2023-10-11 –, Conference room
The Rijksmuseum stores vast amounts of contextual information about its collection in different systems. Most systems support specific processes, for example documentation and research. Integrating the data would have many benefits, providing users with richer experiences. Linked Data could be a technical solution to this challenge, but how does a museum go about implementing integration infrastructure?
We approached Linked Data not as an end-product, but as an essential technology at the center of our collection data infrastructure. This brought additional requirements, in terms of stability, scalability and maintainability. An architecture design resulted in a daunting list of skills that would be required for implementing the infrastructure, revolving around code development, infrastructure and data engineering.
This presentation will discuss how we evolved from a one-and-a-half-member team into a ten-member multidisciplinary engineering team, with experience in academia, cultural heritage and industry. Part of the Research Services department, we have close ties with information specialists that steward data in the source systems. We adapted new practices: Scrum, Infrastructure as Code and DevOps. We collaborate with our website builder in making our Linked Open Usable Data usable for them.
In the future we want this infrastructure to be a sound foundation to build new services upon. We foresee a core team at the museum, which is scaled up on a project basis. Projects could entail adding new source systems, data services or retrieving data from external source.
Since June 2019 Chris Dijkshoorn has been the head of the Collection IT department at the Rijksmuseum. Before that, he worked as data manager at the museum. He implemented the Accurator nichesourcing platform as part of his PhD at the User Centric Data Science group at the Vrije Universiteit Amsterdam. Doing research at the Rijksmuseum, British Museum and the Netherlands Institute for Sound and Vision provided valuable insights in the requirements of institutions regarding Linked Data and crowdsourcing.