tea.mathoverflow.net - Discussion Feed (Is this Theorem known?)

Will Jagy comments on "Is this Theorem known?" (11224)

Will Jagy — Fri, 03 Dec 2010 11:59:08 -0800

Just make a decision ahead of time about how much you can afford to share and how much you intend to keep secret. What I have in mind is this question, which drove me absolutely nuts:

http://mathoverflow.net/questions/12364/integer-subset-that-only-occupies-p-1-2-equivalence-classes-mod-p

The OP never gave any detail, just kept saying that what people wrote was wrong. I was quite surprised to find out he has a Ph.D., all the signs pointed to an undergraduate keeping a big juicy secret close to his chest. I deleted my comments and my irrelevant answer.

Scott Morrison comments on "Is this Theorem known?" (11223)

Scott Morrison — Fri, 03 Dec 2010 11:48:38 -0800

Yes, this is fine.

There is a slight danger with these questions: they could potentially provide "cover" for self promotion that we'd feel uncomfortable about. Certainly if it's asked as "There's this result I need, and which I know how to prove, but I'd really prefer if I can just cite something." then it's certainly okay. I think on the whole it's safe enough to just blanket approve this genre and deal with problematic cases on their merits.

Storkle comments on "Is this Theorem known?" (11222)

Storkle — Fri, 03 Dec 2010 04:08:17 -0800

Does "without success" in this context mean you've talked to people and the theorem you've proven is not known to them? Wouldn't that be success? Anyway, it's obviously appropriate to post such a question.

Pete L. Clark comments on "Is this Theorem known?" (11220)

Pete L. Clark — Fri, 03 Dec 2010 02:08:09 -0800

I think that this kind of question is acceptable on MO. It is absolutely the sort of question that a research mathematician would like to ask to a ready-made community of other research mathematicians, so it seems in the spirit of the site.

There is the point that it is hard to definitively answer "no" to such questions. But in this case the next best thing is to hear "no, I don't think so" from a leading expert, and MO can provide that.

Moreover, reference requests seem to be completely kosher. Isn't this a kind of reference request?

Martin B. comments on "Is this Theorem known?" (11214)

Martin B. — Fri, 03 Dec 2010 00:07:38 -0800

Assume I have proven a theorem and I want to know if it is known or even written up in the literature. Is it appropriate to ask that on MO?
If not, what other methods do you use? I've already tried to use MathSciNet and talked to some people, without success.