Multi-entity models of resource description in the Semantic Web A comparison of FRBR, RDA and BIBFRAME

Purpose - The 1998 International Federation of Library Associations (IFLA) document "Functional Requirements for Bibliographic Records" (FRBR) has inspired a family of models that view bibliographic resources in terms of multiple entities differentiated with regard to meaning, expression, and physicality. The purpose of this paper is to compare how three FRBR and FRBR-like models have been expressed as Semantic Web vocabularies based on Resource Description Framework (RDF). The paper focusses on IFLA's own vocabulary for FRBR; RDF vocabularies for Resource Description and Access (RDA), an emergent FRBR-based standard for library cataloging; and BIBFRAME, an emergent FRBR-like, native-RDF standard for bibliographic data. Design/methodology/approach - Simple test records using the RDF vocabularies were analyzed using software that supports inferencing. Findings - In some cases, what the data actually means appears to differ from what the vocabulary developers presumably intended to mean. Data based on the FRBR vocabulary appears particularly difficult to integrate with data based on different models. Practical implications - Some of the RDF vocabularies reviewed in the paper could usefully be simplified, enabling libraries to integrate their data more easily into the wider information ecosystem on the Web. Requirements for data consistency and quality control could be met by emergent standards of the World Wide Web Consortium for validating RDF data according to integrity constraints. Originality/value - There are few such comparisons of the RDF expressions of these models, which are widely assumed to represent the future of library cataloging.