# Relation to Cryptography.SE

The Cryptography Stack Exchange recently went into private beta and we're trying to figure out what overlap there is with Mathematics. The Mathematics FAQ seems to imply that questions although mathematical in nature should be kept on Cryptography if they are not general math questions, and also fairly explicitly states that research and design questions should also be kept there. What is the Mathematics community view on this?

Additionally, it would probably be prudent to update the Mathematics FAQ once the Cryptography site reaches public beta, if certain questions are to be migrated there.

• Math.SE covers many topics/questions about applications of mathematics that are relevant only in a relatively narrow field of study (crypto, signal processing,...). And I very much like it that way. It sounds awfully restrictive to me to forbid questions about the math underlying RSA or elliptic curve cryptography from MSE. Or may be I misunderstood, and you are not really proposing something like that? – Jyrki Lahtonen Jul 14 '11 at 6:21
• @Jyrki: I'm trying t avoid proposing anything, but rather informing the Math.SE community that there now is a Cryptography site in the SE network and that the Math.SE community might want to take this into account e.g. in their FAQ or when closing questions. It's entirely up to your community what you want to do. – You Jul 14 '11 at 7:00
• Cool. There will undoubtedly be some migration in the future. My concern was basically that some cryptographic math is relatively deep stuff (not the recurring questions about basic congruences in RSA), and those questions might be responded to more quickly at Math.SE. But such concerns can be handled if/when they appear. I'm too new here to understand how the hierarchy of SE works. Good luck! – Jyrki Lahtonen Jul 14 '11 at 7:15

My view is: if there is a better forum for a question (because all the experts hang out there), then the question should be encouraged to be placed there. With that said, can You give us a one-line blurb about what Cryptography.SE is all about? (Also, what sorts of questions do you welcome there, and what sorts of questions you don't? That will certainly help us update the FAQ.)

Would, say, Why is hash function $h$ ($h(w_1 \oplus w_2) = h(w_1) \oplus h(w_2)$) not good? or Encrypting algorithm be good questions for Cryptography? (I also remember some questions from a long time back on the mathematics behind MD5 or DES or something like that, would those be on topic for Cryptography.SE?)

• Looking at area51.stackexchange.com/proposals/15811/… seems useful. – François G. Dorais Jul 13 '11 at 5:08
• @François, see also this, it is still not clear to me what is the scope of cryptography.SE (e.g. are practical cryptography questions that are being posted on Security.SE included in the scope?). – Kaveh Jul 14 '11 at 4:09
• we generally try to avoid migrating questions to beta sites until they show sufficient health to indicate they'll survive. That said, if a question is an extremely strong match to a target site, I support it. – Jeff Atwood Jul 14 '11 at 10:58

I can't find what you said about MSE faq ("The Mathematics FAQ seems to imply that questions although mathematical in nature should be kept on Cryptography if they are not general math questions").

My view is that research-level theoretical computer science questions include research-level cryptography questions, this is stated clearly in cstheory faq (based on ACM's SIGACT).

• If cryptopeople want an SE for themselves, why not? Yes, there will be overlap, but that already happens between Math.SE and CSTheory.SE in suitable areas of discrete math. For example, I briefly visited cstheory, and noticed that about the same number of questions tagged coding theory can be found in the two stacks. Of course, it is pointless to splinter the hierarchy of SE into too small pieces, so in that sense I support you. I'm really familiar with only Math.SE, and can't tell, whether its scope is considered to be too large, average size or too small in comparison to others. – Jyrki Lahtonen Jul 18 '11 at 13:59
• @JyrkiLahtonen: there is not an overlap between cstheory and MSE, cstheroy is research-level only, i.e. topics cover in undergrad course are off-topic there, I guess you mean MO not MSE. Here I are not arguing about crypto.SE, you can find my opinion and argument about the issue on area51. The question here is about where MSE should/would direct crypto questions, and I am saying if they are research-level theoretical crypto questions then they are part of cstheory's scope – Kaveh Jul 18 '11 at 14:13
• I don't know, but I don't think MSE is limited to non research-level questions? True, it may be best to ask such questions at MO. Anyway, according to my best judgement some coding theory questions at cstheory are at the level of advanced undergrad/early graduate/engineer asking for advice in implementation as opposed to being `research level' questions. It may well be that my take on this is somewhat skewed given that 'research' as done in an R&D dept of a telecommunications company differs from academic research. I would have welcomed those question at MSE. – Jyrki Lahtonen Jul 19 '11 at 10:09