I quit (from MathSciNet and ZbMath)

by Orr Shalit

(This is the post that I wanted to write this weekend.)

Several months ago I informed both MathSciNet as well as Zentralblatt that I would like to stop reviewing papers for these repositories. If you don’t know what I am talking about (your PhD thesis advisor should be fired!), then MathSciNet and Zentralblatt are databases that index published papers in mathematics, contains some bibliographic information (such as a reference list for every paper, as well as a list of papers that reference it), and, significantly, has a review for every indexed paper. The reviews are written by mathematicians who do so voluntarily (they get AMS points or something). If the editors find nobody willing to review, then the abstract appears instead of a review. This used to a very valuable tool, and is still quite valuable.

I quit because:

  1. I don’t have time for the voluntary work for free. This doesn’t mean that I don’t do any voluntary work for free – but since I don’t have time for this I have to be very picky about what voluntary work I do.
  2. This service is very useful for old papers that are hard to get a hand on, or that are written in a language that is not English. It used to be a very good way to stay up-to-date with works in the field. Today, the standard is that almost all papers are written in English and are available freely online. The actual added value of having this external review available is significantly lower than it used to be.
  3. I think that we, as a community, are not doing a good enough job of refereeing papers (I feel this as referee, author, and now also as an editor). I think that if we have some time that we are going to spend volunteering for reviewing papers, we shouldn’t split it up between refereeing and reviewing for databases. We should concentrate on refereeing, which is a crucial part of the mathematical eco-system, and not waste it on reviews, which are in a large part redundant.
  4. Reviewing papers has advantages also for the reviewer: it can discipline and focus the reviewer for staying up-to-date and working through current papers. However, in the current system, the papers are reviewed after they appeared in print (or online). This is ridiculously late. I do like to review papers some time, but the appropriate time to do this is after they appear as preprints on the arxiv. Then I can use my blog to post these reviews. Yes, this is not a standard platform, but nothing is perfect.