MELD: A Linked Data Framework for Multimedia Access to Music Digital Libraries

Abstract

We present MELD, the Music Encoding and Linked Data framework, and its application to musicological articles containing dynamic, multi-path explorations of multimedia music digital library resources. MELD uses Linked Data to combine music-related materials including text, audio, video, images, facsimiles, and music scores. To maximise flexibility across this multimedia publication platform, and increase the potential for scholarly reuse of underlying data, we use RDF to model the nature of the resources being deployed, the relationships between them and, separately, how these resources should be visualised and interacted with. Crucially, this gives the ability to relate resources and parts of resources using semantically appropriate terms. Rather than only relating resources by a seconds- or beat-based spine, we encode richer musicological relationships drawing concepts from a Music Encoding Initiative (MEI) document or the Music Ontology, enabling the use of a wide range of music notational concepts. We describe the framework, demonstrating a JavaScript reference implementation and its utility when publishing musicological scholarship. As a case study, we show a MELD-enhanced version of an article written for the British Library on the performance of works by Frederick Delius. The article integrates TEI text, IIIF-served images, MEI notation and recordings of audio and video. We describe the semantic annotations which underpin this realisation, and how they relate users' experiences of moving between the content to the musicological argument being marshalled. Through this example we illustrate how connecting diverse media types using musically-meaningful semantics can support a richer form of interaction with music digital library materials, beyond the current state of the art.