Critical to any healthcare institution and research university is the availability of a viable dynamic library system that provides rapid access to the products of scientific discovery. Historically that access has been through the printed medium, but the electronic age has led to an evolution in the meaning of “access.” At the same time, costs of publishing "hard copy" and online versions of scientific journals and of subscribing to those journals have escalated and led to a crisis of decisions amongst librarians who seek to keep their resources available to their constituents. Both the public and the scientific writer have come more and more to expect that discovery will be available more proximate to the date of acceptance after peer review than had been possible when that availability depended on printed publication of the results. Thus, the whole concept of "access" has changed and all parties are seeking application of more modern means of providing access in the electronic age.
Our own library has bought into the concept espoused by the Public Library of Science (PLoS), which, with a $9 million grant from the Gordon and Betty Moore Foundation, has seemed to offer the latest version of a "free lunch". In fact, the Hardin Library for Health Sciences has for several months been advertising PLoS and encouraging all University of Iowa biomedical scientists to send their best work to PLoS. Such advertising and pushing one journal over others is unprecedented in my experience at this institution or elsewhere. Thus, it is my purpose in this editorial to look more closely at the free lunch and to examine what its forced feeding to the scientific public will do to other meals that are available (but may not be if PLoS is as successful as its benefactors would hope). My concern for this topic has been heightened because I am a member of the American Physiological Society, the Society for Neuroscience, the American Neurological Association, and the American Heart Association whose journals stand to suffer profoundly if there were a rush to a PLoS style of scientific publishing.
It is first important to take a brief look at business models that currently exist in scientific publishing. PLoS typifies the "author pays" model and will charge each author $1,500 for each manuscript published in that journal. Another model, probably best exemplified by commercial journals (for example, those published by Elsevier) could be called a "reader pays" model and recoups costs of publication by means of subscription costs incurred by individual and library subscribers. Yet another model is that of many society publishers who keep subscription costs lower than many commercial journals by charging fees both to authors and subscribers.
Next it is important to look at the contribution that publishers, regardless of their business model, may make to science. It is necessary to reflect on that contribution because some in the lay public rightly point out that the American taxpayer has already provided considerable financial support to NIH-funded research. Some in that public feel that their tax support gives them the right to immediate and free access to all scientific discovery that came at taxpayer expense. However, the publisher’s role in providing access to that discovery is far from minimal or meaningless. The publisher must ensure high quality peer review that is the hallmark of the best and most sought after scientific journals, must ensure the application of ethical standards in developing the data in each study, and must bear the expense of editing, formatting, and disseminating (electronically and/or in printed form) each accepted manuscript. While that cost may vary from one publisher to the next, the American Physiological Society (APS), which publishes 14 journals including the American Journal of Physiology, Physiological Genomics, and Physiological Reviews is probably representative of many scientific societies. The APS incurs a cost of approximately $3,000 per article published in its journals. In addition, the APS bears the cost and responsibility for maintaining an archive of publications that appear in its journals, itself a sizable and expensive undertaking but one that is vital for an historical archive of the field of American Physiology. Costs to such scientific society publishers are defrayed in part by modest ($50) author submission fees, page charges, color figure fees, and subscription fees. Advocates of PLoS emphasize its "open access" but do not readily acknowledge the hidden costs that make such access a possibility. Nor for that matter do they publicly recognize that many societal journals like those of APS already provide immediate or delayed free access to much of the material that they publish.
Beyond access, however, is the actual cost of PLoS. Clearly, author fees alone do not defray its publication costs. Built into the business model that drives PLoS is a "membership fee" that can be paid by institutions (libraries). With that fee, scientific members of that institution’s staff will be given discounts off the $1,500 author fee, not quite a free lunch but beginning to sound like a special deal till one looks more carefully at the membership fees. For PLoS those institutional membership fees will range from $2,000 to $100,000 per year even with the subsidy provided by the $9 million grant mentioned earlier. Either the membership fees or author fees will likely escalate over time as PLoS has to become fiscally sound as an independent entity. In fact, Vivian Siegel, Executive Director of PLoS indicated on July 24, 2004, at a meeting sponsored by the National Academy of Sciences that the $1,500 per article author submission fee was not a self-sustaining business model. Perhaps PLoS expects that the NIH or another federal funding institution will pick up the tab for publication costs at a later date. If so, that would result in fewer funded grants and less science to be published in PLoS or anywhere else. While the membership fees are a means of sustaining open access publishers such as PLoS, they begin to sound a lot like subscription costs. We should all look long and hard at the likelihood that future costs of open access publishing are bound to rise.
If not PLoS, then what could be a solution to escalating library costs and the need for more open access to scientific literature? The APS, together with 56 other scientific societies, is a signatory to a document entitled “Washington DC Principles for Free Access to Science". Full recitation of the principles themselves exceeds the limits of this editorial but the principles include the following that are already in place in the not-for-profit journals that participated in codifying the principles:
• Selected important articles of interest are free online from the time of publication;
• The full text of journals is freely available to everyone worldwide either immediately or within months of publication, depending on each publisher’s business and publishing requirements;
• Journal content is available free to scientists working in many low-income nations;
• Articles are made available free online through reference linking between these journals;
• Content is available for indexing by major search engines so that readers worldwide can easily locate information.
Further among the principles, the signatories affirm their "strong support for the principle that publication fees should not be borne solely by researchers and their funding institutions, because the ability to publish in scientific journals should be available equally to all scientists worldwide, no matter what their economic circumstances."
That principle stands in stark contrast to the author-pays model espoused by PLoS. There the affluent would be rewarded and the less affluent punished for their inability to pay. PLoS has said they would waive fees for authors who could not afford them. However, that seemingly laudable effort could evolve into dispensation for certain authors whose articles are deemed “worthy” of PLoS and could then be used essentially to buy articles for publication in PLoS.
The DC Principles further aver that as not-for-profit publishers, they "believe that a free society allows for the co-existence of many publishing models." The genesis of PLoS actually developed from an opposite philosophy promoted by Dr. Harold Varmus while he was Director of the National Institutes of Health. In 1999, while serving in that position, Dr. Varmus sought to promote a publishing enterprise supported by NIH with the intent of taking over much of the role played by peer-reviewed scientific journals at that time. That effort, E-Biomed, failed but was followed by several iterations on the same theme with the latest being PLoS.
As a scientific community we must examine this effort both for its positive aspects and its negative impact. Its greatest attribute is the promotion of advanced access; but, as noted, this is a principle that has already been promoted and put into place by many not-for-profit journals and is either being considered or has already been instituted by for profit publishers as well. Thus PLoS provides little gain. Its putative attribute of lowering costs to institutions vanishes upon inspection and is replaced by the likelihood that costs would actually escalate, not decrease, under a PLoS business model of publishing. It offers nothing in terms of optimized peer review, a principle that has been the foundation for most scientific journals and all of those of quality for over a century. The very real downside of a movement toward PLoS, or sons and daughters of PLoS, is the potential loss of societal scientific journals like those of the American Physiological Society, the Society for Neuroscience, the American Heart Association, etc. Loss of journals from those august societies would be an overwhelming loss that could never be replaced by PLoS. Science would suffer, the University of Iowa academic community would suffer, and patients denied the possibility of the open coverage of science that we now enjoy would suffer to extreme.
Let us advocate open access; let us advocate publishing that accomplishes advanced access but with the business model that works for that publisher, not one mandated by government; and let us avoid advertising, and thus taking an institutional position on one journal type, PLoS, over others that have served science well for years.
Dr. William Talman
Professor of Neurology and Neuroscience, University of Iowa
Chairman of the Public Affairs Committee, American Physiological Society
