I have several books of standard exercises that were asked in competitive exams in France (typically, ENS/Polytechnique), from my earlier years.

In particular, a lot of them are quite interesting, either due to the techniques used or the ideas necessary to solve them. (They mostly are about sequences, series, limits, asymptotics.) In that sense, having them on Math.SE would be (I feel) a nice thing for students who want to self-study and practice.

While it seems unethical to just copy-paste them along with the solution (in CW), as the authors of the books did come up with solutions themselves, I am wondering if it is within the scope of the website to ask the questions, wait for answers (in order to have several approaches, hopefully), and eventually maybe provide my own solution (not verbatim from the books).

Note that while the answers are due to the books' authors, the questions themselves are not -- they are from previous exams, and were generally reported by the students who had to solve them.

  • 16
    $\begingroup$ I've asked a large amount of such questions, although there were solutions already available either in books (from the Cassini collection essentially) or from the Revue de la Filière Mathématiques. Contributors in here often come up with different solutions, I think it's rather positive. $\endgroup$ Jan 13, 2017 at 7:37
  • 2
    $\begingroup$ Fantastic! I'll start, then. $\endgroup$
    – Clement C.
    Jan 13, 2017 at 20:22
  • 4
    $\begingroup$ I see nothing against it. Obviously cite that the question is a past exam question. However, if it is normal enough it would probably just be a good question regarding ___. It all depends on the subject. Definitely cite it if it isn't a trivial question to come up with (like what is the solution to y'' + 2y' + y = 0?). $\endgroup$
    – user64742
    Jan 13, 2017 at 21:08
  • $\begingroup$ @TheGreatDuck, is that a question on the site already? $\endgroup$
    – Wildcard
    Jan 14, 2017 at 4:25
  • $\begingroup$ @Wildcard idk. I'm just giving an example of a question that might not need citation as it's a textbook example of a linear diff. eq. and not hard to think of. $\endgroup$
    – user64742
    Jan 14, 2017 at 5:03
  • 2
    $\begingroup$ It's a pity that this question (which is really good) is not getting answers. Even though the OP has got his/her answer (through comments), I think the question should be answered by someone just for the sake of it. $\endgroup$ Jan 14, 2017 at 5:31
  • $\begingroup$ For reference: here is the first post/attempt I made (if that looks alright, i plan on following that format). $\endgroup$
    – Clement C.
    Jan 16, 2017 at 13:40

1 Answer 1


If you already have a solution to a question, then you need to make it clear what you are actually asking. The purpose of this site is not merely to post problems - it is to post questions that you would like to know the answer to.

Posting old competition questions has historically been tolerated, when done carefully and in limited amounts, even though it does not really fit into the spirit of asking questions. Searching for "contest" on this meta site pulls up many previous threads.

When someone does post a contest problem, they should clearly indicate the source of the problem and the fact that they know the answer. Some users have higher standards for questions of this sort, and in particular the amount of context needed may be higher. Simply copying the contest prompt verbatim, with no source and no added context, may lead to downvotes or question closures.

  • 7
    $\begingroup$ It may or may not be relevant to note that OP is not asking about contest problems in the traditional sense, but rather about exams conducted for the purpose of admission to certain (selective) institutions of higher education (typically taken by the analogue of 2nd year undergraduate students). While I am not an expert on these types of exams, I think the problems asked there are relatively more interesting as content of the site, as they are more geared towards testing understanding of things learned in the first two years of a tertiary math curriculum (at the top end of the spectrum). $\endgroup$
    – quid Mod
    Jan 15, 2017 at 16:11
  • $\begingroup$ That might make a more interesting set of problems, agreed. I think there is some similarity, though, in terms of going beyond the problem statement (which is all that the exam would be likely to include) to turn the problem into a motivated question that would be interesting for this site. $\endgroup$ Jan 15, 2017 at 16:14

You must log in to answer this question.

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