Project proposal: Rationality Cookbook

by toonalfrink 1 min read21st Nov 201719 comments


In this post, Eliezer warns us that what works for us, might not always work for others:

"You read nineteen different webpages advising you about personal improvement—productivity, dieting, saving money.  And the writers all sound bright and enthusiastic about Their Method, they tell tales of how it worked for them and promise amazing results...
But most of the advice rings so false as to not even seem worth considering.  So you sigh, mournfully pondering the wild, childish enthusiasm that people can seem to work up for just about anything, no matter how silly.  Pieces of advice #4 and #15 sound interesting, and you try them, but... they don't... quite... well, it fails miserably.  The advice was wrong, or you couldn't do it, and either way you're not any better off.
And then you read the twentieth piece of advice—or even more, you discover a twentieth method that wasn't in any of the pages—and STARS ABOVE IT ACTUALLY WORKS THIS TIME."

Say you have been struggling with insomnia for years. One day, you figure out that replacing your white curtains with black ones makes you sleep like a baby. Relieved and full of optimism, you run to your laptop to share the news. "Everyone! I have solved insomnia!"

Most of you will recognize the mistake.

In my opinion, typical mind is one of the most depressing things out there. Not because of the mistake, but because of it's underlying reality that everyone's mind is different. You may solve an internal problem like depression or insomnia for one person, but that doesn't mean you solved it for everyone. Most solutions only work for a small set of people. It makes these problems so much harder.

But we can at least do the next best thing, which is what this project is about.

The basic point is that having a list of 20 plausible solutions is still better than having to generate all of these yourself. These solutions could even be ranked by the percentage of people they worked for. If, then, you work through the list one by one, you would succeed much faster on average. Even if most people's advice doesn't work for you, you may find that obscure weird fix at the bottom of the list from the person that had exactly your problem.

A rationality cookbook is about aggregating solutions, and tallying how often they worked. Here's what it could look like:


  • Melatonin 70% (14/20)
  • Removing stressors 64% (9/14)
  • Exercise 50% (8/16)
  • ...


  • Meditation retreat 100% (3/3)
  • Routine 80% (8/10)
  • Fixing vitamin deficiencies 10% (2/20)
  • ...

Gym akrasia

  • Aversion factoring 63% (5/8)
  • ...

Social Anxiety

  • CoZe 70% (7/10)
  • ...


X y% (a/b) means that solution X was tried by b people and worked for a of them.

(These numbers are totally made up)

How could we implement this? I've thought about a subreddit, but the mechanics are kind of awkward (some people might upvote just because an idea interests them, and downvoting might make people feel sad). Perhaps it would be better to code a simple website of our own. It could require people to fill in a form with the full story of their solution. It could require a certain amount of LW karma to participate. We could make it part of the new lesswrong. It would truly be ours. That has it's benefits.

Thoughts? Any coders that would like to jump into this?