Science
23 July 2010  Vol 329, Issue 5990, Pages 357-480
http://www.sciencemag.org/current.dtl

Policy Forum
Information Access: Prepublication Data Release, Latency, and Genome Commons
Jorge L. Contreras

Preview
Researchers must disclose their data in order to achieve recognition and to enable others to test, validate, and challenge their hypotheses. In doing so, they create bodies of shared knowledge that are analogous to traditional public resources, such as forests and freeways, often referred to as “commons” (1, 2). The rate at which data are added to these information commons, however, varies greatly. The traditional practice has been to contribute experimental and observational data to the commons when, or shortly after, the analysis of that data is published, sometimes years after its initial collection (3, 4). Because of busy schedules, competitive pressures, and other interpersonal vagaries, the sharing of scientific data can be inconsistent even after publication (5, 6). Many traditional data-sharing practices were challenged, with significant and lasting effect, during the race to sequence the human genome.