The growth in the capacity of the research community to collect and distribute data presents huge opportunities. It is already transforming old methods of scientific research and permitting the creation of new ones. However, the exploitation of these opportunities depends upon more than computing power, storage, and network connectivity. Among the promises of our growing universe of online digital data are the ability to integrate data into new forms of scholarly publishing to allow peer-examination and review of conclusions or analysis of experimental and observational data and the ability for subsequent researchers to make new analyses of the same data, including their combination with other data sets and uses that may have been unanticipated by the original producer or collector.
The use of published digital data, like the use of digitally published literature, depends upon the ability to identify, authenticate, locate, access, and interpret them. Data citations provide necessary support for these functions, as well as other functions such as attribution of credit and establishment of provenance. References to data, however, present challenges not encountered in references to literature. For example, how can one specify a particular subset of data in the absence of familiar conventions such as page numbers or chapters? The traditions and good practices for maintaining the scholarly record by proper references to a work are well established and understood in regard to journal articles and other literature, but attributing credit by bibliographic references to data are not yet so broadly implemented.