Findability

Foster improved findability of research data and other research-relevant digital objects from public funding, for example, by assigning unique digital persistent identifiers and publishing descriptive metadata.

To improve the findability of publicly funded research data, the following measures are recommended:

• Require Public Research Organizations (RPOs) to establish open repositories that provide public access to research data.

• Promote the use of common metadata standards while encouraging the development of disciplinary-specific metadata supported by research communities.

• Ensure Persistent Identifiers (PIDs) are assigned to all research objects, including software, workflows, and physical objects.

• Funders and RPOs should establish clear guidelines for trusted repositories, PIDs, and minimum standards for FAIR metadata, with associated costs included in research calls.

• Governments should promote internationally recognised standards and gradually introduce requirements through funding agencies while fostering consensus among researchers.

• Support cross-referencing services to connect local repositories without requiring uploads to external repositories, ensuring accessibility and cost efficiency.

• Invest in a powerful search engine to improve accessibility, addressing gaps in existing tools like B2Find and Google Dataset Search.

Implementing persistent identifiers (PIDs) and standardised metadata for research data poses several challenges. 

• Many disciplines lack common or disciplinary-specific metadata schemas, necessitating community-driven efforts to develop and adopt standards. 

• Building and maintaining FAIR-compliant data catalogues and repositories requires significant infrastructure, resources, and international collaboration.

• Adopting PIDs and metadata standards, as recommended by initiatives like Horizon Europe, is resource-intensive, requiring systematic planning, regular updates, and researcher engagement.

• The lack of robust search engines capable of advanced data discovery further hampers findability.