UC Berkeley News


Free-science movement gains a foothold at Berkeley
Mega-publishers still control most research, but authors, librarians, and politicians are calling for change

| 14 February 2007

When the journal Nature reported last month that a group of scientific-publishing goliaths had enlisted a "PR pit bull" to "take on the free-information movement" - a story that quickly entered the mainstream media through the Washington Post, then replicated itself endlessly via the blogosphere - the move was widely viewed as a declaration of war. The long knives, it appeared, were finally out. In response, the slingshots, too, came out in force.

Berkeley biologists Michael Eisen, right, and Randy Schekman agree in principle that scientific literature should be widely available, but differ in practice. Eisen, a co-founder of the Public Library of Science, wants researchers' findings to be freely accessible online as soon they're published; Schekman, editor of the society journal PNAS, insists that's not economically feasible. (Schekman photo by Bonnie Azab Powell; Eisen photo courtesy PLoS)

"The twisted logic required to come up with the phrase 'Public access equals government censorship' leaves me in uncomfortable awe," read a typical posting, referring to the most inflammatory of the spins suggested by Eric Dezenhall - whose corporate and celebrity clients have included ExxonMobil and Enron's Jeffrey Skilling - at a previously undisclosed meeting arranged by the Association of American Publishers, an industry trade group. "I can't imagine any of the academics I know willingly supporting/falling for this slickster, big-business, fear-mongering approach to where and how to publish," fumed another blogger.

Similar cries of outrage came from an assortment of academics, librarians, and others who believe, in Stewart Brand's famous formulation, that "information wants to be free." At Berkeley, though, one scientist at the heart of the so-called open-access movement - which aims to apply that principle to the highly specialized realm of scientific and medical research - was thrilled to hear that big commercial publishers were circling the wagons to defend their for-profit, subscription-based model.

"I think it's fantastic," exults Michael Eisen, an assistant professor of molecular and cell biology and a genetics researcher at Lawrence Berkeley National Laboratory. "It's just evidence that open access is working."

Eisen, who received a Presidential Early Career Award for Scientists and Engineers at a White House ceremony in 2004, is a co-founder of the San Francisco-based Public Library of Science, a leading advocate for free, online dissemination of scientific research. With Harold Varmus, a Nobel laureate and former director of the National Institutes of Health, and Patrick Brown, a biochemistry professor at the Stanford University School of Medicine and an investigator with the Howard Hughes Medical Institute, Eisen started the nonprofit PLoS (pronounced Ploss) in 2000, intending to change the way research is disseminated - or not - to the wider world.

PLoS's first act was to circulate an open letter to scientists, asking that they pledge to "publish in, edit or review for, and personally subscribe to" only scholarly and scientific journals that provide free, unrestricted rights to their contents within six months of publication. When that effort failed to alter the landscape - despite garnering nearly 34,000 signatures from scientists in 186 countries - the three decided to start publishing their own peer-reviewed journals, beginning with PLoS Biology in the fall of 2003 and, a year later, PLoS Medicine.

For Eisen and others at the forefront of the movement, the issue is less whether information wants to be free than that society needs it to be. Underlying the push for open access, Eisen says, is the idea that "the interests of virtually everybody involved in the creation and dissemination of scientific knowledge are best served by a system in which the end products of scientific research - the papers that scientists write and publish - are effectively in the public domain."

With subscriptions for many scientific journals running to thousands of dollars - and some single-article downloads priced at more than $50 apiece - proponents of open access argue that vital medical knowledge, for example, is prohibitively expensive for individuals and families seeking reliable information, as well as for health workers caring for patients in developing countries. And as rising subscription fees force many libraries to constantly scale back their collections, even university-affiliated researchers can find themselves in the dark.

"When you do an experiment, or think about the results of an experiment, you now need to be considering the accumulated knowledge of hundreds or thousands of genes," explains Eisen, a computational biologist. "You simply can't know everything that's been published that's relevant to an experiment you're doing."

Many scientists, he adds, need not only access to printed text, but the ability to more fully exploit computer technology by manipulating and massaging other researchers' data - as was done, in one prominent example, with the Human Genome Project.

"It's completely ridiculous that I, a publicly funded scientist, am unable to get access to the articles written by my colleagues and to download them onto my computer," says Eisen. "I'm not trying to steal their ideas, I'm not trying to do anything but make that information - which they've spent their lives generating, and the government has spilled billions of dollars into funding - much more robust, much more dynamic, and much more useful. And publishers - because they view the scientific literature as their own private property, and are trying to protect their ability to profit from that literature - are completely inhibiting my ability to use it in this way."

Open-access pros and cons

The so-called STM (science, technical, and medical) publishing business, an estimated $10 billion industry, is increasingly dominated by a handful of companies like Elsevier (with some 1,800 journal titles) and Wiley (at 350 and counting), both of which were represented - along with the American Chemical Society - at the AAP-sponsored meeting with Dezenhall. According to the Nature report, Dezenhall urged them to focus on "simple messages," such as "Public access equals government censorship" - an apparent reference to demands that articles be posted to PubMed Central, a science-information repository created by the NIH - and to "paint a picture of what the world would look like without peer-reviewed articles," a prospect apt to horrify no one so much as editors at the many peer-reviewed open-access journals.

"We're like any firm under siege," an AAP vice president told Nature in explaining the turn to hardball tactics. "It's common to hire a PR firm when you're under siege."

In a recent posting on its own website, Elsevier - whose journals generate upwards of $1.5 billion in annual revenues - warned that the open-access model, "which transfers the cost of publishing from reader to author, could jeopardize the stable, scaleable, and affordable publishing that currently exists."

For Berkeley's library, 'serials crisis' means shrinking access to information
When the 10 UC libraries joined the Public Library of Science as an institutional member in 2004, Beverlee French, systemwide director for shared digital collections, called the move an effort at "directing some of our scarce dollars away from overpriced journals and toward innovation."

With budgets flat and scholarly-journal prices rising far faster than inflation, however, what's known as the "serials crisis" remains a pressing problem here at Berkeley and at universities and research institutions throughout the nation. Chuck Eckman, associate University Librarian and director of collections at Berkeley, warns that without increases to its budget, the campus library faces a shortfall of roughly $1.4 million in 2008 - with a commensurate reduction in journal, book, and digital-resource acquisitions - and a still-larger deficit in 2009.

"There's normal inflation and excessive inflation," Eckman says, referring to skyrocketing prices for serial journals. According to the Association for Research Libraries, serials costs jumped 226 percent between 1986 and 2000, a period when the Consumer Price Index rose by 57 percent.

Beth Weil, head of the Marian Koshland Bioscience and Natural Resources Library, says that the UC system's negotiating power has helped to tamp down prices for scientific journals - UC is among Elsevier's biggest customers, for example - and the campus has switched to somewhat cheaper online-only versions of many serials, though not for such popular titles as Science and Nature.

But Weil, who serves on the PLoS board of directors, believes the solution to the bind facing university libraries - and the scholars and students who rely on them for access to scientific research - lies with researchers themselves.

"The only way out of this serials crisis that we're currently in is to change where faculty publish, and to make sure that faculty are aware of the crisis in scholarly communication," she contends. "A lot of people view this as a library problem - you know, I frequently hear, 'Beth, what are you going to do about this?' Well, there's really a limit to what I can do about it, because it's really in the hands of the people who make decisions about where they publish and how much control they want to have over their work once it's been published."

Eckman hopes that an increase in the scope and volume of "open" resources, including open-access journals and free, repository-based content, will help to restore some balance - a hope echoed by Weil, who fears prices are headed back up.

"When the state of California was in very bad straits," she says, commercial journal publishers were more willing to hold subscription-cost hikes to "reasonable" levels. "But now, nobody's keeping their price increases particularly reasonable," she adds. "So we're running out of rabbits to pull out of the hat."

- BB

Eisen and other open-access advocates dispute such logic, maintaining that whatever threat they pose is to the industry's profits, and not to authors and readers of scientific literature - who, they point out, are often the same people. Under what's known as the "author pays" model, PLoS charges contributors $2,500 to publish in PLoS Biology and PLoS Medicine, with somewhat lower fees for a number of other journals now in its catalogue. Despite the name, most such fees are paid not by the author, but by the agency funding the research. PLoS authors retain the copyright to their work under a Creative Commons license, which permits unrestricted access, copying, and distribution so long as the original author and source are credited.

Commercial publishers, by contrast, hold the copyrights to contributors' work, allowing them to strictly control access and distribution of their journals' contents.

"The way publishers cover their costs and make a profit is by selling the contents of their journals back to the researchers," Eisen says. "The core economics of that is that universities or research institutions pay subscription fees to get access to the research that they've provided for free to the journals in the first place."

Industry economics are putting a heavy burden on universities as well, at least partly due to a move toward consolidation over the past 15 years that's left 37 scholarly publishers, Nature included, in the hands of six corporate entities. "The cost of serials goes up every year," says Beth Weil, head of the campus's Marian Koshland Bioscience and Natural Resources Library, "and the library's budget has not increased since 2001. That is the problem, simply stated." (See sidebar at right.)

And because so much of the money for scientific publishing derives from the federal Treasury - in the form of direct research grants to scientists and indirect funding to publishers for printing their work - the result, Eisen says, "is a business that for decades has basically been allowed to print checks from the government to themselves with very little restraint."

To suppress or not suppress?

By coincidence, on the morning the Nature story broke, digital-publishing consultant Joseph Esposito was on campus at the invitation of the Center for Studies in Higher Education to discuss what he terms "the three-legged stool" of academic publishing - open-access, not-for-profit, and commercial publishers.

Esposito, a former executive at Simon and Schuster, Random House, and Encyclopedia Britannica, said that while all three business models have their uses in academia, open access has thus far "failed to deliver a compelling economic model." He laid out what might be the fundamental divide between the commercial publishers' "fee for access" model and the "fee for services" model proffered by open-access proponents.

"The joke among publishers is, There are 6.2 billion people on the planet, and 12.4 billion authors," he said. "The point of publishing is not to encourage production. The point of publishing is to suppress production."

"Some people say, 'Wait a second, I don't like that,'" Esposito added. "But that's what publishing is."

For open-access advocates, at least, that's precisely the problem.

"When we're trying to make a judgment as a research community and as a society about what's the best way to disseminate scientific information, the question is, What is the system that best serves the goals of science?" Eisen says. "And the goals of science are to produce knowledge about human health and about the world and to make sure that knowledge is used in ways that both maximally advance research and maximally benefit the public."

Gavin Yamey, a physician and senior editor at PLoS Medicine, explains the difference this way: "A publisher is a service provider," he says. "We're like a midwife. We help deliver the baby. We do not keep the baby."

Scientific-publishing economics

Yet even for some who sympathize with the goals of the open-access movement, the economics don't necessarily add up. Randy Schekman, a Berkeley professor of molecular and cell biology and the editor of Proceedings of the National Academy of Sciences, or PNAS, is among them.

"When [PLoS co-founder] Pat Brown first issued his manifesto some years ago calling on everyone to sign a petition to demand that all the scientific literature be made available immediately to files that would be controlled somewhere in Washington, I did not sign it," he recalls. "Because having been an editor for the Journal of Cell Biology for many years, I realized that this would be the death of many of the journals - even those run by some of the scientific societies, or by places like Rockefeller Press," his former journal's parent organization.

"I'm sympathetic to the notion that the literature should be generally available," adds Schekman. "But I'm not sympathetic to the notion that it should be available instantly."

Unlike many society journals, PNAS is mandated by its parent, the National Academy of Sciences, to break even, and its revenues do not help support the society. In order to keep operating, Schekman says, subscription fees are crucial. To provide an incentive to subscribers, PNAS has a six-month embargo on articles it publishes, after which time they become freely available online.

The author-pays model, he believes, is workable for PLoS only as long as it continues to receive major grants from such funders as the Betty and Gordon Moore Foundation, which helped launch the organization with a $9 million gift.

"You have to have a better business plan than they've had at PLoS," he says, in order to prove the viability of open access as an alternative to existing publishing models.

Even so, PNAS has taken steps in the direction of more-open access, including two instituted by Schekman's predecessor as editor, the late Berkeley professor Nicholas Cozzarelli, who served as a PLoS director. For a fee, authors can opt to have their papers made available upon publication. The journal's contents are also immediately accessible at no cost in developing countries.

As for the reluctance of many scientists to publish their work in an open-access journal - an issue identified by proponents as slowing the movement down - Schekman believes it's more a question of bona fides than business models.

"PLoS could do well and then people would clamor to publish there," he says. "I'm not sure it relates to whether a journal is publicly available or not: I think it has to do with the prestige of the journal. . I don't think it's that Cell, Nature, and Science are proprietary and not publicly available that gives them that charm. It's just that they've done a very good job in building up a very selective journal."

That, in fact, is precisely what Eisen is hoping to do. And he thinks it's a matter of time before researchers come around.

"People believe their careers can be made by a publication in Science or Nature, the elite journals in the field," he says. "So the challenge for open access has been to provide scientists with a system that changes the economics of scientific publishing while still providing them with the other things they get from scientific journals."

The movement could get a boost from legislation expected to be reintroduced shortly by Sens. John Cornyn (R-Texas) and Joseph Lieberman (I-Conn.), the Federal Research Public Access Act, or FRPAA, which would require federal agencies that fund more than $100 million in research annually to make articles resulting from that research available on the Internet. Feb. 15 has been declared a "national day of action" on many college campuses in support of the bill.

Meanwhile, the mindset that rules the publishing world, say some observers, is best captured by an article on the availability of medical research in one of the poorest regions of the globe, titled "Impediments to Promoting Access to Global Knowledge in Sub-Saharan Africa."

Should you happen to be practicing medicine in Kenya, say, without a subscription to Library Management, online access to the article - as PLoS Medicine's Yamey delights in pointing out - will cost you $25.