Free For All? The Science of Publishing Research Online

The scientific community has built a solid reputation for being early adopters of technology, including harnessing the efficiencies of the Internet to disseminate scientific articles and journals electronically. Since 1995, more than 100 society and university not-for-profit publishers have linked up with Stanford University's HighWire Press to turn traditional print journals into online resources that can be easily accessed by both the scientific community and the general public.

In March 2004, representatives from 48 of these leading non-profit medical and scientific societies and publishers announced their commitment to providing free access and distribution of published research findings. Via the "Washington D.C. Principles for Free Access to Science," these not-for-profit entities outlined their commitment to work in partnership with scholarly communities, including libraries, with the goal of advancing science by providing both the scientific community and general public with online access to more than 1.6 million articles, with more than 600,000 of such articles available free of charge.

Making information free to individuals who cannot afford to pay for it—as well as developing software to support authors, editors, and reviewers—to fuel the democratization of information is not easy, particularly given a lack of resources to fuel non-commercial online ventures. "The critical aspect for us as not-for-profit society publishers is to combine that democratization with the maintenance of business models that allows us to give back to our society and to our discipline," says Martin Frank, spokesperson for the D.C. Principles and executive director of the American Physiological Society (APS). "The challenge is to continue to maintain the ability to cover the costs of publication, because nothing is free."

The business models used by society publishers differ, with some offering a mix of free and paid content, including incorporating page views that range from $30 to $60 per page to help share publication costs, in an effort to keep institution subscription costs down. Frank explains that commercial publishers recover costs solely and exclusively from subscriptions, seeking a 30% to 40% profit margin to generate the revenues they need. Another platform gaining momentum is the author-page model—the open-access model—whereby authors are asked to pay fees that range from $500 to $1,500 to publish their research. Frank says there's no one true model, and while society publishers do have subscription-based platforms they also provide free content, sometimes immediately or after intervals of six to twelve months, with archival content almost always available at no charge.

"The question we are dealing with in this environment is whether publications should be free to the public immediately or free after a transition interval," says Frank. For instance, APS has an educational journal that is available free online immediately. "We also have another journal called Physiological Genomics, which we are experimenting with to create a mirror image of the open access model," he says. This model gives authors the option of either paying to publish articles to make content available immediately or pay nothing, which puts the content under access control for the first twelve months, making it accessible by subscription only.

The formation of the Washington D.C. Principles evolved to help society organizations determine their role in the evolving Web-based scientific publishing environment. "I've spoken to library groups and I think that most of them commend the societies for trying to keep costs down and support models that allow for access under subscriptions and open free access after a period of time," Frank says. "They recognize that it's a viable model."

Meanwhile, commercial publishers charge considerably more for access to such content, and some non-profits have had to revise their models as well to combat rising costs, which are then passed on to institutions. This is problematic for libraries that subscribe to 1,000 or more scientific journals. Only time will tell which models will ultimately survive. "There are problems and strengths to each," says Frank. "As a society publisher, the issue we deal with is that whatever revenue we make from our journals goes to the publication program and comes back to our society, allowing us to support programs that will develop the next generation of scientists, which in the absence of any excess revenue from our publications program, we'd be unable to do."