HISMET (HIStorical METadata) is a small, focused project — funded under the NWO Open Science XS scheme — that addresses a specific gap in the current state of archival digitisation: while transcription and segmentation pipelines have matured rapidly, topical metadata creation has lagged behind, and where it exists it is overwhelmingly tuned for researcher-side use rather than for the daily workflow of the archives that hold the source.
The project develops tooling for archive-side topical metadata creation, tested on a concrete corpus and evaluated against both manual archival practice and existing automated subject-indexing systems (notably Annif). It supervises four student assistants and serves, in part, as a training project for the next generation of archive-attentive digital humanists.