Public-Access Options Mulled

The White House Office of Science & Technology Policy (OSTP) would like to make all published results from federally funded research freely available online, but it hasn’t yet decided the best way to do that.

Advocates of such a policy argue that the public should have free access to federally funded research because the work was paid for by their tax dollars. Publishers, however, contend that subscriptions or some other source of revenue is necessary to offset the costs associated with the online publishing and archiving of journal articles.

The public-access debate heated up about two years ago when the National Institutes of Health began implementing a mandatory policy that requires all results of research funded by NIH to be deposited in the agency’s online database, PubMed Central, within 12 months of publication. Anyone with an Internet connection can access and search papers in the database free of charge.

The White House is now considering expanding that NIH policy to other federal agencies that fund scientific research. To get as much input as possible in a short amount of time, OSTP launched a six-week online discussion forum in December. The forum was part of the Obama Administration’s open government initiative, an effort to increase transparency, participation, and collaboration between the federal government and the American people.

Most of the discussion took place on OSTP’s blog, which gave stakeholders an opportunity to provide input and react to each other’s comments in real time. Some people, however, chose to submit comments by e-mail.

In general, the comments show that there is “clearly a lot of momentum in favor of increased public access,” says Rick Weiss, senior policy analyst at OSTP. “The question is not whether there is significant opinion in favor of it, the question is how to implement it and what the details would look like.”

Hundreds of comments were submitted by a wide range of stakeholders including academic and government researchers, librarians, publishers, students, teachers, and members of the public. Discussions focused on embargo times (that is, the length of time between publication and public release), concerns about copyright and intellectual property, the value publishers add to journal articles, business models that accommodate public access, and the impact of an enhanced public-access policy on journal publishers.

At about the same time that OSTP began to investigate expanding the NIH public-access policy to other federal science agencies, the House Committee on Science & Technology convened about a dozen stakeholders to examine the state of scholarly publishing and develop consensus recommendations on public access. OSTP was involved in those early discussions and encouraged the group, known as the Scholarly Publishing Roundtable, to formally submit their comments during OSTP’s public-access forum.

The roundtable released its report last month and submitted it to OSTP’s public-access forum for consideration. Two of the members, one who works for the open-access publisher Public Library of Science (PLoS) and another employed by the for-profit commercial publisher Elsevier, did not endorse the report, however. They both submitted their own comments to OSTP, showing just how challenging it is to reach consensus on the issue.

Groups in favor of expanding the NIH public-access policy to other federal scientific agencies applauded OSTP’s efforts to convene a robust, open discussion. In comments submitted to OSTP, the Scholarly Publishing & Academic Resources Coalition (SPARC), an alliance of more than 800 academic and research libraries, emphasized how enhanced access “will promote advances in science and technology, encourage innovation and discovery, and enhance the diffusion of knowledge throughout our society.”

SPARC has long advocated that all federal agencies that fund scientific research be required to make the results of that research freely available on the Internet in a timely manner.

But exactly what constitutes a timely manner is up for debate. Although SPARC is pushing for public access as soon as a paper is published, the group acknowledges that such a policy could have a negative impact on publishers that depend on subscriptions for revenue. In comments submitted to OSTP, the group wrote that “an embargo period of up to six months is an acceptable compromise.”

Other groups, such as the Scholarly Publishing Roundtable, suggest that some flexibility in terms of the embargo period may be in order to accommodate the needs of the different agencies and various fields of science. In the roundtable’s report, an embargo time of zero to 12 months is recommended.

Some publishers, however, say that even 12 months is not long enough for certain fields. “Our publications can have half-lives of seven or more years in terms of citations, though downloading of online versions significantly declines after two to three years,” J. Alexander Speer, executive director of the Mineralogical Society of America, commented on the OSTP blog.

OSTP recognizes that one size may not fit all in terms of the embargo time. “In the biomedical research world, the publishing wheel turns very rapidly,” acknowledges Diane DiEuliis, assistant director of life sciences at OSTP. “But in the social sciences, a paper may come out and be the only paper on that topic for two to three years.”

Several publishers of scholarly journals, including the American Chemical Society, which publishes 38 peer-reviewed journals and Chemical & Engineering News, are concerned that OSTP doesn’t seem to recognize that many publishers already have mechanisms in place to facilitate immediate public access to their published content.

For example, in 2006 ACS began offering open access to research published in its journals through the ACS AuthorChoice program. That program follows what is known as the author-pays model. According to the ACS website, authors pay $1,000 to $3,000 to make their paper accessible online without a subscription. Open-access journals, such as PLoS, use an author-pays model in conjunction with private donations to support publishing costs.

In 2009, ACS published 34,611 journal articles. Of those, 210 were made immediately available without a subscription via AuthorChoice, according to ACS. The lack of participation in the AuthorChoice program may suggest that researchers do not have sufficient funds to put their research out in the open, despite the fact that they may want to.

ACS is urging the federal government to provide funds to authors to participate in programs like AuthorChoice. Howard Hughes Medical Institute and the U.K.’s Wellcome Trust already allow “researchers they fund to use a portion of their grant funds to facilitate immediate open access to their published research through the ACS AuthorChoice program,” John P. Ochs, vice president of strategic planning and analysis in the publications division of ACS, wrote in comments submitted to OSTP.

Some people commenting on the OSTP blog agreed that the federal government ought to subsidize reasonable author fees for research that it supports. “If you use an author-pays model of publishing, the extra costs with publishing manuscripts are minimal compared to the research,” wrote Ivan R. Baxter, a computational biologist with the Department of Agriculture’s Agricultural Research Service.

Allyson Mower, scholarly communications and copyright librarian at the University of Utah, recommended that author fees for publishing be included in future grants. “It seems that the cost of publishing, disseminating, and preserving research results should be taken into account at the grant level. There is a cost, and that cost should become a line item in the allocation of awarded grants,” she noted on the OSTP blog.

Some people questioned what value publishers provide to journal articles. Publishers, including ACS, took the opportunity to clarify exactly what they do. “Publishers, such as the ACS, fund the infrastructure that enables the discovery, registration, certification, finalization, dissemination, and (most recently) preservation of research articles through peer-reviewed journals and the Web platforms that host them,” Ochs noted.

In some cases, scientific societies like ACS also provide interpretive material designed to make journal articles more understandable to broad nonspecialist audiences. Without such material, it is unclear how much public interest there would be in chemistry papers, for example.

The lack of interest in chemistry by the general public was pointed out in a comment by the Royal Society of Chemistry (RSC), which publishes 30 peer-reviewed journals. “Providing public access to this chemistry content will result in negligible usage by the public,” James Milne, editorial director at RSC, wrote in the RSC comment.

Milne questioned the true need for public access to chemistry research articles and “the value of implementing a potentially costly and damaging public-access policy for this field.” RSC and other scientific societies are concerned that such a policy will reduce library subscriptions and ultimately have a detrimental effect on established society publishers and their journals.

ACS estimates that 30–35% of the articles it publishes would be impacted if the NIH policy is expanded to all federal agencies.

The public-access policy team at OSTP is now trying to make sense of more than 500 comments and plans to share its analysis with an interagency working group. OSTP intends to move as quickly as possible to develop policy recommendations, Weiss says.

But it will take some time for OSTP to sift through all the comments. A lot of them “came with very substantive attachments, data sets, journal articles, and other kinds of additional information,” Weiss notes. Although feeling somewhat overwhelmed, he says that overall it was a positive process.

“There have been a handful of times where OSTP has turned to this mechanism of using an interactive blog as a means of getting a wide range of stakeholder input,” he tells C&EN. “This ongoing experiment in open government and collaboration with the public has been remarkably successful. We received huge amounts of useful information that would have taken any agency a long time to pull together.”