I recently encountered an answer that does nothing but refer to the poster's book:

Please look pages 141-147 of the book:

S. Saitoh and Y. Sawano, Theory of Reproducing Kernels and Applications, Developments in Mathematics 44, Springer (2016). 2023.5.4.20:36

This answer was posted by S. Saitoh. Also, literally 100% of this poster's answers across multiple sites in the Stack Exchange network promote the author's books and/or papers, and the account has had multiple spam flags upheld against it on multiple sites.

However, the spam flag on this answer was declined as having no evidence to support it. Why was this not considered spam?

  • 5
    $\begingroup$ Have you checked that pages 141-147 of said book actually do not answer the question (in particular, as none of the regular MO users had flagged the answer so far, and as in your profile I do not see an indication that you are a mathematician)? -- The question owner has accepted the answer, and this happened 24 days after thanking the answerer in a comment. So, it seems to have happened not without taking some time to check. If you insist that the answer is spam (which is possible), then a bit more explanation is needed. Otherwise, the benefit of the doubt is with the poster of the answer. $\endgroup$
    – Stefan Kohl Mod
    Commented Feb 8 at 18:57
  • 3
    $\begingroup$ @StefanKohl I can't verify whether it answers the question or not because the answer contains no information whatsoever other than "go read my $140 book". $\endgroup$ Commented Feb 8 at 19:43
  • $\begingroup$ It seems to me that the only ones who can answer your question, EJo, are the moderators, so I would recommend contacting them directly instead of posting here. $\endgroup$ Commented Feb 8 at 20:44
  • 4
    $\begingroup$ I think a "see this refrence" answers are not high-quality even when they aren't self-citations, and pointing to a book, no link, not available otherwise on the internet is the least-useful version of this. That it is the author's own is merely the bow on top. (I didn't decline the flag) $\endgroup$
    – David Roberts Mod
    Commented Feb 8 at 20:44
  • 3
    $\begingroup$ Looking at the book, equations (2.263) and (2.264) and the subsequent comment look to me (a non-expert) like what the OP of the question on main was asking for. And I am in the fortunate situation that I can get the book from the library, which I what I presume the OP on main did. $\endgroup$
    – David Roberts Mod
    Commented Feb 8 at 20:49
  • 8
    $\begingroup$ Given that the username of the answerer matches exactly with an author of the book, and this is clear at a glance, this is not an undisclosed promotion. I personally would like it if people said things like "see my book", rather than "see the book", but this is a culturally-dependent idiom—it might not be acceptable for some to explicitly pronounce their own work so obviously. $\endgroup$
    – David Roberts Mod
    Commented Feb 8 at 20:52
  • 10
    $\begingroup$ I think that references to the literature are perfectly good answers, and should be encouraged. It doesn't matter if the OP wrote the book/paper as long as it answers the question. If something has a short and snappy explanation it's good to give it, but for more complicated things a reference is often the most useful thing since it will presumably contain a lot of related material as well. $\endgroup$ Commented Feb 9 at 2:08
  • 8
    $\begingroup$ (I also resist the idea that one should restrict one's self to references that are available on the internet. If it's available online that's helpful, but often times the clearest explanations are only available in the library.) $\endgroup$ Commented Feb 9 at 2:14
  • $\begingroup$ @DavidRoberts Disclosure in a username or in a user's profile should not be sufficient, because those can be changed at any moment without bumping the answer. $\endgroup$ Commented Feb 12 at 21:45
  • 2
    $\begingroup$ I agree with everyone saying that this is not the best answer (because it is just a reference with no further information), but it is clearly not spam and flagging it as such is an inappropriate use of the flagging tools. $\endgroup$ Commented Feb 12 at 22:29
  • $\begingroup$ @FedericoPoloni sure, as a long-term thing, but I'm assuming good-faith here, and that the user name is actually the real name of the person who posted it, etc. Let's just say this is miles away from an example I saw yesterday where someone posted a ChatGPT paragraph void of real content, ended with a smiley emoji, and hyperlinked that smiley to some random website with a suspicious url. A professor new to the site helpfully pointing someone to their book and giving specific pages to look at is not a sin, and if said professor returns, maybe we can gently nudge for more info next time :-) $\endgroup$
    – David Roberts Mod
    Commented Feb 12 at 23:02
  • 1
    $\begingroup$ @Andy oh, I agree. But in this case, the book was available on SpringerLink and had a doi. For those that know about such things, it's polite to supply the link. Or if one knows it's not available electronically, this can be mentioned, to save people time hunting around for it. The more information that helps downstream users find the resource with the least friction is my policy, and we as a community can supply it when lacking; but it's nicer when users put in the effort themselves, if they have time. $\endgroup$
    – David Roberts Mod
    Commented Feb 12 at 23:04

2 Answers 2


The answer under consideration is not spam, because it is a valid answer to the question that was asked (and the question was more or less on-topic for this site).

Ideally, we would like some kind of mathematical content in the answer text so people can learn something new at a glance, but not everyone answering questions has that kind of time or energy.


I did not process that flag, but I agree with whoever did. I think "spamming" is often construed overly broadly anyway. In my mind, the first criterion for judgment should be relevancy, with the (potential) self-promotion part being secondary. There is little doubt in my mind that the answer given was very relevant, and I'm quite willing to suppose it was indeed helpful.

I'll link to a Meta StackExchange post on the topic, which may help: https://meta.stackexchange.com/help/promotion.

(Some posters head off the issue at the pass, writing words like "at the risk of seeming to toot my own horn", but such denials are not a requirement. Speaking of self-effacing modesty: nobody should feel constrained about horning in and improving such a post, by editing in critical content from the pages provided, in order to answer the question more directly.)

  • $\begingroup$ Did you put here a different link than you intended? You wrote: "I'll link to a Meta StackExchange post" - but this is actually followed by a link to the help center rather than a link to a post. (And the same content is in the MathOverflow help center.) $\endgroup$ Commented Feb 12 at 14:28

You must log in to answer this question.

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