Online Scholarly Communication
December 21, 2012
Today I received an invitation to review an article from an unfamiliar journal having something to do with “artificial intelligence and computational complexity”. As a geographer specializing in spatial patterns of disease, this seemed like quite a stretch. The invitation was also replete with misspellings and oddly formal language. A visit to the journal web site revealed a willingness to publish papers generated by Scigen, a random paper-generating algorithm developed by some MIT students a few years ago.(1)
While this was among the more extreme examples of “scientific spam” I have received, items of this sort have been appearing in my inbox almost daily, extending to invitations to attend conferences, republish old papers, join editorial boards, and even to start my very own journal. Many who are reading this have undoubtedly had similar experiences; most of you have probably given them no further thought upon pressing the delete key. But these messages have piqued my curiosity, and there is more of a story here than just simple attempts to deceive and defraud. In fact, there is a very wide gradation of marginal publishers out there, ranging from outright scam artists to those with a modicum of hard-earned credibility, most of which are based in non-English speaking developing nations . A list of such publishers is maintained by Jeffrey Beall, a librarian at the University of Colorado.(2) The existence of this list raises important questions about the overall state of academic publishing. Specifically, it represent the dark undercurrent of the Open Source publishing movement. With Open Source publishing, authors retain copyright of their work and the content is freely accessible to all. Open Source publishers make their revenue from fees paid by the author, which typically amount to several thousand dollars per article, as opposed to the traditional model where revenues are generated from sales and subscriptions. But despite its noble intentions of enhancing the sharing of information, reducing publication costs, and empowering authors over publishers, Open Source led to some unintended consequences.
Consider, for example, the options available to one of my graduate students looking to submit her first paper after defending her dissertation. She is reluctant to send to traditional journals from traditional publishers, because she feels – even at some potential detriment to her long-term career path - that as the creator and author of her work, she is entitled to its copyright.(3) She is also sympathetic to the fact that our home institution is shedding thousands of traditional journal titles annually because of spiraling costs. Adding to her burden, I told her of my recent experience where two potential authors for a book project I am working on had to drop out after they could not obtain the rights to their own work at reasonable cost. Open-source journals are therefore much more to her taste, but publication fees here are spiraling as well, and she lacks grant support at present. It is true that many open-source journals offer a “hardship” exemption, but she is sheepish about claiming this. Certainly from an institutional perspective, it is awkward when its employees must claim hardship exemptions to publish the work they were paid to do. Suddenly a $300 fee for an unknown but not-obviously-fraudulent journal originating in Egypt or Indonesia begins to sound like a plausible option.
There are other models available, but none have yet to gain much traction in geography. In the natural sciences, there is ArXIV (pronounced archive, arxiv.org), a repository of over seven hundred thousand self-submitted articles with limited filtering mechanisms to weed out obviously fraudulent or unscientific content . Many of these articles are later submitted to peer-reviewed journals, but this step is increasingly bypassed. There are also journals entirely managed and published by expert volunteers, such as the Journal for Spatial and Information Sciences (www.josis.org). A volunteer-run journal may sound daunting, but consider that “volunteers” already do all of the writing, vetting, peer review, solicitation of reviewers, and much of the editing. That mainly leaves layout, design, and web publication, all of which can be largely automated.
Our intellectual and career development depends on our ability to share our findings with like-minded scholars. I personally do not care exactly how this takes place, as long as the process is fair and open and maintains scientific integrity. As I have touched upon here, these qualities are showing signs of erosion on multiple fronts. It is in our disciplinary interests to remain abreast of these trends and consider creative solutions that ensure the continued vitality of scholarly communication. I welcome any feedback or comments from the AAG membership along these lines.
Research Scientist New York State Cancer Registry
1 SCIgen - An Automatic CS Paper Generator . http:// pdos.csail.mit.edu/scigen.
2 Beall J. Beall’s List of Predatory, Open-Access Publishers. http://metadata.posterous.com/83235355.
3 For an eloquent defense of this reasoning, see Aaronson, S. Review of The Access Principle by John Willinsky. ACM SIGACT News 2007; 38(4): 19-23. www.scottaaronson.com/writings/journal.pdf.