2
$\begingroup$

I have a tendency to go back after my questions are answered for a couple weeks if I see room for improvements. My question is a simple one, it is: What is the best practice amount of time to go back and edit your questions and improve formatting and to clear up your thoughts?

I was thinking two weeks before you move on, but I might not be considering them viable for too long or I might not be considering them for a long enough period of time.

$\endgroup$
3
  • $\begingroup$ Is it just about formating the question? As long as you don't change the question much, I don't think there should be a time limit. $\endgroup$
    – Vincent
    Feb 14 '15 at 1:24
  • $\begingroup$ Formatting and adding extra content to clarify my ideas. $\endgroup$ Feb 14 '15 at 1:31
  • $\begingroup$ I would add that if you are unsure of your question, use the sandbox! :) $\endgroup$
    – James
    Feb 16 '15 at 16:05
4
$\begingroup$

This, like much else on Meta sites, is going to be a matter of opinion, but that doesn't mean I can't offer mine. :-)

My opinion is that as long as edits don't invalidate existing answers, you should not feel compelled to not update a question just because a certain amount of time has passed or even because a certain number of answers have been posted.

Why is that? Well, the first part is obvious; nobody wants to see the answer they put a lot of thought and effort into go to waste and perhaps even receive downvotes because someone edited the question to ask what is, in effect, a different question. (Whether that is because constraints were added that the existing answers don't, and cannot easily, meet, or it was because someone replaced the question entirely. Yes, we've seen that latter happen!)

The second part might not be as obvious. Sometimes, you realize long after the fact that the question wasn't as clear as it could have been. Maybe someone comes across the question after a year or two (or more!) and posts a comment, let alone an answer, that alerts you to the fact that the question, while existing answerers have all taken a similar approach to reading the question, could reasonably be misinterpreted. In that case, it makes sense to edit the question to clarify. But also see above; do your best to not invalidate existing answers. You may also want to leave a comment on each of the existing answers to alert the poster of the answer to the edit, so that they can choose whether to revise the answer in response or not.

Formatting-alone edits are a slightly different kettle of fish. Remember that any major activity on the question (a new answer, or an edit to either the question or an answer, including title or tag edits) bumps the question to the top of the front page. Again, there is no hard-and-fast rule about when this becomes a problem, but I try to think of it sort of like this:

  • Is the question already/still on the front page? Then it is probably okay to bump it.
  • Has the question made it to the network hot questions list? Then bumping it isn't going to have any major adverse effect, and the extra attention it is receiving might warrant fixing up even minor parts to make it as clear as possible.
  • In other cases, try to make edits more substantial than merely changes to formatting, because the edit is going to push more recent content off the front page. If the question is old, hasn't received much attention recently, and is good enough that the only thing you can improve is formatting, it is perhaps the case that the question really is good enough.

If you feel the need to edit the question in such a way that the edit is going to invalidate existing answers, consider whether you really are asking a different question and should post it as a different question instead. Remember that (generally speaking) it isn't a duplicate if a valid answer to one is not a valid answer to the other question. Adding constraints on answers would normally meet that criteria and make the second question not a duplicate, though possibly related.

$\endgroup$

You must log in to answer this question.

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