Together with EOSC-hub, RDA Europe and FAIRsFAIR, OpenAIRE recently organised two workshops titled “Services to support FAIR data; tyring to explore how existing infrastructures can work together and understand how to deliver services that support the creation of FAIR research outputs. These are part of 3-workshop series: the first two already took place in April, and a proposal for the third one has been submitted to the call for workshops of the Open Science Fair Conference in Porto in September 2019.

The  introduction set the context of FAIR in the EOSC and immediately two key current developments were presented: the goals and mode of operation of the EOSC WG on FAIR by Sarah Jones (DCC) and the RDA WG FAIR Maturity model goal and interim results from Kostas Repanas (DG RTD).

Three (3) implementation stories following services within the Research Data Lifecycle illustrated what aspects of FAIR and how services supported them, focussing mainly on interoperability challenges as key to successfully implement FAIR data and FAIR services:

• CoreTrustSeal, how repsository certification can add to FAIRness of data (Mari Kleemola from Finnish Data Archive and CESSDA)
• EOSC-hub services, PIDs in action for interoperable services (Baptise Grenier & Enol Fernandez from EGI)
• PIDs for FAIR data, A PID Graph for allowing discovery of research (Maaike de Jong, FREYA)

and explored the issues of preserving data and make long-term preservation and explicit need, along with the necessity to validate the approach of structuring services with real research communities, to conclude with addressing the gaps in PIDs availability.

The unconference second part of the workshop allowed for interaction and discussion (four groups) around key topics related to services: perceived costs and time investment in making data FAIR, the main challenge(s) to address, the cultural changes required, competing standards, required organisational structures or policies, and needed skills or training to effectively produce, curate, preserve and disseminate research data. The participants were also asked to comment on the implementation priorities, the areas of intervention, any expectations from the EOSC Governance and data infrastructures, and to conclude, the identification of support mechanism to implement services for FAIR data.

The discussion in the different groups explored existing issues around FAIR, such as the too ambitious definition of FAIR and the missing statement on why researchers should be FAIR or adhere to the FAIR principles.

All groups discussed the different approaches to FAIR among disciplines, where in some cases researchers don’t feel the need or the push to share research data outside their community as none in there has ever been tasked with looking at the “outside” the community of practice; in other cases, the uptake of the FAIR principles is easier because it’s in the nature of the community itself. In general terms, the need for specific services, support, skills development and training varies tremendously from one discipline to another, and change remains domain-specific. Nonetheless, increasing the amount of cross- and inter-disciplinary research is a strong driver for greater “harmonisation” of FAIR-like approachers between disciplines. When discussing communities’ approaches, a certain resistance to change emerges, that has to be addressed with an approach demonstrating the results and the benefits.

while key points for next steps include the following: