Fwiw I don’t find this very convincing. If a comment gets highly up voted for instance, it has the consequence of getting more dumb replies, but mostly people just accept that rather than having to caveat their comments and asking not to up vote.
If a comment gets highly up voted for instance, it has the consequence of getting more dumb replies
Also has the consequence of getting more smart replies, simply as a result of getting more replies overall due to being more popular.
If writing good comments resulted in stupid comments in response (a drawback) with no corresponding benefit, then people would write fewer good comments. The reason we don’t observe the latter to be happening isn’t because stupid responses aren’t a disincentive, but because good responses (alongside stuff like increased karma, reputation, the intrinsic desire to communicate deeply-held beliefs, the feeling that Someone is Wrong on the Internet, etc.) are a sufficiently good incentive that overcomes this disincentive.
People don’t ask not to upvote because receiving upvotes is obviously a desirable thing for the comment-writer.[1] To the extent your confusing comment tries to imply otherwise, it appears entirely out of tune with reality.
A comment getting up voted has highs status implications, and gets their ideas seen more. I think that’s the main desirable thing of high up votes, more discussion is really hit or miss in terms of desire ability.
Which is the point of my comment, there’s tons of externalities to the system of up voting and down voting that we just put up with because the system basically works.
And when you change the system by imposing an additional constraint/disincentive on receiving karma, without a corresponding benefit (unlike in the case of bad responses, which come attached with good responses), the balance of externalities changes. Therefore, so does commenters’ behavior.
Moreover, I disagree with the notion that more discussion is neutral on average/in expectancy. That’s only if the original comment writer[1] can’t easily distinguish between productive and unproductive responses or can’t bring themselves emotionally to ignore the latter and focus only on the former. If the person can do that, then they obtain positive value from the responses overall (they get to learn something new or have their misconceptions corrected, for example), even if the majority of them are useless.
In fact, I just did a quick-and-dirty check. I went to my comments feed on GreaterWrong, sorting by “Top” (i.e., in descending karma value order). I then opened up the first 10 comments in the list. Then, I set the “offset” URL parameter to 1000 (i.e., went to the 51st page of the list, showing my 1001st through 1020th highest-rated comments), and opened up the first 10 comments on that page. For each comment, I counted the number of direct replies which I would consider to be dumb (not wrong, but dumb). The counts:
Total number of dumb replies to my 1st through 10th highest-upvoted comments: 2
Total number of dumb replies to my 1001st through 1010th highest-upvoted comments: 1
Given the sample sizes, I’m going to mark this one down as “no difference detected”.
Fwiw I don’t find this very convincing. If a comment gets highly up voted for instance, it has the consequence of getting more dumb replies, but mostly people just accept that rather than having to caveat their comments and asking not to up vote.
Also has the consequence of getting more smart replies, simply as a result of getting more replies overall due to being more popular.
If writing good comments resulted in stupid comments in response (a drawback) with no corresponding benefit, then people would write fewer good comments. The reason we don’t observe the latter to be happening isn’t because stupid responses aren’t a disincentive, but because good responses (alongside stuff like increased karma, reputation, the intrinsic desire to communicate deeply-held beliefs, the feeling that Someone is Wrong on the Internet, etc.) are a sufficiently good incentive that overcomes this disincentive.
People don’t ask not to upvote because receiving upvotes is obviously a desirable thing for the comment-writer.[1] To the extent your confusing comment tries to imply otherwise, it appears entirely out of tune with reality.
Unless you impose a cost big enough to render this claim false, which would be the case if something-like-kave’s-proposal gets implemented
A comment getting up voted has highs status implications, and gets their ideas seen more. I think that’s the main desirable thing of high up votes, more discussion is really hit or miss in terms of desire ability.
Which is the point of my comment, there’s tons of externalities to the system of up voting and down voting that we just put up with because the system basically works.
And when you change the system by imposing an additional constraint/disincentive on receiving karma, without a corresponding benefit (unlike in the case of bad responses, which come attached with good responses), the balance of externalities changes. Therefore, so does commenters’ behavior.
Moreover, I disagree with the notion that more discussion is neutral on average/in expectancy. That’s only if the original comment writer[1] can’t easily distinguish between productive and unproductive responses or can’t bring themselves emotionally to ignore the latter and focus only on the former. If the person can do that, then they obtain positive value from the responses overall (they get to learn something new or have their misconceptions corrected, for example), even if the majority of them are useless.
Or the audience!
I have never found this to be the case.
In fact, I just did a quick-and-dirty check. I went to my comments feed on GreaterWrong, sorting by “Top” (i.e., in descending karma value order). I then opened up the first 10 comments in the list. Then, I set the “offset” URL parameter to
1000(i.e., went to the 51st page of the list, showing my 1001st through 1020th highest-rated comments), and opened up the first 10 comments on that page. For each comment, I counted the number of direct replies which I would consider to be dumb (not wrong, but dumb). The counts:Total number of dumb replies to my 1st through 10th highest-upvoted comments: 2
Total number of dumb replies to my 1001st through 1010th highest-upvoted comments: 1
Given the sample sizes, I’m going to mark this one down as “no difference detected”.