Knowledge graphs as epistemic devices: the experiments in art and technology case study

This paper presents the E.A.T. Knowledge Graph, a linked open data project representing the collaborative networks, experimental artworks, and technological innovations of Experiments in Art and Technology (E.A.T.), a pioneering initiative at the intersection of art, science, and engineering founded in 1966. Focusing primarily on the performance series 9 Evenings: Theatre & Engineering, the project demonstrates how semantic technologies can make complex, distributed cultural practices computationally tractable while preserving their experimental and relational character. Working from archival sources including correspondence, technical diagrams, planning notes, and collaboration records, we developed a knowledge graph using Wikibase as infrastructure. The modeling approach addresses three core dimensions: collaborative networks that crossed disciplinary boundaries, performance artworks that challenged conventional classifications, and composite technological objects assembled from commercial and custom-engineered components. Our methodology emphasizes flexible, iterative ontological development, granular provenance tracking, and progressive enrichment of entity records as new sources are processed, allowing the data model to adapt as archival materials surface unanticipated representational needs. The resulting knowledge graph functions as an epistemic device supporting multiple access modes: exploratory browsing through entity pages, targeted queries revealing collaborative patterns and role-identity crossings, network visualizations, and an interactive hierarchy browser for materials and technologies. Published as linked open data, the project enables cross-institutional reuse and offers a methodological framework for representing complex cultural practices in archival contexts, providing new pathways for discovery, analysis, and interpretation in cultural heritage research.