4
$\begingroup$

MO advises the questioner to add at least one top level tag for his/her question:

Please try use at least one tag corresponding to an arXiv subject area (details about what each tag covers):

ag.algebraic-geometry, at.algebraic-topology, ap.analysis-of-pdes, ct.category-theory, ca.classical-analysis-odes, co.combinatorics, ac.commutative-algebra, cv.complex-variables, dg.differential-geometry, ds.dynamical-systems, fa.functional-analysis, gn.general-topology, gt.geometric-topology, gr.group-theory, ho.history-overview, it.information-theory, kt.k-theory-homology, lo.logic, mp.mathematical-physics, mg.metric-geometry, nt.number-theory, na.numerical-analysis, oa.operator-algebras, oc.optimization-control, pr.probability, qa.quantum-algebra, rt.representation-theory, ra.rings-and-algebras, sp.spectral-theory, st.statistics, sg.symplectic-geometry

For more on how to tag you question effectively, consult this Tagging Guide (via meta.SO)

Should we include a similar advice (under the tag box)?

| |
$\endgroup$
3
$\begingroup$

Why should we require at least one arXiv tag for each question?

(I am aware of the same rule at MathOverflow, but I do not know why they have that rule, either.)

| |
$\endgroup$
  • 1
    $\begingroup$ I think the idea is to make it easier to search the site. If you're interested in [pr.probability], you can just search for that tag, and not also [stochastic-processes], [ito-calculus], [statistical-physics], [martingales], etc. It seems a little less useful here; I don't know if this is because the arXiv tags are less well organized for theory, or because TCS is a smaller field, or both. $\endgroup$ – Peter Shor Nov 23 '10 at 22:59
  • 1
    $\begingroup$ Actually, I think this is a very good question. The cs arXiv tags here are seem to be less well organized for classifying questions in TCS than the math ones are. We possibly could consider creating our own high-level tags (Suresh notices that there's no good high-level tag for [quantum-computing], and the logical one, [quant-ph], is confusing because it doesn't have a two-letter abbreviation attached), but we should be very careful about how we do this. $\endgroup$ – Peter Shor Nov 24 '10 at 0:40
  • $\begingroup$ @Peter: Thank you for the explanation and the idea. I agree that having a coarse and solid categorization is very useful if done correctly. I share the feeling that arXiv tags seem to be less useful on cstheory than MathOverflow. I also agree that our own high-level tags would be useful but are probably not easy to do it right. $\endgroup$ – Tsuyoshi Ito Nov 24 '10 at 0:50
  • $\begingroup$ Clearly, the way to figure out new high-level tags is via discussions on meta. I don't see why it wouldn't work to do it incrementally, starting with fixing what needs to be changed most. We might want to wait a little while until things that need to be fixed become clearer. $\endgroup$ – Peter Shor Nov 24 '10 at 3:30
  • $\begingroup$ I think the decision about using arXiv tags was made before I started using meta. Check the post arXiv vs non-arXiv tags. I haven't checked the original idea for tags on MO or SO. For me the tags are mainly organizational, and the reason for having them is similar to having subject classification and keywords in papers and I think the tags on arXiv serve the same objective. It might not be very big deal right now but it becomes important to have these categories as we get more and more questions. You can ignore them if you don't like them but I think it is useful to have them. $\endgroup$ – Kaveh Nov 24 '10 at 4:23
  • $\begingroup$ The top level tags clarifies the subject of a question and one can skip the questions from subjects he or she is not interested (for example on MO I skip almost all algebraic-geometry questions as I don't even understand the terms used in them so reading them is not really useful). But of course requiring one top level tag does not mean that we should follow the same tags arXiv uses. (It might be better to discuss that in the post I mentioned.) In fact we have already made some modifications to it, like [ds.algorithms] and [ds.data-structures]. $\endgroup$ – Kaveh Nov 24 '10 at 4:27
  • $\begingroup$ I think a good candidate for a top level [quantum-computing] tag would be [qc.quantum-computing]. $\endgroup$ – Kaveh Nov 24 '10 at 4:28
  • $\begingroup$ @Kaveh: By “the post arXiv vs non-arXiv tags,” do you mean the following discussion: arXiv tags vs not? If so, I had read that page but I could not find any reason to require the use of arXiv tags on that page. It seems to me that that page is more about how to name tags rather than which tags to require. $\endgroup$ – Tsuyoshi Ito Nov 24 '10 at 16:53
  • $\begingroup$ @Tsuyoshi: Yes, that was the one. I think I also saw another related discussion on tags but don't remember where. $\endgroup$ – Kaveh Nov 24 '10 at 16:56
2
$\begingroup$

Yes, I think this has already been decided, but we just haven't yet written the relevant FAQ entry.

Please feel free to fix it.

| |
$\endgroup$
  • 1
    $\begingroup$ MO has this text under the tag box on the ask question page, which I think is more effective. $\endgroup$ – Kaveh Oct 8 '10 at 12:42
  • $\begingroup$ @Kaveh: so your question is really about adding text under the tag box. $\endgroup$ – RJK Oct 8 '10 at 13:02
2
$\begingroup$

Just some thoughts. probably shouldn't be a top-level tag, because anybody who is asking a category-theory question on cstheory rather than math overflow should be asking a question related to computer science, and it would help us more to know which area of computer science it's related to, than just whether it involves category theory. Other tags borrowed from the Math arXiv, such as and seem fine to me. (Although maybe we should try to decide whether we want to keep the math arXiv tag or the CS arXiv tag or both.)

| |
$\endgroup$
  • $\begingroup$ I didn't mean we should list the top level math arXiv tags but rather our own top level tags. On [ct.category], I think there are category theory questions which are more suitable on cstheory than MO, category theory seems to be of high interest in (theoretical) programming languages community. About [co.combinatorics] ad [dm.discrete-math], probably we should merge them and make [co.combinatorics] a synonym for the other one (but my feeling is people like [co.combinatorics] more). $\endgroup$ – Kaveh Nov 24 '10 at 16:52
  • $\begingroup$ Yes, but I think any category theory question here should be labeled [pr.programming.languages] as well (if that's the relevant field of computer science). That is, we should not use the tag [ct.category-theory] but [category-theory] so it does not appear to be a top-level tag. Right now, the tag reads [ct.category-theory]. On the other hand, I wouldn't mind having a top-level tag [st.statistics] for this stack-exchange. $\endgroup$ – Peter Shor Nov 24 '10 at 20:16
  • $\begingroup$ It seems to me that category theory for PL is something like probability theory for (parts of) complexity. The PL people may ask category theory questions which are not really PL questions but arise in that context. My feeling is we should keep them, although I would personally prefer to ask a pure category theory or probability theory question fist on MO. $\endgroup$ – Kaveh Nov 25 '10 at 18:39

You must log in to answer this question.

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