I quit the Effective Altruism forum due to a new rule requiring all new posts and comments be basically put in the public domain without copyright, so anyone could e.g. sell a book of my posts without my consent (they’d just have to give attribution). More info. I had a bunch of draft posts, so I’m posting some of them here with minimal editing. In general, I’m not going to submit them as link posts at EA myself. If you think they should be shared with EA as link posts, please do it yourself. I’m happy for other people to share links to my work at EA or on social media. Please share stuff in whatever ways you think are good to do.
I think EA could be over 50% more effective by raising its standards. Community norms should care more about errors and about using explicit reasoning and methods.
For a small example, quotes should be exact. There should be a social norm that says misquotes are unacceptable. You can’t just change the words (without brackets), put it in quote marks, and publish it. That’s not OK. I believe this norm doesn’t currently exist and there would be significant resistance to it. Many people would think it’s not a big deal, and that it’s a pedantic or autistic demand to be so literal with quotes. I think this is a way that people downplay and accept errors, which contributes to lowered effectiveness.
There are similar issues with many sorts of logical, mathematical, grammatical, factual and other errors where a fairly clear and objective “correct answer” can be determined, which should be uncontroversial, and yet people don’t care and take seriously that getting it right is important. Errors should be corrected. Retractions should be issued. Post-mortems should be performed. What process allowed the error to happen? What changes could be made to prevent similar errors from happening in the future?
It’s fine for beginners to make mistakes, but thought leaders in the community should be held to higher standards, and the higher standards should be an aspirational ideal that the beginners want to achieve, rather than something that’s seen as unnecessary, bad or too much work. It’s possible to avoid misquotes and logical errors without it being a major burden; if someone finds it’s a large burden, that means they need to practice more until they improve their intuition and subconscious mind. Getting things right in these ways should be easy and something they you can do while tired, distracted, autopiloting, etc.
Fixes like these alone won’t make EA far more effective by themselves. They will set the stage to enable more advanced or complex improvements. It’s very hard to do more important improvements when frequently making small errors. Getting the basics rights enables working more effectively on more advanced issues.
One of the main more advanced issues is rational debate.
Another is not trusting yourself. Don’t bet anything on your integrity or lack of bias when you can avoid it. There should be a strong norm against doing anything that would fail if you have low integrity or bias. If you can find any alternative, which doesn’t rely on your rationality, do that instead. Bias is common. Learning to be better at not fooling yourself is great, but you’ll probably screw it up a lot. If you can approach things so that you don’t have the opportunity to fool yourself, that’s better. There should be strong norms for things like transparency and following flowcharted methods and rules that dramatically reduce scope for bias. This can be applied to debate as well as to other things. And getting debate right enables criticism when people don’t live up to norms; without getting debate right norms have to be enforced in significant part with social pressure which compromises the rationality of the community and prevents it from clearly seizing the rationality high ground in debates with other groups.
Messages