This question obviously applies to not only mathematics, so maybe I should post it on academia.SE or somewhere else; but then again, mathematical literature has its own specifics related to existing sources of mathematical data on MR, Zentralblatt, arXiv, etc., so I still decided to post it here.

I've got, by now, around 4gb of various mathematical texts in electronic form (pdf, djvu, ps files of papers, books, talk slides, etc.). Clearly I try to keep them under control by sorting them out in subfolders, but sometimes it is hard to figure out where a specific file belongs. So it would be extremely helpful to use some existing classification like the AMS one. Even if none of the existing classifications suits my tastes completely, it is very convenient to use choices made by somebody else already instead of thinking about such choice everytime by myself.

Does anyone know any existing software that would facilitate such a thing? That is, given 4gb of files, look up the titles, tie them to some online classification source, and create a searchable queryable (?) database.

  • 1
    $\begingroup$ en.wikipedia.org/wiki/BibDesk#Organizing $\endgroup$ – Francois Ziegler Nov 3 '18 at 9:07
  • 1
    $\begingroup$ Not exactly what you are asking but I use the papers software. It can look up info from mathscinet and arxiv. You can group papers for various projects, and it can create bibtex files in a convenient way. It is not perfect, but much better than maintaining folders by hand. $\endgroup$ – Thomas Rot Nov 3 '18 at 9:11

I would try out Papers. You'll need to build your own work flow, it's not like you can throw in the entire library and come back later for a complete classification database, but the software will handle the bookkeeping for you.

It can read both pdf and djvu formats and will automatically query a variety of repositories (ACM, ADS, Google Books, Google Scholar, JSTOR, MathSciNet, Project Muse, Scopus, Web of Science) for bibliographic information. The database is fully searchable.

A possible work flow might go like this:

  1. import the entire library into Papers
  2. have Papers search the MathSciNet repository, and other repositories of choice, to automatically add metadata to each paper
  3. add key words to each paper by hand
  4. define sets based on the AMS classification scheme
  5. add papers to sets based on key words

Somewhat time consuming, true, but not too tedious I would think.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.