The debate over “open access” to scientific research results has for years been an academic one, hashed out largely among university libraries and journal publishers. After all, free and unrestricted access to peer-reviewed scientific papers isn’t the sort of thing that typically attracts mainstream attention.
But the issue has been jolted into the spotlight recently, with a new White House policy aimed at making taxpayer-funded research freely available and the suicide of Internet activist Aaron Swartz, who worked to make such information free. This week, the New England Journal of Medicine, a traditional subscription-based publisher that puts many of its articles behind a paywall, published four perspective pieces on the issue.
The essays come down on all sides. Depending on who you ask, open access is: a public good, a way to spur research, inevitable, or a major threat that is eroding the economic model that allows publishers to vet and present the most worthy scientific work. The real answer is likely an amalgam, but what’s already clear is that the push toward open access is marching forward.
Martin Frank of the American Physiological Society cites data in his New England Journal essay that illustrates the explosion: 20,702 articles were published in open-access journals in 2000, compared with 340,130 in 2011 -- nearly a fifth of all articles published that year.
Last week, the White House science adviser, John Holdren, issued a memo instructing federal agencies that fund more than $100 million in research to create a mechanism to make published results free a year after publication. It is an expansion of a policy already in place at the National Institutes of Health, the nation’s largest funder of biomedical research.
Ann Wolpert, director of libraries at the Massachusetts Institute of Technology, and an author of one of the opinion pieces, said in an interview that it was an “extraordinary step forward in the government’s philosophy about the importance of sharing the research that flows from the funding they apply to important problems.”
But not everyone agreed. Michael Eisen, a biologist at the University of California, Berkeley, argues on his blog, “it is NOT junk,” that the policy is not a major victory -- in part because it extends a seemingly arbitrary protection to publishers -- open access with a 12-month waiting period.
One way of understanding the effects of a more broadly-applied open access policy may be to look at what has happened to federally-funded biomedical research. Open access comes with costs; it shifts the burden of paying for peer-review and publishing the work from the journal subscriber to the scientists who did the work.
According to Frank, if all its articles had to be published in open-access journals, Harvard Medical School would pay $13.5 million per year to publish its 10,000 articles. He also notes that universities must, on average, pay $150,000 each to create their own open access repositories. The NIH sets aside $4 million per year to put its articles into the freely available repository, PubMedCentral.
MIT has its own open-access repository, and Wolpert said it contains upwards of 30,000 articles. She believes there are only two countries in the world whose residents have not accessed the database.
Publishing and reviewing scientific work has its costs, and publishers, libraries, and funding agencies are still grappling with what system works best. Central repositories, where researchers deposit their papers, for example, may seem like a great idea. But they can erode the power of publishers -- even open-access ones -- by attracting traffic that normally would have gone to journal websites, helping generate ad revenue.
Frank cites data indicating that the open-access journal, Public Library of Science, lost a fifth of its traffic to the open-access repository where NIH-funded research is deposited.
Dr. Charlotte Haug, editor of the Journal of the Norwegian Medical Association, describes the dark side: a shadowy tier of “vanity presses,” open-access journals that will publish anything for a fee, and may not do the rigorous vetting performed by other journals.
In many ways, the questions facing science journals echo challenges to other publishers, from record labels to newspapers. The challenge is to find an economic model that will allow everyone in the system to thrive -- and to support the continued creation of quality content.
Haug’s argument about the value of a smart editor will be familiar to anyone who has found the Internet both a source of valuable, interesting content, and a time sink that can mean a lot of wasted minutes sorting through endless, unavoidable junk:
“As a reader,” Haug wrote, “I do not want to spend my time reading vast quantities of low-quality research and would be willing to pay someone to do the sort of filtering for quality, relevance, and novelty that journal editors have traditionally done.”