D2.4 2nd Report on FAIR requirements for persistence and interoperability

This document is the second iteration of three reports on the state of FAIR in the European scientific data ecosystem, by the FAIRsFAIR project. This report focuses on providing relevant current information about persistent identifiers, semantic interoperability and technical implementations of the FAIR data principles. The report advises researchers, data-stewards and service providers to co-create and co-develop solutions case-by-case, but with a strong endeavour towards a larger FAIR ecosystem, seeking sustainable and cost-effective solutions.

This report is the second deliverable by the FAIRsFAIR project on technical implementation of FAIR principles. The first deliverable was a landscaping effort - to a general audience - that reviewed and documented commonalities and possible gaps regarding semantic interoperability, and the use of metadata and persistent identifiers across infrastructures. This report builds on the previous work and explores current developments to increase awareness on what good FAIRness means and how it could be promoted in practice. 

The FAIR data principles have varying implications for different stakeholders. Thus, our aim is to provide an explanatory guide to researchers, data stewards and - where possible - service providers on the use of PIDs, metadata and semantic interoperability. We are presenting the information in sections geared towards a specific target audience i.e researchers, data stewards & service providers, with a focus on highlighting the aspects most relevant to the particular stakeholder group. 

In order to achieve wide penetration and the potentially significant benefits of FAIR data, it is important for the development and implementation of FAIR data principles to be driven by researcher needs. Our main conclusions are as follows:

1. A generic solution for achieving FAIRness does not exist. The solutions should be selected and decisions made on a case-by-case basis. The assessment of FAIR data solutions should always start from the user needs but always with respect to the user’s larger research community.

2. Every effort to make something FAIR should balance the investments needed to implement each FAIR principle, and the expected benefits of FAIRness to the scientific community.

3. In order to achieve a FAIR data ecosystem with sustainable PIDs, metadata and semantic artefacts, researchers, data stewards and service providers should work together on technical solutions. 

4. Achieving Interoperability for both humans and machines requires a large investment, but it has promising benefits. Technology can solve a lot of the interoperability problems at a technical level - but this does not solve misunderstandings at the semantic level - humans still need to communicate, agree on terms and vocabularies. It is important to take advantage of existing frameworks to build cohesion.

We welcome comments and feedback. It is possible to comment here: https://docs.google.com/document/d/1h8yAlK8o3SCjG_tgE_fIB66F68aomMe6MuDv5oDK-4U/edit?usp=sharing

We are working on a version which will include changes on the basis of the feedback, to be published in the first quarter 2021.