10

I ran into this today, but reading a bit on meta, I notice that this is actually a common occurrence:

The problem

  • Mel posts a slightly incorrect or incomplete answer to a question
  • Victor posts another, slightly better answer
  • Victor downvotes Mel’s answer and comments on it explaining why it is incorrect/incomplete
  • Other people come along, see Victor’s comment first, take issue with the nitpicky attitude and/or the phrasing, and downvote Victor’s answer in retaliation
  • Some of them might also upvote Mel’s incorrect/incomplete answer out of sympathy

I think we all agree that this behaviour is not in line with StackOverflow’s goals. The ideal user should feel compelled to upvote/downvote solely based on the content of an answer, but many people don’t seem to get it.

The root cause

The two root causes of the problem are obvious:

  1. There is too much emotion connected with a downvote. A downvote is felt to be a “punishment” to the answerer; and some feel that an answer that is only slightly wrong does not deserve such “harsh” punishment. Therefore, they do the same thing back, this time actually intended as a punishment, and it is out of proportion because multiple people do it.
  2. People have the opportunity to retaliate because they can tell which answer is from the same person as the controversial comment.

The problem, continued

The incentive this generates is to stop commenting when you downvote. There is no point in trying to be helpful to the answerer if there is a greater risk of being punished for it. Clearly, this too is against the spirit of Stack Overflow; the idea is to give the answerer a chance to improve their answer, but this prevents it.

The proposed solution

Allow users to comment anonymously, but only once for each downvote, for the express purpose of justifying the downvote. This removes the opportunity to link the comment to another answer and thus prevents this form of retaliation.

8

First, the problem you've presented is just a story. You haven't provided any links to examples to show that it's an actual problem.

Second, if someone is being such a jerk in comments that it's causing multiple people to downvote them then the content of that comment is the real root cause, not the downvote. Allowing anonymous commenting is not a solution to this problem. It will only make things worse. It will allow Victor to be an even bigger jerk, and I for one do not want to see that. There's a good reason that commenting requires a small amount of reputation.

If an answer is only slightly wrong, then it probably doesn't deserve a downvote. A comment is usually enough to let the author know what needs fixing. If an answer is so bad that it can't be easily fixed, then I'll downvote. If it's close but not quite right, I'll leave a comment. I rarely do both at once.

  • 1
    ① Some comments are trying to be helpful, even when they sound criticising. Such comments do not necessarily come from a jerk. ② So according to you, if an answer is slightly wrong, it doesn’t deserve a downvote, but if it is not even slightly wrong and its author has been a jerk somewhere else, then it does? – Timwi Aug 13 '10 at 23:45
  • @Timwi: No, I didn't say anything about people being jerks deserving downvotes for it. – Bill the Lizard Aug 14 '10 at 2:58
  • The problem isn't just 'a story'. I noticed this phenomenon a lot. A few times this happened: I made a comment criticizing the content of someone's post, and then I notice that suddenly several of my answers seem to get downvotes. It's obvious that one person decided to 'punish' me. Also, I believe that some users dislike other users and routinely downvote posts by their 'target'. I got involved in a heated debate with someone in chat once - after that I noticed that my answers were often downvoted soon after posting, despite eventually receiving praise from others.Spite downvoting is real. – kandyman Dec 11 '18 at 11:34
  • @kandyman And that's different from the five point story presented in this post. – Bill the Lizard Dec 11 '18 at 13:25
  • @Bill the Lizard No it's not. The topic is about voting on posts based on reasons other than the content of the post, i.e. voting down for personal reasons like spite, dislike of a user, etc. Or voting up to 'encourage' another user. The point is that the vote system can easily be abused by anyone who takes a personal interest in using it for reasons other than evaluating content in answers. – kandyman Dec 11 '18 at 13:29
  • @kandyman And there's still no evidence that that is happening on any large scale. – Bill the Lizard Dec 11 '18 at 13:51
  • @Bill the Lizard. First, nobody said it was happening "on any large scale", but it doesn't have to be large scale to make it a problem that people are exercised about and something worthy of discussion. People seem to have strong feelings about the feature - that should tell you something. Second, there's equally no evidence that people like the downvote feature on a large scale - it seems you are simply assuming that people agree with the model, but I see no reason to conclude that either. It's not like they have a choice in the matter. – kandyman Dec 11 '18 at 15:07
  • @kandyman People use the downvote feature on a large scale. That's plenty of evidence. – Bill the Lizard Dec 11 '18 at 18:26
  • @BilltheLizard And how can you tell the ratio between appropriate use and inappropriate use? – kandyman Dec 11 '18 at 18:51
  • @kandyman I was a moderator at the time that I wrote this answer. Inappropriate use is a tiny fraction of downvotes (and most of those are reversed automatically). – Bill the Lizard Dec 11 '18 at 18:54
  • I doubt that one moderator can have an accurate perspective on how much abuse goes on and how much of it is reversed. But in any case, the point remains that it is easy to abuse the voting system and people seem to be dissatisfied about that. There are threads about this topic on almost every Stack Exchange I know. They need to do something about it. – kandyman Dec 11 '18 at 18:59
7

The incentive this generates is to stop commenting when you downvote.

STOP ASSOCIATING COMMENTS AND VOTES

Don't assume that my critical comment on your post means I down-voted it. Even if you were down-voted around the same time I left my comment. I may well just be pointing out what no one else bothered to.

Don't feel the need to comment whenever you down-vote. Comment when you have something to say, something to add - comments like, "-1 for the reasons N pointed out" are just noise, and read suspiciously like the author is trying to rub salt in a wound.

Oh, and when you do leave a comment after voting, don't put "+1" / "-1" in front of it, or otherwise reveal that you've voted. That stupidity just reinforces the idea that comments and voting go hand in hand. If you pull that crap and then get down-voted in revenge, you pretty much asked for it.

Finally: don't use comments to whine about the down-votes you're getting, or solicit comments from the people who down-voted you. Flag any comments you see like that.

  • 3
    Case in point - stackoverflow.com/questions/3467518/… - check "TheCloudlessSky"'s comment. – ChrisF Aug 13 '10 at 15:39
  • 2
    +1 because meta is sarcasm. – dmckee --- ex-moderator kitten Aug 13 '10 at 15:46
  • 1
    Firstly, even if I don’t explicitly state that I downvoted and even if you rant about this here, people will still associate the comment with a downvote. Secondly, even if they stopped associating it, your answer still doesn’t address the problem whereby people downvote an unrelated answer. – Timwi Aug 13 '10 at 23:47
  • 1
    @Timwi: down-votes are anonymous, and reasons for down-voting are personal. You aren't going to stop people from down-voting whatever they want for whatever reasons they want to. There are systems in place for limiting extreme abuse, but any attempt to force people to comment or shield commenters from revenge votes are likely to make things worse: see the discussions linked by Tobias, Peter, and ChrisF on your post above. And I know I can't stop people from making incorrect assumptions - my goal is to prevent behaviors and changes to the system that will validate these assumptions. – Shog9 Aug 14 '10 at 0:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .