The lecture will address the limitations of cultural archives as data models for humanities research. Digital collections and platforms often prioritize the cultural data of certain countries, capital cities, and institutions, due to a lack of data transparency and limited availability of previously digitized content with lower representation. The results of my empirical research support the idea that large-scale analysis of archival catalogues should include an understanding of the socio-technical contexts of data creation, such as metadata. An awareness of lengthier structures, such as contextual data, data on object biographies, para-data or historical interpretation may add depth and meaning to subsequent analysis.
The lecture will situate the "collections as data" movement as a practice which requires a deep understanding of documentary, legal, technological and interpretational circumstances of data production for humanities research. In the knowledge environment where AI generated images may further reduce cultural diversity, understanding how data models are produced is an important prerequisite for studying cultural diversity and change.
