I know this isn't an issue specific with math.se, but this is one of the community where I feel like it is widespread.

Take as an example my last question. I am basically asking "Is the result X new?". From the comments it is pretty obvious that the answer should be "no, it isn't new. It's a consequence of little Fermat's theorem". Except nobody actually consider posting it as an answer I could accept: only in (not always very nice) comments.

And it is not just one person who did that: it looks like everybody is seeing this as normal (when it really isn't).

When this situation occurs, is there a way I can (nicely) react?

  • 5
    $\begingroup$ A duplicate of math.meta.stackexchange.com/q/1559/42781? Anyway, see for the example the answer at math.meta.stackexchange.com/a/36082/42781, in particular the section "What can we do?" (I say this knowing that I am potentially providing an answer in a comment. ;) $\endgroup$ Commented Feb 6 at 17:59
  • 22
    $\begingroup$ That question was closed as being "opinion-based". The commenters may have been wary of posting an answer to a question that they expected to see closed, as answering such questions can have unpleasant consequences. It's safer to answer in a comment. $\endgroup$ Commented Feb 6 at 21:00
  • $\begingroup$ As the linked question was closed 4 minutes after this post, I see a recurring reason behind the recurring situation: posting answers is risky. (From the linked duplicate meta question) $\endgroup$
    – peterwhy
    Commented Feb 6 at 21:03
  • 2
    $\begingroup$ People finding comments mean has been a problem across many StackExchange websites, especially Stack Overflow. I think it's partially caused by the design of these sites where they're meant to be "strictly Q&A" and try to prevent any other interactions between users. For instance, if one were to type out a long proof asking for any errors, someone would type in the comments a one-line counterexample that breaks the entire proof apart. While this would be helpful for others, it can be humiliating for the OP, esp. if the comment seemed rude. $\endgroup$ Commented Feb 7 at 3:27
  • 2
    $\begingroup$ @Accelerator I am talking about a general lack of intellectual charity. Like if someone writes that any number can be written as a product of prime factors you can assume they are talking about strictly positive integers, unless there is something else that suggest it isn't the case. Or someone asking for a specific info to be added an someone else complaining (to the OP) that that info is useless. $\endgroup$ Commented Feb 7 at 5:28
  • 1
    $\begingroup$ I think I see what you mean. IDK if there's a proper way to react. Whenever I'm chatting with someone online and they say something borderline rude, I tend to ignore it. But if they keep doing it, I set my boundaries and tell them to be nicer. In your case, it's probably better to just move on if you find their comments condescending. $\endgroup$ Commented Feb 8 at 6:17
  • 4
    $\begingroup$ "Thanks! This indeed solves my question. Please write it as an answer. I'll accept it.": this is the way I generally handle such cases. $\endgroup$
    – Dominique
    Commented Feb 13 at 14:55

2 Answers 2


What I do: I leave a comment asking them to write it as an answer, so people can upvote and accept their answer. Hopefully this gives them some incentive to do so. If you want, you can also let them know that the site works better if answers are posted in the 'Your Answer' box, not in the comments, so the question can be treated as answered.

If you notice they have a pattern of answering in the comments and don't respond to gentle encouragements, or just don't feel up to gently teaching people how to use the site properly, another option is to flag their comment as 'no longer needed' without leaving a comment of your own, and optionally write your own answer with the answer to the question. I'm not saying you have to do that if it doesn't feel right, I'm just saying that it is another option on the menu you can consider if you like.


If someone answers in a comment, then you should feel justified in using their comment to write a self-answer.

In your answer you should acknowledge the comment, and (if possible) expand on it somewhat, by adding details or examples. In the example given, you could fill in the details about how it follows from Fermat's little theorem. Then accept your own answer. It is a good idea to add a comment to your answer explaining "As there was no answer written, but sufficient information in comments, I am writing a self answer." You might even link to the meta discussion where it approves of this practice. https://stackoverflow.com/help/self-answer

If you are concerned that you will be accused (unfairly) or reputation farming, then mark your answer as a "community wiki". This encourages others (including those who wrote the original comment) to edit or improve, and means you won't gain or lose reputation. But this is optional, not required.

See also https://stackoverflow.blog/2011/07/01/its-ok-to-ask-and-answer-your-own-questions

  • 7
    $\begingroup$ I agree here, except that I emphasize that you should feel no obligation to mark it as a community wiki. Comments are ephemeral and might disappear. And there are certain mechanisms that return questions with no answer to the front page on occasion --- this is distracting when the question actually has been answered in the comments. $\endgroup$
    – davidlowryduda Mod
    Commented Feb 13 at 22:28
  • $\begingroup$ I agree that getting someone to write up their comment properly should be the first choice. I've once tried to summarise comments in an answer to my own question, so I could have future reference to it. However, it seemed some people took a view that I was trying to appropriate their knowledge to score points on the site, and I received downvotes. I wouldn't do it again. $\endgroup$
    – demim00nde
    Commented Feb 14 at 11:54
  • 6
    $\begingroup$ @demim00nde yes, that is unfortunate. This is why I suggest adding a comment - and try to be philosophical about reputation, upvotes, downvotes etc. If you have learned something you have "won", no matter how many downvotes you get. $\endgroup$
    – James K
    Commented Feb 14 at 13:44
  • $\begingroup$ This to me is an incredibly wise answer, as it forces one to confront one's own learning curve. You want an answer to a question, you are going to have to learn to write it on your own at some stage. $\endgroup$ Commented Feb 19 at 23:37

You must log in to answer this question.

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