UC Berkeley

Abstract: Editor's Summary A critical element in the products of research projects is the data set, largely treated as a marginal appendage to the written record. With the requirement for data management plans in grant proposals to the National Science Foundation and National Institutes of Health, the issue of managing research data sets is gaining attention. Sharing primary data enables a greater return on the original investment, expanded discovery and fertilization of ideas across disciplines. The major impediment is lack of an infrastructure to archive non‐textual data sets, in addition to hurdles including locating data, deterioration, format standardization, permission for use and suitability. It is essential that institutional policies recognize data sets as a key product of intellectual work worthy of inclusion in a bibliography. This must be followed by action to promote their preservation, metadata enrichment, cross‐lingual interoperability, geographic and temporal coding and data provenance and to contend with changes over time.