It has stimulated some good discussion about alternative reasons that this phenomenon exists.
I had neglected this important consideration and I will retract my downvote until I have thought about this more. I still think this post shows off some bad cognitive habits, and I’m afraid that it getting many upvotes would both incentivize bad cognitive habits and reflect poorly on Less Wrong. Thus currently my policy is “downvote if it gets above 15, upvote if it gets below 0, else do nothing”.
It was not an attempt to construct a model of human psychology, which would be more fit for a PhD thesis than a 550 word post.
I didn’t mean to imply that. I was trying to say “this is not how one should generally go about constructing a model of any aspect or set of aspects of human psychology” but thought that sounded too clunky.
I agree that a <1000 word post shouldn’t go into lots of details, but if you’re trying to keep it short then I think it’s a bad idea to put forth a hypothesis unless it’s sufficiently clear that it’s a particularly good or interesting one. I think you could have spent the words you did on your hypothesis much more effectively by proposing some plausible hypotheses and then explicitly asking Less Wrong what they thought. I would consider upvoting the post if you did this, sexy title be damned, but I realize that would be a fair bit of work for you even if you agreed it would be better.
Interesting. Okay, thank you for the feedback. One thing I’m going to think about is signal-to-noise ratio vs. putting ideas out there.
My first inclination is that putting out a larger volume of potentially correct work and letting it go through trial-by-fire and be discussed is superior to waiting until an argument is fully bulletproofed, caveat’ed, and so on. But there’s probably some tuneups I could have made to write it more strongly—I’ll sleep on it tonight, re-read your comments tomorrow, and give it more thought. Thanks for your replies.
A large volume of potentially correct work that you seek discussion on might do better in the discussion area. I found this one valuable for the term Egalité Irréfléchie, which does have a nice ring to it; but otherwise a bit tedious: Everything between the second paragraph and “I used to wonder...” sounded completely superfluous; the last half sounded painfully belabored on its own points.
The redundancy of the text was balanced by the lightness of the support. Where lukeprog would’ve added 60 citations to peer-reviewed papers, you merely said “there’s plenty of research on this.”
I’ve enjoyed your writing before, but I think this one wasn’t quite ready for the main LW posting area.
My first inclination is that putting out a larger volume of potentially correct work and letting it go through trial-by-fire and be discussed is superior to waiting until an argument is fully bulletproofed, caveat’ed, and so on.
Definitely depends on what your goals are. If you’re interested in getting feedback for your ideas while stimulating discussion then doing what you did except with more of a “but that’s my take on it and I’m not completely satisfied with it yet, what does Less Wrong think?” approach will get more useful feedback. Posting things somewhat-haphazardly will get you more feedback on background/meta stuff, like the things I’ve focused on in my comment replies, which can be useful but can also backfire in non-obvious ways. Your reputation might get hurt to some extent, which will cause you to get less upvotes and attention in the future when you want people to really pay attention to your well-thought-out ideas. I think this downside is very easy to underestimate. I tentatively think (though I haven’t done a quantitative analysis and there are many other possible explanations) that I used to get a lot more karma per comment before I started posting about things that had way too much inferential distance, or that pattern-matched to things people believe for silly reasons. That was about 4 months ago.
I had neglected this important consideration and I will retract my downvote until I have thought about this more. I still think this post shows off some bad cognitive habits, and I’m afraid that it getting many upvotes would both incentivize bad cognitive habits and reflect poorly on Less Wrong. Thus currently my policy is “downvote if it gets above 15, upvote if it gets below 0, else do nothing”.
I didn’t mean to imply that. I was trying to say “this is not how one should generally go about constructing a model of any aspect or set of aspects of human psychology” but thought that sounded too clunky.
I agree that a <1000 word post shouldn’t go into lots of details, but if you’re trying to keep it short then I think it’s a bad idea to put forth a hypothesis unless it’s sufficiently clear that it’s a particularly good or interesting one. I think you could have spent the words you did on your hypothesis much more effectively by proposing some plausible hypotheses and then explicitly asking Less Wrong what they thought. I would consider upvoting the post if you did this, sexy title be damned, but I realize that would be a fair bit of work for you even if you agreed it would be better.
Interesting. Okay, thank you for the feedback. One thing I’m going to think about is signal-to-noise ratio vs. putting ideas out there.
My first inclination is that putting out a larger volume of potentially correct work and letting it go through trial-by-fire and be discussed is superior to waiting until an argument is fully bulletproofed, caveat’ed, and so on. But there’s probably some tuneups I could have made to write it more strongly—I’ll sleep on it tonight, re-read your comments tomorrow, and give it more thought. Thanks for your replies.
A large volume of potentially correct work that you seek discussion on might do better in the discussion area. I found this one valuable for the term Egalité Irréfléchie, which does have a nice ring to it; but otherwise a bit tedious: Everything between the second paragraph and “I used to wonder...” sounded completely superfluous; the last half sounded painfully belabored on its own points.
The redundancy of the text was balanced by the lightness of the support. Where lukeprog would’ve added 60 citations to peer-reviewed papers, you merely said “there’s plenty of research on this.”
I’ve enjoyed your writing before, but I think this one wasn’t quite ready for the main LW posting area.
Definitely depends on what your goals are. If you’re interested in getting feedback for your ideas while stimulating discussion then doing what you did except with more of a “but that’s my take on it and I’m not completely satisfied with it yet, what does Less Wrong think?” approach will get more useful feedback. Posting things somewhat-haphazardly will get you more feedback on background/meta stuff, like the things I’ve focused on in my comment replies, which can be useful but can also backfire in non-obvious ways. Your reputation might get hurt to some extent, which will cause you to get less upvotes and attention in the future when you want people to really pay attention to your well-thought-out ideas. I think this downside is very easy to underestimate. I tentatively think (though I haven’t done a quantitative analysis and there are many other possible explanations) that I used to get a lot more karma per comment before I started posting about things that had way too much inferential distance, or that pattern-matched to things people believe for silly reasons. That was about 4 months ago.