I don’t know how to say this without sounding elitist, but my guess is that people who prolifically write LW comments and whose karma:(comments+posts) ratio is less than around 1.5:1 or maybe even 2:1 should be more selective in what they say. Around this range, it would be unwarranted for mods to rate-limit you, but perhaps the bottom 30% of content you produce is net negative considering the opportunity cost to readers.
Of course, one should not Goodhart for karma and Eliezer is not especially virtuous by having a 16:1 ratio, but 1.5:1 is a quite low ratio. You get 1.5 karma if 0.75 high-karma users or 1.5 low-karma users weakly upvote your comment. Considering that it takes barely a second to upvote and comments get tens of views, this is not a high bar for the average comment.
Caveats:
If your absolute number of comments + posts is low, the risk of making new users participate less outweighs the risk of comments clogging up the frontpage.
In some contexts votes might underestimate quality, like if your comment is on some esoteric topic that few people engage with.
Another circumstance where votes underestimate quality is where you often get caught up in long reply-chains (on posts that aren’t “the post of the week”), which seems like a very beneficial use of LessWrong, but typically has much lower readership & upvote rates, but uses a lot of comments.
Aiming to improve the quality of threads by telling users to post/comment less in a public forum seems to me like it’s not going to be very helpful in the long term. If users self-select by simulating how much karma their post/comment gets (perhaps via extrapolating the average karma ratio of their previous posts) it might work for a subset, but will fail for other cases:
Namely:
you’ll miss people who would have benefited from the advice but predict the quality of their comment wrong (“yes I have a low average but this next post will be my big break”)
you’ll affect people who the advice wasn’t directed to / who wouldn’t have needed it but implemented it anyway (“it must be my low quality thoughts that are the problem, better leave all writing to Eliezer” or “looks like I have an average score of {TOO_LOW}, better not post this {ACTUALLY_INSIGHTFUL_INSIGHT}.”)
As you said yourself, it only takes a few seconds to upvote—so it does to downvote. I would expect the system to converge to lower quality comments being voted down more so the observed quality of the comment will roughly align with the votes on the comment.
If the system does not allow you to filter out comments that are below some downvote/upvote ratio, then perhaps the system needs to be tweaked to allow this kind of filtering—but the solution is not telling users to post less. (And if someone is serially creating low quality content, this person can usually be approached individually, rather than trying to optimize for the public forum.)
This seems worthwhile to care about for two reasons:
There’s a real risk of driving away people who would have something valuable to contribute but who end up self censoring before they post anything (and, AIUI, the bar for LessWrong is already really high).
There’s a risk of people optimizing for karma ratio rather than quality of the comment they’re posting. The mental shortcut to take is “Now I have enough karma to make a low effort comment” which is probably not the effect you want.
(edit: improved the first paragraph to better articulate what I mean to say.)
Don’t fully disagree, but still inclined to not view non-upvoted-but-neither-downvoted things too harshly:
If I’m no exception, not upvote may often mean: ‘Still enjoyed the quick thought-stimulation even if it seems ultimately not a particularly pertinent message’. One can always down vote, if one really feels its warranted.
Also: If one erratically reads LW, and hence comments on old posts: recipe for fewer upvotes afaik. So one’d have to adjust for this quite strongly.
This ratio is indeed something moderators pay attention to and influences decision about e.g. rate limits, though it’s not directly triggering rate limits if I recall currently.
Are all users’ post/comment counts and karma available in bulk? I’d be curious what the distribution looks like. I’m also curious what my ratio is for recent comments—the fact that it’s 2.3 over more than a decade doesn’t tell me much.
I often have to argue at work against the tyranny of metrics (I more often have to argue against the ignorance of not enough metrics, but that’s not important to this point). This is a classic example of something people should probably look at and consider often, but it’s highly context- and individual-dependent whether a change is warranted.
Karma isn’t consistent nor rigorous enough to set targets for.
edit: I’m actually more curious about variance in the supply side. I vary pretty widely in how much I vote based on topic and mood, not on any unidimensional quality estimate. Showing use users their recent vote/read ratio could encourage voting.
[Continuing to sound elitist,] I have a related gripe/hot take that comments give people too much karma. I feel like I often see people who are “noisy” in that they comment a lot and have a lot of karma from that,[1] but have few or no valuable posts, and who I also don’t have a memory of reading valuable comments from. It makes me feel incentivized to acquire more of a habit of using LW as a social media feed, rather than just commenting when a thought I have passes my personal bar of feeling useful.
Note that self-karma contributes to a comments position within the sorting, but doesn’t contribute to the karma count on your account, so you can’t get a bunch of karma just by leaving a bunch of comments that no one upvotes. So these people are getting at least a consolation prize upvote from others.
Right at the top of that range, I agree, I probably comment too much for the amount I have to contribute. talking to people is fun though. a flipside to this: if you agree with Thomas, consider downvoting stuff that you didn’t find to be worth the time, to offset participation-trophy level upvotes.
I don’t know how to say this without sounding elitist, but my guess is that people who prolifically write LW comments and whose karma:(comments+posts) ratio is less than around 1.5:1 or maybe even 2:1 should be more selective in what they say. Around this range, it would be unwarranted for mods to rate-limit you, but perhaps the bottom 30% of content you produce is net negative considering the opportunity cost to readers.
Of course, one should not Goodhart for karma and Eliezer is not especially virtuous by having a 16:1 ratio, but 1.5:1 is a quite low ratio. You get 1.5 karma if 0.75 high-karma users or 1.5 low-karma users weakly upvote your comment. Considering that it takes barely a second to upvote and comments get tens of views, this is not a high bar for the average comment.
Caveats:
If your absolute number of comments + posts is low, the risk of making new users participate less outweighs the risk of comments clogging up the frontpage.
In some contexts votes might underestimate quality, like if your comment is on some esoteric topic that few people engage with.
Another circumstance where votes underestimate quality is where you often get caught up in long reply-chains (on posts that aren’t “the post of the week”), which seems like a very beneficial use of LessWrong, but typically has much lower readership & upvote rates, but uses a lot of comments.
Aiming to improve the quality of threads by telling users to post/comment less in a public forum seems to me like it’s not going to be very helpful in the long term. If users self-select by simulating how much karma their post/comment gets (perhaps via extrapolating the average karma ratio of their previous posts) it might work for a subset, but will fail for other cases:
Namely:
you’ll miss people who would have benefited from the advice but predict the quality of their comment wrong (“yes I have a low average but this next post will be my big break”)
you’ll affect people who the advice wasn’t directed to / who wouldn’t have needed it but implemented it anyway (“it must be my low quality thoughts that are the problem, better leave all writing to Eliezer” or “looks like I have an average score of {TOO_LOW}, better not post this {ACTUALLY_INSIGHTFUL_INSIGHT}.”)
As you said yourself, it only takes a few seconds to upvote—so it does to downvote. I would expect the system to converge to lower quality comments being voted down more so the observed quality of the comment will roughly align with the votes on the comment.
If the system does not allow you to filter out comments that are below some downvote/upvote ratio, then perhaps the system needs to be tweaked to allow this kind of filtering—but the solution is not telling users to post less. (And if someone is serially creating low quality content, this person can usually be approached individually, rather than trying to optimize for the public forum.)
This seems worthwhile to care about for two reasons:
There’s a real risk of driving away people who would have something valuable to contribute but who end up self censoring before they post anything (and, AIUI, the bar for LessWrong is already really high).
There’s a risk of people optimizing for karma ratio rather than quality of the comment they’re posting. The mental shortcut to take is “Now I have enough karma to make a low effort comment” which is probably not the effect you want.
(edit: improved the first paragraph to better articulate what I mean to say.)
Don’t fully disagree, but still inclined to not view non-upvoted-but-neither-downvoted things too harshly:
If I’m no exception, not upvote may often mean: ‘Still enjoyed the quick thought-stimulation even if it seems ultimately not a particularly pertinent message’. One can always down vote, if one really feels its warranted.
Also: If one erratically reads LW, and hence comments on old posts: recipe for fewer upvotes afaik. So one’d have to adjust for this quite strongly.
This ratio is indeed something moderators pay attention to and influences decision about e.g. rate limits, though it’s not directly triggering rate limits if I recall currently.
*Quickly checks my ratio*
“Phew, I’ve survived the Kwa Purge”
Are all users’ post/comment counts and karma available in bulk? I’d be curious what the distribution looks like. I’m also curious what my ratio is for recent comments—the fact that it’s 2.3 over more than a decade doesn’t tell me much.
I often have to argue at work against the tyranny of metrics (I more often have to argue against the ignorance of not enough metrics, but that’s not important to this point). This is a classic example of something people should probably look at and consider often, but it’s highly context- and individual-dependent whether a change is warranted.
Karma isn’t consistent nor rigorous enough to set targets for.
edit: I’m actually more curious about variance in the supply side. I vary pretty widely in how much I vote based on topic and mood, not on any unidimensional quality estimate. Showing use users their recent vote/read ratio could encourage voting.
[Continuing to sound elitist,] I have a related gripe/hot take that comments give people too much karma. I feel like I often see people who are “noisy” in that they comment a lot and have a lot of karma from that,[1] but have few or no valuable posts, and who I also don’t have a memory of reading valuable comments from. It makes me feel incentivized to acquire more of a habit of using LW as a social media feed, rather than just commenting when a thought I have passes my personal bar of feeling useful.
Note that self-karma contributes to a comments position within the sorting, but doesn’t contribute to the karma count on your account, so you can’t get a bunch of karma just by leaving a bunch of comments that no one upvotes. So these people are getting at least a consolation prize upvote from others.
Right at the top of that range, I agree, I probably comment too much for the amount I have to contribute. talking to people is fun though. a flipside to this: if you agree with Thomas, consider downvoting stuff that you didn’t find to be worth the time, to offset participation-trophy level upvotes.