Strategic Alignment and Governance

On 13 November 2025, the Advisory Group convened for the first time, marking a strong and focused start to the project’s strategic governance. LI-MA presented recent progress and priorities, aligning the collective agenda around linked open data; introducing our four-level artwork model developed by LI-MA, and discussing sustainable identifiers and scalable services for the sector. The discussion confirmed a deliberately iterative approach: starting with the simple media artworks, scaling towards complex media artworks, and embedding collaboration with national and international partners. 

The Advisory Group, comprising Elsbeth Kwant, Strategic Advisor at the Royal Library of the Netherlands, and Robert Sanderson, Senior Director for Digital Cultural Heritage at Yale University (USA), embraced its role as strategic sounding board, quality guardian and connector, reinforcing LI-MA’s leading and pioneering position. Feedback was unequivocal. As one member observed: “I think digital art does have a luxury here. Others are doing valuable work, but not at this level.” The remark underscored the project’s technical depth, international relevance and momentum, and its potential to set a benchmark for the wider field.

From Modelling Breakthroughs to Collective Scrutiny

Building on this strategic alignment, LI-MA convened leading experts on 14 November 2025. The Expert Meeting built directly on two earlier project updates, “Project Connecting Media Art: Modeling & Mapping” and “A Breakthrough in Data Modelling”. While these articles outlined the project’s aims and conceptual foundations, this meeting marked a deliberate shift in approach. This was not a presentation to the field, but a working conversation with it.

Participants reflected a broad and influential cross-section of the Dutch heritage sector: Netwerk Archieven Design en Digitale Cultuur (NADD), Netwerk Digitaal Erfgoed (NDE), funder Pica, the University of Amsterdam, Eye Filmmuseum, the Cultural Heritage Agency of the Netherlands (RCE), a Wikidata specialist, and LI-MA.

Ahead of the meeting, all reviewed two core documents: “The Four-Level Artwork (FLA) Technical Data Model” and “The Abstract Data Model”, and a set of questions around persistent identifiers (PIDs). The brief was explicit: be critical, identify weak spots and help refine the model so it can function beyond LI-MA’s own collection. That mandate was fully met. Discussions were lively, precise and at times unapologetically complex – reflecting the reality of media art itself. 

Sustainable Identifiers for Media Art

A central theme of the meeting was the implementation of sustainable identifiers for media art – essential if collections want to be truly connected, citable and machine-readable over the long term. The group addressed fundamental questions: at which levels should PIDs be assigned; how should multiple versions and derivatives be handled; and who ultimately bears responsibility when identifiers are permanent, but rights and metadata may change over time.

LI-MA will continue to model all levels (Work, Expression, Manifestation and Item) internally with unique identifiers, while prioritising the most stable levels, particularly Expression, as public landing points. Relationships between versions, files and iterations will be explicitly modelled, making complex works traceable without flattening their richness (this will be addressed in a follow-up project).

Crucially, responsibility for the content remains with the source holder, while the technical creation and distribution of sustainable identifiers and linked data may be delegated to a technical broker. In this role, LI-MA acts as a technical intermediary, supporting institutions in implementation while accountability remains with the source holder.

A Demanding But Powerful Data Model Tailored to Media Art

LI-MA’s FRBR/WEMI-based data model – designed to describe media artworks across the interconnected levels of Work, Expression, Manifestation and Item – was confirmed as a powerful but demanding instrument. The meeting highlighted the need to further explain, using more examples. 

Participants also agreed on the importance of integrating elements beyond the WEMI stack, including the diverse types of documentation and contextual materials characteristic of media art. At the same time, a clear distinction will be maintained between shared artwork records spanning multiple collections and institution‑specific data. 

From Dataset to Linked Open Data

Another focal point was the ambition to register a media art dataset with the NDE Dataset Register and to publish information as linked open data (LOD). The consensus was to start with what is stable, meaningful and reusable. Initial LOD publication will therefore focus on the information from the Expression layer, the information about the artwork, mapped to widely adopted vocabularies such as schema.org, while richer structures remain within internal systems.

Wikidata was identified as a strategic complement to institutional LOD, particularly for international visibility and reuse. Names /persons (of media art artists) stood out as a crucial linking element across collections and domains. 

Importantly, the dataset discussion acknowledged real-world constraints: legacy databases, limited capacity and divergent contracts. The roadmap responds with a stepwise approach, using LI-MA’s own collection as a testbed, and includes a case study with partner the Van Abbemuseum to practically test synchronisation, rights management, and sustainable identifiers strategies.

A Shared Roadmap

Beyond technical outcomes, the meeting reinforced a strongly collaborative approach. Connecting Media Art is firmly positioned as a sector-wide effort, with LI-MA acting as a leading, pioneering and actively engaged driving force. In this role, as the national hub and repository safeguarding over 70 media art collections across the Netherlands, LI-MA provides the shared infrastructure and expertise underpinning this collective roadmap for sustainable digital art preservation.

Building on this foundation, kKnowledge sharing is embedded through the Data Workshop for Digital Art (SBMHK and LI-MAs Digital Art Data Lab), sector meetings and LI-MA’s annual international symposium Transformation Digital Art. The symposium provides a platform for exchange with partners such as Zentrum für Kunst und Medien (ZKM) and Rhizome, contributing to a connected, future-proof media art ecosystem.

From Modelling to Implementation

Next steps are already defined. A follow-up of the Expert Meeting: a focused, hands-on technical working session on the data model and schema.org mapping will take place in January 2026, followed by a lecture at Transformation Digital Art in March 2026 (sign up too!).

At these moments, LI-MA will lead the sharing of insights, the testing of assumptions and the development of a genuinely federated, collaborative infrastructure for media art. Together, these mark the shift from modelling to implementation, and from individual initiatives to shared infrastructure.  

The final LI-MA roadmap, including technical documentation, will be shared at the conclusion of Connecting Media Art in April 2026.

Header: Jan van Munster, Cirkels (1972), 2'38". In collections: Stedelijk Museum Amsterdam, LI-MA.