We’ve been working to develop a model of “data sharing as publication” in our work with Open Context. In our view, publication helps communicate some of the need for quality and standards alignment that makes effective data dissemination something more formal than implied by the term “sharing.”
We’re definitely not alone in this assessment, and other groups are also experimenting with various models of data publication. One group is the California Digital Library (CDL), a unit that runs many of the University of California’s leading scholarly communications and data preservation efforts.
The CDL recently began a project developing a platform to support a new kind of outlet for scientific data dissemination. They’re using the term “Data Journals” (see this link to a presentation introducing the idea) to describe this new outlet, since “journals” are a familiar and accepted part of researcher communications.
Like more conventional journals, a Data Journal would provide a disciplinary focus for data dissemination. In addition, because the CDL has expertise in the technical infrastructure of publication and data curation, a Data Journal would benefit from the same library curation, indexing, and search services that are so essential in conventional scholarly communications. That means we will be able to reliably track citations and impact of datasets, just as we can track such metrics for conventional publications.
This goal aligns very well with work we’re undertaking (with support from the Alfred P. Sloan Foundation) to develop editorial workflows to improve data quality and align datasets with standards that may be expected in a given domain. Because of these converging and complementary goals, we’re joining with the CDL and will be piloting the development of a Data Journal for archaeology. The Data Journal will help communicate and set expectations about the nature and quality of data. It will also increase the future impact of datasets by adding relevance to them, increasing their discoverability, and contributing to their longevity. Finally, and perhaps most importantly, we’re eager to develop the concept of a Data Journal because it will help make “data” a first class-citizen in the world of research communications.