Suppose I write a code-golf question. How long should I wait before accepting the shortest answer? One week? Two days? More? I checked the FAQ, but couldn't find any information on this topic. Is there a general rule or is it just when the question asker feels like it?
\$\begingroup\$
\$\endgroup\$
1
-
5\$\begingroup\$ It's up to you, and if you're willing to change the accepted answer if a better one comes along it doesn't really matter. However, some people seem get put off or even offended if you accept an answer too early, so I guess a week is a good rule of thumb. \$\endgroup\$– Martin EnderCommented Aug 18, 2014 at 11:24
Add a comment
|
2 Answers
\$\begingroup\$
\$\endgroup\$
5
The way I run my own golfing challenges is to accept the shortest one at any given time (and reassign the accepted answer should a better one come in). Yes, it's more work because you have to monitor for new answers, but I think that's the best way to encourage people to keep finding better solutions.
It's best to wait a week before accepting for the first time, though.
-
4\$\begingroup\$ I'd like to do that as well, but some people always seem to complain about OPs who accept the first answer that comes along (even if they'd update it later on), so much that they sometimes downvote the question for it. Hence, I'm a bit cautious with that and usually wait a week until the first accepted answer. Alternatively, I guess you could just always explicitly state that you'll be updating the accepted answer. \$\endgroup\$ Commented Aug 18, 2014 at 11:26
-
4\$\begingroup\$ I do agree that a week-long wait is reasonable and that, combined with the continual update of the best answer after that, should be standard practice. (Updated the answer.) \$\endgroup\$ Commented Aug 18, 2014 at 11:27
-
\$\begingroup\$ If a question has an accepted answer I tend to think that the OP is not following it actively anymore, hence I'm discourage from making an effort in trying to write a shorter one \$\endgroup\$– pqnetCommented Aug 29, 2014 at 3:18
-
\$\begingroup\$ @pqnet Why? It's easy to reassign the accepted answer, so having an accepted answer should not be read to imply that the OP has "moved on". \$\endgroup\$ Commented Aug 29, 2014 at 3:24
-
2\$\begingroup\$ well, if you are used to other stackexchange sites, in which you usually accept an answer when you are satisfied, you subconsciously think that when a question has an accepted answer the OP is satisfied, when it hasn't he's not. It may not apply for codegolf, but still a marked answer give the "we have a winner" feeling \$\endgroup\$– pqnetCommented Aug 29, 2014 at 3:31
\$\begingroup\$
\$\endgroup\$
In addition to the accepted answer, I'd like to bring up another point: you don't necessarily have to accept an answer to a challenge at all. If you want to promote competition within individual languages, which is normally considered a good thing (so that the use of golfing languages doesn't drive away people who would prefer to answer in something more widely used), placing an accepted answer checkmark can discourage people because it looks like their languages are being passed over in favour of the ones which are more suited to the challenge. So you should at least consider just leaving no answers as accepted.
In a few special cases, notably answer-chaining (and to a lesser extent popularity-contest), the sort order of the answers is relevant, and placing an accepted answer checkmark will disturb it. So I'd be much stronger in these cases, and say that in questions where the sort order matters, you should not place the checkmark (as opposed to other questions, where you should merely just consider not placing it).