Good morning. I recently posted a question on Math Stack Exchange here and got a fantastic solution. The problem in question is one that I've had for some time, and both myself and research professor have been searching for proof of the claim both through papers and using mathematical techniques.
It would appear that a particular solution to the problem has been found, though I must go through the proof with my professor to verify its truth. If it is true, what is the protocol for citing the proof and giving the author credit? Do stack exchange contributors post solutions knowing that their work may not be credited? I would never want to plagarize, but the information is useful for the greater problems that we have been trying to solve.


There are actually two questions on MathOverflow that ask how to cite answers found on MathOverflow. I would recommend following their advice.

One of them is a fairly old question, whose answer is somewhat outdated. Summarizing:

  1. Even though you may not be legally obligated to show where you got your answer (for example, when MO points you to a theorem already in the literature), it is a good idea — it never hurts to be generous and honest. (And if you're copying part of an answer, not citing it would be plagiarism.)

  2. There is a little button for each answer that says "cite" and will give you a citation. You might have to reformat this to make it fit in your journal's bibliography style.

  • So it would seem that using "cite" to obtain code used for the bib would be sufficient? That seems to be what I'm getting out of this question as I observe the answers I'm getting. Thank you for your help. – Lalaloopsy Aug 28 '18 at 14:16
  • 2
    The authorities at MathOverflow seem to think this is sufficient. You might want to replace the author's MO tag with his full name in your citation, if they give it in their profile. – Peter Shor Aug 28 '18 at 14:23
  • I reached out to the author to let him/her know and hopefully from there i can get info necessary. Thank you for your help. – Lalaloopsy Aug 28 '18 at 14:24
  • 1
    Whether it's sufficient depends on the policies of publication venue. And to follow strict citation patterns, you should add the authors real name to the MO tag (if they are different), not replace, because the real name doesn't show up anywhere on MO. – JeffE Aug 28 '18 at 16:50

Perhaps the person who posted it actually found it elsewhere and can provide a reference to you. Alternately, if you are willing to be publicly known here, you can find a way to exchange emails and hence give proper credit to the individual(s).

In the past, I've exchanged emails with members on this site using a mod as a go-between, but that implies that you can reach the mod via email and that the mod actually knows the other person's email. I note that the mods themselves aren't able to get people's email addresses from the site and it requires someone with higher authority. But if it is possible, the correct way is to ask a mod or administrator to give your email address to the other person, rather than to ask for theirs.

However, the rules are the same. If you use someone else's work, you need to cite them and an anonymous citation may not be enough unless they tell you on the site that you may do that. In that case, keep a record (pdf) of the exchange. The chatroom rather than the comment stream may be a better place to explore such things. (https://chat.stackexchange.com/rooms/2496/the-ivory-tower, or something similar at the math site: https://chat.stackexchange.com/?tab=site&host=math.stackexchange.com.)

But see the answers for the questions that this duplicates, also. I would, myself, worry about a citation that used an alias as author, rather than a true-name.

  • 2
    I would, myself, worry about a citation that used an alias as author, rather than a true-name.Why? Would you hesitate to cite a paper if you thought it might be published under a pseudonym? – JeffE Aug 28 '18 at 16:51
  • 1
    @JeffE, not if the person published frequently under that pseudonym, but for a one-off, yes, I'd worry about how it would be viewed. Among other things, you don't know very well if the citation you are using was, itself, plagiarized. I'll note that Bourbaki was a case in point. It was a well established pseudonym for a group and therefore could be well trusted even when the members were anonymous. They established a body of work that established the credential. – Buffy Aug 28 '18 at 16:58
  • No, I don't know if any given paper is plagiarized, but that's also true for papers published under the authors' True Names (or more accurately, presumably published under the authors' True Names, since we can't actually tell). But more importantly, at least in my corner of the world, it's the content of papers that determines whether they are trustworthy, not their authorship. – JeffE Aug 28 '18 at 20:39

Not the answer you're looking for? Browse other questions tagged or ask your own question.