Abstract:

In this presentation, we show how semantic data structuring can be rethought through situated curating within two Wikibase platforms, ParVO and ecHoImages, developed for research in arts education. Tested with arts students, these platforms support the development of decyclopedization, a concept that challenges encyclopedic neutrality, and explore situated knowledge as an archival method. Decyclopedizing becomes a mode of (un)learning, using playful, subversive tactics within the Wikimedia ecosystem to reveal marginal, contextual, and alternative forms of knowledge.

These practices contest assumptions of universality in dominant data models and highlight how relational, partial, and context-dependent forms of knowing transform data conception. By foregrounding error, fiction, and experimentation in pedagogy, the platforms expose the relational logics that shape the visibility and legitimacy of knowledge, rather than presenting closed results.

Building on recent work in situated curating, the project reframes heritage not as a static object but as a living, relational process emerging from artistic engagement with ruins. In these spaces, human, ecological, and more-than-human actors intersect, and we raise a central question: What kinds of platforms and semantic models do we need for arts education grounded in situated curating?

Here, data practices become politically and contextually situated, with knowledge emerging from relational processes rather than pre‑defined taxonomies. This approach challenges traditional archiving and knowledge-organization regimes, requiring semantic structures capable of handling contingent, situated forms of knowledge. As data must remain shareable and reusable, the work aligns with FAIR and CARE frameworks.

Because arts-based, transdisciplinary research operates across evolving questions and heterogeneous systems, the project emphasizes knowledge-graph visual interfaces that treat data as capta, constructed through narrative and relational inquiry.

Methodologically, ParVO and ecHoImages are developed as experimental semantic infrastructures using Wikibase, where vocabularies, entity types, and relations evolve alongside research. Manual, collaborative data entry by students and researchers foregrounds cataloguing as an interpretive practice. We integrate UI/UX testing into the research process, especially in ecHoImages’ React interface, which enables users to follow material trajectories across artistic, ecological, and industrial contexts.

The project draws on feminist and intersectional epistemologies, infrastructural critique, critical data studies, and digital humanities to examine how infrastructures embed values. Situated curating thus serves as a conceptual anchor for ethically and contextually informed data design.

In conclusion, usability tests reveal that students simultaneously learn content and reflect on data practices, engaging with data as capta and inscribing their own narratives into the platforms. Through the lens of situated curating in extreme or ruined environments, the team develops operational vocabularies spanning the arts, architecture, archaeology, and the natural sciences, acknowledging human and more-than-human entanglements. Ultimately, these projects propose situated curating as a new transdisciplinary field, opening future possibilities for digital platforms working in education with linked open data.

Keywords:

Digital Humanities, Situated Curating, Arts Education, Archive, Data Visualisation.