2
$\begingroup$

Happened to me recently, and I wonder whether this is acceptable kind of behaviour, and not basically stealing ideas to get points.

In that particular case the person who asked the question appears to be a perfectionist who felt it's appropriate not only to supply an answer, but also remove accept from a similar answer by me and accept their own answer. I felt a bit, hmm, used, though.

Perhaps it ought to be clarified somewhere.

$\endgroup$
4
  • 1
    $\begingroup$ That looks like plagiarism. -- Which question do you refer to? $\endgroup$
    – Stefan Kohl Mod
    Jun 27, 2021 at 8:54
  • 2
    $\begingroup$ I won't use such a strong word in this case. My answer was "X follows from Y, a well-known useful hammer". The own answer went on unpicking Y and extracting what's needed from there. mathoverflow.net/questions/385579/… $\endgroup$ Jun 27, 2021 at 10:40
  • 1
    $\begingroup$ It may have been a mistake for the questioner to accept your answer in the first place, but I would forgive that mistake. $\endgroup$
    – user44143
    Jun 29, 2021 at 11:40
  • 1
    $\begingroup$ I'm not sure the right reaction was to downvote both the question and the newly accepted answer, especially given that I see no comment related to this issue in the given post. $\endgroup$
    – YCor
    Jul 1, 2021 at 14:13

2 Answers 2

5
$\begingroup$

From a general Stack Exchange platform viewpoint, this is allowed but considered bad form. A questioner can accept whatever answer they prefer, including their own. It is not wrong, per se, to accept the most complete and detailed answer, even if that answer was written after seeing another answer. Moreover, there is nothing wrong with someone (the original questioner or somebody else) writing a longer and more detailed answer that reiterates information that was already posted in a correct answer but goes into greater detail. Especially on a site like Math Overflow, where precision and clarity of exposition may be particularly important, this can actually be a valuable service.

However, changing the accepted answer to one that copied a significant amount of information from the first correct answer is, while permitted, rather bad manners. Doing this robs the first person to correctly answer of acknowledgement of their correct response. I think this is especially the case when the new accepted answer was written by the questioner themself.

$\endgroup$
3
  • 4
    $\begingroup$ If the new answer is mostly the same, but provides improvements or useful additions, it would have been far more appropriate to make those small changes to the original answer. $\endgroup$ Jun 27, 2021 at 13:31
  • 3
    $\begingroup$ @RayButterworth, surely in a non-CW case it is inappropriate to edit someone else's answer to make changes other than those that correct some unintended error? As a general matter, if I see an answer and think that it should have more details, my impression is that the community (and the particular answerer) will not welcome my meddling directly in it, and that it is much more welcome for me to write a new answer explicitly crediting the old one. (At least, that is what I have always done; I hope it is right.) $\endgroup$
    – LSpice
    Jun 29, 2021 at 12:34
  • 1
    $\begingroup$ @LSpice, agreed. My comment was too strongly stated. We don't have a real example here, and my impression from the question, especially with the word "perfectionist" is that the new answer was substantially the same, but with better wording and perhaps fuller details. $\endgroup$ Jun 29, 2021 at 13:31
3
$\begingroup$

IMO, The key point is to appropriately credit the person who gave the original answer in the body of the new answer, as providing the key idea and any other parts of the new answer they can claim credit for. It's not reasonable to expect people to look at timestamps to figure this out.

This done, the issue of "stealing" reputation points is a minor one. Once you get a few upvotes, the accepted answer bonus is a small portion of the total boost for the answer. Writing a new answer bumps the question, which might lead to you getting one or two more upvotes, refunding the lost rep.

$\endgroup$

You must log in to answer this question.

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