I know there's been pretty substantial discussion regarding Moderators being able to accept unaccepted answers, but the obvious issue with that is Moderator workload, purpose, and expertise. For an example just check out Moderators accepting answers on user's behalf after a certain time period

I'd propose that another direction be taken, after a period of say 6 months, the question is then opened up to users with a particular reputation level (you pick), and that after a certain number of these super-users vote to accept a particular answer, that the question be closed off as accepted.

This would address the workload issue by distributing it, it would address the expertise issue but requiring essentially peer review, and it would attack the growing problem of unanswered questions.

I believe an incentive could be awarded to those super-users for being part of an accepted answer to encourage the gradual cleanup of unanswered questions, maybe even a new badge of some sort.


EDIT: As I've mentioned in response to comments below, sometimes, the most upvoted answer isn't the best practice, as a user, I'd like to know what the pro's do more than what the masses do, and so in limiting this particular ability to those with higher rep would tend to lean in favour of a best-practice rather than popularity.

EDIT 2: I see that some have already decided that this is a duplicate simply because the question was asked, but my contention is that none of the others have proposed viable methods, I think this does, and it does it differently.

EDIT 3: The question of what if the OP still wants it open came up below, I would propose that the OP then have the ability to reverse the community selected answer, and lock it from further community closing's.

EDIT 4: Based on the discussion below, it's clear that this isn't just a repeat of past questions and that there are some good ideas here, I BEG those who marked this question as a duplicate to read on and reconsider so that we can continue to have a conversation on the merits, or at the very least point out why you don't believe it deserves to be heard out. I don't believe you gave it a fair shake.

Final Edit: Long live the popularity contest!

  • 6
    We already have it. It's called voting.
    – Antony
    Dec 17, 2013 at 16:14
  • @Antony Voting doesn't close a question, while it may help filter out the garbage, it doesn't help speed up my ability to find the best practice, and not necessarily the popular one.
    – oucil
    Dec 17, 2013 at 16:16
  • @Bart with respect, he doesn't propose any method for doing so, I've suggested a number of qualifications and benefits of the approach.
    – oucil
    Dec 17, 2013 at 16:17
  • @oucil The principle is so fundamental that I hardly think it matters. This has been proposed in various ways, using various metrics or rep level or whathaveyou, but it's one of those no-go-s. Acceptance is up to the OP. That's it.
    – Bart
    Dec 17, 2013 at 16:19
  • @oucil accepting an answer also doesn't close a question Dec 19, 2013 at 23:09

2 Answers 2


Only the OP knows whether the question is actually answered. Here is an older question of mine with many good answers, none of which solve my problem. The question needs to be open and is still looking for a correct answer. I would have to just re-ask it if one of the answers that doesn't solve the problem were to be accepted.

  • That's another reason why a threshold would be needed be flagged as answered. If there are no good answers, then the super-users wouldn't select one.
    – oucil
    Dec 17, 2013 at 16:19
  • @oucil you can set the threshold as high as you want. That still doesn't mean it helped the OP.
    – djechlin
    Dec 17, 2013 at 16:22
  • Then limit the ability to questions with a certain number of responses. If a substantial portion of the community believes it's been answered, I contend 99% of the time the OP is just lazy. In your case, if it was closed on you, you should have the ability to re-open and lock from further community closes.
    – oucil
    Dec 17, 2013 at 16:28
  • @oucil too complicated, too little benefit. Nothing wrong with a question without accepted answers. We used to show user acceptance rate on their question, so answerers were at least warned they would be less likely to have an answer accepted.
    – djechlin
    Dec 17, 2013 at 16:30
  • @oucil furthermore upvotes just don't do it. Accepting the answer with the most upvotes automatically is just what the community thinks "would probably" solve the OP's problem. In the case of "lazy" askers, there's as good of a chance they left out a key detail. Their involvement is crucial to determine an accepted answer.
    – djechlin
    Dec 17, 2013 at 16:31
  • I think a best practice according to the pros would save me time sorting through the thousands and thousands of unselected answers out there, I don't think that's invaluable, but to each their own right ;)
    – oucil
    Dec 17, 2013 at 16:32
  • @oucil why not just look at the ones with the most upvotes, instead of the ones with most upvotes and some other designation saying it had the most upvotes?
    – djechlin
    Dec 17, 2013 at 16:33
  • @oucil also answers saying "this would probably help" or "you should try this" or in some cases (e.g. a security violation) "DON'T do it that way" are highly upvoted, for better or for worse. But they are not accepted. You can't say this would be more helpful when it's a strictly lossy operation.
    – djechlin
    Dec 17, 2013 at 16:34
  • Because sometimes the best practice / answer was late to the game, this site for all it's intent is still very beholden to the first set of answers no matter how valid. Sometimes the best answer, and or best practice doens't have the highest number of votes.
    – oucil
    Dec 17, 2013 at 16:35
  • @oucil oic, I understand your proposal better now. This is less crazy. Isn't this just equivalent to giving "superusers" mega-votes worth something like 10 votes each?
    – djechlin
    Dec 17, 2013 at 16:37
  • @oucil so that violates an interesting principle of SO. More rep gives you moderation power but the Q/A platform is kept as democratic as possible. (I say "as possible" because if I have two answers, one from a new user and one from Jon Skeet saying roughly the same thing, I'm probably going to trust Skeet more and accept his answer. C'est la vie.) It strikes me as very risky form a community standpoint to violate this rule. E.g. as a new user why upvote at all when the club of more experienced users just pick the best answer themselves?
    – djechlin
    Dec 17, 2013 at 16:39
  • I guess you could potentially do it that way, but it wouldn't necessarily help clean up the system of all the unanswered questions out there, the method of a "tipping point" answer based on a different set of upvotes not tied to the traditional votes would hit both birds with one stone.
    – oucil
    Dec 17, 2013 at 16:39
  • Choosing your own answer adds reputation, which allows you to do more, the incentive to accept your own is already there. Using a separate tally to "Community Accept" keeps the popularity contest out of the equation and out of the general public eye. I wouldn't even make it visible to anyone not capable of voting on it.
    – oucil
    Dec 17, 2013 at 16:44

Accepting an answer is up to the user and the user alone.

Indicating what the community thinks is correct it made through voting.

Both mechanisms work great and should not be changed.

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