Interpretation and use of scientific datasets by those that are not engaged in the creation or production of those datasets is pivotal for enabling science that is driven by data. RDA has made significant progress in addressing this issue through the initial Data Type Registries (DTR) WG, which has now finished, but much remains to be done, hence this proposal for a follow-on working group. The fundamental effort here is to describe scientific datasets in a human-and-machine-readable fashion, enabling humans and software to parse and understand the semantics, context, and assumptions behind the data. We reference all such descriptions “data type records”, regardless of the standard or best practices standing behind those descriptions. Data types complement traditional descriptive metadata records, providing re-usable descriptions of dataset structure and semantics aimed mainly at supporting data processing, while at the same time providing an additional attribute that can be used for a certain kind of discovery. The initial DTR WG focused on developing an infrastructural component that would manage such data type records. The new WG, DTR2, will focus on aiding data producers come up with useful data type records.
Please see the attached PDF for the full case statement.