I agree that the compulsory author-pay model is appalling, it undermines our core academic values. There is, however, a seldom discussed alternative: that of voluntary author-pay. Editorial decisions do not depend on the author’s ability to pay. This method has been used for decades by SIAM and is now also used by the ACM, to no detriment to the quality of their journals; in fact, SIAM Journal on Computing and the J. of the ACM are the two most coveted venues for papers in theoretical computer science.
It is true that the journals of these societies are not open access, but they are low-cost, they cost far less than comparable journals published by Elsevier or Springer; and the example of these society-owned journals illustrates the point that voluntary author-pay does not necessarily corrupt the journal. For these (bottom-line-conscious) societies, journal quality is not a market issue; it depends on the editors’ intended prestige for the journal and their ability to attract top papers. I can even imagine an economic incentive to maintain top quality; that way, a larger fraction of authors is likely to possess grants that can pay the voluntary charges. I would suggest to our colleagues to support open-access journals that adopt the voluntary author-pay policy (assuming they have a reputable editorial board and a clear intention to maintain quality).
The voluntary author-pay model should not be confused with Springer’s beautiful sounding “open choice” model, one of the worst distortions of the “open access” idea. Under “open choice,” authors can buy open access to their papers, so you can have papers in the same issue of a journal one of which is open access and the other is not. I am puzzled how our academic community is willing to put up with such blatant discrimination based on the authors’ financial means.
I disagree with your statement that the only open-access options for your papers are the arXiv and some institutional repositories, none of which provide peer review. There is an increasing number of peer-reviewed free journals (free to readers as well as to authors) you can choose from (such as the Annals of Mathematics, the LMS Journal of Computation and Mathematics, the Electronic Journal of Combinatorics, …) All these journals would continue to qualify for the “free” designation if they solicited voluntary page charges.
The discussion of who really pays for an open-access journal tends to pay little attention to the fact that the bulk of the cost of subscription-based journals is and has always been born by the universities in terms of the time their employees spend as authors, referees, and editors of such journals. By any accounting, this hidden expense is far greater than the (considerable) cost of hiring a (part-time) editorial assistant and the maintenance of a website. So when we discuss the “distorted market” that is developing to spread the cost of open-access publishing, we are talking about a small fraction of that cost – the bulk is spread in exactly the same way as it always was. While maintaining the Annals of Mathematics is surely a non-negligible budget item at Princeton, the overall cost to the community is far less than the cost of say Inventiones, a Springer journal of comparable quality sold at an exorbitant price. And the Annals, being open access, gets far wider distribution. If our colleagues, ever eager to create new journals, could convince more universities to follow Princeton’s example, the community overall would benefit while our scholarly principles would remain intact.
In summary, I believe open access is compatible with the principles of scholarship; the question for us academics is not whether to pursue open access but how. Models promoted by the biggest publishers are inherently suspect but alternatives exist.
Laci (Laszlo Babai)
May 14, 2009