Poster

Jan Krasni (TU Berlin)

ERIHAD.OS (Open Science Ethical Research Infrastructure for Hateful, Antidemocratic, and Discriminatory Data) addresses a problem that research data management meets wherever data cannot be released: how can research data be made findable, interoperable, and reusable when making them accessible is ethically problematic or legally critical (or rather impossible)? The project works with collections of antisemitic, racist, extremist, and comparable material, but the underlying problem is general. A growing class of holdings is curated, documented, scientifically valuable, and permanently non-disseminable.

Interviews with ten European research and monitoring organisations produced a finding that cuts against the common assumption. The binding constraint is rarely data protection law on its own. It is the aggregation of several structurally independent closure regimes, above all platform terms of service and copyright, alongside criminal law, contractual commitments to affected communities, and institutional sovereignty over holdings. No single Open Science instrument dissolves that complex nexus of regulations. ERIHAD.OS therefore shifts the question from making data accessible to making data examinable.

That shift has an architectural consequence. ERIHAD.OS explores Trusted Research Environment approaches in which researchers analyse sensitive collections within controlled environments, while the data remain with the holding institution and access is governed according to legal, ethical, and institutional requirements. The project follows the data-visiting model developed in European federated-analysis infrastructures, notably RAISE (certified institutional nodes, persistent identifiers for analyses rather than for datasets) and the Personal Health Train (containerised analysis tasks travelling between data-holding stations). For research data management this reverses the usual object of curation: what is registered, cited and preserved is the analysis, not (always) the corpus.

The poster presents the project's approach to metadata and discovery across distributed collections, tiered and attribute-based access, anonymisation and data minimisation, provenance and documentation, and a governance layer that deliberately leaves final release decisions with the data-holding institutions. It also marks the limits: what a proof of concept can demonstrate within a two-year project, and what would require infrastructure commitments beyond it.

ERIHAD.OS is coordinated by the Center for Research on Antisemitism at Technische Universität Berlin and is funded through the OSCARS Open Call, which has received funding from the European Union's Horizon Europe research and innovation programme under grant agreement No 101129751.