Data Discovery in Practice. Text+ and DataGEMS between Community and Scalability.

The European Open Science Cloud (EOSC) aims to improve access to and reuse of research data across disciplines. Within this framework, both national infrastructures and projects funded by European research initiatives play a crucial role in shaping the evolving field of research data management (RDM). This contribution explores the intersections and distinctions between Text+, a German NFDI consortium dedicated to language- and text-based research data, and DataGEMS, a European project launched in 2025 under the umbrella of EOSC, which aims to offer a domain-independent platform for data discovery and management. Existing data discovery systems have been criticized for offering limited search functionalities, inadequate metadata, and a lack of support for complex queries, hindering efficient data discovery and reuse; challenges that both Text+ and DataGEMS seek to address in their respective ways. Text+ supports research in the humanities and social sciences through the curation of structured, language-based resources such as corpora, lexical data, and digital editions. Its approach is shaped by established disciplinary practices, long-term community involvement, and a focus on sustainability and interoperability. Text+ operates within a national infrastructure context, with clearly defined data types and usage scenarios. In contrast, DataGEMS addresses more general challenges in data discovery by offering exploratory search and additional data analysis capabilities. These include the use of natural language queries, pattern-based search, and guided workflows for dataset exploration and comparison. The system is designed to handle diverse data types—structured, unstructured, historical, and real-time—and aims to support users with varying levels of technical expertise. Initial use cases focus on meteorology, education, and language data. While Text+ targets a domain-specific community, DataGEMS aims to provide a scalable solution across a wide array of disciplines. Despite differing scopes and target audiences, both initiatives share core objectives such as enhancing data FAIRness and improving accessibility. Particularly in the context of language data, potential areas for technical and conceptual integration can be identified. For instance, Text+ could explore how DataGEMS' profiling and content-aware search mechanisms might support new forms of data discovery, while DataGEMS could benefit from language data domain expertise developed within Text+. The comparison illustrates how specialized and generic approaches to RDM can complement one another, and a collaboration could further explore the challenges that may arise when attempting integration across different levels of abstraction or user expectations. By providing a flexible and content-aware data discovery layer, DataGEMS can therefore act as a communicative interface between European-level infrastructures like EOSC and domain-specific initiatives such as Text+. DataGEMS can offer a platform where curated language resources from Text+ can be made more visible and reusable in broader research contexts. Conversely, the structured, high-quality metadata and standards from Text+ can inform and enrich the discovery mechanisms within DataGEMS. This bidirectional potential highlights the role of DataGEMS not only as a technical solution, but as a mediator between domain-driven and cross-domain RDM efforts. As both infrastructures evolve, identifying interfaces for collaboration—particularly around metadata enrichment, discovery workflows, and user engagement—can strengthen data reuse practices across disciplinary and infrastructural boundaries.