MSC vs. ArXiv (and some interesting info on mathjobs)

One of my pet peeves is how annoyingly the AMS’s math subject classification is for people working in quantum algebra and quantum topology. The MSC has 97 different major subjects and my field is not one of them, and instead appears many times a subheading. In the new 2009 classification there’s at least the following: 16T, 17B37, 18D10, 20G42, 33D80, 57R56, 58B32, 81R50, and 81T45. Here I’m only counting things that are obviously quantum algebra and quantum topology (for example I didn’t list subfactors, quantum computation, knot invariants, etc.) By way of contrast, on the ArXiv there are only 32 categories, yet one of them (math.QA) contains the vast majority of work in my field (of course, many of those are cross-posted).

This mini-rant of mine came up at dinner at an AMS meeting in Waco (more on the excellent “fusion categories” special session later). Someone pointed out an interesting side-effect of this issue that I hadn’t thought of. One of the awesome things about mathjobs is that rather than simply having a large paper stack of applications, the people on hiring committees can instead sort the applications automatically in many different ways. It makes a lot of sense that mathjobs has this feature, but none of us who were on the applying side of things had ever considered it. Here are a few examples of things you might want to search for: look at people applying from a specific school, find everyone who has a recommendation letter from Prof. X, and (relevant to this post) sort by AMS subject classification.

This means that choosing the right AMS subject classifications is actually somewhat important. If you choose poorly then someone who might be interested in hiring you might never actually find your application among the hundreds they’re looking through. So if you’re in a situation like mine it’s worth asking a professor or two which AMS subject classifications they’d be most likely to look through.

Since then I’ve been wondering whether it might be a useful for mathjobs that the data they ask for also include which arxiv classifications applicants have posted preprints under, as that’s the search that I would want to use if I were on a hiring committee. What do people think? Mathjobs is very responsive to requests, so if people think this makes sense I may send them an email.

19 thoughts on “MSC vs. ArXiv (and some interesting info on mathjobs)

  1. Actually, I think sorting by MSC is the canonical way of sorting MathJob applications. When I’m looking at (postdoc) applications, I’m not sitting down and reading 700 applications, rather, I sort them by MSC and go through all the 11’s (and maybe a few 14’s). I expect the majority of postdoc committees also divide the work of reading applications up using the MSC. If you feel that you fall in the middle of two main classifications, then you should really decide on a school by school basis which number you wish to be listed under. The goal, ultimately, is to have someone who is interested in your research read your proposal.

  2. I can’t speak to the relevance of MSC or arXiv classifications to job applications, but I have gripes with both classifications.

    For me, the arXiv classifications are useful for seeing the papers that you want to see in your RSS feed (I subscribe to six classifications), but I find the restriction on the number of cross-listings artificial. I don’t understand the comment on the arXiv that two listings are usually sufficient and more than three are frowned upon. Recently I posted a paper with a collaborator on the two lists where I would look for it, only to hear from my collaborator that he didn’t see the announcement because I hadn’t included it in the two lists where he would expect to find it. I would guess that this situation isn’t terribly unusual.

    As for the MSC, the only thing I’ve used it for so far is to fill out job applications (finding this frustrating for reasons similar to Noah’s)
    and to say what kind of papers I’d like to review (finding this frustrating because it is unclear to me where to place the papers that I normally read).

  3. Here’s the ArXiv’s explanation of why they don’t use the MSC classification. It’s a combination of two reasons: the MSC has several old fashioned topics which wouldn’t have enough preprints to sustain interest and the math.QA problem. So it’s possible that the lack of math.QA in the MSC is the only major problem that would be fixed this way.

  4. @3: Your advice of varying which MSC based on the school makes a lot of sense for people falling between two classifications (Number theory and algebraic geometry being a common one, or say algebraic geometry and combinatorics). I’m not sure it’d be useful for the .QA problem (though to the extent that it’s a topology vs. algebra question then it would be useful to vary by school).

  5. If you’re going to put enough thought into things to figure out which MSC class to use for each school, would it make more sense to just email the people at the school who might be interested in your resaearch?

  6. In an ideal mathematical universe, one classification
    scheme would serve all purposes. But arXiv started mainly to promote quick access to new work (e.g. physics,
    where written work has a brief shelf-life). The categories
    can and do adapt over time. MSC aims at long-term
    continuity in the MathSciNet and Zentralblatt databases,
    from 1940 on. Some items get mothballed over time,
    other new areas get somehow integrated without changing
    the overall structure. It’s a compromise, but does allow
    for many kinds of database searches, taking advantage of
    not just titles but also author summaries or reviews,
    reference lists. MSC is inherently conservative but is
    receptive to suggestions (some of mine like 17B08 and
    20F55 were implemented, others not).

    I find daily mailings from arXiv quite handy, centering
    on RT (plus some AG, GR, QA, CO overlap). But I find
    MathSciNet essential for other purposes. MathJobs could
    profitably allow for both kinds of subject-area listings.
    Anyway, having dealt with the old paper system while on
    search committees, I find MathJobs infinitely better for
    access anytime, anywhere. We used to fill four file
    drawers with paper in a secure room, but papers got lost,
    misfiled, etc. For our mid-sized department, everyone
    is invited to participate in hiring, but an elected committee
    decides. Everyone gets assigned to some MSC areas like
    11, 14, etc., but can look at anything else they want to.
    Some extra tags like QA and RT would probably help a bit.
    But nobody can look carefully at 600 files under time limits
    placed here by the administration on deciding interviews.

  7. Ben writes “would it make more sense to just email the people at the school who might be interested in your resaearch?”

    Just to give a data point, at every school I applied to, I e-mailed the faculty member who I thought was mostly likely to care about my research. If there was no reasonable faculty member, I didn’t apply.

    Now, this method makes a lot more sense for someone doing a targeted search, as I was, than to someone who sends out 80 applications. But, if that is you, I recommend you follow this strategy.

  8. Speaking for myself, I feel *incredibly* awkward sending emails like that, but in these days of the application pools for postdocs heading to 4 figures, it’s a sad necessity.

  9. Advisers and other faculty (some better known than
    others, to be sure) have a responsibility to write some of
    these emails to colleagues elsewhere. It’s not just up to
    the applicant. I still get emails from colleagues pushing
    their students, even after my formal retirement. It’s most
    important to contact people who could be mentors.

    There are many differences in the process of hiring people
    as postdocs or on tenure-tracks. As the system has
    evolved over decades, there are some research-only
    fellowships or institute visits, decided with rigid deadlines.
    There are some longstanding 2-3 years teaching postdocs at prestigious places, also with deadlines. There are
    even more ad hoc arrangements to hire 1-3 year visitors
    based on teaching needs. Salary, teaching load,
    duration are factors along with the time frame for
    hiring, based on institutional pressures. August,
    anybody?

    Tenure-track hiring at Ph.D.-granting universities is
    more structured, but unpredictable in timing. Then there
    are the 1000+ other kinds of colleges, universities. Not
    easy, but good luck to those of you caught up in all this.
    The good news is that most “mathematicians” really do
    want to hire the best people. The bad news is that there
    may not be enough “mathematicians” running the process.

  10. Advisers and other faculty (some better known than
    others, to be sure) have a responsibility to write some of
    these emails to colleagues elsewhere.

    Yes, but students shouldn’t count on this. Not all advisers have the same philosophy about this stuff.

  11. I strongly recommend the practice of emailing potentially interested people at universities you care about, and I applied it liberally myself. Faculty tend to be very good at ignoring emails, and can ignore yours easily if they want (I know I do…sorry!) So it’s not an imposition IMHO to write. On the other hand, seeing the process from the other side now, I can vouch that it’s a miracle when files DON’T fall through the cracks. There’s a small hiring committee, which most likely doesn’t include the people close to your
    interests in the faculty, and most likely no one will alert them to your file even if you explicitly mention people in your cover letter (unless the committee is intrigued by your file already) – at UT eg we had over 700 files last year for each of postdoc and tenure-track.. On the other hand if you let someone know, they can take a look at the file if interested, or mention it to a committee member, and then it has a chance. And as the “someone” mentioned anonymously in the post, I can vouch for the fact that some files were missed by me until a late stage due to being listed under silly MSC headings..

  12. Hi, thanks all for the interesting topic and discussion. We finally got
    some time this summer to get the ArXiv classifications implemented in
    the mathjobs. An applicant can now select a MSC and any of the
    second-level research interests (including ArXiv’s and those entered by
    a dept), for example, 13 and math.QA. A search committee member can
    sort by the MSC or search for both. Thanks again for the idea.

Comments are closed.