5:51 AM
There is 97181 questions on the site, 75586 of them has a top-level tag, which means that about 22% of the questions is missing a top-level tag.
If we add (graph-theory), (linear-algebra) and (set-theory), then it is 80315 and there are about 17% of questions with none of those tags.
Depending how seriously we take the recommendations on Mathoverflow Meta, one could say that at least one fifth or one sixth of the questions on MO is incorrectly/insufficiently tagged.
8
Here are some comments to start off on question 3. Feel free to edit. What are top level tags, and why should I try to use them? The top level tags are tags like nt.number-theory and at.algebraic-topology which begin with a two letter code. They roughly follow their arXiv counterparts, from whi...
17
The tags with a two-letter prefix are exactly those tags that correspond to a math subject class on arXiv. They serve as top-level tags. As a rule, each question should have at least one of these top-level tags. The formatting is mainly in place to distinguish them from usual tags; that it can ...
Looking at this chat message from François G. Dorais, it seems that he considers top-level tags one of the things which could help to keep the tagging system on MathOverflow in a good shape.
Oct 16 '17 at 4:06, by François G. Dorais
@MartinSleziak Yes, the tag management on MO is very different than on Mathematics. There are several reasons for that. For example, the systematic use of broad area tags borrowed from the arxiv on day one brought a lot of stability to the general tagging system here. Mathematics explicitly decided not to use such tags in the early days, which I think was a wise choice since a great deal of users on Mathematics are unfamiliar with the arxiv classification.
Oct 16 '17 at 4:07, by François G. Dorais
In the early days of MO, there was a lot of tagging discussion which helped draw guidelines for what other tags might be useful and appropriate for the site. These days, however, only a handful of users are interested in tagging issues, probably since the system works well enough for everyone.
I should probably include also the links which I use to check the number of questions with top level tags. The links are rather long, so I can only post them as a multiline message. (Which has the disadvantage that the link is not clickable in the transcript and has to be copied and pasted if somebody wants to use it.)
https://mathoverflow.net/questions/tagged/ag.algebraic-geometry+or+nt.number-theory+or+co.combinatorics+or+at.algebraic-topology+or+gr.group-theory+or+dg.differential-geometry+or+pr.probability+or+rt.representation-theory+or+ct.category-theory+or+fa.functional-analysis+or+ac.commutative-algebra+or+lo.logic+or+gt.geometric-topology+or+ca.classical-analysis-and-odes+or+gn.general-topology+or+ra.rings-and-algebras+or+mg.metric-geometry+or+ap.analysis-of-pdes+or+mp.mathematical-physics+or+st.statistics+or+ds.dynamical-systems+or+cv.complex-variables+or+ho.history-overview+or+oa.operator-algebra…
8 hours later…
2:05 PM
Since I am mentioning links, I'll add also links which I use to find questions having deprecated tags and no top-level tag, like I did here.
I use links similar to the links in the post Which “non-arxiv” tags are de facto top-level tags?, I just change the main tag.
https://mathoverflow.net/questions/tagged/abstract-algebra+-ag.algebraic-geometry+-nt.number-theory+-co.combinatorics+-at.algebraic-topology+-gr.group-theory+-dg.differential-geometry+-pr.probability+-rt.representation-theory+-ct.category-theory+-fa.functional-analysis+-ac.commutative-algebra+-lo.logic+-gt.geometric-topology+-ca.classical-analysis-and-odes+-gn.general-topology+-ra.rings-and-algebras+-mg.metric-geometry+-ap.analysis-of-pdes+-mp.mathematical-physics+-st.statistics+-ds.dynamical-systems+-cv.complex-variables+-ho.history-overview+-oa.operator-algebras+-na.numerical-analysis+-sg…
So the pessimistic way of looking at the deprecated tags is to say that there are almost 900 questions with deprecated tags which have to be retagged at some point, if MO should get rid of them.
« first day (1892 days earlier) ← previous day next day → last day (2000 days later) »
Transcript for
Oct26
Oct '1827
Oct28
MO editors' lounge
Talk related to editing, tagging and related matters for the s...