Universal Access to Medieval Manuscripts

The corpus of digital facsimiles of medieval manuscripts is expanding rapidly due to large-scale digitisation and the widespread adoption of IIIF. With the development of Handwritten Text Recognition, the stage is set for unprecedented exploration of this prominent category of heritage objects. Yet digitised manuscript images are still largely considered as the final stage of datafication, rather than the starting point for structured data creation. As a result, increased image availability has not fundamentally transformed data curation practices, which remain typically bespoke, fragmented, and distributed across multiple specialized tasks, of which HTR is just one. This situation severely inhibits the operationalisation of medieval manuscripts for research and impact. This workshop aims to lessen the gap between manuscript digitisation and datafication by facilitating structured data retrieval directly from the digital reproductions. In an intensive 3-day hackathon, teams composed of computer scientists, heritage professionals and domain experts in Digital Manuscript Studies will test and further develop technologies for the automated identification of the contents of medieval manuscripts. We will test methods based on textual and visual clues to determine the most performant approaches while also considering the workflows and day-to-day practices of the heritage professionals who would be using these tools. The hackathon will be embedded in a conversation involving both policy makers in the heritage domain and digital infrastructures for Humanities research to ensure optimal impact of the workshop’s outcomes.