More writeups!

February 6th, 2020
meta, transparency
A kind of post I would like to read more of is "I did X, and here's how it went." You tend to see this most with research, but I've enjoyed reading them on all sorts of things. This is one of the main kinds of post I write, and I would encourage others to give it a try!

Here's an arbitrary selection of writeups I've enjoyed:

Major reasons I like writeups include:

  • Exposure to problems that I didn't know were important to people.

  • Detailed looks at how people handled problems.

  • They're generally very concrete and written after the fact, which makes it harder to write something that's neat but wrong.

  • When I'm thinking of doing something and find a writeup by someone who did something similar it's fantastically useful.

When you voluntarily make something public, you do risk that people will irresponsibly beat you up over failings. I think it's generally worth it to go ahead anyway, tell the whole story, and help build a norm of sharing things so others can learn.

Referenced in: 5x5 Go Board

Comment via: facebook, lesswrong, substack

Recent posts on blogs I like:

AI incompetence often comes from misalignment

Sometimes I see people say “I’m not worried about AI risk because AIs are really bad at things.” I think this is a misunderstanding.

via Thing of Things April 27, 2026

You should try contra dancing

a story of middle school Ben • a not-very-illuminating description of the mechanics • flow, joy, and community • the antidote to the rest of life • how to try contra

via benkuhn.net April 24, 2026

On AI writing in 2026

I use AI to write a little bit: I ask it for high level feedback on blog post drafts, make mechanical edits, and sometimes use it to brainstorm options for wording at a paragraph level. It’s unusual that I accept its wording or changes without modificatio…

via Home April 16, 2026

more     (via openring)