A workflow to publish Collections as Data: the case of Cultural Heritage data spaces

Cultural Heritage institutions have been making their digital collections available for the public for several decades. Advances in technology such as Artificial Intelligence and Machine Learning have provided a new context in which digital collections can be analysed using computational methods. Initiatives such as Collections as data and the FAIR data principles, have emerged to provide best practices and guidelines for publishing digital collections suitable for computational use. These initiatives are complemented with the CARE principles to strengthen ethical considerations in data governance and reuse. In parallel, experimental Labs have been implemented in Galleries, Libraries, Archives and Museums (GLAM) in order to reuse the digital collections.

Data spaces have emerged as an innovative way to publish and reuse digital collections. Based on previous work undertaken in the context of the GLAM Labs Community, this workflow provides a set of steps to publish Collections as data. It aims to guide and encourage cultural heritage institutions, step by step, in publishing their collections, so that they are suitable for computational use. It is important to note that the checklist items are iterative in nature and there is no particular order to apply them. Each institution can choose which items to use. Priorities depend on the purpose, the context, the content, the intended use or target users of the dataset.

This workflow has been developed in the context of the common European data space for cultural heritage. It builds on the International GLAM Labs’ Checklist to Publish Collections as Data in GLAM Institutions. Our intention is to further enhance and refine this workflow based on community feedback and contributions to facilitate the publication of Collections as data.