February 17, 2008

On eprints at Harvard and Full Monty open-access

I'm still trying to figure out the consequences of Harvard's Arts and Science faculty voting last week to push open-access publication of faculty work. This is fundamentally different from the occasional individual boycott of subscription-based journals. Harvard's faculty move is closer to Congress's push for a mandate that all grant-funded articles etc. be accessible to the public within a year of original publication. It is from these institutional moves that the publishing world will change. There is a simple, digestible explanation for the open-access moves related to grants (the public pays, so the public should be able to read) and the Harvard A&S faculty (we're established enough not to have to worry about the reputational economy of subscription journals). What flows from that is not necessarily clear, but we can reasonably assume that something will flow.

Reputational economies and the refereeing process

There are two broader issues here that need to be untangled. One is the reputational economy of academe, which is partly tied to the referee process and partly to post-publication reputational measures, such as citations. As physics has shown with arXiv, a discipline can survive quite nicely with a much fuzzier boundary between working paper and publication. But maybe that's because of the established reputation of physics. Similarly, I think history, classics, math, and other disciplines that have relatively high intellectual status (if not in resources) have nothing to fear from loosening up the refereeing process.

But what about other disciplines, including education? Education research already has a number of unrefereed publications that receive a lot of attention, largely because of differential access to publicity. Unlike medicine, where the top-reputed journals have publicists that distribute press releases (and you will see those regularly reported in the press), education has a different distribution of publicity. If you look at the indispensable Fritzwire, you'll see oodles of announcements for think-tank-based research symposia, and the ability to hire publicity folks does have an impact on what gets reported. As one colleague in another institution explained, when I asked why his work received far less attention in his area than the think-tank-based work of X and Y, which I thought was of lower quality, "Sociology departments don't usually hire publicists."

This is not to say that all think-tank-funded research is of poor quality, or that articles in refereed journals is of high quality: you don't know until you read the stuff. Nor am I suggesting that think tanks fire their publicists or stop doing the legwork to get attention. My point is rather that given the existing visibility of nonrefereed work in education, in addition to the status issues in education already, I suspect that faculty in education will be far more reluctant to let go of a peer-refereed model. Even though the notion of peer refereeing is historically and geographically bounded (see Einstein versus the Physical Review for one example), it is wrapped up in status issues. For Harvard's A&S faculty to vote for an open-access preference is one thing. For even Harvard's education faculty to go the same route? We'll see.

Economic models for open access

Since EPAA is described by John Willinsky as a "zero-budget journal," I'm living the tensions involved in open-access.  We don't charge either readers or authors for anything, though I have no compunction about asking authors to review other manuscripts as part of a reviewing ecology, and I've shifted the submission checkoff to alert authors that very long manuscripts or manuscripts with a number of tables may involve some paid preparation of an article post-acceptance. (I haven't yet asked authors to pay for such preparation, but it's a recent move.) Apart from the administrative issues involved, I am not philosophically inclined towards allowing advertising on EPAA. Maybe I should, but I and many editorial board members would be uncomfortable with that. But as a result, the burden of making the journal work is largely on volunteer labor, or labor borrowed from other tasks. Even if I were to accept advertising into EPAA, I suspect that we would not receive much revenue from it, and it may not be worth the headaches involved.

The most visible open-access journal system, the Public Library of Science, relies on publication fees charged to authors, starting right now at $1250. Here is the PLoS explanation of publication fees:

It costs money to produce a peer-reviewed, edited, and formatted article that is ready for online publication, and to host it on a server that is accessible around the clock. Prior to that, a public or private funding agency has already paid a great deal more money for the research to be undertaken in the interest of the public. This real cost of "producing" a paper can be calculated by dividing your laboratory's annual budget by the number of papers published. We ask that-as a small part of the cost of doing the research-the author, institution, or funding agency pays a fee, to help cover the actual cost of the essential final step, the publication. (As it stands, authors now often pay for publication in the form of page or color charges.) Many funding agencies now support this view.

For largely grant-funded disciplines, that's doable. For others? Not possible, either because an institution will not pay publication fees or because an author may be an independent scholar.

Here's the bottom-line concern: For journals in non-grant fields that are currently subscription-based and where there is paid staff who work on the journal, the transition to subscription-free work is fraught with risk, and I suspect that forcing all currently-operating journals to go subscription-free would result in the closure of hundreds of journals. I don't think anyone wants that to happen, but there is no secure economic model for open-access journals right now. We'll see the development of hybrids for some time (such as the Teachers College Record in education research), and that will work to some extent. And my guess is that a number of journals would have no problem with open-access for a substantial number of country-specific domains, to help scholars in countries that do not generally have institutional subscriptions to expensive journals. But that's different from the "Full Monty" open-access journal.

Where to go from here

Of the two issues, my guess is that the reputational-economy question is easier to answer. I suspect citation harvesting will be the basis of future reputation economies in academic publication. Google Scholar is incomplete and inaccurate, but so is ISI's Web of Science, and as long as academics don't treat bibliometrics as carved in stone, things should work out (or at least the problems are of a much lower magnitude than other problems we face). Unlike David Rothman, I do not see online comment forums and rating algorithms working, in part because few researchers can afford the time to invest in such forums or devices. For institutions that care about research, they will still use external reviews at promotion gates, and that will supplement other information.

The economic model of "full Monty open-access" is going to be harder to achieve. Maybe I should state what I would love, as an editor: for someone to figure out how to provide me great copyediting and compositing. Make it so I don't have the headaches of economic administration and post-acceptance detail work, and I'll probably swing towards accepting advertising or a sliding-scale manuscript-processing fee. That's going to be a bit of a challenge, since I have very particular ideas about how an article should look. But a clearinghouse that manages advertising, moderate manuscript-processing and publication fees, copyeditors and compositors, and has a quality-control mechanism for the copyeditors and compositors would do me a huge favor. And if this finicky editor will accept it, and if you can make it work economically, you just might make open-access work on a sustainable basis.

Listen to this article
Posted in EPAA on February 17, 2008 11:05 PM |