5

Maybe this should be at the higher level. I'm looking at how quickly some questions have accepted answers within minutes of posting. Is there a way to program to now allow an accept until X hours or Y answers have passed/posted?

This runs both ways, even when mine is the accepted one, I'd caution the asker to wait, and move the accept to the better, unposted, answer to come....

  • 1
    See Also: How does accepting an answer work? – C. Ross Aug 14 '13 at 17:46
  • Thanks. It says the system requires 15 minutes, but no min on the number of answers. My mild opinion is an hour is more reasonable. Although, if I have something to add, I'll still do so regardless of the accepted answer award. Earlier, on Math.SE, I spent 10 minutes composing a first answer. In those 10 minutes, 5 answers went up and one, accepted. Last look, I was +33 vs the accepted +20. No big deal, just a popular question on a high traffic board, I guess. – JTP - Apologise to Monica Aug 14 '13 at 19:52
  • It's a balance between getting people to choose the best, and getting them to accept at all. I think some wouldn't come back an hour later to accept if they had an answer in ten minutes, especially on the faster sites like SO. – C. Ross Aug 14 '13 at 21:24
  • @JoeTaxpayer This issue has been raised on the main meta before, and a minimum wait of 15 minutes was imposed. Maybe a longer wait time is in order for a smaller site like money.SE? – John Bensin Aug 14 '13 at 21:33
  • John & Ross, thanks, I see the points on both sides. And I see the 15 min was probably debated and chosen as optimum. – JTP - Apologise to Monica Aug 15 '13 at 0:00
  • @JoeTaxpayer Don't let the fact that other answers are posted discourage you; I don't think it does, but I figured I'd state it for the record. Also, I don't think blocking accepts until a certain number of answers has been posted is a good idea because questions with a narrower focus may only get one answer. Maybe that could be combined with a time limit. – John Bensin Aug 15 '13 at 13:36
  • @JoeTaxpayer I don't think it could hurt to post another answer on the main meta question, though, possibly drawing attention to the fact that 15 minutes is a very short window for sites with considerably less traffic than SO. – John Bensin Aug 15 '13 at 13:38
  • @JoeTaypayer That highly-voted-but-not-accepted answer at Math SE earned you a shiny gold Populist badge. – Chris W. Rea Aug 15 '13 at 15:32
  • 1
    The accepted answer is exactly that. It's the answer the user has chosen to accept, it's not always going to be the right answer or the most-right answer. That's just the nature and part of the reason for up votes. – Paul Oct 11 '13 at 13:15
4

The thing to remember is that the accepted answer can be changed, and if a user is engaged and familiar enough with the system to accept one in the first place, they'll probably be likely to change it if something compels them to do so.

Fifteen minutes (the current barrier) is sufficient on sites where a question is likely to receive several answers in the first few minutes. It's designed to not put off other potentially better answers by the presence of the check mark.

On smaller sites, it's possible that a few answers may come in within that time, but that's not quite always the case. If the limit was put off to an hour, it's very likely that the person wouldn't remember to accept at all, as presumably they're off to do something with the information they received in the answer. Hence, it's a bit of a balance.

We want the value to be consistent across the network otherwise we run into the pains of documenting where it isn't the same, which is a major headache. Even though it's possible, I don't think we'd want to start making sites unique in this regard, and the 15 minute window seems to be working out well in most situations.

If a clear pattern emerges that questions get one, and only one answer and it's clear that the accepted answer is the culprit, then we really should be looking at UI changes so that folks are more strongly encouraged to answer anyway if they have more to add.

I won't argue that this doesn't happen from time to time, but not really in enough of a frequency to make us think about bumping the minimum network wide.

  • 1
    In my experience, a question receiving one fast answer is forever at a disadvantage, especially if that answer is very low quality and/or not useful. This has been a problem on all SE sites for as long as I can remember. – corsiKa Jan 8 '14 at 20:57

You must log in to answer this question.

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