Problem addressed
The current “bibliographical” approach to (FAIR) data remains mostly at the level of catalogue entries: providing some limited bibliographic (e.g. DC), discovery (subject list) and management (e.g. CERIF) metadata. It does not provide a sufficiently detailed description of the datasets for most reuse or data integration requirements. This adds cost to reuse, and greatly hinders the possibility of using data across different domains, because it does not supply necessary information about, for example, data structure, data provenance and process, and guides to harmonising the differences in domain semantics.