I remain of the opinion that posts made in good faith should not be voted below one karma without a pretty good reason. The original Doing EA Better was just too massive to facilitate discussion of specifics, and splitting it up to facilitate more specific discussion seems reasonable. I do not see a good reason for this to have negative karma.
While I haven’t voted either way on this post, I think it is one of the least well done portions of the original larger post. Ozy’s response, quoted in the comment above, shows how this post is harmful on its own terms.
I recall at least one, possibly both, of the other segments being in the negative at certain points in time before settling to weakly positive karma. My memory could be wrong, but that suggests that the early negative vote isn’t primarily a function of this particular segment being problematic.
A charitable interpretation would be that it’s a symptom of there being no separate way to ‘upvote and disagreevote’ something that a Forum user thinks is important but still disagree with. −5 from 17 (at time of writing) does seem unbalanced though, especially given the original DEAB post was highly upvoted, and one of the most common suggestions was for the authors to break it up into smaller chunks
I think the unfortunate absence of disagreevote on posts is a good bit of what we are seeing. Given the single vote type, I’m more okay with downvote-to-disagreevote on posts that have a decent amount of karma. But negging a substantial post sends an implied message that the content was inappropriate or unwelcome, which comes across as somewhat unfriendly at best.
(For comments, negging serves a more useful purpose in rank ordering comments, and getting a light neg on a five-line comment just doesn’t have the same sting as on a substantial post.)
Personally, I don’t think it’s a problem if a substantial post has negative karma. Someone could read a post, agree that it was well-written and detailed, but still think it was bad and that they’d want to see fewer posts like it. A downvote seems like the right response there.
Overall, I think there’s a tendency for people to upvote things more often when they are very long, and that this is one factor that pushes the average Forum post to be too long. This makes me especially wary about norms like “negging a substantial post is somewhat unfriendly”.
That said, I’d be quite happy to see disagree-voting added to posts, since many people would find it useful (including me!).
Agree on not upvoting for length; I meant “substantial” to exclude shower thoughts and similar material that should clearly should be a shortform.
I think a net karma of (say) +5 conveys pretty effectively that the community doesn’t think much of the post and “want[s] to see fewer posts like” it. The difference between that and a net karma of −5 is that the latter comes across as a sanction. Although people ideally wouldn’t take karma on their posts personally, people are also human and are prone to do so. And it’s well-documented that people tend to view taking away something they had (say, $100) as significantly more negative than acquiring the same thing.
So, if posts were movies, ending up with +5 (where the median post gets much more) is somewhat like bombing at the box office and being panned by the critics. Everyone gets the message loud and clear. Getting −5 is inching closer to receiving a Razzie. That feels like potential overdeterrence to me unless there are strong reasons for the award.
Also, a downvote is a blunt instrument, and it’s worth thinking about the message that assigning a net negative karma to a reasonably well-written, medium-or-higher effort post sends to newer and casual users. I fear that message will often come across as “Better not write something the majority disagrees with; the same thing might happen to you.” Any karma system almost inevitably incentivizes widely-acceptable bromides already, and I think assigning net negative karma to reasonably high-effort posts absent strong reason risks intensifying that tendency. (To be clear, I have standard downvoted at least two posts already in the negative this week, one for a click-baity title and one for excessive promotion of a for-profit enterprise, so I am not suggesting strong reasons do not occur.)
In the end, I don’t think any marginal increase in signalling that is achieved by net negative (vs. very low positive) karma is worth the downsides of net negative karma in most cases of good-faith, rule-compliant, reasonable-effort posts.
I agree that +5 and −5 will feel more different to most people than +5 and +15.
I think this reflects a common dilemma with karma systems, which is that people tend to use them in one of two ways:
Voting based on how they feel about content, without regard for its current karma
Voting so that they bring content closer to the karma score they think it should have
There are many cases where I’ve seen a comment at, say, −10, and I’ve had the thought “I dislike this comment, but −10 seems too harsh”, and I’ve had to choose whether to upvote or downvote (or leave it alone).
My behavior in those cases isn’t consistent — it depends on the context, my mood, etc.
I expect that method (2) leads to fewer pile-ons and reduces echo chamber effects. But it also creates a weird dynamic where people are upvoting things they think are bad and vice-versa to make a more complicated point (what would Aaron Hamlin say?).
If someone were deciding how to vote on my post, I think I’d want them to just express their feelings regardless of what other people had done, because that result would feel more “true” to me and give me more information about what readers actually thought.
I’m not sure there is a right answer in the end, and I’m definitely not confident enough to try to push people in one direction or the other (to the point of calling it “unfriendly” to downvote posts below zero, or, say, “dishonest” to vote against one’s feelings).
I remain of the opinion that posts made in good faith should not be voted below one karma without a pretty good reason. The original Doing EA Better was just too massive to facilitate discussion of specifics, and splitting it up to facilitate more specific discussion seems reasonable. I do not see a good reason for this to have negative karma.
While I haven’t voted either way on this post, I think it is one of the least well done portions of the original larger post. Ozy’s response, quoted in the comment above, shows how this post is harmful on its own terms.
I recall at least one, possibly both, of the other segments being in the negative at certain points in time before settling to weakly positive karma. My memory could be wrong, but that suggests that the early negative vote isn’t primarily a function of this particular segment being problematic.
A charitable interpretation would be that it’s a symptom of there being no separate way to ‘upvote and disagreevote’ something that a Forum user thinks is important but still disagree with. −5 from 17 (at time of writing) does seem unbalanced though, especially given the original DEAB post was highly upvoted, and one of the most common suggestions was for the authors to break it up into smaller chunks
I think the unfortunate absence of disagreevote on posts is a good bit of what we are seeing. Given the single vote type, I’m more okay with downvote-to-disagreevote on posts that have a decent amount of karma. But negging a substantial post sends an implied message that the content was inappropriate or unwelcome, which comes across as somewhat unfriendly at best.
(For comments, negging serves a more useful purpose in rank ordering comments, and getting a light neg on a five-line comment just doesn’t have the same sting as on a substantial post.)
Personally, I don’t think it’s a problem if a substantial post has negative karma. Someone could read a post, agree that it was well-written and detailed, but still think it was bad and that they’d want to see fewer posts like it. A downvote seems like the right response there.
Overall, I think there’s a tendency for people to upvote things more often when they are very long, and that this is one factor that pushes the average Forum post to be too long. This makes me especially wary about norms like “negging a substantial post is somewhat unfriendly”.
That said, I’d be quite happy to see disagree-voting added to posts, since many people would find it useful (including me!).
Agree on not upvoting for length; I meant “substantial” to exclude shower thoughts and similar material that should clearly should be a shortform.
I think a net karma of (say) +5 conveys pretty effectively that the community doesn’t think much of the post and “want[s] to see fewer posts like” it. The difference between that and a net karma of −5 is that the latter comes across as a sanction. Although people ideally wouldn’t take karma on their posts personally, people are also human and are prone to do so. And it’s well-documented that people tend to view taking away something they had (say, $100) as significantly more negative than acquiring the same thing.
So, if posts were movies, ending up with +5 (where the median post gets much more) is somewhat like bombing at the box office and being panned by the critics. Everyone gets the message loud and clear. Getting −5 is inching closer to receiving a Razzie. That feels like potential overdeterrence to me unless there are strong reasons for the award.
Also, a downvote is a blunt instrument, and it’s worth thinking about the message that assigning a net negative karma to a reasonably well-written, medium-or-higher effort post sends to newer and casual users. I fear that message will often come across as “Better not write something the majority disagrees with; the same thing might happen to you.” Any karma system almost inevitably incentivizes widely-acceptable bromides already, and I think assigning net negative karma to reasonably high-effort posts absent strong reason risks intensifying that tendency. (To be clear, I have standard downvoted at least two posts already in the negative this week, one for a click-baity title and one for excessive promotion of a for-profit enterprise, so I am not suggesting strong reasons do not occur.)
In the end, I don’t think any marginal increase in signalling that is achieved by net negative (vs. very low positive) karma is worth the downsides of net negative karma in most cases of good-faith, rule-compliant, reasonable-effort posts.
Thanks for continuing to engage so thoughtfully!
I agree that +5 and −5 will feel more different to most people than +5 and +15.
I think this reflects a common dilemma with karma systems, which is that people tend to use them in one of two ways:
Voting based on how they feel about content, without regard for its current karma
Voting so that they bring content closer to the karma score they think it should have
There are many cases where I’ve seen a comment at, say, −10, and I’ve had the thought “I dislike this comment, but −10 seems too harsh”, and I’ve had to choose whether to upvote or downvote (or leave it alone).
My behavior in those cases isn’t consistent — it depends on the context, my mood, etc.
I expect that method (2) leads to fewer pile-ons and reduces echo chamber effects. But it also creates a weird dynamic where people are upvoting things they think are bad and vice-versa to make a more complicated point (what would Aaron Hamlin say?).
If someone were deciding how to vote on my post, I think I’d want them to just express their feelings regardless of what other people had done, because that result would feel more “true” to me and give me more information about what readers actually thought.
I’m not sure there is a right answer in the end, and I’m definitely not confident enough to try to push people in one direction or the other (to the point of calling it “unfriendly” to downvote posts below zero, or, say, “dishonest” to vote against one’s feelings).
I upvoted this post due to this comment. I don’t see a good reason for this to have negative karma either.