Speakers
Description
The Text+ Registry is a unified cataloguing system, designed to enhance cross-domain discoverability of scientific resources. It addresses a core challenge in the humanities: resources developed across different projects and institutions adhere to divergent metadata standards, lack common access points, and exhibit limited interoperability. The Registry tackles these issues by promoting FAIR data practices across three resource domains — collections, lexical resources, and editions.
The system is grounded on the architectural principles A) metamodeling, B) model-driven design, and C) information layering:
A. Metamodeling enables the concurrent governance of multiple data models without requiring system-wide redesign upon the integration of new resource types.
B. Model-driven design derives application components from formal models, ensuring technology independence and scalability.
C. The layering approach preserves full provenance while producing enriched resource descriptions: metadata ingested from heterogeneous sources such as data centers, research information systems, domain-specific catalogues and databases are stacked as transparent, clearly attributed layers.
Manual curation by domain experts complements automated aggregation without compromising source integrity. Integration with authority files (e.g. GND), the use of controlled vocabularies, an instance of the Research Software Directory (RSD) and SSHOMP create a semantic network connecting textual resources with software tools and institutional actors. A Search API exposes harmonised metadata to external services via standardised interfaces, while DataCite mappings further extend cross-domain discoverability.
The poster showcases the technical implementation addressing harmonisation challenges, practical ingestion workflows, and the transformation of heterogeneous information into rich resource descriptions — positioning the Registry as a component of an interconnected research infrastructure within the NFDI and beyond.