FAIR Digital Object Ecosystem Testbed

This poster describes the development of a testbed for FAIR Digital Objects, consisting of an ecosystem of interacting services to demonstrate mandatory and optional FAIR use cases and to identify gaps in the specifications. Preprocessing data for research, like finding, accessing, unifying or converting, takes up to 80% of research time spans. The FAIR (Findability, Accessibility, Interoperability, Reproducibility) principles aim to support and facilitate the reuse of data, and are therefore tackling this problem. A FAIR Digital Object (FAIR DO) is one way to capsule research data resources of all kinds (raw data, metadata, software, ...) so they are following the FAIR principles. A FAIR DO ecosystem can be regarded as a set of services to enable the creation and use of such FAIR DOs. Besides basic functionality like PID management and PID record validation, it may also offer assistive services, i.e. for the automated building of a search index to allow reverse-searching of PIDs. To establish a FAIR DO ecosystem, it must reach a certain level of maturity in order to be used productively. To identify gaps in specifications and concepts during development and use, as well as to demonstrate necessary and optional use cases, we developed a testbed for FAIR use cases as a part of the HMC project, which is easy to set up and run on everyday computers. Currently, the testbed enables PID record management and validation using a PIT service implementation following the RDA PID Information Types (PIT) Working Group Recommendations and an externally hosted Data Type Registry following the RDA Data Type Registry Working Group Recommendations . It also features automated indexing of PID records (proof-of-concept), and provides an implementation of the Collection API specification, which was published by the corresponding RDA Research Data Collections Working Group . The most important gap identified, and the most difficult to close, is the design and specification of the profiles that determine the content of PID records. Software in use by researchers will need to use these contents to determine whether and to what extent it can use a FAIR Digital Object, so this gap needs to be strongly considered.The testbed development has been supported by the research program ‘Engineering Digital futures’ of the Helmholtz Association of German Research Centers and the Helmholtz Metadata Collaboration Platform