4
$\begingroup$

I recently migrated away a Mathematica question. Now, our faq does not explicitly allow software questions so I assumed that it was off topic, and sent it away.

BUT--it seems like this question may be better suited for Physics.SE, since it deals with higher-end physics that most Mathematica.SE community members will not know.

Do we allow such questions? If not, should we allow them? Maybe add a line to the faq to that end.

$\endgroup$

3 Answers 3

3
$\begingroup$

The FAQ includes "Computational questions" in the "kinds of questions should not be asked here".

I generally try to sort out if SciComp or Stack Overflow would be best. Or I encourage the author to be clear if it is the physics or the computation that (s)he is asking about.

I'm wishy-washy about the one you've linked to. In a lot of ways it looks like a "save me having to read the docs" question, but I know from personal experience that that it is not always clear from the docs exactly where the limits of a complicated package may lie.

Any way, it is definitely not a physics question, but it is a question about physics tools. On Stack Overflow programming tools are on topic, but I don't know if we have a really clearly stated policy on this.

$\endgroup$
1
  • 1
    $\begingroup$ Perhaps it's not stated clearly enough, but as far as I know, it's always been the case that questions about the tools used to do physics, not about the physics itself, are off topic here. $\endgroup$
    – David Z
    Dec 11, 2012 at 19:12
3
$\begingroup$

I would prefer to allow both computational as well as software questions as long as their origin is purely physical, or, in other words, as long as it can be assumed that someone completely disconnected from physics (say, in financial maths) could not answer them.

For example, numerical integration sticks up its ugly head basically everywhere and is hence better of on CompSci, but nobody who is not a physicist will even think about Feynman diagrams and ways on how to calculate them. Hence the former should be migrated on sight, whereas the latter should stay here (and made explicitly on-topic in the FAQ).

To restate the criterium:

If it is sensible to assume that someone who is not part of a physics department, or similar institution focused on physics, will encounter a given, clearly computational/tools-related, question in his day-to-day work, then the question should be migrated to CompSci (or Mathematica).

$\endgroup$
4
  • 1
    $\begingroup$ I completely agree with this. On Mathematica, there does not even exist a string-theory or a quantum-field-theory tag, so this question should have stayed here for example. $\endgroup$
    – Dilaton
    Dec 11, 2012 at 17:24
  • 1
    $\begingroup$ I could get behind this. How does the phrase "tools unique to physics" work for you? $\endgroup$ Dec 11, 2012 at 18:16
  • $\begingroup$ @dmckee I didn’t use tools unique to physics in the last paragraph and only included ‘tools-related’ after computational as I don’t think the question in question was a computational question but rather a question about tools (in this case, Mathematica packages). $\endgroup$
    – Claudius
    Dec 11, 2012 at 18:19
  • 1
    $\begingroup$ I agree in general, though I would add the caveat that if someone asks for a package in such a specific environment as Mathematica, for which there is already a SE site, then it should be migrated. $\endgroup$
    – user10851
    Dec 12, 2012 at 1:35
0
$\begingroup$

I think about the physics questions involving computation should be decided case by case by giving the people knowledgable in the physics topics concerned some time to notice the question and see what they think about it, since they are the experts.

$\endgroup$
1
  • 1
    $\begingroup$ As I've mentioned in chat, what's off topic is off topic, even if the community can answer it. But, it'll be nice if the community chimes in here :) $\endgroup$ Dec 11, 2012 at 16:32

You must log in to answer this question.

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