Linked Data for libraries Benefits of a conceptual shift from library-specific record structures to RDF-based data models

摘要

Purpose - The purpose of this paper is to provide recommendations for making a conceptual shift from current document-centric to data-centric metadata. The importance of adjusting current library models such as Resource Description and Access (RDA) and Functional Requirements for Bibliographic Records (FRBR) to models based on Linked Data principles is discussed. In relation to technical formats, the paper suggests the need to leapfrog from machine readable cataloguing (MARC) to Resource Description Framework (RDF), without disrupting current library metadata operations. Design/methodology/approach - This paper identified and reviewed relevant works on overarching topics that include standards-based metadata, Web 2.0 and Linked Data. The review of these works is contextualised to inform the recommendations identified in this paper. Articles were retrieved from databases such as Emerald and D-Lib Magazine. Books, electronic articles and relevant blog posts were also used to support the arguments put forward in this paper. Findings - Contemporary library standards and models carried forward some of the constraints from the traditional card catalogue system. The resultant metadata are mainly attuned to human consumption rather than machine processing. In view of current user needs and technological development such as the interest in Linked Data, it is found important that current metadata models such as FRBR and RDA are re-conceptualised. Practical implications - This paper discusses the implications of re-conceptualising current metadata models in light of Linked Data principles, with emphasis on metadata sharing, facilitation of serendipity, identification of Zeitgeist and emergent metadata, provision of faceted navigation, and enriching metadata with links. Originality/value - Most of the literature on Linked Data for libraries focus on answering the "how to" questions of using RDF/XML and SPARQL technologies, however, this paper focuses mainly on answering "why" Linked Data questions, thus providing an underlying rationale for using Linked Data. The discussion on mixed-metadata approaches, serendipity, Zeitgeist and emergent metadata is considered to provide an important rationale to the role of Linked Data for libraries.