Implementing FAIR in data sharing: who are the actors and what are their responsibilities?

Despite the fact that the implementation of the FAIR principles (Findable, Accessible, Reusable, Interoperable) has become necessary in new research projects to meet the requirements of funding organisations and respond to some funders calls, many institutions do not consider data sharing through implementation of the FAIR principles as a research output, when evaluating researchers. However, this is an essential aspect to ensure reproducibility and re-usability of research and therefore, it should be highly encouraged, recognized as such, and even rewarding this kind of behavior could be considered. . 

In order to make data FAIR, various steps have to be defined to ensure FAIRification. The larger the community, the larger the need for a stepwise FAIRification procedure approved by the whole community. In this chain, each of the stakeholders has to fulfill its functions on long term approaches (governance, policies, data stewardship  research). Several institutions define various stakeholders for FAIR data sharing within the whole disciplinary community. To improve the realisation of the FAIR principles, it is now essential to describe in a common way: i) each of the actors having a possible impact on the FAIRification of research data as well as their roles in these FAIRification processes, ii) the FAIRification landscape at the international level in which each of the stakeholders is located by identifying the resources and networks useful for the different communities and iii) each of these stakeholders (concerning not only FAIRification, but also raising awareness of FAIR, governance of data sharing, training and evaluation of practices and their recognition / encouragement). Based on the ongoing work of the SHAring Reward & Credit (SHARC) Interest Group, this poster proposes i) a description of institutional FAIR landscape, of available resources and a vision of the stakeholders,.ii) a diagram of the links and responsibilities of each actor in an iterative FAIRification process. Of course the variation in the research and data science organisation between institutions has to be taken into account when implementing FAIR principles but this approach and tentative clarification of actors is a first step that can then be adapted to each national or disciplinary context.