Preprint ecosystems

See allHide authors and affiliations

Science  29 Sep 2017:
Vol. 357, Issue 6358, pp. 1331
DOI: 10.1126/science.aaq0167

In 1991, an electronic system through which interested parties could access non-peer-reviewed physics papers was launched by Paul Ginsparg at Los Alamos National Laboratory. This evolved into the “preprint” server arXiv, which has played a central role in information sharing in physics, mathematics, and related fields over many years. Recently, other preprint servers have been or are being developed to serve other fields including biological sciences, chemistry, and medicine (see the News story on p. 1344). Given the different cultures associated with these disciplines, and other external factors, policies and practices must evolve as these servers become integrated into their communities.


“…policies and practices must evolve as these servers become integrated…”

An overarching issue regarding preprints is that they have not been peer reviewed. Peer review serves as an important check on the quality of research. Like essentially all practicing scientists, I have experienced having manuscripts scrutinized by peer reviewers, leading to the identification of important weaknesses and suggestions for improving experimental design and presentation. Similarly, as a reviewer, I have tried to assess manuscripts thoroughly and fairly and have sometimes identified concerns related to even the most basic aspects of experimental design and interpretation. The majority of manuscripts posted on arXiv do go on to be submitted to peer-reviewed journals for subsequent publication.

Before starting in my current role at Science, I participated in a meeting to discuss preprint servers in biology (see http://science.sciencemag.org/content/sci/352/6288/899.full.pdf). Here, a major motivation is the unpredictable time between the submission of a manuscript to a peer-reviewed journal and its publication. Delays in publication can have important implications, particularly for early-career scientists for whom having the product of their efforts available for wide examination can be an asset when competing for jobs or grant funding. Preprints have the potential to address this concern to some degree. The U.S. National Institutes of Health released a policy earlier this year that explicitly encourages the use of preprints and other “interim” research products.

The Science family of journals accepts the submission of a research paper for which a preprint of the submitted version is posted on not-for-profit servers such as arXiv and bioRxiv, as we support mechanisms that relate to the rapid communication of findings within the scientific community. We encourage authors to discuss with our editors any postings to other servers, and we encourage our editors to become involved in discussions about preprint-related issues as they evolve. Interactions with the press related to preprints has become one challenging area. When preprints of papers that are of potential interest to the press and the public are available, reporters on occasion approach authors prior to peer-reviewed publication. If a paper is under consideration in a Science-family journal, we leave it up to authors as to how they respond to media inquiries, but do note that media coverage could be taken into account by editors when considering novelty and make it difficult to embargo the paper if accepted for publication. We believe that giving reporters access to papers that will soon appear in our journals, on an embargoed basis, leads to more complete and accurate reporting of important science stories.

The ecosystems of science have changed tremendously over the quarter century since arXiv began. New technologies, processes, demands, and the scope of data—all of which can vary substantially from discipline to discipline—pose new challenges for accelerating the communication of research progress. As initiatives with preprint servers continue to evolve, the Science family of journals will attempt to continue to navigate this landscape for the benefit of the scientific enterprise and for society.

View Abstract

Stay Connected to Science

Navigate This Article