12
$\begingroup$

I've seen that several times: an answer makes OP want to change the question in a way that would damage the answer. Obviously this should not be done. In fact I think it is better to delete the question altogether than do that. But occasionally I see it done anyway.

So I believe such behavior must be strongly discouraged. Is it written anywhere that it is a bad thing to do?

Related questions: Should I edit this question or make a new question? and Extending questions and answers gradually

Found another related question later: Should I delete my own question if it turns out trivial after a comment?

(I am on the verge of accepting the answer but something still leaves me unsure)

$\endgroup$
  • 11
    $\begingroup$ I think in general an edit to a question which invalidates an answer should be reverted, and if the edit is redone by the OP, the question should be locked. An exception though is the case that the answer reveals that an important condition has been forgotten in the question, and the question becomes entirely trivial or uninteresting for this reason. $\endgroup$ – Stefan Kohl Jul 22 '16 at 21:53
  • 8
    $\begingroup$ We see this all the time: I ask a question, but incorrectly. A trivial answer shows it was incorrect. So I fix it to be the question it should have been. $\endgroup$ – Gerald Edgar Jul 23 '16 at 13:34
  • 2
    $\begingroup$ This question contains an implicit answer. I think that that should be entered explicitly as an answer, so that people can vote on it. $\endgroup$ – Neil Strickland Jul 26 '16 at 15:09
18
$\begingroup$

As Gerald Edgar said, the most common reason for this situation is that the answer reveals to the OP that they posed the question in a form that did not properly capture what they intended to ask. I think that it is often OK to edit the question in this context, provided that the edited question explains the history in a clear and prominent way, with due credit to the person who provided the original answer.

$\endgroup$
  • 6
    $\begingroup$ Including, for instance, the original form of the question for reference, and the adapted one with clear delineation. $\endgroup$ – David Roberts Jul 26 '16 at 2:24
  • 1
    $\begingroup$ I agree, except that if the answer points out a really trivial part that was forgotten, then it's not necessary in each case to credit the person of the original answer. $\endgroup$ – domotorp Jul 28 '16 at 4:22

You must log in to answer this question.

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