2026-10-29 –, LUMC01
Using an existing, established PID system, a library consortium and government bodies are working together to create open data that can be attributed and which is linked and future-proof.
Two organisation from different disciplines have joined forces to enhance messy metadata into clean, reliable and connected data. Rather than dealing with variations like abbreviations, renamings, or typos, a single persistent identifier links datasets across data portals and over time. Replacing free text with GND URIs transforms open governmental metadata into real Linked Open Data — machine-actionable, traceable, and interoperable.
This high-level view is all about how different fields can work together to break down walls between people's ways of thinking. This is a real-life example of what works, what breaks systems, and the common pitfalls of public-sector cooperations.
This session examines how persistent identifiers (PIDs) are used to improve the quality, interoperability, and long-term usability of open government data in Germany. It focuses on a collaboration between the German National Library (DNB) and the national open data portal GovData, which promotes usage of the integrated authority file - Gemeinsame Normdatei (GND) - into public sector metadata workflows.
Open government data is frequently limited by inconsistent and ambiguous metadata, particularly in free-text fields such as publisher or organizational names. Variations in spelling, abbreviations, and naming conventions reduce discoverability and hinder reuse. This session addresses the problem by demonstrating how the use of stable identifiers, specifically GND URIs, enables the transformation of heterogeneous metadata into structured, machine-readable Linked Open Data.
The presentation provides an overview of the conceptual and technical approach taken in this collaboration. It outlines how authority data from the library domain can be applied to administrative datasets, including the mapping of public sector entities to existing authority records and the handling of cases where identifiers are missing or incomplete. The session also discusses the integration of PID-based workflows into existing data infrastructures and publication processes. Aside from technical aspects, the session focuses and highlights organizational and cultural challenges encountered during the collaboration. Differences in terminology, data practices, and professional perspectives between library and government stakeholders required alignment and iterative coordination. The presentation reflects on these experiences, identifying key factors that enabled progress as well as common pitfalls that affected implementation.
The session contributes to ongoing discussions on Linked Open Data and cross-sector interoperability by providing a concrete, replicable use case. It demonstrates how established authority systems can be reused beyond their original domain and how cooperation between institutions can support more consistent and reliable metadata practices. Also, how a transfer of knowledge between domains increases efficiency and provides benefits on all sides.
Participants will gain a clear understanding of the existing limitations of free-text metadata in open data portals and the benefits of adopting PID-based approaches. They will also learn about practical steps for integrating authority identifiers into their own systems, focusing on considerations for governance, data modeling, and workflow design. The session is relevant, among others, for professionals working with open data, metadata management, digital libraries, and public sector, by offering actionable insights for improving reliability and data interoperability across domains.
