I think this is a very good post that captures a lot of useful information.
I admit, though, that I did not read all of it, and I merely went through to see what might be new to me / how you categorized the different things.
Slight tangent:
I worry about the practicality of writing rationality stuff and how the medium affects the practice. Like, to what extent does writing something like this have an impact on allowing people to actually learn / execute on such skills?
EX: There seem to be pedgagoical techniques even in the writing medium that we could take more advantage of, e.g. Eliezer’s old Advanced Epistemology 2012 sequence which gave readers puzzles to think about, or, if we allow the full power of interactivity, stuff like what what Coursera does with homework / quizzes.
One thing that seems important is the state that you’re in when you read stuff / what the framing is. EX: If you’re reading this stuff on your downtime and not at an opportunity where you can actively practice stuff, then the chance of this sort of info cashing out into real actionables is definitely lower.
(Of course I’m not saying there’s no value in having stuff like this primed / available / integrated into your models. Just that there’s a difference between that (which I think is what happens to most people when they read stuff like this) and actually going out into the world and doing stuff.)
It’s often hard to do this when you’ve already got time going into trying to do other things; the right opportunities in your day that allow you to be actively engaged or doing deliberate practice seem limited.
So there seem to be a few things here:
1) One is about how we probably could improve the whole platform on which we’re trying to teach rationality and extend it past these ridiculous essays (of which many of us are guilty of writing) and actually try applying more of the techniques regarding effective learning that we write about: stuff in the general vicinity of optimizing for reader consumption / takeaways.
2) Two is about how creating the affordances in 1) might not even be enough because doing basically anything well requires a dedicated level of attention and effort that most people have a limited budget of. Realistically, I expect only a few people who read a massive post like this to have it affect their real-life actions. There’s still an intention-action gap here that we can’t fully bridge via text. But it’s not just this gap; it’s also about how people don’t always have enough of a budget to integrate this stuff because other things demand their limited attention.
3) Conflicting endgames: There’s definitely a whole bunch of value in charting out and trying to summarize existing research into a nice synthesis. And this post is an example of that, so maybe the point isn’t that we hope people can learn all of them from this post alone. But given that most productivity / research synthesis posts on LW look like this, I worry that we might conflate posts like these with posts that are designed to try and teach you a skill, which might look quite different. (I think operationalization, answering the question of “What does this abstract skill look like in practice?” is a huge part of it, and it’s still largely neglected ’round these parts.)
I think this is a very good post that captures a lot of useful information.
I admit, though, that I did not read all of it, and I merely went through to see what might be new to me / how you categorized the different things.
Slight tangent:
I worry about the practicality of writing rationality stuff and how the medium affects the practice. Like, to what extent does writing something like this have an impact on allowing people to actually learn / execute on such skills?
EX: There seem to be pedgagoical techniques even in the writing medium that we could take more advantage of, e.g. Eliezer’s old Advanced Epistemology 2012 sequence which gave readers puzzles to think about, or, if we allow the full power of interactivity, stuff like what what Coursera does with homework / quizzes.
One thing that seems important is the state that you’re in when you read stuff / what the framing is. EX: If you’re reading this stuff on your downtime and not at an opportunity where you can actively practice stuff, then the chance of this sort of info cashing out into real actionables is definitely lower.
(Of course I’m not saying there’s no value in having stuff like this primed / available / integrated into your models. Just that there’s a difference between that (which I think is what happens to most people when they read stuff like this) and actually going out into the world and doing stuff.)
It’s often hard to do this when you’ve already got time going into trying to do other things; the right opportunities in your day that allow you to be actively engaged or doing deliberate practice seem limited.
So there seem to be a few things here:
1) One is about how we probably could improve the whole platform on which we’re trying to teach rationality and extend it past these ridiculous essays (of which many of us are guilty of writing) and actually try applying more of the techniques regarding effective learning that we write about: stuff in the general vicinity of optimizing for reader consumption / takeaways.
2) Two is about how creating the affordances in 1) might not even be enough because doing basically anything well requires a dedicated level of attention and effort that most people have a limited budget of. Realistically, I expect only a few people who read a massive post like this to have it affect their real-life actions. There’s still an intention-action gap here that we can’t fully bridge via text. But it’s not just this gap; it’s also about how people don’t always have enough of a budget to integrate this stuff because other things demand their limited attention.
3) Conflicting endgames: There’s definitely a whole bunch of value in charting out and trying to summarize existing research into a nice synthesis. And this post is an example of that, so maybe the point isn’t that we hope people can learn all of them from this post alone. But given that most productivity / research synthesis posts on LW look like this, I worry that we might conflate posts like these with posts that are designed to try and teach you a skill, which might look quite different. (I think operationalization, answering the question of “What does this abstract skill look like in practice?” is a huge part of it, and it’s still largely neglected ’round these parts.)
Interesting thoughts. I agree that straight informational text like this probably isn’t the most pedagogically effective.