11/08/2026
1. Building the M-PATCH Metadata Database:
Laying the Foundations for Linked Data Research
M-PATCH is being developed around a prototype corpus of 189 Latin hagiographical texts from the Long Tenth Century. Researchers at Ghent University use this corpus to test annotation tools and generate data that support the development of the project's Linked Data infrastructure. At the same time, these data help researchers study how texts changed as they were copied, edited and transmitted over time.
To make this possible, the project requires detailed and highly structured data not only about the texts themselves, but also about the manuscripts in which they survive and the editions through which they have been studied. Researchers need to know where and when manuscripts were produced, how texts circulated, and how editors selected and used manuscript evidence. This information is essential for understanding textual variation across the corpus.
No existing resource brought together all this information in a structured and reusable way. As a result, the project team set out to build its own database from scratch.
In April 2025, Steven Vanderputten started compiling and organising the data. What began as a spreadsheet gradually grew into a much larger resource than originally anticipated. Today, it contains information on 189 texts, more than 1,600 manuscripts and over 800 editions.
As the dataset expanded, managing the information in Excel became increasingly difficult. In July 2026, GhentCDH developers Joren Six and Miel Peeters migrated the data to a relational database using Mathesar, an open-source interface for PostgreSQL databases.
The new database makes it easy to explore connections between texts, manuscripts, and editions. Researchers can identify which manuscripts hold particular texts, trace where manuscripts were produced and later kept, examine which sources were used by editors, and investigate relationships across the corpus as a whole. The database also includes links to catalogues, IIIF manifests, secondary literature, and other digital resources.
What started as a spreadsheet has now become one of the core research infrastructures of the M-PATCH project. It will serve as the foundation on which the project's Linked Data environment will be built.
The database is still being refined, and a user interface is currently under development. In addition, work continues on improving the datasets as well as their relationships. To support this effort, several student assistants are helping the team verify metadata, check geographic information, retrieve IIIF manifests, and validate links between records. Loes Adriaenssens, Servaas Gevaert, Oshin Ghiotto, Korneel Huybrecht, Tibo Moreel and Florine Zeegers are currently adding and cross-checking metadata. Soon, even more students will join the project to work on manuscript transcription. Although development is ongoing, most of the core structure is now in place. With the database continuing to grow and improve, M-PATCH is building one of the most extensive structured datasets for the study of medieval hagiography. This resource will provide the foundation for the project's Linked Data infrastructure and future large-scale analyses of textual transmission.
