Weekly Review, 2021-01-17
Weekly reviews are where I aggregate the best of my week. My goal is to make my information consumption more active, while sharing interesting ideas, articles, podcasts, and videos with others.
This week I wrote about a lesson I learned in high school and how I'm applying it now, updated empty notes I linked to last week, talk about the new [[Idea Queue]], and share some notes on [[Balajis.com]].
Note of the Week - [[Skills beat Fitness]]
If you only read one thing from this review, it should be this. I talk about the most important lesson I learned from years of training for baseball - Skills beat Fitness - and how I am using that lesson now.
Last Week, I linked to [[Things You Should Never Do, Part 1]] without having added any notes to the page! I updated it this week with notes from the original article by [[Joel Spolsky]]. It's about a fundamental law of programming and the effects it has on code bases everywhere. Check it out: [[Things You Should Never Do, Part 1]]
The [[Idea Queue]] is a new page where I'm going to keep a list of future ideas I want to work on. This list will constantly change and I may not follow through with many of them, but I thought sharing them would be a cool way for people to give feedback before I even start working on them.
One idea on the [[Idea Queue]] is [[Imaginary Podcast Preparation]]. The plan is to build notes on individuals as if I were going to have them on a podcast. This week, I started to compile notes from [[balajis.com]].
For now, this page just has a collection of notes on ideas I've first learned about from Balaji.
My favorite note from Balaji so far is [[How to Gradually Exit Twitter]].
Some others include
Last week, I wrote about my notes on how I differentiate between [[complicated vs complex]] ideas. This week, I added quick thoughts on how I think about a new term - your [[explainability threshold]].
Article of the Week - [[Seeing Like An Algorithm]]
This article explores ideas behind the For You Page on [[TikTok]] and why it is so effective. The main point is [[Algorithm Friendly Design]], which is a design starts with algorithms in mind from the start. This enables the creation of huge datasets meant for training machine learning algorithms with a specific business driven purpose.
This is also what originally led me to [[A Big Little Idea Called Legibility]].
My main takeaways from this article:
- Most apps we use today are at their limits of good reccomendations because of constraints on design and lack of signal from users on specific pieces of content (e.g. Google, Facebook, Yelp, Twitter)
- [[Algorithm friendly design]] is going to become a design pillar for all future large scale apps. Any large scale consumer facing companies will begin to suffer as new apps with algorithm friendly design create much better user recommendations because they have a much better dataset as a result of design.
[[Algorithm friendly design]] (or the same idea under a different name) will probably become a popular consulting job or the strongsuit of new app building workshops.
I tried to do a better job of putting some notes on any page I linked to this week. I have trouble deciding if I should put more content on notes, or have a lot of pages that are only referred to by linked notes.
Either way, if you've come this far, I hope you found something you enjoyed.
My Linked Notes
One last thing
If you liked these notes, hit me on Twitter!