DCAT-R (Data Catalog Vocabulary for RDF Repositories) extends [[[DCAT3]]] with classes and properties specifically designed for describing and managing RDF repositories from an internal, application-oriented perspective. While DCAT 3 focuses on the external perspective - cataloging datasets and describing services for discovery - DCAT-R adds an intra-service perspective, providing the structural vocabulary foundation for building RDF application frameworks and services.
The vocabulary introduces a four-level hierarchy - Service, Repository, Dataset, and Graph - where each level refines a DCAT 3 class for RDF-specific needs. It classifies graphs into four disjoint types (Data, Manifest, System, and Working) to distinguish user data from operational infrastructure and configuration. Additionally, DCAT-R provides a hierarchical directory mechanism for organizing large-scale datasets and a manifest-based configuration system that separates distributed repository metadata from instance-local service settings. Since all DCAT-R classes extend their DCAT 3 counterparts, every valid DCAT-R description remains a valid DCAT 3 description, ensuring interoperability with existing DCAT-aware tools and catalogs.
This is version 0.1.0 of the DCAT-R vocabulary specification. Comments and feedback are welcome via GitHub Issues.
This vocabulary builds upon DCAT 3, developed by the W3C Dataset Exchange Working Group. The authors acknowledge contributions from the RDF and Linked Data community.
This project is funded through NGI Zero Core, a fund established by NLnet with financial support from the European Commission's Next Generation Internet program.