useful to get more lead info like their email address, I’ve added a question to collect their email, where I will send them a follow-up.
Step 3: Set up a frequency cap (how often in a given period someone can comment and get this autoresponder.
And that’s it! Click “Activate” and now your Facebook Post autoreply but is running and ready to collect leads from everyone who comments on your Facebook posts.
And that’s it!
You’ve created your Facebook auto responder and connected it to your post(s).
If you want, you can boost your Facebook post for more visibility, or you can leave it to gather comments from organic traffic.
Either way, whoever comments on your post will become a new lead — and it only took a couple of minutes.
See this whole lead generation chatbot flow in action by leaving a comment on this post.
Who Will See Your Facebook Auto Responder?
Unfortunately, it can be tough to get out of donkey territory in terms of views and engagement for your business on Facebook.
But there are ways to improve your chances of unicorn posts!
Using a Facebook auto responder helps to boost engagement, increase reach, and promote brand awareness for your business.
This great tool also guarantees leads by using Facebook Messenger automation to collect user information and convert more commenters into customers.
Now’s the time for you to head over to MobileMonkey and create a Facebook auto responder for yourself.
Watch your engagement boom and your contact lists grow!
Be a Unicorn in a Sea of Donkeys
Get my very best Unicorn marketing & entrepreneurship growth hacks:
2. Sign up for occasional Facebook Messenger Marketing news & tips via Facebook Messenger.
About the Author
Larry Kim is the CEO of MobileMonkey — provider of the World’s Best Facebook Messenger Marketing Platform. He’s also the founder of WordStream.
You can connect with him on Facebook Messenger, Twitter, LinkedIn, Instagram.
Do you want a Free Facebook Chatbot builder for your Facebook page? Check out MobileMonkey!
Originally posted on Mobilemonkey.com | https://medium.com/marketing-and-entrepreneurship/convert-commenters-into-leads-with-a-facebook-auto-responder-for-page-posts-e37641ad7ec8 | ['Larry Kim'] | 2020-03-18 09:36:00.860000+00:00 | ['Entrepreneurship', 'Facebook Chatbot', 'Marketing', 'Chatbots', 'Facebook'] | Title Convert Commenters Leads Facebook Auto Responder Page PostsContent Convert Commenters Leads Facebook Auto Responder Page Posts Larry Kim Follow Mar 18 · 10 min read really difficult get view engagement Facebook page post day Last year Facebook post engagement dropped 50 YoY order gather lead business can’t publishing donkeylevel Facebook post time Here’s thing Facebook’s news feed algorithm give visibility post engagement active engagement like comment Facebook post people see here’s unicorn tactic higher Facebook post engagement help increase leads… Facebook post auto responder Facebook tool help business engage customer expand reach importantly convert commenters lead Today want inspire 5 idea use Facebook bot auto responder convert commenters lead we’ll make comment guard together Facebook Auto Responder concept Facebook auto responder simple It’s actually Facebook auto reply bot automatically sends message user comment Facebook post set one bot go different name — Facebook auto responder auto reply bot comment guard — serve purpose Turn people comment post named contact message privately personally process simple Build Facebook auto responder Set compelling Facebook post bot Wait user comment user comment bot automatically send message best part user responds bot’s message user added Facebook Messenger contact list bot request user information email phone number user added business’s contact list Piece cake There’s secret Facebook auto responder get unicorn engagement gather ton lead First you’ll want make Facebook post compelling people want comment great way guarantee comment Offering service Hosting contest Asking openended question Giving discount code Entertainment user interaction create post offering discount code example encourage user comment order receive code set Facebook auto responder bot request user’s email giving discount code Boom you’ve increased engagement encouraging lot user comment post collected ton lead using Facebook auto responder tool 5 Ways Use Facebook Auto Responder Get Leads Increase Engagement There’s awesome example Facebook auto responder wild here’s idea inspire create own… B2B Services Businesses aren’t always interacting everyday Facebook user there’s reason Facebook auto responder limited use B2C company Companies provide B2B service get much use Facebook tool Here’s example Search Engine Journal question directed marketer people know marketing well — otherwise would left behind marketing tactic first place Users comment post order receive SEJ’s list outdated marketing tactic Here’s result First send message content original post asking user type “ghosts” order receive list typed “ghosts” and… bot sent list complete ghost GIF Offering B2B service like white paper list etc exchange connecting Facebook Messenger exchange value collect lead Converting commenters lead post like simple asking user email emailing list instead Contests great way get folk engaged post host contest Businesses host contest time prize include giveaway discount free product thing aren’t even related company People love contest type post tend get lot engagement — especially entering contest simple commenting post That’s using Facebook auto responder contest great way convert user lead Check example contest demo bot responding user commented contest entry post user comment name favorite football team Facebook auto responder sends message one thing mindful post involve contest giveaway specifically asking people comment steering clear Facebook’s caution around engagement bait Facebook rule posting engagement bait simply mean post goad user interacting It’s important mindful Facebook’s Newsfeed Guidelines setting Facebook auto responder bot otherwise post might get removed Open Ended Questions topic avoiding engagement bait here’s example post could set Facebook auto responder without specifically asking people comment like share post Asking openended question allows people share thing love Consider creating post asks user openended question type whatever want answer Examples “What’s favorite food” “what’s favorite color” “what’s dog’s name” etc Check posted simple question MobileMonkey Island Facebook group wasn’t set Facebook auto responder simply posing question look 309 comment people sharing they’re post Facebook business page set Facebook auto responder would around 309 potential lead Discount Codes Offering discount code definitely boost engagement leading greater post visibility helping convert user contact Here’s example another MobileMonkey demo bot Notice post explicitly ask user comment post order receive discount code say “Let know below” way user know comment post isn’t direct violation Facebook’s engagement bait rule User Interaction we’ve covered engagement mean lead order get engagement post need compelling better way interest user making part fun Facebook post promote user interaction tend well Posts like include guessing game silly quiz riddle fun topic Take example tweet mine It’s simply photo chalkboard riddle written riddle silly fun sparked user comment…a lot commenters got answer right riddle question mark end it’s riddle — it’s simply statement Regardless silly riddle got lot engagement user online tweet Facebook post set Facebook auto responder would still gotten ton engagement — engagement would generated lead riddle Facebook post Facebook auto responder bot would reached anyone commented offering answer folk see tweeting answer 100 reply don’t see would converted lead could reach fact let’s make exact Facebook post auto reply bot together MobileMonkey Comment post see final comment autoresponder action Build Facebook Auto Responder 3 Steps Using MobileMonkey first step building bot sign free MobileMonkey chatbot builder you’ll get started creating Facebook Messenger funnel bot follow dialogue get started sign MobileMonkey connect Facebook page follow along step Facebook tool MobileMonkey go FB Comment Guards Step 1 Name comment guard choose post want add comment guard dropdown recent Facebook Page post MobileMonkey’s Free Forever Unlimited Contacts plan make comment guard trigger every comment every post Facebook Page It’s called Catch Comment Guard turn every commenter lead followup using MobileMonkey upgrade MobileMonkey PRO see chatbot price select specific post bot respond Step 2 Write conversation bot Facebook post commenters first question bot say people soon comment ask question match whatever included Facebook post Twitter riddle example I’m going ask “would like know answer riddle” give people thumb emoji enter conversation follow question answer riddle soon commenter click thumb emoji say anything bot lead MobileMonkey Facebook automatically provides business user name basic info like located gender language useful lead qualifying information it’s useful get lead info like email address I’ve added question collect email send followup Step 3 Set frequency cap often given period someone comment get autoresponder that’s Click “Activate” Facebook Post autoreply running ready collect lead everyone comment Facebook post that’s You’ve created Facebook auto responder connected post want boost Facebook post visibility leave gather comment organic traffic Either way whoever comment post become new lead — took couple minute See whole lead generation chatbot flow action leaving comment post See Facebook Auto Responder Unfortunately tough get donkey territory term view engagement business Facebook way improve chance unicorn post Using Facebook auto responder help boost engagement increase reach promote brand awareness business great tool also guarantee lead using Facebook Messenger automation collect user information convert commenters customer Now’s time head MobileMonkey create Facebook auto responder Watch engagement boom contact list grow Unicorn Sea Donkeys Get best Unicorn marketing entrepreneurship growth hack 2 Sign occasional Facebook Messenger Marketing news tip via Facebook Messenger Author Larry Kim CEO MobileMonkey — provider World’s Best Facebook Messenger Marketing Platform He’s also founder WordStream connect Facebook Messenger Twitter LinkedIn Instagram want Free Facebook Chatbot builder Facebook page Check MobileMonkey Originally posted MobilemonkeycomTags Entrepreneurship Facebook Chatbot Marketing Chatbots Facebook |
790 | Stephen King Writes Non-Fiction Too | Stephen King Writes Non-Fiction Too
On Writing: A Memoir of the Craft does a masterful job of describing the writing craft, while weaving in King’s personal story.
Many may know this fact by now, but Stephen King, the decorated novelist, once wrote a non-fiction book. Although I am fairly late to the game, considering he published On Writing: A Memoir of the Craft in 2000, I am happy to have finally discovered this gem.
By way of disclaimer, this is the first Stephen King book I have ever read. I hate to admit it, but horror, sci-fi, and all other things Stephen King have never captured my intrigue as genres of choice. Similar to those who enjoy sherry or port, these genres have staunch devotees, but they have yet to appeal to my epicurean palate.
After reading On Writing, however, I may wade further into King’s uncharted waters. Part memoir, instruction manual, and writing guide, this book does a masterful job of describing the writing craft. It weaves in King’s personal story — from his upbringing, early struggles, and addiction to alcohol and drugs, to the undying and tenacious pursuit of perfecting his art. Through pedagogical vignettes, King describes the basics of writing (he used to be an English teacher in Maine, after all).
One of my main takeaways — and something I have experienced in my short time trying to write — is that writing is far more than pure creative expression. It is lightyears from the romantic ideal that some may paint in their minds. Rarely does it entail creative sparks and jolts from muses that whisper inspirational words into your ears at random moments.
Writing, as King describes it, is more akin to building a house. The act of writing is manual labor, requiring dedication, persistence, perseverance, and grit. One cannot approach it self-consciously. It is unkind to thin-skinned, soft flowers. Writing demands truth and a toolbox. If you are not capable of bleeding onto to a page and charging forward without concern for what others may think or say, it is not the activity or occupation for you.
Also, if you want to improve your writing, you must read a lot. Through reading you experience good and bad writing. You learn more words. All elements of a story will start to speak to you more clearly; everything from its structure and plot, to the character development and syntax.
Of course, you cannot call yourself a writer unless you actually write. Constantly and consistently. King is big on this point. He demands that you dedicate time every day to writing. King himself writes 2,000 words a day, even on holidays. His readers should aim for 1,000 per day at a minimum. I am ashamed to say I rarely meet that goal — it is difficult to stay consistent.
As with any craft, people learn what works for them. For King, he likes to write with the door closed, and then rewrite with the door open. The rewrite belongs to anyone who wants to read it. In those moments, he attacks the blank page — without adverbs (which he despises) and never lightly (oops, adverb). He never tells you anything if he can show you instead and he always speaks the truth that bleeds from his writing vein.
And with practice, he believes that you can in fact make a good writer out of a merely competent one. There is hope for most of us at least. To start, I recommend reading On Writing and The Elements of Style by William Strunk Jr. and E. B. White. As King said in his Second Foreword to On Writing, “This is a short book because most books about writing are filled with bullshit.” | https://sebastian-stone.medium.com/stephen-king-writes-non-fiction-too-49181ca82f05 | ['Sebastian Stone'] | 2018-12-15 18:44:47.008000+00:00 | ['Books', 'Nonfiction', 'Memoir', 'Stephen King', 'Writing'] | Title Stephen King Writes NonFiction TooContent Stephen King Writes NonFiction Writing Memoir Craft masterful job describing writing craft weaving King’s personal story Many may know fact Stephen King decorated novelist wrote nonfiction book Although fairly late game considering published Writing Memoir Craft 2000 happy finally discovered gem way disclaimer first Stephen King book ever read hate admit horror scifi thing Stephen King never captured intrigue genre choice Similar enjoy sherry port genre staunch devotee yet appeal epicurean palate reading Writing however may wade King’s uncharted water Part memoir instruction manual writing guide book masterful job describing writing craft weave King’s personal story — upbringing early struggle addiction alcohol drug undying tenacious pursuit perfecting art pedagogical vignette King describes basic writing used English teacher Maine One main takeaway — something experienced short time trying write — writing far pure creative expression lightyears romantic ideal may paint mind Rarely entail creative spark jolt mus whisper inspirational word ear random moment Writing King describes akin building house act writing manual labor requiring dedication persistence perseverance grit One cannot approach selfconsciously unkind thinskinned soft flower Writing demand truth toolbox capable bleeding onto page charging forward without concern others may think say activity occupation Also want improve writing must read lot reading experience good bad writing learn word element story start speak clearly everything structure plot character development syntax course cannot call writer unless actually write Constantly consistently King big point demand dedicate time every day writing King writes 2000 word day even holiday reader aim 1000 per day minimum ashamed say rarely meet goal — difficult stay consistent craft people learn work King like write door closed rewrite door open rewrite belongs anyone want read moment attack blank page — without adverb despises never lightly oops adverb never tell anything show instead always speaks truth bleeds writing vein practice belief fact make good writer merely competent one hope u least start recommend reading Writing Elements Style William Strunk Jr E B White King said Second Foreword Writing “This short book book writing filled bullshit”Tags Books Nonfiction Memoir Stephen King Writing |
791 | 25 Quotes by Toni Morrison That Will Inspire You to Write | Toni Morrison (1931–2019) born on February 18, 1931, in Lorain, Ohio, and named Chole Anthony Wolford at birth.
Morrison was a highly acclaimed American novelist, book editor, essayist and college professor. Her writing focused on her perception of the black experience within black communities, specifically the black female experience.
In 1970 her first novel, The Bluest Eye, was published, and in 1977, The Song of Solomon helped her gain national recognition when she won the National Books Critics Circle Award.
She won the Pulitzer Prize in 1987 for her novel Beloved. She was the first African American woman awarded the coveted Nobel Prize for Literature in 1993.
All writers have days when we need a little writing inspiration. It doesn’t matter if you’re writing a novel, website content, a blog post or a Facebook ad; we all experience times when the words won’t flow, and the little tricks we use to move us in the right direction aren’t working.
Those are the times when I need my comfort food.
My comfort food consists of writers that I read, respect, and admire the most. They provide a unique blend of the right insightful words that can clear my mind, motivate me, and lift me high above the crippling funk called writer’s block.
So, at times, if you too struggle with writing, here are twenty-five quotes from one of the best that will inspire you to write. | https://medium.com/the-brave-writer/25-quotes-by-toni-morrison-that-will-inspire-you-to-write-8c63b263f552 | ['Cheryl Brown'] | 2019-12-11 18:07:19.085000+00:00 | ['Writers On Writing', 'Writing Tips', 'Creativity', 'Motivation', 'Writing'] | Title 25 Quotes Toni Morrison Inspire WriteContent Toni Morrison 1931–2019 born February 18 1931 Lorain Ohio named Chole Anthony Wolford birth Morrison highly acclaimed American novelist book editor essayist college professor writing focused perception black experience within black community specifically black female experience 1970 first novel Bluest Eye published 1977 Song Solomon helped gain national recognition National Books Critics Circle Award Pulitzer Prize 1987 novel Beloved first African American woman awarded coveted Nobel Prize Literature 1993 writer day need little writing inspiration doesn’t matter you’re writing novel website content blog post Facebook ad experience time word won’t flow little trick use move u right direction aren’t working time need comfort food comfort food consists writer read respect admire provide unique blend right insightful word clear mind motivate lift high crippling funk called writer’s block time struggle writing twentyfive quote one best inspire writeTags Writers Writing Writing Tips Creativity Motivation Writing |
792 | Risks and Benefits of Artificial Intelligence on Climate Change | Artificial intelligence — a potential solution?
To start off, let's listen to Jack Ma. We have a bright future with AI relieving some of the many burdens for humans. It seems as though the market is on the rise and that AI will be a part of our society more and more. According to this site, in 2020 the AI market is expected to grow by 54%, likewise in 2021. This could lead to some advancement in many fields, one of which is climate change.
Firstly, many use machine learning (ML) and AI synonymously, I know this is frowned upon by some. Although after reading this article I got the sense that ML is a part of AI but without the ability to perceive, and behave. Also in this article, ML is referred to as a part of a “broader AI”. So, for simplicity ML is in this article seen as a form of AI.
The massive amount of data that exists on climate change warrants the use of AI and is in fact used to some extent already. For instance, pictures of coral reefs are analyzed using ML to determine their status. Also, data on temperature and humidity is used to determine the wellbeing of forests.
One of the sectors that dump the most greenhouse gas into the atmosphere is the electric power sector, to reduce this we need green energy. One problem with many renewable sources though is reliability. For instance, solar power needs the sun, wind power needs wind, and so on. However, using ML to predict the weather, or adjust the propellers of windmills according to the wind, can decrease some of the fluctuations and maximize the output in renewable energies.
There is also hope that the entire electrical grid will become a “smart grid” in the future. Making it possible for the grid to (without human involvement) detect outages and redirect electricity to avoid large blackouts. Also, it could detect massive storms and solar flares and adapt accordingly. It would also be easier to include renewable energy into a smart grid. This would make electricity more reliable and efficient, which is good for the economy and the environment.
There is a lot of data related to climate change that are used to model and “predict” the future. But by using AI or ML to analyze data, predictions could be done with higher accuracy and maybe at a lower cost. One example of an area that needs better modeling is the Antarctic ice sheets that lack the accuracy needed.
There are other areas where AI or ML can be used, including urban planning and optimizing energy use in buildings. It can also be used in discovering new materials, which might speed up the process and lead to new findings that could for instance create solar fuels to store solar energy.
So, it seems as AI could be used in many areas, all from smart sustainable cities to finding new materials to help us adapt to climate change. | https://medium.com/predict/risks-and-benefits-of-artificial-intelligence-on-climate-change-e3d05a455e40 | ['Oliver Zatch'] | 2020-12-20 20:47:54.919000+00:00 | ['Sustainability', 'Artificial Intelligence', 'Climate Change', 'Future'] | Title Risks Benefits Artificial Intelligence Climate ChangeContent Artificial intelligence — potential solution start let listen Jack bright future AI relieving many burden human seems though market rise AI part society According site 2020 AI market expected grow 54 likewise 2021 could lead advancement many field one climate change Firstly many use machine learning ML AI synonymously know frowned upon Although reading article got sense ML part AI without ability perceive behave Also article ML referred part “broader AI” simplicity ML article seen form AI massive amount data exists climate change warrant use AI fact used extent already instance picture coral reef analyzed using ML determine status Also data temperature humidity used determine wellbeing forest One sector dump greenhouse gas atmosphere electric power sector reduce need green energy One problem many renewable source though reliability instance solar power need sun wind power need wind However using ML predict weather adjust propeller windmill according wind decrease fluctuation maximize output renewable energy also hope entire electrical grid become “smart grid” future Making possible grid without human involvement detect outage redirect electricity avoid large blackout Also could detect massive storm solar flare adapt accordingly would also easier include renewable energy smart grid would make electricity reliable efficient good economy environment lot data related climate change used model “predict” future using AI ML analyze data prediction could done higher accuracy maybe lower cost One example area need better modeling Antarctic ice sheet lack accuracy needed area AI ML used including urban planning optimizing energy use building also used discovering new material might speed process lead new finding could instance create solar fuel store solar energy seems AI could used many area smart sustainable city finding new material help u adapt climate changeTags Sustainability Artificial Intelligence Climate Change Future |
793 | You are Rediscovering Yourself. And the epiphanies of lockdown are many | Everything around you bears witness to life but you do not always see it that way. Once you begin to notice it, the ordinary turns every moment into a scavenger hunt for of joy replete with the promise of more. A rainbow appears in your kitchen, bright blue with a yellow handle bursting with multicolor packaging. What looked like rubbish a blink ago doesn’t anymore: Your gaze has transformed it into art.
Sheltering in place during the pandemic is a refresher course about what matters most.
The tap tap tap of fingers on a keyboard as minds think out loud in machine and human language is the reassuring soundtrack to your days. The enthusiastic gurgle of the kettle forecasts the imminent arrival of gentleness in liquid format. Soup simmering on the stove splashes bright yellow over the kitchen tiles, smelling of fresh mint and love. Every cauldron of caring kitchen magic is a three-course meal: Anticipation as an appetizer, soup as an entrée, gratitude as dessert.
Stay put for a while and the aching beauty of ordinariness will make itself known. Contentment is a lot more accessible than you thought. This realization smarts. All that time spent chasing shiny and getting frustrated with how elusive it remained could have been joy.
It’s not too late to tune in. | https://asingularstory.medium.com/you-are-rediscovering-yourself-d5dd25c0af5 | ['A Singular Story'] | 2020-04-16 16:53:24.616000+00:00 | ['Society', 'Mental Health', 'Self', 'Psychology', 'Social Media'] | Title Rediscovering epiphany lockdown manyContent Everything around bear witness life always see way begin notice ordinary turn every moment scavenger hunt joy replete promise rainbow appears kitchen bright blue yellow handle bursting multicolor packaging looked like rubbish blink ago doesn’t anymore gaze transformed art Sheltering place pandemic refresher course matter tap tap tap finger keyboard mind think loud machine human language reassuring soundtrack day enthusiastic gurgle kettle forecast imminent arrival gentleness liquid format Soup simmering stove splash bright yellow kitchen tile smelling fresh mint love Every cauldron caring kitchen magic threecourse meal Anticipation appetizer soup entrée gratitude dessert Stay put aching beauty ordinariness make known Contentment lot accessible thought realization smart time spent chasing shiny getting frustrated elusive remained could joy It’s late tune inTags Society Mental Health Self Psychology Social Media |
794 | Productivity Spelled Backwards Is Self-Discovery | Productivity Spelled Backwards Is Self-Discovery
When you work on the right things, productivity takes care of itself
Public domain image via FreeVintageIllustrations
Earlier this morning, I was sipping a delicious cup of coffee and reading one of my favorite weekly newsletter from Austin Kleon.
He linked to a wonderful little blog post of his called 15 years of blogging (and 3 reasons I keep going). And as someone who’s been blogging for 3 years (and aspiring to many more than 15), I was understandably intrigued…
His first two reasons for blogging were very good, but it was his third reason that struck me:
Because I like it.
He went on to say something more profound than it might appear:
It is very easy to be disciplined when you like what you’re doing.
This kindled a thought of my own:
Every question about personal productivity is just a self-discovery question in disguise.
The more I think about what it means to be productive, the less I find myself interested in techniques, systems, hacks, or even habits, and the more I’m drawn to self-discovery.
The meaning of self-discovery
For me, self-discovery is the ongoing journey to learn more about yourself — your genuine interests, desires, dislikes, needs, preferences, aspirations, and values.
But it’s a surprisingly hard journey. And it’s hard, I think, because it’s so easy to end up unconsciously inheriting preferences and values from other people — family, friends, culture, society, etc.
Of course, who we are and what we value are always social or inherited to some extent. But the real question is one of intentionality:
Do we passively absorb these values and preferences or consciously and intentionally seek to discover them?
The meaning of productivity (and procrastination)
If you struggle with productivity, maybe the problem isn’t how so much as what…
Maybe it’s not your productivity system or anti-procrastination app or even your mindset that’s the problem. Maybe it’s that you’re not working on the right things?
Maybe — just maybe! — you don’t actually know what you want. You think you do because you’re paying attention to what the people you admire want and — with less than full consciousness — you’re copying that.
Maybe you’re absorbing other people’s values instead of discovering your own?
If this was true, what might we expect your work to be like?
I’d guess you wouldn’t feel especially motivated or interested in what you were working on.
You’d start procrastinating quite a bit.
Then you’d start second-guessing yourself and beating yourself up for not being as passionate or motivated or focused as you “should” be.
And of course, all this self-flagellation would lead to even less motivation and more procrastination in the long run — and perhaps eventually, despair.
The virtues of following your nose
Whenever I really get to know someone whom I admire for being prolific, creative, and productive, the same theme keeps popping up: They just really like what they do.
And when I look even more closely, they have this wonderful tendency to follow their nose.
Like a floppy-eared hound dog, they stick their nose to the ground and doggedly follow the scent — blissfully unaware of what the maddening crowd of productivity gurus and creative consultants are whooping and hollering about.
Let me be a little less metaphorical and more direct:
The most productive people seem to practice self-discovery as a way of life.
It’s not even a skill or a habit—though perhaps it started that way. It’s just who they are. It’s what they do. They follow their nose and productivity seems to follow.
Which gets me back to the original idea: | https://medium.com/curious/productivity-spelled-backwards-is-self-discovery-34d62f7c147 | ['Nick Wignall'] | 2020-09-12 22:10:34.727000+00:00 | ['Life Lessons', 'Productivity', 'Inspiration', 'Entrepreneurship', 'Creativity'] | Title Productivity Spelled Backwards SelfDiscoveryContent Productivity Spelled Backwards SelfDiscovery work right thing productivity take care Public domain image via FreeVintageIllustrations Earlier morning sipping delicious cup coffee reading one favorite weekly newsletter Austin Kleon linked wonderful little blog post called 15 year blogging 3 reason keep going someone who’s blogging 3 year aspiring many 15 understandably intrigued… first two reason blogging good third reason struck like went say something profound might appear easy disciplined like you’re kindled thought Every question personal productivity selfdiscovery question disguise think mean productive le find interested technique system hack even habit I’m drawn selfdiscovery meaning selfdiscovery selfdiscovery ongoing journey learn — genuine interest desire dislike need preference aspiration value it’s surprisingly hard journey it’s hard think it’s easy end unconsciously inheriting preference value people — family friend culture society etc course value always social inherited extent real question one intentionality passively absorb value preference consciously intentionally seek discover meaning productivity procrastination struggle productivity maybe problem isn’t much what… Maybe it’s productivity system antiprocrastination app even mindset that’s problem Maybe it’s you’re working right thing Maybe — maybe — don’t actually know want think you’re paying attention people admire want — le full consciousness — you’re copying Maybe you’re absorbing people’s value instead discovering true might expect work like I’d guess wouldn’t feel especially motivated interested working You’d start procrastinating quite bit you’d start secondguessing beating passionate motivated focused “should” course selfflagellation would lead even le motivation procrastination long run — perhaps eventually despair virtue following nose Whenever really get know someone admire prolific creative productive theme keep popping really like look even closely wonderful tendency follow nose Like floppyeared hound dog stick nose ground doggedly follow scent — blissfully unaware maddening crowd productivity guru creative consultant whooping hollering Let little le metaphorical direct productive people seem practice selfdiscovery way life It’s even skill habit—though perhaps started way It’s It’s follow nose productivity seems follow get back original ideaTags Life Lessons Productivity Inspiration Entrepreneurship Creativity |
795 | This Foodie Startup Grew to $27 Million While Saving the World | Connecting the Right Parties and Scaling Up
There is something Uber, Airbnb, and Couchsurfing all have in common. Something that the team of Too Good To Go presumably observed and implemented. As a matter of fact, successfully implemented.
Uber connects people who own a car with individuals who would like to go somewhere. Airbnb connects people who own a summer cottage with travelers. Couchsurfing connects the locals who have a couch at their home with brave travelers who are not afraid to spend the night on a stranger’s couch…
And Too Good To Go connects restaurants and shops that have remaining food with hungry individuals.
The platforms connect parties who would mutually benefit from a business transaction.
And all these platforms connect parties that would otherwise not be able to meet. Or at least not that easily. And definitely not that safely.
A key common feature of these platforms is the rating system. Each party can tell their opinion about the other and give a score, usually ranging from one to five. Ratings keep the affected people accountable for their actions and for the quality of the service. Surely, I wouldn’t choose a Uber driver or an Airbnb host with only one star…
Too Good To Go implemented something similar. Buyers can rate the restaurants by the quality and quantity of the food they get. And also by the friendliness of the staff. That’s a big success factor!
The network effect
Without users from both sides, these companies would be useless. Imagine Uber without drivers or Airbnb without accommodation providers…
So the more users that are on both sides, the more valuable the company is.
Too Good To Go seems to be doing an amazing job here. In just one year, their outreach grew exponentially. Today, they have more than 49,000 partners and 20 million users, and they are still available only in Europe! | https://medium.com/better-marketing/this-foodie-startup-grew-to-100-million-while-saving-the-world-a1cd40fb3bb0 | ['Esther Brhlik'] | 2020-07-27 16:00:07.401000+00:00 | ['Marketing', 'Innovation', 'Sustainability', 'Food', 'Entrepreneurship'] | Title Foodie Startup Grew 27 Million Saving WorldContent Connecting Right Parties Scaling something Uber Airbnb Couchsurfing common Something team Good Go presumably observed implemented matter fact successfully implemented Uber connects people car individual would like go somewhere Airbnb connects people summer cottage traveler Couchsurfing connects local couch home brave traveler afraid spend night stranger’s couch… Good Go connects restaurant shop remaining food hungry individual platform connect party would mutually benefit business transaction platform connect party would otherwise able meet least easily definitely safely key common feature platform rating system party tell opinion give score usually ranging one five Ratings keep affected people accountable action quality service Surely wouldn’t choose Uber driver Airbnb host one star… Good Go implemented something similar Buyers rate restaurant quality quantity food get also friendliness staff That’s big success factor network effect Without user side company would useless Imagine Uber without driver Airbnb without accommodation providers… user side valuable company Good Go seems amazing job one year outreach grew exponentially Today 49000 partner 20 million user still available EuropeTags Marketing Innovation Sustainability Food Entrepreneurship |
796 | Apache Spark-3.0 Sneek peak | Apache Spark has remained strong over the years and now is coming back with one of its major releases with its ongoing goal of Unified Analytics to blend both Batch and Streaming world into one. Let’s see some of the features of it.
Improved Optimizer and Catalog Delta Lake (Acid Transactions) + Linux Foundation Koalas: Bringing spark scale to Pandas Python Upgrade Deep Learning Kubernetes Scala Version upgrade Graph API — Graph and Cypher Script. GPU Support and along with Project Hydrogen Java Upgrade Yarn Upgrade Binary Files
Improved Optimizer and Catalog:
i) Pluggable Data Catalog: (DataSourceV2)
Pluggable catalog integration
Improved pushdown
Unified APIs for streaming and batch
eg: df.writeTo(“catalog.db.table”).overwrite($”year” === “2019”)
ii) Adaptive Query Execution
Make better optimization decisions during query execution
eg: It interprets the size of the table and automatically changes from Sort Merge Join into a Broadcast join and so on.. if one of the tables is small
Dynamic Partition Pruning speeds up expensive joins
Based on the dimension table(Small table) filter query fact table(Large table) will also be filtered making the joins easier and optimal
Delta-Lake:
Delta Lake has been open-sourced for quite some time and has gained its popularity, given its ease of implementation and up-gradation with any existing Spark Applications. I believe, this is a next-generation of Data Lake, which helps overcome Data Swamp as well as the limitations of Lambda and Kappa Architecture. Now with Linux foundation backing up this program will step-up a notch.
Here are some of the features which help us move one step closer towards Unified Analytics.
ACID transactions
Schema enforcement
Scalable metadata handling
Time Travel
Note: More details related to DeltaLake will be updated once I resume on my upcoming daily posts soon — ( Follow me or the hashtag #jayReddy meanwhile)
Koalas: Bringing spark scale to Pandas:
Koalas have been released recently and it is a big add-on for Python developers both for Data Engineers as well as Data Scientists for its similarities between DataFrames and Pandas. Now they can scale up from a single node environment to the distributed environment without having to learn Spark Dataframes separately.
Integrated into Python data science ecosystem. e,g: numpy, matpotlib
60% of the DataFrame / Series API
60% of the DataFrameGroupBy
15% of the Index / MultiIndex API
80% if the plot functions
90% of Multi-Index Columns
Python Upgrade:
Python is expected to completely move out from Version 2 to Version 3.
Deep Learning:
Request GPUs in RDD operations. i.e, you can specify how many GPUs to use per task in an RDD operation, e.g., for DL training and inference.
YARN+Docker support to launch my Spark application with GPU resources. So you can easily define the DL environment in your Dockerfile.
Kubernetes:
Host clusters via Kubernetes are the next big thing it could be on-premise or cloud. The ease of deployment, management and the spin-up time is going to be far exceeding compared to the time taken by other orchestrating containers such as Mesos and Docker Swarm.
Spark-submit with mutating webhook confs to modify pods at runtime
Auto-discovery of GPU resources
GPU isolation at the executor pod level
spark-submit with pod template
Specify the number of GPUs to use for a task (RDD stage, Pandas UDF)
Kubernetes orchestrates containers and supports some container runtimes including Docker. Spark (version 2.3+) ships with a docker file that can be used for this purpose and customized to specific application needs.
Scala Version upgrade:
Scala 2.12
Graph API — Graph and Cypher Script:
Spark Graph Api has a new add-on.
A graph along with Property Graph and Cypher Script.
Cypher query execution, query result handling, and Property Graph storing / loading. The idea behind having a separate module for the API is to allow multiple implementations of a Cypher query engine.
Graph query will have its own Catalysts & it will follow a similar principle as SparkSQL.
GPU Support along with Project Hydrogen:
NVIDIA has the best GPU and it has by far surpassed any other vendors. Spark 3.0 best works with this. (NVIDIA RTX — 2080) is something to watch out for.
Listing GPU Resources
Auto discover GPU
GPU Allocation to a Job and Fall-back
GPU For Pandas UDF
GPU Utilisation and Monitoring
supporting heterogeneous GPU (AMD, Intel, Nvidia)
Java Upgrade:
With every new JDK version release from the Java community, we can see it moving one-step closer towards functional programming.
The release of the Java-8 version was the beginning of it, starting from Lambda Expressions.
Here’s an example of a variable declaration:
Prior to Java 10 version:
String text = "HelloO Java 9";
From Java 10 and higher versions:
var text = "HelloO Java 10 or Java 11";
Yarn Upgrade:
GPU scheduling support
Auto-discovery of GPU
GPU isolation at a process level
Here’s the configuration setup to support GPU’s from Spark or Yarn 3 version onwards
GPU scheduling
In resource-types.xml
configuration>
<property>
<name>yarn.resource-types</name>
<value>yarn.io/gpu</value>
</property>
</configuration>
In yarn-site.xml
<property>
<name>yarn.nodemanager.resource-plugins</name>
<value>yarn.io/gpu</value>
</property>
Binary Files:
Another file format added to support unstructured data. you can use it load images, videos and so on…. The limitation is that it cannot perform a write operation.
val df = spark.read.format(BINARY_FILE).load("Path")
Now that you know a glimpse of the next major Spark release, you can check out the Spark 3.0 preview version.
If you liked this article, then you can check out my article on
Note: Delta-Lake and Koalas can either be part of Spark-3.0 or remain as a separate entity as part of Databricks. | https://towardsdatascience.com/apache-spark-3-0-sneek-peak-284da5ad4166 | ['Jayvardhan Reddy'] | 2019-11-25 04:03:05.397000+00:00 | ['Big Data', 'Artificial Intelligence', 'Data Engineering', 'Data Science', 'Apache Spark'] | Title Apache Spark30 Sneek peakContent Apache Spark remained strong year coming back one major release ongoing goal Unified Analytics blend Batch Streaming world one Let’s see feature Improved Optimizer Catalog Delta Lake Acid Transactions Linux Foundation Koalas Bringing spark scale Pandas Python Upgrade Deep Learning Kubernetes Scala Version upgrade Graph API — Graph Cypher Script GPU Support along Project Hydrogen Java Upgrade Yarn Upgrade Binary Files Improved Optimizer Catalog Pluggable Data Catalog DataSourceV2 Pluggable catalog integration Improved pushdown Unified APIs streaming batch eg dfwriteTo“catalogdbtable”overwrite”year” “2019” ii Adaptive Query Execution Make better optimization decision query execution eg interprets size table automatically change Sort Merge Join Broadcast join one table small Dynamic Partition Pruning speed expensive join Based dimension tableSmall table filter query fact tableLarge table also filtered making join easier optimal DeltaLake Delta Lake opensourced quite time gained popularity given ease implementation upgradation existing Spark Applications believe nextgeneration Data Lake help overcome Data Swamp well limitation Lambda Kappa Architecture Linux foundation backing program stepup notch feature help u move one step closer towards Unified Analytics ACID transaction Schema enforcement Scalable metadata handling Time Travel Note detail related DeltaLake updated resume upcoming daily post soon — Follow hashtag jayReddy meanwhile Koalas Bringing spark scale Pandas Koalas released recently big addon Python developer Data Engineers well Data Scientists similarity DataFrames Pandas scale single node environment distributed environment without learn Spark Dataframes separately Integrated Python data science ecosystem eg numpy matpotlib 60 DataFrame Series API 60 DataFrameGroupBy 15 Index MultiIndex API 80 plot function 90 MultiIndex Columns Python Upgrade Python expected completely move Version 2 Version 3 Deep Learning Request GPUs RDD operation ie specify many GPUs use per task RDD operation eg DL training inference YARNDocker support launch Spark application GPU resource easily define DL environment Dockerfile Kubernetes Host cluster via Kubernetes next big thing could onpremise cloud ease deployment management spinup time going far exceeding compared time taken orchestrating container Mesos Docker Swarm Sparksubmit mutating webhook confs modify pod runtime Autodiscovery GPU resource GPU isolation executor pod level sparksubmit pod template Specify number GPUs use task RDD stage Pandas UDF Kubernetes orchestrates container support container runtimes including Docker Spark version 23 ship docker file used purpose customized specific application need Scala Version upgrade Scala 212 Graph API — Graph Cypher Script Spark Graph Api new addon graph along Property Graph Cypher Script Cypher query execution query result handling Property Graph storing loading idea behind separate module API allow multiple implementation Cypher query engine Graph query Catalysts follow similar principle SparkSQL GPU Support along Project Hydrogen NVIDIA best GPU far surpassed vendor Spark 30 best work NVIDIA RTX — 2080 something watch Listing GPU Resources Auto discover GPU GPU Allocation Job Fallback GPU Pandas UDF GPU Utilisation Monitoring supporting heterogeneous GPU AMD Intel Nvidia Java Upgrade every new JDK version release Java community see moving onestep closer towards functional programming release Java8 version beginning starting Lambda Expressions Here’s example variable declaration Prior Java 10 version String text HelloO Java 9 Java 10 higher version var text HelloO Java 10 Java 11 Yarn Upgrade GPU scheduling support Autodiscovery GPU GPU isolation process level Here’s configuration setup support GPU’s Spark Yarn 3 version onwards GPU scheduling resourcetypesxml configuration property nameyarnresourcetypesname valueyarniogpuvalue property configuration yarnsitexml property nameyarnnodemanagerresourcepluginsname valueyarniogpuvalue property Binary Files Another file format added support unstructured data use load image video on… limitation cannot perform write operation val df sparkreadformatBINARYFILEloadPath know glimpse next major Spark release check Spark 30 preview version liked article check article Note DeltaLake Koalas either part Spark30 remain separate entity part DatabricksTags Big Data Artificial Intelligence Data Engineering Data Science Apache Spark |
797 | A Simple Framework For Writing Blog Posts | One of the most frequent conversations I have with people who want to be bloggers is that it takes them so long to write a blog post. Several hours, or even days, to come up with a thousand words they can hit publish on.
For a long time, I was at a loss. If you can write more than 10 words per minute, it really shouldn’t take you more than an hour or two to write a blog post. Maybe longer if you have to do a lot of research or it’s really long and in depth, but even a beginning writer should be able to get a thousand words out in two hours, tops.
It finally occurred to me that this was one of those instances where something that seemed really easy to me actually isn’t for everyone. I’ve been writing for so long that I’d internalized the ‘how.’ I didn’t think about it, I just did it.
We all have stuff like that. It’s probably what you should be writing about, in fact. Because if something is easy to you, you can bet there are people out there who wish it was easy for them.
During a Ninja Writers call someone asked if I had a system and I was like…well, yeah, I guess so. And I did my best to describe it. I expected them to laugh at me. Honestly. But they ate it up.
I wrote it up that day and the rest is history. The PSST Framework was born.
Yesterday, we talked about choosing the stories you’ll use for your Blog-Your-Own-Book Challenge project. Today, we’re going to talk about exactly how to get each of those posts ready to write.
I hope what you’ll learn will help make sure you actually write your BYOB project all the way to the end. I also think that this system is something that will make you a better blogger overall.
We’re going to talk about a simple framework I use as a pre-writing exercise for my blog posts. The PSST Framework does a couple of things. It helps me make sure that I write a solid blog post that does what I want it to. And it helps me to write it faster.
PSST stands for: Premise, Story, Subheads, and Take Away. Before we get started with each of those sections, I want to talk about three ways you can organize these posts.
In a Notebook
This is the simplest, lowest tech way to organize your posts. Just get a notebook — they’re cheap right now, in anticipation of school starting. Dedicate a page for each of your 31 blog posts.
On Dabble
If you want a digital program for organizing your BYOB project, I’ve been enjoying using Dabble. You can read about how I have mine set up in the post below. I’ve included information about how I use the PSST Framework in Dabble.
In Drafts
Another alternative is to start a draft for each of your 31 posts. You can work on the PSST format right in the post itself. It might be a little more difficult to keep track of your posts this way, unless you don’t start any other new drafts from now until September.
Let’s take a look at each section of the PSST Framework.
Here’s a screenshot from my Dabble plot grid.
Screenshot:author
In Dabble, you can’t change the name of the first column, so in this case ‘Premise’ is actually ‘Book Scenes.’ I’ve also added a last column for the opt-in I want to include in each post, which is helpful because one purpose of these posts is to grow my email list.
For subheads, I’ll need to click on the card to pop it open so I can see what my subheads actually are.
Screenshot: Author
I want you to notice that there are really only a few words in each section. Just enough to remind me about what I want to write.
Premise
This is just another word for ‘title.’ The first step in the PSST Framework is to write the premise of your blog post in a single sentence that will stand as a working title.
Think about it like a working title, actually — because I don’t want you to get caught up in making it perfect right now. And as an added bonus, you’ll need it to be very clear so that you remember what you want to write about later.
Titles should be very concise and clear. Like newspaper headlines. In fact, my favorite exercise for deciding if a title is good or not is to read it to someone and ask if they can figure out what it’s going to be about. If they can’t, I haven’t been clear enough.
Story
Once you have your premise down, your next step is to think about what story you’re going to tell. Even the most teacher-y blog post starts with a story that helps to ground the reader in what you’re about to teach them and why they should care.
A sentence should be enough. Just enough so that you can remember the story you want to tell. You might want to go a little more in depth if you’re afraid you won’t remember.
For instance, in my post ‘what are teeny tiny goals’ I noted that I’ll tell a story about how teeny tiny goals changed my life. I know which story I want to tell and I’m confident I won’t forget. But if I wasn’t so sure, I might add another sentence or two.
I’m a Teacher Writer and like most Teacher Writers, the story is the hardest part for me. I get so excited about just sharing my idea, I forget I’m a storyteller first. And that the story is important to the reader.
(I had to go back in and add a story to this post, after I’d written it. And I actually wrote ABOUT adding a story in this post. See what I mean?)
Subheads
This is the hardest part. You’ll have to think about your premise and what it is you want to tell your reader. What do you want them to know?
Don’t worry about writing perfect subheads that you’ll actually use in your story. Just write a few words or a sentence that will remind you of the sections you want to include in your post.
This is a loose outline of your post. Here are the subheads for my post about teeny tiny goals.
Screenshot: Author
I’m quite sure those won’t be my actual subheads — but they could be. When I look at them, I can see the structure of my post, with a story at the top and a take away rounding the whole thing out.
I’ve given myself enough information so that writing this will be significantly easier than just looking at a blank page. All I need to do is fill out each section with two or three paragraphs.
I haven’t over thought the whole thing. This doesn’t have to be a detailed outline that you could turn into your sophomore English teacher for a grade. Just a few words is enough, I promise.
Take Away
The last part of your post is the take away. What do you want your readers to come away from your post with? It might be any of these:
Solidarity — just the knowledge that they aren’t alone. This is common with personal essays, in particular.
The ability to go out and do a thing. This is common for how-to or DIY posts.
A particular feeling or emotion — hope, escape, desire, etc.
Information that they came for. For instance, if you were writing a book review, the take away would be a recommendation for or against the book.
A tool or other resource tied to your post.
Sometimes the take away is explicit. You’ll end with a paragraph that tells the reader exactly what you hope they’ll do next. Other times it’s more subtle. A personal essay will rarely actually end with ‘I hope you know you’re not alone’ even though that actually is the take away.
Optional: The Opt-In
An opt-in is something that you offer your reader in exchange for their email address. I’m going to talk about this more in the next week, but for now know that you can add the opt-in to your framework if you want to.
I hope you will for BYOB, because the goal here is to build an audience for your book in advance of publishing it. Your opt-in gives people a way to stay in touch and be notified when you publish or when you have something new to offer.
Once you’ve been doing this a while, you’ll have a collection of opt-ins that you can offer whenever it’s appropriate in a post. You can see that I have FRED offered as an opt-in twice in my first three BYOB posts.
FRED is a downloadable blank calendar readers can use as an accountability tool.
And if you click on that hyperlink, you can download it and get on my email list. See how that works?
That’s it.
You have a couple of weeks to go through the framework on all 31 of your blog posts. If you do that, the process of actually writing them will be so much easier and faster. You’ll increase your odds of actually following through.
(HINT: That was my take away for this post!) | https://medium.com/the-write-brain/a-simple-framework-for-writing-blog-posts-de5e64022f58 | ['Shaunta Grimes'] | 2020-07-12 22:37:01.920000+00:00 | ['Byob', 'Writing', 'Blogging', 'Creativity', 'Productivity'] | Title Simple Framework Writing Blog PostsContent One frequent conversation people want blogger take long write blog post Several hour even day come thousand word hit publish long time loss write 10 word per minute really shouldn’t take hour two write blog post Maybe longer lot research it’s really long depth even beginning writer able get thousand word two hour top finally occurred one instance something seemed really easy actually isn’t everyone I’ve writing long I’d internalized ‘how’ didn’t think stuff like It’s probably writing fact something easy bet people wish easy Ninja Writers call someone asked system like…well yeah guess best describe expected laugh Honestly ate wrote day rest history PSST Framework born Yesterday talked choosing story you’ll use BlogYourOwnBook Challenge project Today we’re going talk exactly get post ready write hope you’ll learn help make sure actually write BYOB project way end also think system something make better blogger overall We’re going talk simple framework use prewriting exercise blog post PSST Framework couple thing help make sure write solid blog post want help write faster PSST stand Premise Story Subheads Take Away get started section want talk three way organize post Notebook simplest lowest tech way organize post get notebook — they’re cheap right anticipation school starting Dedicate page 31 blog post Dabble want digital program organizing BYOB project I’ve enjoying using Dabble read mine set post I’ve included information use PSST Framework Dabble Drafts Another alternative start draft 31 post work PSST format right post might little difficult keep track post way unless don’t start new draft September Let’s take look section PSST Framework Here’s screenshot Dabble plot grid Screenshotauthor Dabble can’t change name first column case ‘Premise’ actually ‘Book Scenes’ I’ve also added last column optin want include post helpful one purpose post grow email list subhead I’ll need click card pop open see subhead actually Screenshot Author want notice really word section enough remind want write Premise another word ‘title’ first step PSST Framework write premise blog post single sentence stand working title Think like working title actually — don’t want get caught making perfect right added bonus you’ll need clear remember want write later Titles concise clear Like newspaper headline fact favorite exercise deciding title good read someone ask figure it’s going can’t haven’t clear enough Story premise next step think story you’re going tell Even teachery blog post start story help ground reader you’re teach care sentence enough enough remember story want tell might want go little depth you’re afraid won’t remember instance post ‘what teeny tiny goals’ noted I’ll tell story teeny tiny goal changed life know story want tell I’m confident won’t forget wasn’t sure might add another sentence two I’m Teacher Writer like Teacher Writers story hardest part get excited sharing idea forget I’m storyteller first story important reader go back add story post I’d written actually wrote adding story post See mean Subheads hardest part You’ll think premise want tell reader want know Don’t worry writing perfect subhead you’ll actually use story write word sentence remind section want include post loose outline post subhead post teeny tiny goal Screenshot Author I’m quite sure won’t actual subhead — could look see structure post story top take away rounding whole thing I’ve given enough information writing significantly easier looking blank page need fill section two three paragraph haven’t thought whole thing doesn’t detailed outline could turn sophomore English teacher grade word enough promise Take Away last part post take away want reader come away post might Solidarity — knowledge aren’t alone common personal essay particular ability go thing common howto DIY post particular feeling emotion — hope escape desire etc Information came instance writing book review take away would recommendation book tool resource tied post Sometimes take away explicit You’ll end paragraph tell reader exactly hope they’ll next time it’s subtle personal essay rarely actually end ‘I hope know you’re alone’ even though actually take away Optional OptIn optin something offer reader exchange email address I’m going talk next week know add optin framework want hope BYOB goal build audience book advance publishing optin give people way stay touch notified publish something new offer you’ve you’ll collection optins offer whenever it’s appropriate post see FRED offered optin twice first three BYOB post FRED downloadable blank calendar reader use accountability tool click hyperlink download get email list See work That’s couple week go framework 31 blog post process actually writing much easier faster You’ll increase odds actually following HINT take away postTags Byob Writing Blogging Creativity Productivity |
798 | Fix broken Docker Development on Apple Macbook M1 | Docker Is Not Supported on Apple M1 MacBooks/Mini
Docker is based on Golang and needs virtualization support. Getting Go to support the new architecture is not a problem and will be solved soon.
The big problem is virtualization support, which was always a problem for Docker on windows where Windows Home had no support and not all chips used for PC had the required chipsets.
Docker has written a blog post that explains all the problems that need to be solved: | https://medium.com/better-programming/macbook-m1-breaks-docker-development-14566ab6fa2e | ['Andreas Heissenberger'] | 2020-12-01 15:58:30.517000+00:00 | ['Docker', 'Startup', 'Apple', 'Software Engineering', 'Programming'] | Title Fix broken Docker Development Apple Macbook M1Content Docker Supported Apple M1 MacBooksMini Docker based Golang need virtualization support Getting Go support new architecture problem solved soon big problem virtualization support always problem Docker window Windows Home support chip used PC required chipsets Docker written blog post explains problem need solvedTags Docker Startup Apple Software Engineering Programming |
799 | 🍷 Everyone Was Drunk that Day | E S S A Y
🍷 Everyone Was Drunk that Day
Or an old lady
Illustration by Rolli
War and Peace was overdue again so I carried it back to the library with both hands.
It was October, but it was snowing.
Everyone was drunk that day or an old lady. Three drunks asked me for money, though I only had a pair of quarters on me. The first drunk needed antibiotics for his dog. The second was distraught over his dog needing hip surgery. The third said she had twelve pups and they’d all had puppies. I gave her one of the quarters.
“Riffraff,” said an old lady into her scarf, as she passed. I wasn’t sure if she meant the pup-woman or me. | https://medium.com/pillowmint/everyone-was-drunk-that-day-8975c6ec49c1 | ['Rolli', 'Https', 'Ko-Fi.Com Rolliwrites'] | 2020-07-02 21:09:21.729000+00:00 | ['Humor', 'Nonfiction', 'Life', 'Books', 'Coronavirus'] | Title 🍷 Everyone Drunk DayContent E 🍷 Everyone Drunk Day old lady Illustration Rolli War Peace overdue carried back library hand October snowing Everyone drunk day old lady Three drunk asked money though pair quarter first drunk needed antibiotic dog second distraught dog needing hip surgery third said twelve pup they’d puppy gave one quarter “Riffraff” said old lady scarf passed wasn’t sure meant pupwoman meTags Humor Nonfiction Life Books Coronavirus |