The question Example of function which is not computable was recently edited by the original poster to completely change the meaning. In particular, the accepted answer is no longer reflective of the question and is misleading as an answer to the current question. I'm not sure what our practice is for this sort of situation.

  • 4
    $\begingroup$ As far as I know, OP means original poster, not original poser. Do correct me if my edit was inappropriate (and wrong). $\endgroup$
    – Asaf Karagila Mod
    Jun 6, 2011 at 15:49
  • 1
    $\begingroup$ Note that I edited my answer correspondingly. $\endgroup$ Jun 6, 2011 at 16:10
  • 1
    $\begingroup$ @Quinn: that does help. My concern was mostly that the first sentence of your answer no longer matches the question, through no fault of yours. I think similar things have come up before. Personally, I think the new question should have been posed as a new question, rather than an edit. $\endgroup$ Jun 6, 2011 at 16:24
  • $\begingroup$ Yes, I agree. At the very least, the OP should make it clear when an edit changes the meaning of the question completely. In particular, the content of the original question should remain. $\endgroup$ Jun 6, 2011 at 16:37
  • $\begingroup$ @anegligibleperson has alrady done the same thing. He change the question from $\dfrac{\partial^4y}{\partial x^4}=c^2\dfrac{\partial^4y}{\partial t^4}$ to $\dfrac{\partial^4y}{\partial x^4}=c^2\dfrac{\partial^2y}{\partial t^2}$ in math.stackexchange.com/questions/192436 when there are completely solved answer. Please feel free to scold him. $\endgroup$ Sep 9, 2012 at 22:56
  • 1
    $\begingroup$ @doraemonpaul: I don't think a scolding is in order, because the person might not realize they can ask a second question. But I undid the edit and asked them to do so (I hope I did so without any negativity). $\endgroup$ Sep 10, 2012 at 1:24
  • 5
    $\begingroup$ I think changing questions in a significant way should be strongly discouraged. It wastes peoples effort, leaves answers that are no longer relevant and leaves the answerer in bad light. $\endgroup$
    – copper.hat
    Sep 12, 2013 at 23:06

2 Answers 2


Depending on the situation, I think we should either:

1) Roll back the change, and ask OP to add a new question, linking to the old.


2) Make the edits which ensure that the original question still remains. In this current case, that is easily done. In the general case, when it might not be possible to just make additions to the question, we request OP to create a new question and rollback any changes which change the meaning of the question substantially, pointing OP to this thread.

We should also encourage people who answer to not immediately change their answer and if they do so, they only add instead of taking away.

  • 2
    $\begingroup$ In the past, I have stumbled across the question after a new answer (to the new question) has been posted. In these situations, I have edited the question to note both the original and the new questions being asked. I agree with your analysis - this has seemed to work pretty well. $\endgroup$
    – davidlowryduda Mod
    Jun 6, 2011 at 21:42
  • $\begingroup$ This seems like pointless arguing that does nothing but clutter up the site. If the user accepted an answer, it's over. Learn to cope with the fact that your good intentions don't always get rewarded. $\endgroup$ Mar 8, 2016 at 1:09
  • 1
    $\begingroup$ I suggest also considering adding text to the top of the answer to make it explicit that the question was changed and, possibly, indicate what the difference is so anybody can understand why the answer doesn't match. $\endgroup$ Feb 18, 2019 at 2:34

I think that an OP shouldn't modify her/his question after the answers, and according to the answers!

I can understand that an OP wishes to improve her/his answer, but not to modify the question so to incorporate the answers!

I am new to this site, and today it happened to me that a question was edited after my answer, and some comments that said the same thing I said, and was modified on the base of the answer and the comments. This is the thread:


The OP asked how to prove the statement in the title, and in my answer I said that he couldn't prove it, the statement was false, because it was false for infinite sets.

At this point the OP changed her/his question, and added 'for finite sets'to the statement.

So my answer became stupid and useless, moreover the question became a duplicate (the original original question was not a duplicate), then the question was closed and the thread deleted, I lost my upvotes.

What a mess!

Waste of time for people who participated to the discussion, and deletion of a question and of answers that could be useful to leave visible, because this confusion about finite/infinite sets and cardinality and onto/one-to-one function is frequent, in my experience (I am an economist and a moderator of a site of mathematics in my country).

The reason why I write this post, anyway, is not my upvotes, but the interest of the site to have a good functioning.

  • $\begingroup$ @ Martin Sleziak Thank you for editing, I sometimes don't see minor typos as the light of the computer makes me a little dyslexic, it is not sloppiness! $\endgroup$ Dec 23, 2022 at 21:57

You must log in to answer this question.

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