RKD STUDIES

The transformation of the DttG database into an interactive tool demonstrates the transition from a personal archive into a publicly accessible research resource. Art historical data often begins its life in solitary form: spreadsheets on individual laptops, or informal collections of notes, images, and transcriptions. While these formats are convenient at first, they are difficult to query in sophisticated ways and rarely shared beyond the individual project. As a result, valuable data often remains inaccessible, and duplication of effort is common. Researchers may spend months transcribing or translating primary sources that others have already studied but never disseminated. The DttG project sought to unlock this potential by building a system that supports more nuanced exploration of data while enabling it to be shared widely.

A central principle in the development of the DttG database was openness at both the level of data and infrastructure. The web-based application provides public access to the dataset, and the Django codebase and database schema are released under an open license and hosted on GitHub.¹ This allows other researchers to examine, reuse, or adapt the framework for their own projects, which will be discussed in more detail in section 3.4. The approach follows the precedent of earlier RKD digital initiatives such as Counting Vermeer, where the software for weave analysis was made freely available alongside the research outcomes.² Publishing the DttG code encourages transparency in how the system functions, provides a tested framework that other projects can build upon, and increases sustainability, since a wider community of developers can contribute to its maintenance.

Alongside the open release of the application itself, the project also follows open-data principles. The dataset can be exported in standard CSV format, allowing researchers to work with the raw data outside the web interface, whether for advanced statistical modelling, visualisation, or integration with other datasets. This flexibility is essential in digital humanities work, where tools may evolve but the underlying data itself must remain reusable and enduring. Making the dataset downloadable also supports verification and reproducibility: scholars can replicate queries, test results, or apply alternative analytical methods. Appropriate care was taken to respect intellectual property and privacy concerns, particularly for partner museum data. As a result, while the functionality is built-in to support images of objects and cross-sections, at present these have not been implemented yet.