11
$\begingroup$

I have noticed something a bit confusing to me during my first few days here at MathOverflow. In the "about" section it says users can "improve posts by editing or commenting." On multiple occasions I have seen an answer given as a comment to the original question.

When is it more appropriate to answer a question in the comments section, rather than as an official answer?

$\endgroup$
  • 1
    $\begingroup$ One possibility is that the question has been closed, and so cannot accept any more answers, but can still accept comments. $\endgroup$ – Joel Reyes Noche Oct 22 '13 at 2:46
  • 14
    $\begingroup$ Note that I just answered your question in a comment. I did this because I felt that my answer did not address your question completely. $\endgroup$ – Joel Reyes Noche Oct 22 '13 at 2:46
  • 5
    $\begingroup$ Another reason for answering in comments is that official answers bump up the question in the active list, but comments don't. Some people get annoyed if questions are bumped often, and so some choose to answer in comments if they feel that their answers are not too important. $\endgroup$ – Joel Reyes Noche Oct 22 '13 at 2:48
  • 18
    $\begingroup$ Another case is when you think the question might be a bit trivial, so you give a trivial proof/counterexample in a comment, and either the OP realizes the question needs an edit, or the question is made obsolete. $\endgroup$ – Daniel Moskovich Oct 22 '13 at 3:16
  • 9
    $\begingroup$ If the answer can fit in a comment, then that's a reasonable indication that it might not be worth adding as a full-blown answer. $\endgroup$ – David Roberts Oct 22 '13 at 5:29
  • 2
    $\begingroup$ If, for whatever reason, you do not wish to accrue any points, either for this particular answer or in general, you might post an answer as a comment. $\endgroup$ – Gerry Myerson Oct 22 '13 at 6:12
  • 9
    $\begingroup$ To complement @GerryMyerson's remark: if avoiding points is the only motivation you might consider instead to give the answer as an anwser proper and make it Community Wiki. See meta.mathoverflow.net/questions/493/… and the links for issues with questions answered in comments only. $\endgroup$ – user9072 Oct 22 '13 at 10:14
  • 1
    $\begingroup$ Great, thank you for the information. $\endgroup$ – Max G. Oct 22 '13 at 12:26
15
$\begingroup$

Max G.: that's a good question. I think in many cases the decision is based not on policies or guidelines, but an inner sense that the answer (and thus the question itself) is too easy and that the commenter is reticent, either about the quality of the question and whether it deserves to be answered or even open, or about acquiring points in such a cheap and easy way. (These possibilities were covered by David Roberts and Gerry Myerson.) I can attest I've felt this way on many occasions.

$\endgroup$
  • 1
    $\begingroup$ I think there is some danger in automatically inferring that the commenter thinks the question is poor quality; it is also quite likely the answer is very incomplete. $\endgroup$ – user25199 Oct 22 '13 at 8:27
  • 1
    $\begingroup$ Also, it is not clear that such points are cheap and easy - answers to a poor question are unlikely to gain many votes. $\endgroup$ – user25199 Oct 22 '13 at 8:27
  • 12
    $\begingroup$ I often sketch a one-sentence answer in the comments if I don't have the time to write out a full answer (I am very slow to write them), but I can still regard the question as a good one and deserving of a full answer. $\endgroup$ – David Roberts Oct 22 '13 at 9:07
  • 3
    $\begingroup$ @Carl Those are good points, but I don't equate an easy (easily answered) question with a poor question. If the question is very easy, I might answer in a comment (with an option of expanding later into an answer) until I see that others find it worthwhile for MO; i.e., until then I might be hedging my bets by answering quickly in a comment. $\endgroup$ – Todd Trimble Oct 22 '13 at 11:51

You must log in to answer this question.

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