In the summer of 1991, Paul Ginsparg, a researcher at the Los Alamos nuclear laboratory, set up an email system for about 200 string theorists to exchange papers they had written. The World Wide Web was a mere infant—it had been opened to the public on Aug. 6 of that year. The string theorists weren’t particularly interested in making their research widely available (outsiders would have a tough time following the conversation anyhow). Ginsparg’s archive was a way for the theorists to communicate with one another.
For a short while, it would remain an insular tool for exchanging the latest theories of quantum gravity. But the novel system of communication would become the basis for a new model of academic publishing. Some wags would later joke that it was string theory’s greatest contribution to science.
By 1996, Ginsparg would write: “Many of us have long been aware that certain physics journals currently play NO role whatsoever for physicists. Their primary role seems to be to provide a revenue stream to publishers, a revenue stream invisibly siphoned from overhead on research contracts through library systems.” The arXiv, as it came to be known, was by then used widely in physics; some mathematicians and computer scientists had also started using it. Ginsparg had increasingly turned from doing physics to running the archive. (In 2002, he even received a MacArthur “genius grant” for his work on the arXiv .)
Since April 2008, researchers with funding from the National Institutes of Health have been required to submit their articles to a site called PubMedCentral, one of the arXiv’s offspring. After an embargo period (up to 12 months post-publication), the articles are openly accessible. During the embargo period, journals would have the option of restricting access to subscribers and charging nonsubscribers on a per-article basis (about $30). This experiment in open-access publishing is now on the verge of ending altogether or becoming the new status quo, depending on which politicians win an important legislative battle.
The Federal Research Public Access Act, reintroduced today by a bipartisan assortment of politicians, would broaden the open-access requirement to nearly all federally funded research. The rationale is that taxpayers, having paid once for the research, shouldn’t have to pay again to read what was done. Today’s bill is a response to the Research Works Act, which was introduced in December. The Research Works Act would roll back NIH’s open-access policy and prohibit the government from imposing any similar policies in the future.
The invisibly siphoned revenue stream that Ginsparg referred to comes from institutional subscriptions, which don’t come cheap. A year’s print subscription to Cancer Genetics, say, will run you (without discounts) $5,010 per year. (Individuals can subscribe for $280.) Cancer Genetics, along with 2,637 other journals, is published by Elsevier, a multinational conglomerate that made $1.1 billion last year on $3.2 billion in revenue—a 36 percent profit margin. This is typical of the industry. It helps that the “referees” who peer-review journal articles perform the job for free. (Almost 5,000 scholars are now boycotting Elsevier in protest of price-gouging and other practices, in a movement started by a British mathematician on Jan. 21.) Erik Engstrom, Elsevier’s current CEO, made $3.2 million in 2010; his predecessor Ian Smith got more than $1.7 million as a parting gift when he left after eight months on the job.
A journal article serves many purposes. One of them is to make money for publishers. Scientists and other academics publish in scholarly journals as a credentialing mechanism and, secondarily, to tell people about their work. Journals used to be crucial for both of these reasons, but in a world where academics could just post a paper up on their own websites, the primary purpose of a journal article is its professional validation. That’s why it makes some sense that the authors of a journal article should pay for the privilege of that validation, via peer review, rather than readers paying for the privilege of reading.
That is the reasoning behind the Public Library of Science (PLoS), a nonprofit group of seven journals that launched in October 2003. The PLoS journals weren’t the first “open-access” journals, but they have become the standard-bearers of the rapidly growing movement. PLoS journals charge authors between $1,350 and $2,900 per article, which goes to cover overhead. The work is then freely available to all on the Web. These fees are paid for out of research grants directly, rather than, as in the old system, being siphoned through university libraries. For those who can’t pay (for instance, scientists from poor countries), the costs are waived.
Such “open-access” models blur the current legislative debate a bit. Since articles published in open-access journals are freely available from the get-go, the legal requirement that they be made accessible after some waiting period becomes moot. But it is a spur for old-fashioned journals, which stand to lose if their archives are made freely available, to change their business model.
There is little doubt that author-pays models will be less lucrative than the subscription-based models, because they do not allow for the same rates of growth—it’s easier to grow a subscriber base than an author base. But it does seem the fees can cover production costs, even though the old guard tries to argue otherwise. Allan Adler of the Association of American Publishers, which has been leading the lobbying push against public-access mandates, says he doubts the open-access business model is “sustainable.” However, PLoS brought in more than it spent in 2010, and its CEO, Peter Jerram, made $432,640 in 2010—it’s not a shoestring operation, even if it doesn’t come with millions.
The open-access movement has been gathering steam. Harvard adopted an open-access policy in 2008. The policy requires faculty to grant their institution a nonexclusive right to freely distribute their scholarly articles. Cornell, Dartmouth, MIT, and the University of California-Berkeley followed in September 2009; as did Princeton in September 2011. But the university policies allow their researchers to apply for waivers from the open-access requirement if publishers won’t let them make their papers available. The current NIH rule and the broader Federal Research Public Access Act have no such loophole.
The open-access movement has strong momentum. After a hacker was arrested in July 2011 for breaking into JSTOR, an online archive of journal articles, the company opened up first some of its archive from before 1923 to the public, then later granted limited open access to more recent articles. In England, the Royal Society made its historical archives, including its Philosophical Transactions, first published in 1665 and thus the world’s oldest peer-reviewed publication, open-access in October. More recent publications were also made more available, albeit after (at most) a one- to two-year post-publication embargo. Google Scholar has wide coverage and frequently gives the public access to full text, even of subscription-gated papers, via researchers’ websites (though it omits PDFs over 5 megabytes, irking researchers in disciplines like archaeology that rely on larger image files). JSTOR’s future in the world of Google Scholar is tenuous.
Of course, most scientists already get unfettered access to the journals they need through their institutions. But the current ecosystem of publishing still is not particularly healthy for them. Scientists joke about things like the minimum publishable unit (also least publishable unit, or, for short “publon”). Maximizing the number of publications while minimizing their intellectual content doesn’t serve any broader interest. But it’s the inevitable result when the number of publications (which is objectively verifiable) becomes disproportionally important in relation to the quality of insight. Academic administrators have grown increasingly concerned with the “impact factor” of journals—i.e., how often the journal is cited. This, in turn, has led to pressure on researchers to cite for the sake of citing.
The progress of science won’t turn on the publishing model. Journal articles are the shadow of science, not science itself. But by taking power away from journal publishers, open-access (and public-access mandates) should make for a healthier scientific ecosystem. It won’t immediately fix the “publon” effect, but charging for publication should exert at least a slight pressure on scientists to actually have something to say.
Elsevier and other commercial publishers have an incentive to encourage the publication of as many papers as possible, regardless of the quality. In a statement, Elsevier says laws like FRPA “could undermine the sustainability of the peer-review publishing system.” These claims are easily mocked.
The shell game here is the oldest one in politics: an attempt to pass off the parochial interest of the few (journal publishers) as a broader societal benefit. The debate in Congress cuts across ideological lines—the competing bills have Republican and Democratic co-sponsors in both the House and the Senate. It should be mentioned here that Rep. Carolyn Maloney, a sponsor of the Research Works Act, got $15,750 in donations from the Elsevier and its executives in the last two years (out of a total of $119,300 that the company and its executives spent on congressional races). The bills are likely to be held up in Congress for quite a while. The White House, in the meanwhile, is conducting its own review of the issue.
Smaller journals will suffer in coming years, as they give way to informal sharing among colleagues and lower-margin open-access replacements. Top-tier publications like Nature and Science will survive; in fact, the publishers of both journals have publicly said they oppose the Research Works Act. They will survive because they have acquired such stature that a result is no longer published in Nature or Science because it’s important; it’s important because it was published in Nature or in Science.
Whatever the White House ends up saying, and even if Congress remains gridlocked, the movement toward open publishing now seems irreversible. In 1996, Ginsparg said that it wasn’t a question of if, but when “commercial publishers accustomed to large pre-tax profit margins” would find themselves unable to compete with a “global raw research archive” combined with “high-quality peer-reviewed overlays.” The answer to his question seems clear: now.