home page

Public Access to Public Research: A Radical Idea Grows Respectable

Historic monument: The November 1999 Molecular Biology of the Cell was the first complete issue of a scholarly journal placed into PubMed Central for online public access.
Historic monument: The November 1999 Molecular Biology of the Cell was the first complete issue of a scholarly journal placed into PubMed Central for online public access.

In Washington, DC, they broke out the bubbly February 22 at the Dupont Circle offices of SPARC, the Scholarly Publishers and Academic Resources Coalition. “A magnum of prosecco in the office,” laughs SPARC executive director Heather Joseph. “It was kind of fun.” The occasion for celebration was a directive issued that afternoon by the White House Office of Science & Technology Policy (OSTP) to all federal agencies that fund at least $100 million a year in research and development. The directive requires investigators funded by those agencies to make available for online public access within a year or so of publication all journal papers, including their data sets and supplementary material that were supported by taxpayer dollars. It was a milestone in a struggle over access to the scientific literature that began in the 1990s and in which ASCB played an important role.

Until the OSTP order, the only federal agency that required public access posting was the National Institutes of Health (NIH). As mandated by a 2007 congressional appropriations bill, NIH has required all grantees since April 2008 to deposit for online access through PubMed Central (PMC) an electronic copy of the final version of any published peer-reviewed paper that draws on NIH-funded research. The new OSTP directive will extend a similar mandate to 19 additional federal agencies. Included for the first time are the National Science Foundation (NSF), the Department of Energy (DoE), and the Department of Agriculture, all major players in biology research. Each must now come up with its own public access program.

The OSTP directive is a significant victory for the “open access” movement, says Joseph who champions the cause at SPARC and before that as Publications Manager for ASCB. “It’s been a long haul to get the concept of open access understood,” says Joseph, “and debunk the fears that grew up around it, which unfortunately many of the commercial publishers are happy to perpetuate. We had to get people to understand that you can have a healthy journal publishing market using an open access model, that subscription access can co-exist peacefully [with open access] but, given the choice, researchers tend to prefer to have as many of their colleagues as humanly possible get access to their work and for them to have access to the work of as many of their colleagues as humanly possible.”

The new directive from the Office of Science and Technology Policy extends the NIH’s 2008 pioneering open access policy to 19 additional federal agencies.
The new directive from the Office of Science and Technology Policy extends the NIH’s 2008 pioneering open access policy to 19 additional federal agencies.

Corks, however, were not popping all over the scientific publishing world in honor of OSTP. Many of the big commercial scientific publishers such as Macmillan (the Nature Group) had become more or less resigned to the new rules, especially after President Obama signed the 2010 renewal of the America COMPETES Act, which authorized the extension of public access to other federal agencies. The wide acceptance of public access to federally funded research was revealed in 2011 by a public relations disaster around a short-lived bill called the Research Works Act (RWA). Supposedly promoted by another commercial scientific publisher, the bill would have gutted the NIH open access program by defunding it. Instead, RWA provoked across-the-political-spectrum outrage from “information wants to be free” Internet techies to Macmillan Publishing.  

The new OSTP rule was greeted more warmly, albeit without the high spirits and with a splash of anxiety by other nonprofit scholarly journal publishers such as the American Institute of Physics (AIP) in College Park, MD. “Overall, I’m pleased with the balance and flexibility that’s indicated in the document,” says Executive Director and CEO Frederick Dylla. “I’m a little nervous because you’re dealing with a bureaucracy and you’re often not dealing with the same set of folks, year after year.” The current federal fiscal standoff and the OSTP directive that any new public access program be funded through existing allocations also pose questions about long-term stability in Dylla’s mind.

Dylla, who has been working in advance of the expected OSTP rule with DoE and NSF on a tagging system to identify federal funding on paper submissions, believes that these sort of “devil in the details” problems will make or break the wider federal public access policy. Many of the policy details of the OSTP directive are not widely appreciated, Dylla contends, especially since the news media have overplayed the open-the-gates aspects. “Unfortunately the headlines focus on ‘the government is going to open up the pay wall.’ But if you actually read the (OSTP) memorandum, it’s much more nuanced.”

First, OSTP acknowledges the real value that publishers add to scientific publishing, he says. The savings from dropping printing on paper are minor compared with the continuing expenses of managing the peer review process, setting editorial standards, performing the multi-layered typesetting required in modern HTML manuscripts, and building a 24/7 online access platform and permanent archive. “The cost varies from $1-4,000 per article depending on the journal,” says Dylla. “Somebody has to pay for that.”

What Dylla likes about the OSTP order is its encouragement for public and private partnerships plus its flexibility in implementing public access. For example, Dylla points out that the famous NIH rule of posting within 12 months is not ironclad in the OSTP plan. “If you read that section, it’s very carefully worded. It says agencies should use a 12-month embargo as a guideline and that agencies need to look at differences among fields and disciplines.” A 12-month embargo, says Dylla, “is not such a problem in a fast-moving, well-funded field like biomedicine but there’s plenty of evidence that for humanities, social sciences, or mathematics, a 12-month rule can be problematical.”

As ASCB’s publications director, Heather Joseph found that journal usage patterns were front-loaded: “The graph was a backwards hockey stick.”
As ASCB’s publications director, Heather Joseph found that journal usage patterns were front-loaded: "The graph was a backwards hockey stick."

The final devilish details are still months and months away, he says. “Is every publisher going to like what is put in place? Probably not. Will this move along as quickly as some people want? Probably not.” But Dylla believes that public access and peer-reviewed journal publishing must learn to co-exist. “This is one of the most important things that scientific societies do for science and for society.”

Whatever is to be read in the new OSTP ruling, it comes as a result of a long campaign toward public access in which ASCB made history. In the earliest days of the public access debate in 1999, ASCB was the first scholarly publisher to turn over the entire contents of a journal, the November 1999 issue of Molecular Biology of the Cell (MBoC), to NIH for free online access through PMC. The Proceedings of the National Academy of Science (PNAS) signed on soon after but only to post its research articles at first.

MBoC helped to work out the technical bugs with NIH on transferring and posting the contents accurately. “We were really the guinea pigs in terms of quality assurance,” Joseph remembers. When PMC opened for business in February 2000, it had one issue each from MBoC and PNAS. Today PMC has 2.6 million articles from 1,221 full content, 248 partial content, and 2,014 “selective deposit” journals.

The open access wars began in 1999 when Harold Varmus, Nobel laureate, ASCB member, and then the Director of the NIH, unveiled a radical proposal. In the Internet age of global science, Varmus declared that it was feasible, ethical, and practical to give the public, including scientists, free and immediate access to all biomedical research already paid for by federal money. Varmus called his theoretical NIH public repository “E-Biomed.” Later Varmus envisioned a broader mission for a “Public Library of Science,” or PLoS. In the end, PubMed Central (PMC) became the NIH depository.

From the outset, the original Varmus proposal was greeted with shock and awesome anger. The shock came from scholarly journal publishers who feared that their subscription income would evaporate. The anger came from the big commercial scientific publishers who imagined their business models collapsing.

At ASCB, Joseph recalls, shock was replaced by curiosity. “The ASCB being a journal publisher,” she recalls, “was, of course, interested in protecting our journal but was more interested in serving the interests of our members. So we immediately hiked up to the NIH campus and sat with Dr. Varmus, asking him to talk a little bit more about what he was doing.” Varmus explained the potential and equity of free online access but began to back off on including “all” materials and on “immediate” availability.

Meantime, the ASCB was doing statistical research. MBoC had been online since October 1997, and by analyzing usage patterns, the Society believed that it had discovered a workable business model. Usage was frontloaded, mostly during the first month after online publication with a bit less in the second month before it dropped off. “The graph looked like a backwards hockey stick,” says Joseph. “ The true value of the articles seemed to be concentrated in the first two months. People were coming back to the articles but it was clear that it was for citation purposes.” The hockey stick graph was repeated month after month.

As NIH Director, Harold Varmus first made open access a public issue and then turned to starting peer-reviewed, open access journals.
As NIH Director, Harold Varmus first made open access a public issue and then turned to starting peer-reviewed, open access journals.

“So we felt pretty safe making the bet that the value to our subscribers was in the first few months,” she recalls. Calculating that libraries would see the value and not cancel, then ASCB Executive Director Elizabeth Marincola and MBoC editors Keith Yamamoto and David Botstein offered the full MBoC contents to NIH with only a two-month subscriber embargo. It was a gamble that paid off, says Joseph, in financial and scientific returns.

However, by the fall of 2000, only eight journals had joined the NIH repository. That December, Varmus who had left NIH to become president of the Memorial Sloan-Kettering Cancer Institute, was at the ASCB Annual Meeting to unveil his idea for PLoS as an online, open access library for scientific research. He also proposed a scientific boycott of the big commercial journals, which were resisting any talk of open access, tooth and nail. Varmus asked ASCB members and other scientists to sign a pledge refusing to peer review or edit for or to submit or subscribe to any journal that would not post federally funded research papers in an online public repository such as PMC within six months of their initial publication.

Gary Ward, a University of Vermont microbiologist, was in the audience, listening spellbound. Open access as described by Varmus spoke to many of Ward’s frustrations about the commercial journal system. The boycott seemed the perfect response. “I realized that scientists actually had a lot of power. Without us submitting, without us reviewing, or without us sitting on editorial boards, the system couldn’t continue.” Ward signed the pledge and then watched the boycott fall apart. “The petition got a lot of people energized and got the discussion going,” Ward recalls, “but it really revealed how entrenched the for-profit publishers were. A lot of the people who signed the petition discovered that they didn’t have a lot of options. One by one, scientists started violating their pledges.”

Varmus was back at the 2002 ASCB Annual Meeting with a new plan and new allies, Patrick Brown of Stanford University and Michael Eisen of the University of California, Berkeley. They announced the conversion of PLoS from a science library into a journal publisher. With a $9 million grant from the Gordon and Betty Moore Foundation behind them, Varmus et al. unveiled two flagship, peer-reviewed, open access online journals, PLoS Biology and PLoS Medicine. (PLoS became PLOS in 2012.) The idea was to widen the choices for publishing high-impact research in fully open access journals.

Still with the Varmus departure, the NIH open access initiative was left in limbo until the new NIH director, Elias Zerhouni, appointed in 2002, took up the cause once again. Under Zerhouni, NIH aggressively pushed for a significant expansion of the repository. A largely resistant scientific publishing community pushed back. In February 2005, under Zerhouni, NIH issued a voluntary policy to deposit authors’ manuscripts into the PMC repository. Eventually, after many hearings and meetings, not without drama, among government officials and various stakeholders, Congress decided in 2007 to convert the voluntary policy into a mandate for all investigators funded by NIH. With the new OSTP directive, it moves to the other federal open access programs.

The public access was a policy orphan when Elias Zerhouni became NIH Director in 2002. When he left in 2008, the NIH had opened the doors to real online public access.
The public access was a policy orphan when Elias Zerhouni became NIH Director in 2002. When he left in 2008, the NIH had opened the doors to real online public access.

Back in 2007, the publishers were not amused, but the times, the Internet, and the public demand for health information were changing. The official NIH policy notice issued in January 2008, at the start of Zerhouni’s last year as NIH Director, spelled it out for all NIH-grantees. Any peer-reviewed paper accepted for publication after April 2008 was subject to the new mandatory deposit rule. Grantees who would or could not transfer the public deposit right would not be eligible for future funding. To many longtime observers, this was a major achievement for science, for scientists, and for access to the research that is paid by taxpayers.

Six months later, the dam burst in U.S. financial markets and, in the economic tsunami that followed, some observers wondered if scholarly publishing might go the way of Lehman Brothers. Yet despite PMC and the Great Recession, NIH figures showed that between 2007 and 2011, the number of biological and agricultural science journals increased by 15% and their subscription prices by 26%. Medical and health science titles rose by 19% and prices by 23%. This was not a shrinking business. Something else changed; the public was using public access. Now on a typical weekday, 700,000 users will access PMC. Looking at IP addresses alone, PMC says that 25% of users are from universities, 17% from companies, and 40% from the general public.

If public access is now mainstream, the arguments over “open access” continues.  After hearing Varmus address the ASCB Annual Meeting, Ward became a leading advocate within the ASCB Council, especially during his term as Treasurer from 2002 to 2008. Looking back, Ward believes that there were numerous turning points in the open access wars but singles out two—the NIH 2008 policy change and ASCB’s 1999 decision to put MBoC into PMC with a daring two-month embargo. “That really set the parameters of the debate because other publishers would say that there’s no way we could offer our content for free after 12 months because it would bankrupt us if all our subscriptions were free. We could go, ‘Wait a minute, MBoC has been doing this after just two months with all of its content—not just the NIH-funded stuff—and we’ve still got our subscribers’.”

Of the new OSTP directive extending the public access rule, Ward is happy to see the wider reach across federal agencies but also that for the first time, the new rule brings data sets into public access. “For a practicing scientist, that’s a pretty big deal. In many cases, papers just summarize the data. As data sets get bigger and bigger, whether it’s image data or high throughput screening data, it would be really nice as a reader to go back and look at the original data. The OSTP policy for the first time addresses that issue.”

On ASCB’s Council, Gary Ward was a strong advocate for open access: “I realized that scientists actually had a lot of power.”
On ASCB’s Council, Gary Ward was a strong advocate for open access: “I realized that scientists actually had a lot of power.”

What worries Ward is that the OSTP directive is an extension of an executive order. “If the next president who is elected doesn’t believe in open access, it could disappear overnight,” Ward says. For that reason, he favors a legislative solution currently embodied in a bill before both houses called the Fair Access to Science & Technology Research Act (FASTR). The FASTR bill has bipartisan support in both houses of Congress, says Ward, pointing out that one of the Senate sponsors is the well-known conservative John Cornyn (R-TX). 

FASTR has differences in language from the OSTP directive, which would tailor its applicability to a smaller number of federal agencies, 11 versus 19 for OSTP, says Ward. But FASTR would shorten the allowable embargo to six months. “Many of us have been arguing for a long, long time that a year is way too long,” Ward argues. “If you’re a scientist, you can’t wait 12 months. Six months is still too long. And going on MBoC’s experience, two months is adequate. But going from 12 months to six months is a big step forward.”

But for Ward, the key difference in FASTR is that it would allow public access and public reuse of data without copyright restrictions. “The poster child for this is text mining,” Ward explains. “That’s a huge lost opportunity that we can’t go in and electronically mine the text of the corpus of scientific literature.” FASTR addresses that, says Ward. “Agencies must come up with a way to license work that’s funded by the government in such a way that citizens who pay for that work can now reuse it.”

–John Fleischman

Created on Tuesday, March 5, 2013

facebook twitter1 youtube linkedin