It’s getting increasingly easy to farm karma on LessWrong by making almost substantive-less posts that just articulate a position most people agree with. Imo this is a really huge problem, karma matters, it shapes what people see and it sets incentives.
As your good deed for this week, find something to strong-downvote that you agree with. I’d guess most people do that far too infrequently, maybe never.
I call these “nod posts”. They’re posts that make you nod in agreement rather than prompting you to say something like “Oh, I didn’t realize that before but I’m glad that I do now!”
I think there is often a fair amount of value in reading nod posts though. They can deepen your understanding of a topic, remind you of something important, give you more frames and mental models to utilize, amongst other things. I agree that such posts are overrated though, in terms of the karma they get and, relatedly, how much priority they get in the feed.
Just registering my wish for more investigation (theory and practice) of pluralistic filtering / karma schemes. Something in the genre of eigenmorality or something. I don’t know how to just pick, or iterate towards, good + feasible voting norms or ideals / shared intentions. (Cf. https://tsvibt.blogspot.com/2025/11/forum-poweruser-forum.html )
My LW upvoting policy is that every once in a while I go through the big list of everything I’ve read, grep for LessWrong posts, look through the latest ~50 entries and decide to open them and (strong) up/downvote them based on how they look, a few months in retrospect.
Feel free to send complaints to the mods. I don’t super think that stuff has been getting worse on this dimension (when I look at top karma posts I really don’t see stuff that largely just validates what people believe), but I might be totally missing something (and I haven’t looked much at middling-karma posts).
If it’s a huge problem we can change the karma algorithm. Before then, the mods are very active users themselves and have huge vote strength which means we can shift total karma of posts quite a bit (if everyone on the LW mod team strong-votes on something that’s like 50+ karma in any direction). We also can take it into account for curation which sets the tone of the site quite a bit, and can use it to counteract some cultural trends. We can also change the UI of voting to make various things different (like doing things that makes it more likely that people vote on stuff retroactively after a few days or weeks to get more signal from voting).
I was reading it as a response to my thing, which Rafael complained about yesterday, probably among other things. (I still owe a response to that comment, I’ll get to it in the next few days hopefully.) I agree that that post of mine got more karma than it deserved, and certainly agree that karma is somewhat divorced from value on lesswrong (and also everywhere else on the internet). My current opinion is that I’m happy that I wrote and published it—good use of a few hours. Again, I’ll read that comment soon-ish, see if I change my mind ¯\_(ツ)_/¯
It’s definitely not a response to your post! I mean, yes I do think the post is an example of the pattern, but your posts in general are not, and it’s also not the most egregious example, the front page is not exactly filled with takes on this particular topic.
Im terms of being directly causally linked to me getting annoyed enough to write the short form, the other post I just complained about playes a much bigger role, and is a much more central example. But I’d still dispute that the shortform is a response to that. This is a complaint I’ve had for months and could have written at any point.
Don’t really want to name examples because it would single out individual authors. That’s why I didn’t include any in the first place.
I don’t think it is good to have your vote as a function of the current karma value, except in very special cases. One is negative karma quick takes, another one is with the impossible condition that most people in LW uses a distributed cooperation scheme that can make the karma value converge.
Otherwise, if everyone try to vote to make the post the “fair valuation” (which can be different per person):
It will just make you try to change your vote every so often
It makes algorithmic adjustments work less well. I would hope that LW eventually uses some impression-weighted karma value for the frontpage (unsure if currently being used), rather than everyone trying to use their own internal algorithm, which would make such adjustments inaccurate.
I mostly agree, with an exception for posts/comments that are already negative or close to zero.
I see a comment I think is useless/unpleasant/bad, and I think about downvoting it. I see its currently sitting at −10 and dont vote because I dont want to pile on even harder against that person. But, if the same comment had been at +20 i would have thought ‘what, really? Well, downvote’.
If I like something I dont let its current karma effect my voting.
It is, although it was unintentional karma farming, I had no expectations that this would go above, idk, 7.
The issue is (was?) that it’s difficult to make a more high-effort post without singling out users, which I don’t want. This is why I didn’t say anything like this until now. But this is now solved with habryka greenlighting a private channel. If I’ll say anything else about this, I’ll do it there.
fwiw I think most peopel karma farming are also doing it approximately as unintentionally as you did here. I agree with you there’s a problem here but, like, the problem lives in the distributed system.
I refrained from upvoting your comment despite agreeing with it.
Relatedly, I think the agreement vote button helps me upvote low-substance comments I agree with less. It’s a convenient outlet for the instinct to make my support known. Posts don’t have an agreement button though.
I think in my favorite world, we made it feel more like reacts were the way to signal support for low-substance comments, and Approval Vote was more of a deliberate choice.
I think there’s value in making the same argument in different ways, iterating and trying to find the best version of it. If someone does an especially good job arguing for a position that I’ve already seen argued before, I don’t mind upvoting it.
It’s getting increasingly easy to farm karma on LessWrong by making almost substantive-less posts that just articulate a position most people agree with. Imo this is a really huge problem, karma matters, it shapes what people see and it sets incentives.
As your good deed for this week, find something to strong-downvote that you agree with. I’d guess most people do that far too infrequently, maybe never.
How do you know it’s getting increasingly easy?
Qualitative/intuitive impression, I didn’t apply any formalized metric.
I call these “nod posts”. They’re posts that make you nod in agreement rather than prompting you to say something like “Oh, I didn’t realize that before but I’m glad that I do now!”
I think there is often a fair amount of value in reading nod posts though. They can deepen your understanding of a topic, remind you of something important, give you more frames and mental models to utilize, amongst other things. I agree that such posts are overrated though, in terms of the karma they get and, relatedly, how much priority they get in the feed.
Just registering my wish for more investigation (theory and practice) of pluralistic filtering / karma schemes. Something in the genre of eigenmorality or something. I don’t know how to just pick, or iterate towards, good + feasible voting norms or ideals / shared intentions. (Cf. https://tsvibt.blogspot.com/2025/11/forum-poweruser-forum.html )
I like niplav’s upvoting policy:
Feel free to send complaints to the mods. I don’t super think that stuff has been getting worse on this dimension (when I look at top karma posts I really don’t see stuff that largely just validates what people believe), but I might be totally missing something (and I haven’t looked much at middling-karma posts).
Is this a problem that mods can solve? I mean if I 100% convinced you, what would you do about it?
If it’s a huge problem we can change the karma algorithm. Before then, the mods are very active users themselves and have huge vote strength which means we can shift total karma of posts quite a bit (if everyone on the LW mod team strong-votes on something that’s like 50+ karma in any direction). We also can take it into account for curation which sets the tone of the site quite a bit, and can use it to counteract some cultural trends. We can also change the UI of voting to make various things different (like doing things that makes it more likely that people vote on stuff retroactively after a few days or weeks to get more signal from voting).
FYI I think this is fairly obviously a problem, or at least, a real distortion (but, I don’t have a strong sense that it’s easy to do anything about).
Gotcha, I’ll at least consider writing a more comprehensive argument. Don’t mind including examples in a private context.
Not sure I disagree, but could you give some example?
I was reading it as a response to my thing, which Rafael complained about yesterday, probably among other things. (I still owe a response to that comment, I’ll get to it in the next few days hopefully.) I agree that that post of mine got more karma than it deserved, and certainly agree that karma is somewhat divorced from value on lesswrong (and also everywhere else on the internet). My current opinion is that I’m happy that I wrote and published it—good use of a few hours. Again, I’ll read that comment soon-ish, see if I change my mind ¯\_(ツ)_/¯
It’s definitely not a response to your post! I mean, yes I do think the post is an example of the pattern, but your posts in general are not, and it’s also not the most egregious example, the front page is not exactly filled with takes on this particular topic.
Im terms of being directly causally linked to me getting annoyed enough to write the short form, the other post I just complained about playes a much bigger role, and is a much more central example. But I’d still dispute that the shortform is a response to that. This is a complaint I’ve had for months and could have written at any point.
Don’t really want to name examples because it would single out individual authors. That’s why I didn’t include any in the first place.
Lol. I liked your post, but did not upvote it. I think a fair valuation of that post is 35 karma.
I’ve also observed a trend where putting in a lot of effort tends to get less engagement than low effort posts.
But still don’t feel like that’s a recent trend.
I don’t think it is good to have your vote as a function of the current karma value, except in very special cases. One is negative karma quick takes, another one is with the impossible condition that most people in LW uses a distributed cooperation scheme that can make the karma value converge.
Otherwise, if everyone try to vote to make the post the “fair valuation” (which can be different per person):
It will just make you try to change your vote every so often
It makes algorithmic adjustments work less well. I would hope that LW eventually uses some impression-weighted karma value for the frontpage (unsure if currently being used), rather than everyone trying to use their own internal algorithm, which would make such adjustments inaccurate.
I mostly agree, with an exception for posts/comments that are already negative or close to zero.
I see a comment I think is useless/unpleasant/bad, and I think about downvoting it. I see its currently sitting at −10 and dont vote because I dont want to pile on even harder against that person. But, if the same comment had been at +20 i would have thought ‘what, really? Well, downvote’.
If I like something I dont let its current karma effect my voting.
Fwiw I made a post/question/poll about this once (like whether you should vote based on total karma).
This post is perhaps ironically an example of itself. :P
It is, although it was unintentional karma farming, I had no expectations that this would go above, idk, 7.
The issue is (was?) that it’s difficult to make a more high-effort post without singling out users, which I don’t want. This is why I didn’t say anything like this until now. But this is now solved with habryka greenlighting a private channel. If I’ll say anything else about this, I’ll do it there.
fwiw I think most peopel karma farming are also doing it approximately as unintentionally as you did here. I agree with you there’s a problem here but, like, the problem lives in the distributed system.
I refrained from upvoting your comment despite agreeing with it.
Relatedly, I think the agreement vote button helps me upvote low-substance comments I agree with less. It’s a convenient outlet for the instinct to make my support known. Posts don’t have an agreement button though.
I think in my favorite world, we made it feel more like reacts were the way to signal support for low-substance comments, and Approval Vote was more of a deliberate choice.
I don’t understand how reacts can be used to signal support for low-substance comments. Like using thumbs up instead of upvote?
Related to the thread, I also want posts to have agreement karma.
I think there’s value in making the same argument in different ways, iterating and trying to find the best version of it. If someone does an especially good job arguing for a position that I’ve already seen argued before, I don’t mind upvoting it.