Thursday, January 21, 2010

Paying for arXiv

from: Chronicle of Higher Education, January 21, 2010: Wired Campus

Cornell Library Proposes New Model to Keep arXiv Going

By Jennifer Howard

Cornell University Library announced today that it wants the top institutional users of to help pay for the online scientific repository. "Keeping an open-access resource like arXiv sustainable means not only covering its costs, but also continuing to enhance its value, and that kind of financial commitment is beyond a single institution's resources," Oya Rieger, Cornell's associate university librarian for information technologies, said in a statement describing the new strategy.

The experiment is shaping up to be a test of how well multiple institutions can band together to support critical scholarly resources. For scientists in physics, mathematics, quantitative biology, statistics, computer science, and related fields, arXiv has become an indispensable clearinghouse for the latest research. The brainchild of a physics professor, Paul Ginsparg, the repository holds nearly 600,000 e-prints of research articles, many of which appear there before they make their way through the formal journal-publishing process.

It costs Cornell about $400,000 a year to maintain arXiv, according to Anne R. Kenney, university librarian at Cornell. The library's annual budget runs in the $40- to $50-million range. Some 200 institutions account for about 75 percent of the download traffic on arXiv, and it's that group that Cornell hopes will pony up first. The suggested contribution for the heaviest users is $4,000. Ms. Kenney says that most of the top 25 have said they will participate.

Calling arXiv "a lifeline" for areas of the world with limited access to scholarly publishing resources, Ms. Kenney emphasized that arXiv will continue to be open access. Individual users will not be charged to submit or to review its contents.

One major user of arXiv's resources is the Massachusetts Institute of Technology. Ann J. Wolpert, director of libraries there, confirmed that MIT will answer the call for money, even though it comes at time when resources are tight. "It's just not reasonable to think that one institution could carry that burden in perpetuity," she said. "The way Cornell has been approaching this is eminently sensible and fair."

Ms. Kenney emphasized that the call for contributions is a short-term fix that will buy time to look into longer-term solutions. "It opens the door to seeing something like arXiv as being a public good worthy of support beyond one institution," she said. That could lead to federal money or -- perhaps the best-case scenario -- an endowment.

MIT's Ms. Wolpert also sees the arXiv pitch as a harbinger of things to come. "The call for support comes at a time when we're all looking for ways to sustain those shared resources that have become so critical to the academy these days," she said. "I think the library is looking to work with faculty in these key disciplines in our institutions for a longer-term, more sustainable, more integrated solution to the way scholars communicate."

One critical concern for libraries is how to avoid paying for the same content in multiple versions. Could repositories like arXiv replace journals altogether? That's a delicate question to pose to a librarian.

Ms. Kenney said that one of arXiv's strengths has been how well it has coexisted with more-traditional publishing. "I don't know what the future holds," she said. "There has been a critical role that formal publishing provides in terms of the vetting of materials. I can tell you that it has been gratifying to have a pretty healthy relationship between arXiv and the formal scholarly literature."

Ms. Wolpert cautioned against taking a narrow, either-or view of publishing options in the Internet age. Research moves along a publishing continuum "that is now managed in separate pieces, but that we hope with some thought and care and leadership from faculties and [scholarly] societies and libraries will become a more integrated system," she said. It will be worth watching to see how much closer the arXiv experiment gets us to such a system.