There is a 5 minute grace period after voting on a question or answer after which the vote is locked until the question or answer is edited. I like this concept since is keeps people from toggling their votes and encourages them to think before voting.

I think having a similar system in place for accepting answers would also be beneficial. Sometimes I'll have an answer accepted, and 5 days later (or sometime much longer) the acceptance is rescinded in favor of another answer. Occasionally I'm on the flip side of that as well. This is fine is the change is due to a new answer being posted after the acceptance, but sometimes this isn't the case.

I propose the following system: Once an answer is accepted, the grace period (1-2 days?) begins, after which the acceptance is locked in place. The lock is removed only if a new answer is posted.

Variant 1: Remove the lock if the accepted answer is edited.

Variant 2: Remove the lock if any answer is edited.

Variant 3: If answer A is accepted and then pre-existing answer B is edited, then the acceptance can only be moved from A to B; it cannot be removed from A without going to B and it cannot be given to pre-existing answer C either.


  • 6
    It's not too uncommon for a wrong answer to get accepted. Changing the accepted answer when you realize that should be possible. – CodesInChaos Nov 4 '11 at 21:15

No. We should not do this.

Very often answers are posted after the fact which are better than the best answer. Very often, an answer will appear to work, but will have some flaws.

You need to be able to change which answer is the accepted answer. I see no benefit in preventing the user from changing the accepted answer. It will just make people more reluctant to accept answers.

Accepting an answer effectively means "I am not actively seeking more answers to this question" and indicates to those looking to answer unsolved questions that the question does not need an answer. Discouraging people to mark answers accepted is counterproductive and harmful.


I disagree. The person asking the question should be able to accept any answer they want. They should also be able to rescind an acceptance.

An example of why: I ask a question. I get an answer that looks like it answers my question, so I accept it without checking that it actually does. A few days go by before I actually get around to testing the answer (long weekend, say), and I find it doesn't answer my question. If the answer acceptance is locked in place I have no way of rescinding the acceptance, even though the answer didn't actually answer my question.

  • 3
    Just curiosity: why would you want to accept an answer so quickly, is there any advantage to that? – Gert Arnold Nov 4 '11 at 21:23
  • @GertArnold - No, but we know people would. – Paul Bellora Nov 4 '11 at 21:26
  • 1
    This is precisely the situation that I would like to avoid. When I come upon a question with an accepted answer, I like to think that it's correct. Prematurely accepting answers is already somewhat prevented by the waiting period, and I see this as a way to encourage that further. – PengOne Nov 4 '11 at 21:31
  • this has happened to me far too often. i might accept an answer when progress is made only to find out that either it really didn't do anything, or it made it even worse (when originally it looked promising due to a new error message; this may make you think the old one is finally fixed) – tehdoommarine Aug 30 '12 at 2:46

You must log in to answer this question.

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