In support of the WorldFAIR project and other activities to improve the implementation of the FAIR principles, the Cross-Domain Interoperability Framework (CDIF) Working Group is producing a suite of recommendations and guidelines. The CDIF will include modules for various aspects of the FAIR principles, including discovery (this module), data integration, semantic harmonization, data access, and the supporting technology.
This draft of the Discover module is released for public review and feedback: https://doi.org/10.
Colleagues are invited to send any comments to Simon Hodson <simon@codata.org>, Stephen Richard <smrtucson@gmail.com>, Arofan Gregory <ilg21@yahoo.com> by Monday 15 January 2024.
Executive Summary
The Cross-Domain Interoperability Framework (CDIF) provides a set of patterns and recommended practices for research infrastructures servicing cross-domain research areas to support broad reuse of FAIR-data offerings not only within a community but also between communities. The frame-work is intended to provide guidance based on existing practice, not specific to particular domains. The idea is to identify capabilities and component services, along with the information payload–with appropriate models and interfaces–to act as a lingua franca within and between domain and infrastructure boundaries. The framework includes modules for various aspects of interoperability and reuse, including discovery (this module), data integration, semantic harmonization, data ac-cess, and the supporting technology.
Discovery in the context if CDIF is based on providing documentation (a.k.a. metadata) for re-sources—not only file-based data sets, but other resources required for research, data analysis, and policy development. The approach to discovery is based on providing information to support search applications— text-based information like titles and abstracts, as well as information about authors, keywords, spatial and temporal coverage, and publication dates. In the emerging FAIR paradigm, the basic discovery documentation also needs to provide information to support initial assessment of a resource with information about variables reported in the data, measurement methods, processing workflows, data quality, format, and access.
This specification lays out a set of content items required or recommended to support these basic discovery goals, as well as procedures that resource creators can use to make documentation (metadata) describing resources they offer accessible to search aggregators using the internet. Recommendations include implementation of the metadata schema using JSON-LD and the Sche-ma.org vocabulary, broadcast of metadata using standard Web site maps, and embedding metada-ta in resource HTML landing pages. Appendices provide a recommended implementation using Schema.org JSON-LD, with examples, and mapping to Signposting and RDA PID Kernel metadata schemes.