23
$\begingroup$

I am beginning to use the computer algebra system Magma heavily. I am spending a lot of time with its documentation (which is wonderfully organized and clear), but am very much hoping that there is a forum on the StackExchange network appropriate for questions of the form, "how would you do such-and-such a calculation in Magma?" since some of what I want to know has so far been very hard to extract from the documentation.

I have used Math.SE in the past for a question about algorithm design which I implemented in Magma, but the questions I want to ask now are not about algorithm design, just very nuts-and-bolts questions about the language and functionality of Magma. I asked my first such question on Math.SE last night; it hasn't attracted any answers or comments, and perhaps this is because it's less than 12h old, but maybe also because it is not really a math question at all but just a question about the use of a particular CAS. We have a tag here, and questions of this type have been asked and answered before, but not so frequently, and they have also sometimes been closed as off-topic.

A propos of this meta question, I see that StackOverflow has tags for both GP-PARI and Sage. I have never used StackOverflow before since I am not a programmer, but this does make it seem like StackOverflow is an appropriate place for questions about execution of tasks in specific CAS's. On the other hand it doesn't have a magma-cas tag. It has a magma tag but it's referring to this rather than the CAS, although some of the questions under the tag are about the CAS. I think only one such question (under the magma tag, about the CAS) has been answered.

Thoughts?

$\endgroup$
8
  • 5
    $\begingroup$ You could consider the site Computational Science It is focused on numerics, but I once talked with a mod there (on area 51, in the context of a proposal mention on that meta thread, IIRC) and they said they are in principle open to questions on more "algebraic" software too. Indeed, there are a few pari/gp questions there. (Searching for magma also gives hits, but it seems the same as on SO, something else.) $\endgroup$
    – quid Mod
    Nov 24, 2015 at 18:31
  • 1
    $\begingroup$ @quid - thanks, I'll try that. $\endgroup$
    – Ben Blum-Smith
    Nov 25, 2015 at 0:09
  • 1
    $\begingroup$ I generally say that one should ask a question where it will be best answered. In principle, magma questions could be on topic here. From what I can tell, we have very few magma experts here (more than one, but very few). But I'm also not sure if there is another site where more magma experts hang around. So for a bit, you might end up asking the same sort of question on a few sites and seeing where it sticks the most. $\endgroup$
    – davidlowryduda Mod
    Nov 25, 2015 at 4:47
  • $\begingroup$ Your title asks about here and "StackOverflow". The latter seems strange. Certainly your questions are off-topic at "MathOverflow", though. $\endgroup$
    – GEdgar
    Nov 25, 2015 at 19:04
  • $\begingroup$ I was going to reply saying Matlab questions here are routinely closed. But checking that, I find lots of Matlab questions that were not closed. $\endgroup$
    – GEdgar
    Nov 25, 2015 at 19:06
  • 1
    $\begingroup$ @GEdgar I am virtually certain that SO is actually meant (and I do not see what is strange about it). For one thing note "I have never used StackOverflow before since I am not a programmer" For another, the description of the situation is accurate for SO and not MO (that has a magma tag referring to the CAS and not something else). $\endgroup$
    – quid Mod
    Nov 25, 2015 at 19:19
  • $\begingroup$ @GEdgar - yes I meant StackOverflow. It didn't even occur to me until I searched for previous questions like mine and discovered the one I linked in the OP, which was, "where should I ask questions about PARI?" The accepted answer (from Willie Wong) was StackOverflow and that's when I discovered SO has tags fro PARI and Sage. $\endgroup$
    – Ben Blum-Smith
    Nov 25, 2015 at 20:19
  • $\begingroup$ This seems (to some extent) related: meta.math.stackexchange.com/questions/17119/… $\endgroup$
    – Martin Sleziak
    Nov 30, 2015 at 18:13

1 Answer 1

Reset to default
5
$\begingroup$

My short answer is that being on-topic requires some intended application of the mathematically useful software.

Magma Computational Algebra System is a purchasable product which is promoted as "a large, well-supported software package". As such general Magma feature questions should not be directed at StackOverflow, Math.SE, or SciComp.SE (aka Computational Science), but rather to the vendor (however academically oriented it might be).

What could well be on-topic at Math.SE or SciComp.SE are Magma-related questions about performing mathematical/computational tasks. For some discussion, at least as it related to Math.SE, see this previous Meta question:

Do we actually "welcome questions about: • Software that mathematicians use”?

As a SciComp.SE participant, I think I'm accurate in stating that a similar line is drawn there regarding what is on-topic. See for example:

How appropriate are software package-specific installation questions?

Although SciComp.SE tilts in the direction of numerical computation, there is a big tent philosophy regarding algebraic/symbolic computation. See for example Where do we want to draw the distinction between math questions and computational science questions?.

$\endgroup$
2
  • $\begingroup$ I understand the distinction you are drawing (between "general feature questions" and "questions about performing mathematical / computational tasks") in a vague, general way, but can you help me understand it in application to my case? Here is a question I asked, fairly typical of the type I want to ask: scicomp.stackexchange.com/questions/21420/… . To you is this on-topic? $\endgroup$
    – Ben Blum-Smith
    Dec 2, 2015 at 13:02
  • 1
    $\begingroup$ Yes, to me it is on-topic (and I created the tag magma-cas at SciComp.SE). It would be improved if you could give an example of the ring $R$ and corresponding polynomial ring $R[X_1,\ldots,X_n]$ you have in mind, and what you would do with the $R$-subalgebra once you have it. $\endgroup$
    – hardmath
    Dec 2, 2015 at 13:37

You must log in to answer this question.

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