The later part of this past week was dominated, for me, by discussions of the article published in Science about a “sting” operation directed against a small subset of open access journals that purports to show that peer-review is sometimes not carried out very well, or not at all. Different versions of a “fake” article, which the authors tell us could easily be determined to be poor science, were sent to a lot of different OA journals, and it was accepted by a large number of them. This has set off lots of smug satisfaction amongst those who fear open access — I have to suspect that the editors of Science fall into that category — and quite a bit of hand-wring amongst those, like myself, who support open access and see it as a way forward out of the impasse that is the current scholarly communications system. In short, everyone is playing their assigned parts.
I do not have much to say in regard to the Science article itself that has not been said already, and better, in blog posts by Michael Eisen, Peter Suber and Heather Joseph. But by way of summary, let me quote here a message I posted on an e-mail list the other day.
My intention is not so much to minimize the problem of peer-review and open access fees as it is to arrive at a fair and accurate assessment of it. As a step toward that goal, I do not think the Science article is very helpful, owing to two major problems.
First, it lacked any control, which is fundamental for any objective study. Because the phony articles were only sent to open access journals, we simply do not know if they would have been “caught” more often in the peer-review process of any subscription journals. There have been several such experiments with traditional journals that have exposed similar problems. With this study, however, we have nothing to compare the results to. In my opinion, there is a significant problem with the peer-review system as a whole; it is over-loaded, it tends toward bias, and, because to is pure cost for publishers, no one has much incentive to make it better. By looking only at a sliver of the publishing system, this Science “sting” limited its ability to expose the roots of the problem.
The other flaw in the current study is that it selected journals from two sources, one of which was Jeffrey Beall’s list of “predatory” journals. By starting with journals that had already been identified as problematic, it pre-judged its results. It was weighted, in short, to find problems, not to evaluate the landscape fairly. Also, it only looked at OA journals that charge open access article processing fees. Since the majority of OA journals do not charge such fees, it does not even evaluate the full OA landscape. Again, it focused its attention in a way most likely to reveal problems. But the environment for OA scholarship is much more diverse than this study suggests.
The internet has clearly lowered the economic barriers for entering publishing. In the long run, that is a great thing. But we are navigating a transition right now. “Back in the day” there were still predatory publishing practices, such as huge price increases without warning and repackaging older material to try and sell it twice to the same customer, for example. Librarians have become adept at identifying and avoiding these practices, to a degree, at least. In the new environment, we need to assist our faculty in doing the same work to evaluate potential publication venues, and also recognize that they sometimes have their own reasons for selecting a journal, whether toll-access or open, that defy our criteria. I have twice declined to underwrite OA fees for our faculty because the journals seemed suspect, and both time the authors thanked me for my concern and explained reasons why they wanted to publish there anyhow. This is equally true in the traditional and the OA environment. So assertions that a particular journal is “bad” or should never be used needs to be qualified with some humility.
At least one participant on the list to which I sent this message was not satisfied, however, and has pressed for an answer to the question of what we, as librarians, should do about the problem that the Science article raises, whether it is confined to open access or a much broader issue.
By way of an answer, I want to recall a quote (which paraphrases earlier versions) from a 2007 paper for CLIR by Andrew Dillon of the University of Texas — “The best way to predict the future is to help design it.” Likewise, the best way to avoid a future that looks unpleasant or difficult is to take a role in designing a better one.
That the future belongs to open access is no longer really a question, but questions do remain. Will open access be in the hands of trustworthy scholarly organizations? Will authors be able to have confidence on the quality of the review and publication processes that they encounter? Will open access publishing be dominated by commercial interest how will undermine its potential to improve the economics of scholarly communications? If libraries are concerned about these questions, the solution is to become more involved in open access publishing themselves. If the economic barriers for entering publisher have been lowered by new technologies, libraries have a great opportunity to ensure the continuing, and improving, quality of scholarly publishing by taking on new roles in that enterprise.
Many libraries are becoming publishers. They are publishing theses and dissertations in institutional repositories. They are digitizing unique collections and making them available online. They are assisting scholars to archive their published works for greater access. And they are beginning to use open systems to help new journals develop and to lower costs and increase access for established journals. All these activities improve the scholarly environment of the Internet, and the last one, especially, is an important way to address concerns about the future of open access publishing. The recently formed Library Publishing Coalition, which has over 50 members, is testament to the growing interest that libraries have in embracing this challenge. Library-based open access journals and library-managed peer-revew processes are a major step toward address the problem of predatory publishing.
In a recent issue brief for Ithaka S&R, Rick Anderson from the University of Utah calls on libraries to shift some of their attention from collecting what he calls “commodity” works, which many libraries buy, toward making available the unique materials held in specific library collections (often our “special” collections). This is not really a new suggestion, at least for those who focus on issues of scholarly communication, but Anderson’s terminology makes his piece especially though-provoking, although it also leads him astray, in my opinion. While Anderson’s call to focus more on the “non-commodity” materials, often unique, that our libraries hold is well-taken and can help improve the online scholarly environment, I do not believe that this is enough for library publishing to focus on. Anderson’s claim that by focusing on non-commodity documents will allow us to “opt out of the scholarly communication wars” misses a couple of points. First, it is not just publishers and libraries who are combatants in these “wars;” the scholars who themselves produce those “commodity” documents are frustrated and ill-served by the current system. Second, there is very little reason left for those products — the articles and books written by university faculty members — to be regarded as commodities at all. The need for extensive investment of capital into publishing operations, which is what made these works commodities in the first place, was a function of print technology and is largely past.
So libraries should definitely focus on local resources, but those resources include content created on campuses that has previously been considered commodities. The goal of library publishing activities should be to move some of that content — the needs and wishes of the faculty authors should guide us — off of the commodity market entirely and into the “gift economy” along with those other non-commodity documents that Anderson encourages libraries to publish.
If libraries refocus their missions for a digital age, they will begin to become publishers not just of unique materials found in “special” collections, but also of the scholarly output of their constituents. This is a service that will grow in importance over the coming years, and one that is enable by technologies that are being developed and improved every day. Library publishing, with all of the attendant services that really are at the core of our mission, is the ultimate answer to how libraries should address the problem described only partially by the Science “sting” article.