[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[xmca] Open Access journals

I just want to make clear that I'm all in favor of Open Access journals, under some financial models for supporting operating costs.

There certainly are some very well respected OA journals, of which PLoS ONE is probably the best known (in biology, started by senior researchers). So far as I know these are almost exclusively in the natural sciences, where most research is grant-funded and where granting agencies, government and private, are moving steadily to a policy of paying publication charges so that the articles will then be available free online to all.

That model assumes that funding agencies pay, and the costs can be substantial ($1000-2000 per article). In addition to PLoS, which publishes several journals of this kind, Springer publishing in the UK has a division that also uses this model, and makes a profit. I am not sure, but I think PLoS is essentially non-profit.

I can't see many social science researchers paying this kind of money to publish. ScienceDomain, whose quality is in question for now, is aiming at this market, but also and mainly at natural sciences, with its rate of $500 per article.

There is an alternative Open Access model, the repository model (also known as the Green OA model) in which you publish in a regular journal but the copyright agreement allows you to deposit a pdf of the article in an institutional repository, where it can be accessed online for free. Some government-sponsored research now requires a depository copy and the publishers are being forced to agree (though they are also lobbying in Congress against the policies). Of course you can also just put the pdf on your own website and let Google do the rest of the work.

It's not at all clear to me what costs for OA would add up to hundred or thousands of dollars per article. Certainly not the cost of uploading pdfs to the cloud. I also can't see justifications based on indexing costs (which have been claimed). If the journal has no subscriptions, but runs entirely on publication charges, then it's going to have some basic operating costs, even if it's not doing print publishing, but these journals do not pay peer reviewers, who do most of the work, and in general I don't think they pay editors very much. There are also obvious economies of scale to be realized; fewer journals publishing more articles would make the operating cost per article less.

If you operate 20 online journals, and each publishes 100 articles per year, even at $500 per article, you are generating $1 million a year in revenue. If you are based in India (and I am guessing this is true for ScienceDomain), that ought to buy a lot of copyediting, typesetting (really necessary anymore??), and cloud storage. Plus a small staff, small subsidies to editors, and the rest comes in free labor from us. It ought to be possible to run such an operation for a lot less than $500 per article.

Now it's a widely known secret that the average journal article is only read by about 2-3 people and most articles never get cited. That means that you can't rely on small per-download charges for adequate revenue. UNLESS of course you only publish really GOOD research, which tends to get read and cited by lots of people, each of whom could afford a modest charge like $1 to download. (yes, there are international issues; the current models mostly have sliding scale charges already)

The other major question is the quality of the review process. Looking at ScienceDomain for 2 journals (the BERJ clone and their Physical Review clone) they each have a couple of senior academics on their editorial teams, plus a lot of people who are very likely not qualified, and who certainly would not command the degree of widespread confidence by their respective research communities to be in charge of a journal. Editors choose reviewers and evaluate their reviews. Editorial Boards are supposed to remind the editor once a year to keep standards high (and we sometimes actually do reviews, too). It's a reputation system, and it can be overly exclusive and sometimes too "in-group" oriented, but it's a lot better than a system in which no one has any idea whether "peer-reviewed" means anything better than a random draw. Blind review is not supposed to mean that the blind are reviewing the blinded. Someone is supposed to be able to see quality differences.

But who pays? And how much?


Jay Lemke
Senior Research Scientist
Laboratory for Comparative Human Cognition
Adjunct Professor, Department of Communication
University of California - San Diego
9500 Gilman Drive
La Jolla, California 92093-0506

New Website: www.jaylemke.com 

Professor (Adjunct status 2011-2012)
School of Education
University of Michigan
Ann Arbor, MI 48109

Professor Emeritus
City University of New York

xmca mailing list