I’d like to throw out some more bad ideas, with fewer disclaimers about how terrible they are because I have less reputation to hedge against.
Inline Commenting
I very strongly endorse the point that it seems bad that someone can make bad claims in a post, which are then refuted in comments which only get read by people who get all the way to the bottom and read comments. To me the obvious (wrong) solution is to let people make inline comments. If nothing else, having a good way within comments to point to what part of the post you want to address feels like a strict win, and given that we already have pingbacks I think letting sufficiently good comments exist alongside the post would also be good. This could also be the kind of thing that a poster can enable or disable, and that a reader can toggle visibility on.
Personal Reputation
I don’t have great models for how reputation should or does work on LessWrong. The second of these is testable though—I’d be curious to see what happened if prominent accounts, before commenting, flipped a coin, and in half of all cases posted through a random alt. Of course it may not be a bad thing if respected community figures get more consideration, but it would just be interesting to know how much of an effect it had. There are loads of obvious ways to hedge against this, all themed around anonymisation at different levels, but I think here it’s less of a ‘can’ and more of a ‘should’, so I’d be curious to hear anyone else’s thoughts on that.
Curated Comments
I agree that there are comments that are epistemically not so great. There’s some underlying, very complicated question about ‘who gets to decide what comments people should read’, and I have some democratic instinct which resists any centralisation. But it does feel like some comments are notably higher-effort, or particularly at risk of brigading. I reckon a full prediction market-style moderation system would be a mess, but it seems like it wouldn’t be that hard if, when someone made a comment, they could submit it for curation as ‘a particularly carefully considered, relevant, and epistemically hygienic response’ which, if approved, would be bumped above non-curated comments, with some suitable minor penalty for failed attempts or notes of feedback.
Debate as a model
In formal debate (or at least the kind I did) you distinguish between a point of information and a point of order. When you try to lodge a point of information, the opposing speaker can choose whether they’d like to be interrupted, and you’re just interjecting some relevant facts. A point of order, though, is made to the chair, when there’s a procedural violation, and it can very much interrupt you. I’m not sure how you’d extend this to lesswrong but it feels like a useful distinction in a similar context.
I very strongly endorse the point that it seems bad that someone can make bad claims in a post, which are then refuted in comments which only get read by people who get all the way to the bottom and read comments. To me the obvious (wrong) solution is to let people make inline comments. If nothing else, having a good way within comments to point to what part of the post you want to address feels like a strict win, and given that we already have pingbacks I think letting sufficiently good comments exist alongside the post would also be good. This could also be the kind of thing that a poster can enable or disable, and that a reader can toggle visibility on.
You can link to comments, so that is an easy technical solution. As ever , it’s mainly a cultural problem: if good quality criticism were upvoted, it would appear at the top of the comments anyway, and bit be buried.
I’d like to throw out some more bad ideas, with fewer disclaimers about how terrible they are because I have less reputation to hedge against.
Inline Commenting
I very strongly endorse the point that it seems bad that someone can make bad claims in a post, which are then refuted in comments which only get read by people who get all the way to the bottom and read comments. To me the obvious (wrong) solution is to let people make inline comments. If nothing else, having a good way within comments to point to what part of the post you want to address feels like a strict win, and given that we already have pingbacks I think letting sufficiently good comments exist alongside the post would also be good. This could also be the kind of thing that a poster can enable or disable, and that a reader can toggle visibility on.
Personal Reputation
I don’t have great models for how reputation should or does work on LessWrong. The second of these is testable though—I’d be curious to see what happened if prominent accounts, before commenting, flipped a coin, and in half of all cases posted through a random alt. Of course it may not be a bad thing if respected community figures get more consideration, but it would just be interesting to know how much of an effect it had. There are loads of obvious ways to hedge against this, all themed around anonymisation at different levels, but I think here it’s less of a ‘can’ and more of a ‘should’, so I’d be curious to hear anyone else’s thoughts on that.
Curated Comments
I agree that there are comments that are epistemically not so great. There’s some underlying, very complicated question about ‘who gets to decide what comments people should read’, and I have some democratic instinct which resists any centralisation. But it does feel like some comments are notably higher-effort, or particularly at risk of brigading. I reckon a full prediction market-style moderation system would be a mess, but it seems like it wouldn’t be that hard if, when someone made a comment, they could submit it for curation as ‘a particularly carefully considered, relevant, and epistemically hygienic response’ which, if approved, would be bumped above non-curated comments, with some suitable minor penalty for failed attempts or notes of feedback.
Debate as a model
In formal debate (or at least the kind I did) you distinguish between a point of information and a point of order. When you try to lodge a point of information, the opposing speaker can choose whether they’d like to be interrupted, and you’re just interjecting some relevant facts. A point of order, though, is made to the chair, when there’s a procedural violation, and it can very much interrupt you. I’m not sure how you’d extend this to lesswrong but it feels like a useful distinction in a similar context.
You can link to comments, so that is an easy technical solution. As ever , it’s mainly a cultural problem: if good quality criticism were upvoted, it would appear at the top of the comments anyway, and bit be buried.