Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.


    When I first joined MO about two years ago, I remember remarking the preponderance of algebraic geometry questions on the then fledgling site. I believe that this was perhaps even discussed either in meta or in comments to some of the earlier questions. The obvious correct answer was that this could be explained by the fact that the MO founders worked in this topic. I also remember thinking that perhaps this would even out eventually as MO penetrates the mathematical community. Two years on I think it is perhaps a good time to ask this question here in meta, not that it has anything to do per se with the running of the site. I hope you can indulge me :)

    I have been keeping track, not very scientifically, but just occasionally by clicking on the "tags" and seeing what the distribution looks like and I've noticed that the ratio ag.algebraic-geometry to mp.mathematical-physics has been constantly hovering around 10:1 for as long as I remember, whereas naively I would have expected the ratio to decrease with time. I'm not sure what to read in this. I am a mathematical physicist and yet I have not asked a single question tagged as mp.mathematical-physics. I suppose this could be explained by the fact that if I have a question on mathematical physics I usually ask people I know (or perhaps these days also on TP.SE) and I only ask on MO questions which seem to me to be likely answerable on MO.

    I am sorry to say that I have not kept track of other tags since, as I said, I was not conducting a scientific study: just once in a while keeping track of how many mathematical physics questions are on the site. (It is then natural to compare this with the top tag and hence that's why I have observed the constancy in the ratio.) But I am curious to known whether other people have remarked something similar for other tags and whether one can deduce anything from this about the penetration (or lack thereof) of MO in the mathematical community.


    The number and diversity of topology questions and answers has certainly picked-up. The arrival of the Thurstons and Goodwillie and such certainly helped with that.

    I suppose there'd be a good way to quantitatively do your tag test -- perhaps something like a month-by-month analysis of the tag density or questions-with-tags weighted by number of answers. This could be done with the dump.

    We've talked about this before in some regards. Subjects that aren't well represented here sometimes don't get the most kind welcome (numerical analysis for one) since the people vetting the questions tend not to be particularly sensitive to the nuances in the field. On the other hand, the massive battery of algebraic geometers here makes it seem like near any question in that subject area will get a fairly kind treatment.

    Another primitive measure is the tag badge page:

    Which seems to give a pretty clear indication that MO is primarily working well for algebraic geometry and number theory questions.

    • CommentAuthorWill Jagy
    • CommentTimeNov 7th 2011
    Jose, my expectation was that things would not really even out to the same ratios as real-world publishing pages or some other measure offsite. Those proportions took centuries to stabilize. There have been enough instances where people began on MO, asked or answered questions in a less-represented field, were dismayed by lack of response (or negative response) and gave up. So, in the medium run, a decade, say, I had in mind more of a PageRank equilibrium, number of posts on MO in a topic heavily influenced by the previous number of posts in that topic and the number of steady MO users interested in that topic.
    • CommentAuthorNilima
    • CommentTimeNov 8th 2011 edited
    I'm also curious about the relative distribution of activity between tags, and whether it evolves. My impression (without any data or formal rationale) is that in sub-disciplines which already have efficient mechanisms by which to get answers to technical questions, MO less a technical resource than a broad-education one. The difference is akin to that between asking one's collaborator a technical question and attending a colloquium by someone in a different area. One learns and is informed about activity in other fields in a colloquium. To get answers to something nitpicky in one's own field, efficiently and without fuss, one likely goes elsewhere.
    A blast from the past, before I noticed this whole meta thing:

    Steve, thanks! That was the post I remember reading, but couldn't find. I guess I was looking for it in meta and not on the main site.