Probably one of the largest problems I regularly see with the Stack Exchange model is that most people asking questions don't really understand how the system works. Often questions never get an answer marked as being the answer, which isn't a big deal because the highest voted answer still gives meaningful feedback to anyone viewing the question.

The slightly more rare, but arguably worse situation is when an asker understands enough to mark an answers as the answer, but then goes away and someone else comes along and posts another much better answer. If the question is kind of on the fringe, it can take quite some time for the vote count to swing in favor of the new, better answer and the OP has long since left. This leaves a question in a less than desirable state where the inferior answer has both more votes and is marked as the accepted answer.

One possible solution to this would be to allow the accepted answerer to reassign the accepted answer to an answer they feel is better than their own. People that answer questions tend to be much more active on the site and have a vested interest in trying to make sure that we have the best possible answers to questions. Additionally, if the answer's answer has been marked as accepted, they clearly had a solid understanding of the question and the details needed for the solution, so it seems like they should be qualified to make changes to the accepted answer.

This came up in response to this question where the answer provided by Falcon Momot is head and shoulders above my own answer. As a work around, I've made mention of his answer in mine to make sure people take a look at it, but really, it is a much better answer for the Accepted Answer on that question, it just happened to come in 6 months late.

  • An alternative solution to the problem of inferior but accepted answers being shown first forever is at meta.stackexchange.com/q/268666/215590 – PolyGeo Jul 8 '16 at 19:13
  • @PolyGeo - yes, I'm aware of that one, this is a different situation as it is an area where both answers get consistently upvoted and it's a rare enough situation that it could take an exceptionally long time for the better answer to make up for the six month head start. – AJ Henderson Jul 8 '16 at 20:19

I would say no. Acceptance of an answer means that the OP found it useful at the time.

I agree, sometimes somebody else posts a better answer than yours. You've done the noble thing in editing your own answer to point readers to the new answer. Vote it up and leave it there.

There are plenty of cases where an OP does change their acceptance. If they don't, it doesn't harm the site as a whole.

The majority of cases that I've seen where the scenario you describe happens, the question, first answer and acceptance all occur within about five minutes. It is a drive-by question and the OP has gone away without waiting for a good answer, just an adequate one. You won't stop people doing that.

|improve this answer|||||
  • 2
    I suppose I can see what you are saying, but that also somewhat makes me question the value of even making public notice of the accepted answer then. If it is just a rep reward for whoever happens to actually provide assistance to the OP, then it doesn't seem like something that needs to be publicly visible, other than simply an indication that the question was solved. I suppose it may also make a difference between knowledge questions and problem fix questions as well. You want a confirmed problem fix answer even if it isn't the "best", but for knowledge questions, there is no "worked". – AJ Henderson Jul 8 '16 at 15:51

I once answered a quite popular question and my answer was accepted a few hours later. Unfortunately, comments started to pile up saying it was not the right answer. I did my best to edit it, but the only way to do it appropriately was to plagiarize another answer with far more upvotes than mine. The community members heavily upvoted the other answer while downvoting mine.

I asked the OP to change the acceptance, but he refused to do it. Then, I posted a Meta question, but there was nothing a moderator can do about it.

I think a moderator should be able to change the acceptance only when the accepted answerer requests for it with reasonable explanation. I believe moderators can decide if it is a legitimate request based on their experience and knowledge. It will do no harm to the community and any of the members except that the accepted answerer will lose 15 reputation points (s)he earned.

As a few communities become older, some old answers might not be up to date and chances are higher that a new answer to an old question could be better and more useful to current and future members than an old accepted answer. Enabling moderators in certain communities to change the acceptance for old questions, e.g., three to five years old, doesn't sound like a terrible idea.

|improve this answer|||||
  • 2
    I'm not sure I see what having a moderator involved in the process would add. Moderators are not subject matter experts on everything in their community, but the accepted answerer would be presumed to be since the OP accepted their answer. I can't see any way an answerer would abuse this either unless they were performing deliberate sabotage (in which case I suppose a mod should probably be able to undo the move). – AJ Henderson Jul 8 '16 at 18:33
  • @AJHenderson I agree moderators can't be experts on everything, but I can safely bet that they can judge whether your request for changing the acceptance is reasonable or not. I think it would be better than allowing the accepted answerer to make changes to the accepted answer. – Rathony Jul 8 '16 at 18:37

You must log in to answer this question.

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