This question already has an answer here:

Sometimes an accepted answer is actually incorrect.

In this case, it is better to have a functionality to let any user to submit the question to the board to be reviewed by users with higher reputation with expertise on the topic and then rejudge it.

This is extremely important since some "correct" answers are not correct at all.

Thanks. Just throw some thinking.

marked as duplicate by Servy, Cody Gray, Doorknob, BinaryMisfit, LittleBobbyTables Aug 21 '13 at 15:28

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • 10
    If an answer is incorrect, downvote and comment. – Time Traveling Bobby Aug 21 '13 at 14:45
  • @M.NightDemonbobby But the questioner doesn't necessarily will review it. And if the upvotes are more than 1, let's say 5, one downvote still make it 4, which still looks like a correct answer. – zs2020 Aug 21 '13 at 14:46
  • 2
    For an unrelated question, I was just reading Information Security's Help Center, and noticed something strange that is related to this question: "Mistakes - Sometimes an answer may be voted up highly despite being wrong. Downvote and comment as usual, but if it doesn't get resolved through voting, you can flag it." What an unusual policy. Do the moderators there really adjudicate technical accuracy? – Cody Gray Aug 21 '13 at 14:59
  • @CodyGray That should probably be fixed, I'd think, but that's an issue for that site's meta... – Servy Aug 21 '13 at 15:00
  • I'm not convinced it's a mistake, so I don't know if it needs to be "fixed". This is something the moderators edited into the standard FAQ. But yeah, it might be something to bring up on their Meta. I'm just hijacking the comments on this oft-asked duplicate because it is tangentially related and on my mind. – Cody Gray Aug 21 '13 at 15:01
  • 1
    For those who downvoted without giving a comment, you guys are not constructive at all. Shame on you. – zs2020 Aug 21 '13 at 15:21
  • 2
    @sza And for those that downvoted and did give a comment, or upvoted an existing comment, or just didn't bother to comment because they saw an existing comment? – Servy Aug 21 '13 at 15:27
  • @sza you are aware votes signify agreement here, not necessarily assessment on quality (at least they should be) – Cole Johnson Aug 21 '13 at 15:27
  • @ColeJohnson Suggesting a change that is very frequently suggested in similar forms tends to result in lots of downvotes. While many likely do disagree with the proposal, there's also a pretty noticeable lack of research here. – Servy Aug 21 '13 at 15:29
  • @ColeJohnson It's ironic, if a change that is very frequently suggested in similar forms should be considered as a new feature. – zs2020 Aug 21 '13 at 15:31

Take a look at the tag. People generally post code snippets, and not their whole project (Android project tend to have many files). Only the original poster could adapt the code from the answers. No one else can. OP should test the code before accepting the answer. If there are multiple answers, upvote the correct one. Downvote the wrong answer. There isn't more you can do.

It is however OP's job to only accept the answer when it's tested. I think it is a bad idea to only let high reputation users accept the answer, because:

  • They will probably just approve it. They will trust OP has accepted the right answer.
  • It will result in large queues. We already have problems controlling the close votes queue.
  • The high reputation users don't know everything. If you throw a Haskell question at a C++ person; how would he know what the right answer is?

There are at this moment 5,552,433 question on Stackoverflow. 4,315,026 questions have answers with a score of 1 or more. Let's say 50% of these questions have an accepted answer; 2,157,513 answers will have to be (arguably unnecessary) reviewed.

It is a waste of time, because almost all accepted answers will be approved. Trust OP, or downvote/comment and move on. Only if an answer is dangerously wrong you'll need to flag it.

  • I am not saying to let the high reputation users accept the answer, I was talking about for some question/answer to be reviewed by them only when people propose the answer as incorrect. For example, you may come across a marked correct answer but you don't think it is correct, and the OP is not active at all, you then can submit the request about reviewing it to the board. Does it make sense? – zs2020 Aug 21 '13 at 15:19
  • Speaking of the close votes queue, it's over 70,700... – Cole Johnson Aug 21 '13 at 15:23
  • @sza In that case it is better to contact OP directly with a comment or a PM. Bullet-point 3 is still corect. High reputation users cannot adapt the code snippets, because they cannot access (other files/dependancies of) the project. Flag the answer when it is dangerously wroing, otherwise downvote/comment and move on. I believe it has to be this way. – FalconC Aug 21 '13 at 15:23
  • 2
    @FalconC we have PM in SO? Pretty sure we don't. – gitsitgo Aug 21 '13 at 15:25
  • @FalconC Of course it should be reviewed by the high reputation users with expertise on it... Isn't that obvious? – zs2020 Aug 21 '13 at 15:26
  • @gitsitgo You can email users when they have entered their email address. I have given it the wrong name, sorry. – FalconC Aug 21 '13 at 15:27
  • @sza You mean users with a gold badge in a programming languages? That is possible for a limited amount of questions, where theory is more important than the code. If you prefer, you can ask OP how the accepted answer worked out. Also, it is possible that users will flag accepted answers just because their answer was not accepted. – FalconC Aug 21 '13 at 15:32

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