4

In this question a factually incorrect answer has been accepted as the correct answer.

I've browsed some of the other questions, and generally the situation is either a wrong answer is popular (but not accepted), or an incorrect answer is accepted but is close enough to being correct that it can be fixed with some editing.

In this specific case, the question is about a specific number (2^32) and the correct answer is 16,777,216, it's really black and white.

Is the downvote/comment approach still the correct approach here? I mean, with SO serving as documentation these days, having answers like this accepted seems counter to the purpose of the site.

|improve this question|||||
  • 3
    Yep, downvote and comment. – Bart Mar 8 '13 at 15:32
  • The feature request to allow moderators to delete wrong answers has been soundly declined (and rightfully so). The same question links to a few discussions on the topic. – David Robinson Mar 8 '13 at 15:33
  • 1
    @DavidRobinson That is not exact: Moderators can delete an answer, even if it was accepted; they will never delete an accepted answer just because it is wrong. What it can be done for those answers is down-vote them, if they are totally wrong; that doesn't require a moderator. The feature request you linked is about allowing moderators to remove the tick shown for accepted answers, which is different from deleting an accepted answer. – kiamlaluno Mar 8 '13 at 16:26
12

Is the downvote/comment approach still the correct approach here?

Yes, that is the correct approach, in my opinion. You generally should not edit such an answer, unless the incorrect nature of the answer seems to be an error of typing or the like. In your case, it seems that the incorrect answer is exactly what was meant by the poster, so editing it to fix it would violate the principle of not changing the meaning of a post with editing.

A significant problem with editing a question's content for correctness is that it circumvents the voting system as a way to judge content. Yes; the voting system can be wrong. It's an imperfection, because people are involved, and the style and other human-oriented factors can affect it. But overall, it works moderately well as a way for a discerning reader to gauge posts.

Editing for correctness instead of voting/commenting basically lets a single person override the whole system, to a degree. It's that one person saying, "the community is wrong, and I am right." Maybe that's true; but there is no good way to judge that.

What if the editor, for example, is the one who is wrong? How do we judge that? Again; we already have a system in place for that.

I do not have any desire for "furthering incorrect information", but this is the Internet. Users with short attention spans are liable to accept lots of incorrect information. For example, what about a user who skips the "correct" answer because it seems like too much work, and instead goes for a really brittle, bad solution that they 'like' better?

Mostly, I think it all comes down to the principle that the voting system works pretty well on a broad level, and we need to be careful about individual users trying to override that, because there is no system to make sure they are right/wrong. Posting their own answer is the best way to do that, and have it judged. Perfect? not at all... but not so bad in aggregate.

|improve this answer|||||
  • What is the rationale for furthering incorrect information? Many users with short attention spans only look at the top answer and move on. Do we really want that? – Dan Dascalescu Feb 27 '14 at 2:24
  • @DanDascalescu I added some info to the answer above, responsive to your concerns. I started to comment, but it was too much, and I think it expounds on my opinion in general here. :) – Andrew Barber Feb 27 '14 at 14:32
4

In addition to downvoting/commenting incorrect answers, also consider some other options as well.

  1. If the OP of the question is still active on the site, considering pinging them via a comment on the question. It is possibly they might not see it, or care, but at least you made an attempt to see if they can change the accepted answer.
  2. Up-vote competing correct answers, or if one does not exist yet, try to leave one yourself.
|improve this answer|||||
  • That still won't help prevent visitors with short attention spans from running with the wrong answer after they land on SO via a Google search. Correct answers that are below the accepted or top voted one won't be read. – Dan Dascalescu Feb 27 '14 at 2:25
  • @DanDascalescu This question was about what should be done based on the current guidelines of the site. There's literally nothing else that can be done given the fact that the OP of the question has sole control of the accepted answer, and moderators will not delete answers based on technical merit. The only thing left to do is downvote and try to get the OP of the question to change the accepted mark. – psubsee2003 Feb 27 '14 at 9:19

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