Follow us on


The 'free access' debate

Print edition : Jan 30, 2004 T+T-

Open access to scientific journals is beneficial to scholars and has wide support as a concept, but it needs viable revenue models and great commitment among its promoters.

LAST October, two biologists at the University of California, San Francisco (UCSF) circulated an e-mail asking biologists all over the world to boycott all journals of Cell Press, which publishes some of the most prestigious journals in biology including Cell, Molecular Cell and Neuron. The call was to protest against the high charges of online access to these journals. According to the researchers, Elsevier, the owner of Cell Press, charges $90,000 a year to provide access to six Cell Press journals over and above the $8 million that the university pays for e-access to its other titles. Last heard, while Elsevier has not relented, other universities are following suit.

According to reports, Cornell University is set to cancel its subscriptions to several hundred scientific journals published by Elsevier in January. Harvard University and the University of California, Santa Cruz (UCSC) too have taken steps to cancel their subscriptions. The current practice of many large publishers - true of Elsevier as well - is to offer `bundled' subscriptions that cover a large number of its publications along with e-access. Even though the bundling includes a very large number of indifferent and low-impact journals, institutions around the world, including many in India, have entered into such agreements with Elsevier to gain access to a large number of journals. Elsevier publishes over 1,200 titles, double as many as its close competitor. But the price rise in these bundling schemes seems to have outpaced inflation, resulting in such drastic moves.

Similar moves are evident in the United Kingdom as well. A House of Commons committee was constituted recently to conduct an inquiry into scientific publishing and the growing academic backlash as a result of increasing cost of journal access. Specifically, the committee will look into the fast emerging concept of "open access" to journals online, which would allow anyone to access electronically scientific literature free of cost and into whether the British government should support the trend towards free access to information arising from public-funded research.

The timing of these developments suggests that the moves derive their impetus from the launch of PLoS Biology, an open-access online journal, by the Public Library of Science (PLoS), a non-profit organisation committed to making scientific and medical literature a freely available public resource. It has been spearheading the campaign for open access to research journals in biomedical sciences since 2000. PLoS' founders include Nobel laureate Harold E. Varmus, former director of the National Institutes of Health (NIH) and currently president of the Memorial Sloan-Kettering Cancer Centre in New York; the inventor of the DNA-microarray Patrick Brown of Stanford University, California; and computational and evolutionary biologist Michael Eisen of the Lawrence Berkeley National Laboratory, California.

PLoS came into the limelight when in October 2000 a group of leading biologists, led by Varmus, circulated an open letter resolving that, beginning September 2001, the signatories of the letter would boycott journals that did not grant unrestricted and free online access to their research articles within six months of their initial publication date (Frontline, February 1, 2002). The signatories numbered more than 30,000 from 180 countries - although perhaps only a few of them have kept their pledge of boycott finally. And since most journals did not accede to PLoS' call for open access, PLoS decided to become a publisher itself. With start-up money to the tune of $9 million from the private Gordon and Betty Moore Foundation, the first online issue of PLoS Biology came out on October 13, 2003. Interestingly, the Editor of PLoS Biology is the former editor of Elsevier's Cell, Vivian Siegel. PLoS plans to bring out PLoS Medicine in 2004.

PLoS is not the first to publish an open-access journal. Since 1999, the London-based BioMed Central (BMC) has been serving as a portal for free access to electronic versions of biology and medical journals, which today number nearly 150. Also, publishers like Oxford University Press and Britain's Institute of Physics have been experimenting with open-access journals. Besides the fact that BMC is a commercial enterprise, most of the journals it hosts are low-impact and low-visibility journals and, except for its publication Journal of Biology, others have yet to attain the status to attract high-quality research papers. PLoS Biology has set its aims high even with the first issue by trying to compete with elite, high-end biology journals such as Nature, Cell, Science and New England Journal of Medicine.

"We take on a new role as publishers to demonstrate that high-quality journals can flourish without charging for access," write Brown, Eisen and Varmus in the editorial in the inaugural issue of the journal. "Our aim is to catalyse a revolution in scientific publishing by providing a compelling demonstration of the value and feasibility of open-access publication. If we succeed, everyone who has access to a computer and an Internet connection will be a keystroke away from our living treasury of scientific and medical knowledge," they state. "Freeing the information in the scientific literature from the fixed sequence of pages and arbitrary boundaries drawn by journals or publishers... opens up myriad new possibilities for navigating, integrating, mining, annotating and mapping connections in the high-dimensional space of scientific knowledge."

Indeed, the benefits of open access to results of research cannot be denied. As Varmus and his associates point out, open availability and freedom to use the complete archive of published DNA (deoxyribonucleic acid) sequences in databases like GenBank, EMBL and DDBJ resulted in transforming a collection of individual gene sequences into something far richer. If the decision to deposit published raw sequence data in an openly accessible central repository had not been made in the 1980s, this would not have been possible. More recently, the opening of access to the entire human genome sequencing data has enabled rapid and deeper analyses to bring the biological community closer to understanding the nature of human life in a holistic and integrated fashion.

Publishing a scientific journal, especially of the quality that PLoS Biology aims for, is a costly affair. Editorial staff, administration, the peer review process, quality production standards and so on cost a great deal. So who pays for PLoS Biology? The business model of the traditional print journals (which now provide online access as well) is based on subscriptions and toll charges for site access. But this is fundamentally incompatible with the PLoS philosophy of open access and so PLoS Biology has turned this `reader pays' model on its head by levying a `dissemination fee' of $1,500 a paper if accepted for publication. BMC too charges $500 to publish a paper. It recoups its costs through charges for access to value-added products of research publications, like 1,000 Authors, which is a compilation of the 1,000 best research papers published in its journals. The question is whether this `author pays' model is viable and sustainable.

PLoS has argued that the "relatively small" cost of unrestricted dissemination of a paper, and hence its visibility and impact, should be treated as one of the fundamental costs of doing research and should be an integral part of the funds given by the research supporting agency, university or the host institution. The authors of the editorial estimate that the cost of publication amounts to less than 1 per cent of a research project funding. This cost will not be a new expense to the authors' institutions, which will not have to pay any longer the high costs of publishing through subscriptions. "By simply changing the way we support the scientific publishing enterprise, the scientific community and public would preserve everything we value in scientific publishing and gain all the benefits of open access," PLoS argues. Already several funding agencies that have announced policies supporting open access include private funders such as the Howard Hughes Medical Institute in the United States, the Wellcome Trust in the U.K., the Max Planck Institute in Germany and the Centre National de la Recherche Scientifique (CNRS) in France.

But there has been considerable scepticism about the long-term sustainability of the open-access journal based on the `author pays' model. PLoS' main contention that open-access publishing would cost significantly less than the current system is still in the realm of debate. Ira Mellman, Editor of Journal of Cell Biology, published by Rockefeller University, has opined that that the PLoS cost estimate is four to six times lower than the actual cost. In a journal aiming to attract the best of research in the field, about 90 per cent of papers submitted would be rejected, thereby implying a high cost for editorial work and peer review.

According to Martin Blume, the Editor-in-Chief of the American Physical Society, which brings out the highly recognised physics journals Physical Review, Physical Review Letters and Reviews of Modern Physics, staff requirements and peer review account for a major portion of a journal's costs and bringing out only an online edition of a journal will at best bring down the cost by 20 per cent. From the perspective of a developing country like India, $1,500 (about Rs.70,000) a paper is still very high and, assuming an average project funding of between Rs.25 lakhs and R.30 lakhs, the publication cost works out to much higher than 1 per cent.

Journal of High Energy Physics

The correct perspective for the present has perhaps been voiced by the Association of Learned and Professional Society Publishers: "... the various proposals... raise complex economic, logistical and sociological questions... Much more information needs to be gathered through experimentation and analysis. ALPSP, therefore, welcomes the establishment of journals with different economic models for open access in order that the benefit to scholars and the long-term stability and viability of these models can be assessed."

But this open-access publishing road to access freely all scientific literature will be a painfully long process, feels Stevan Harnad, a Professor in Cognitive Sciences at Quebec University in Montreal, Canada, and a champion of the self-archiving initiative. Self-archiving refers to scientists depositing their research in open electronic archives, which could include one's own website. When these archives conform to standards created by the Open Archives Initiative, then search engines and other tools can treat the separate archives as one and the archives become interoperable. That means the contents of all such archives can be harvested, integrated, navigated and searched seamlessly, as if they were all in one global "virtual" archive.

The practice of widely circulating pre-prints in the physics community is a very old one. With the coming of the Internet, this assumed its electronic version in the form of an e-print archive. Therefore, from the perspective of the physics community, quite used at home with the pre-print culture, open access is not such a serious issue. But other disciplines, notably biological sciences, have been slow or even reluctant to adopt this practice. This has perhaps only become worse with issues such as intellectual property rights (IPRs) looming large over biological research, which is increasingly becoming corporate-owned. Self-archiving initiative is an effort to convince all scientists to adopt the e-print archive mode to disseminate one's research.

Consider the following statistics that point to the formidability of the task of moving from `fee access' to `free access'. There are currently 24,000 research journals worldwide, publishing about 2.5 million articles a year. Of these, about 600 are open-access journals, publishing about 75,000 articles a year. "So, what about access to the 2.425 million articles, for which there is no suitable open-access journal today? Should researchers wait for 23,400 more open-access journals to be created one by one?" asks Harnad. The faster track to access all these freely is for the authors of these 2.425 million papers to self-archive them in their own institutions' website, he says. Fifty-five per cent of all research journals, according to him, while all not yet ready to become open-access journals, are ready to support self-archiving by the authors of the papers appearing in them. "There are already three times as many articles that are made open access every year through self-archiving than through open-access publishing," he says. "So why are we talking only about open-access journals and why is open access being identified with open-access journals only?" he asks.

Self-archiving has one inherent shortcoming. That is, how is one to judge the quality of e-prints posted on the archive. Scientists prefer to read refereed papers rather than un-refereed ones. While this is the case in physics, where the readership consists of groups of specialists and the pre-print culture has made readers judge for themselves, this is unlikely to work in life and medical sciences. There is thus the need for a quality filter, or a peer review service, for papers being hosted on an e-print archive. So one still needs publishers of journals who will not publish but provide the services of peer review and editing and hand the papers over to be posted on a self-archive, much like a midwife handing over the babies after delivery. But this `midwife' model of e-publishing will also entail costs, and that too substantial, because, as mentioned earlier, it is the peer review and editorial work that costs a fair share for a publisher. While self-archiving is an attractive proposition, and will cost comparatively much less, the issues of quality filtering and peer review and paying the publisher for those services in a viable revenue model need to be addressed and solutions sought.

Nevertheless, there is clear evidence of the open-access movement catching on in recent times. The year 2003 particularly witnessed several open-access initiatives worldwide, and the concept has been articulated in various international fora, including those of the United Nations, the most important being the recently concluded World Summit on Information Society (WSIS). One of the first global statements was the Budapest Open Access Initiative (BOAI) by the Open Society Institute founded by George Soros. The statement was originally signed by 16 scientists from various disciplines in February 2002. It set out the basic definition of open access and advocated the twin complementary strategies of `self-archiving' and `open-access journals'. The statement said: "Achieving open access will require new cost recovery models and financing mechanisms, but the significantly lower overall cost of dissemination is a reason to be confident that the goal is attainable and not merely preferable or utopian."

More recently, in June 2003, the Bethesda Statement on Open Access Publishing was released following a meeting of some of the key backers of open-access publishing in the U.S. The statement endorsed the principles of the open-access model and recognised that "publishing is a fundamental process of the research process and the costs of publishing are a fundamental cost of doing research". The scientist-signatories indicated their support to open access by agreeing to publish selectively in, reviewing for and editing for open-access journals and journals that are effectively making a transition to open access. In October 2003, representatives of major European research institutes met in Berlin and issued a declaration - called the Berlin Declaration to Knowledge in the Sciences and Humanities - in support of open-access publishing of scientific and scholarly research. The declaration, which has been now signed by the German, French, Belgian and Greek scientific agencies, and countries like Croatia are likely to follow suit, is seen as a milestone in the campaign for open access.

"This is a very important step in favour of open access that is likely to trigger a paradigm change all over the world concerning scientific publishing," says Francis Muguet of the Ecole Nationale Superieure de Techniques Avancees (ENSTA), Paris, who chaired the Civil Society Working Group on Scientific Information as part of the WSIS. In a kind of culmination of the above initiatives, the WSIS of the U.N. - the first part of which concluded in Geneva on December 12 and whose second part will be held in Tunis in November 2005 - adopted a broad set of principles and a plan of action that included the Summit's endorsement of open access to scientific literature. Signed by 176 countries, which participated in the summit, the documents constitute a guide to the future development of information and communication technologies (ICTs).

The Declaration Principles (para 28) states: "We strive to promote universal access with equal opportunities for all to scientific knowledge and the creation and dissemination of scientific and technical information, including open-access initiatives for scientific publishing." Apparently, even the inclusion of this was not straightforward. Given the fact that the concept of `open-access' is still debated and there is no consensus within the scientific community or a proven model, earlier draft versions had contained only a passing reference to the concept. However, a sustained campaign to incorporate it in stronger terms by `open access' activists over the past year resulted in its explicit reference in the Principles. The breakthrough was achieved, according to Muguet, on November 3 during PrepCom 3A to the Summit, thanks to Croatian support.

The corresponding Plan of Action statement (para 10(i)) says: "Encourage initiatives to facilitate access, including free and affordable access to open-access journals and books, and open archives for scientific information." Interestingly, however, in para 22(b), it is not as categorical. It says: "Promote electronic publishing, differential pricing and open-access initiatives to make scientific information affordable and accessible in all countries on an equitable basis." This, the activists, see as a "diluted" or "half-baked" statement. Muguet would like to see this rewritten. The lack of consensus was reflected in the position taken by leading scientific organisations, including the United Nations Educational and Scientific and Cultural Organisation (UNESCO), the International Council for Science (ICSU), the Third World Academy of Sciences (TWAS) and the European Centre for Particle Physics (CERN). In their joint submission to the WSIS, they have spoken of "equitable access" only.

In developing countries like India, however, no initiative towards open access is evident. "There are clear and simple arguments that open access has many advantages and yet many developing country scientists are not embracing it," points out Subbiah Arunachalam, a strong advocate of open access at the M.S. Swaminathan Research Foundation (MSSRF), whose research has shown that open-access papers are cited more often than restricted access papers. "Institutions are not even promoting institutional e-print archives. I want the government, which funds more than three-fourths of research in the country, to insist that those who receive funds from the government should make their results available through open-access archives and open-access journals," he says.

A more important issue in the Indian context is the following. While the journals of the Indian Academy of Sciences, which are of international quality and are all open access, there are a whole lot of indifferent journals coming out from other academies and institutions. "The time may be opportune... to revamp science journals in India... in order to enhance their competitiveness on the international scene," P. Balaram, Editor of the journal Current Science, pointed out in a recent editorial. "Issues of access will be irrelevant unless there is a concerted attempt at quality improvement," he wrote.