Call it a chain reaction. Following the leads of the physics, mathematics, economics, and biology communities, the American Chemical Society (ACS) announced yesterday that it will start a preprint server for chemistry papers, tentatively titled ChemRxiv.
The site will be modeled after arXiv, the decades-old server that's provided a home for preliminary research in physics, mathematics, and computer science, and bioRxiv, begun 3 years ago by Cold Spring Harbor Laboratory Press, which does the same for biology. It will be the first preprint server begun by a professional scientific society, groups that have historically been concerned about the impact of free preprint servers on the revenue they derive from traditional journal publishing. ACS is chemistry's dominant professional organization and one of its leading publishers.
The emergence this year of ASAPbio, a loose coalition of high-profile biologists who have embraced preprints, drove the society to action, says Kevin Davies, vice president of the society's publications division in Washington, D.C. There was also an element of disciplinary peer pressure: "It would be somewhat bizarre," he says, "if chemistry was left too much longer as the sole major scientific field that lacked a major preprint server." ACS has invited other organizations to become co-organizers or sponsors of the new service, which will likely be a prime topic of conversation at the society's fall meeting later this month in Philadelphia, Pennsylvania.
Chemistry has long had a culture that prizes the close holding of data and scientific results until peer review and final publication. Many editors would reject papers that had appeared as preprints, counting that appearance as prior publication. That culture helped doom an effort 16 years ago by Elsevier, the for-profit publishing giant and a major rival to ACS, to start its own chemistry preprint server through its ChemWeb subsidiary. (In May, Elsevier bought the Social Science Research Network, the leading site for social science discussion papers.)
Davies believes a professional organization provides a natural fit for a preprint server. Although ACS has not formally polled its membership, he adds, most of its advisers and journal editors support ChemRxiv. (Science and Nature, among many others, now publish papers that first appeared as preprints.) The initiative was first proposed by Laura Kiessling, a chemistry professor at the University of Wisconsin, Madison, and editor-in-chief of ACS Chemical Biology, during a society retreat in January.
Twenty of ACS's 50 journals now routinely accept papers that first appeared as preprints, whereas another dozen require the authors to discuss the fact with an editor. The rest still reject such studies. Each journal will have the authority to decide whether to participate in the preprint service, says Davies, who predicts that some areas of chemistry will be slower than others to embrace the idea.
"It's always heartening to see other disciplines belatedly joining the late 20th century," says Paul Ginsparg, arXiv's founder and a physicist at Cornell University. "And it's refreshing to see more experimentation in this space."
Ginsparg worries that the publishing side of ACS could make some chemists with papers intended for non-ACS journals less likely to participate in the new service, fearing its appearance on the service would bias editors against their work. (That dynamic may be one reason why Nature Precedings, intended as a preprint server for the life sciences, struggled for 5 years and finally closed in 2012.) In biology, many authors also avoid preprints out of the fear that editors would consider their study previously published and reject it, says John Inglis, executive director of the Cold Spring Harbor Laboratory Press in New York. But bioRxiv has grown steadily, he notes, fueled by the rising emphasis on transparency combined with the creeping pace of peer review.
ChemRxiv’s success will hinge on recruiting prominent members of subdisciplines and avoiding competition, Ginsparg says. ACS does not expect to derive revenue from ChemRxiv, Davies says, and the society is optimistic that the server can be supported without adding staff.