Considerations for FAIR data catalogs from DAPHNE4NFDI

Large-scale photon and neutron (PaN) experiments generate vast amounts of valuable data, yet implementation of the FAIR principles (findable, accessible, interoperable and reusable) remains incomplete. Metadata catalogs are a key tool for addressing this challenge. They do not store data itself but provide structured searchable descriptions that make datasets discoverable and reusable. This article, developed within the DAPHNE4NFDI consortium of Germany's National Research Data Infrastructure (NFDI), reflects the discussion on the role of metadata catalogs in PaN science such as SciCat, ICAT and SampleDB from a scientific point of view, and presents the status of current implementations. Integration of standards such as NeXus, sample databases and electronic laboratory notebooks is also addressed. Catalog implementations are categorized by purpose as raw data, institutional data and public data, and examined by practical aspects including authentication and authorization, persistent identifiers, and metadata enrichment. Drawing on case studies such as the RefXAS X-ray absorption spectroscopy database, we summarize lessons learned and present recommendations for minimizing fragmentation, enabling interoperability, and harmonizing identity and persistent-identifier policies. Looking ahead, semantic technologies and knowledge graphs will play a central role in building a federated FAIR-compliant catalog infrastructure that supports both facility and university researchers.