Between March and November 2019, FAIRsFAIR ran a Europe-wide survey amongst repository managers, research data managers, policy makers, service providers, and data stewards, and undertook desk research and individual interviews in order to assess the prevailing FAIR policy and practice landscape, Specifically, the aim was to ascertain:
The findings from this work are detailed with preliminary recommendations in the reports D3.1 FAIR Policy Landscape Analysis and D3.2 FAIR Data Practice Analysis, both published in December 2019 and made available to the research and other stakeholder communities for perusal and comment.
The just-published reports D3.3 Policy Enhancement Recommendations and D3.4 Recommendations on practice to support FAIR data principles represent refinements to the initial recommendations and are currently available for public comment. Further revisions to these living documents will reflect the public feedback received as well as the forthcoming work of other projects funded under the INFRAEOSC-05-2018-2019 call and other relevant initiatives.
FAIRsFAIR’s 2019 study of the data policy landscape detailed in D3.1 FAIR Policy Landscape Analysis revealed that the policies of funding bodies, publishers, journals and research performing organisations (RPOs) reflect the priorities outlined in the Turning FAIR into Reality Report of the European Commission to some extent. However, given its importance within the FAIR ecosystem, there is still much that needs to be done to foster and harmonise policy to support the aims of the European Open Science Cloud and realise the vision of the TFiR report.
An overview of the practical recommendations for policy enhancement presented in the current report appears below.
The need for training: Researchers and data stewards should receive practical guidance on how to implement FAIR within different domains – particularly as regards describing data using appropriate metadata standards, data tags and ontologies. A commitment is needed from all stakeholders to support and meet training needs relating to Open Science.
Structured data markup schemas: Policies should be described consistently using a structured data markup schema and consisting of an agreed set of rules which supports both human and machine readability.
Assignation of persistent identifiers (PIDs): Policies should be clearly versioned, assigned with PIDs, and registered in the metadata records of registries such as FAIRsharing.org and the Data Policy Standardisation and Implementation Interest Group of the Research Data Alliance (RDA).
Clearer definitions of data and expectations around sharing: Working with research communities to define data outputs, policymakers should adopt standard descriptions to ensure that definitions provide clarity on the range of outputs that should be considered and what might be considered “FAIR enough”. In addition, standardised exceptions for not sharing data should be developed and added to the metadata schemas used by repositories.
Harmonisation of requirements for research data management (RDM) and data management plans: In consultation with all stakeholders, data management planning should be supported across the entire research lifecycle so that data can be “born FAIR” and kept “FAIR enough” over time.
Clarification of eligible RDM and data sharing costs: Building upon previous work on defining cost types, funding bodies and research performing organisations should be assisted in implementing these in new grant applications and in monitoring and reviewing RDM costings to assess the effectiveness of current cost models.
Data citation requirements: Funding bodies and publishers should clarify their data citation requirements and provide clear guidance on how to meet these requirements in a standardised way.
Development of sustainable business models: To ensure the costs of making and keeping data FAIR over time are split more equally between stakeholders, funding bodies, RPOs and repositories should assess and report on the costs of making and keeping data FAIR to build up a picture of how the costs might change over time.
To download a copy of the report from ZENODO click here.
In the report D3.2 FAIR Data Practice Analysis the writers identified four main themes around which to propose recommendations aimed at helping research support professionals and their institutions, repositories, infrastructures, and journals take practical actions to realise a FAIR ecosystem.
The current report builds on the initial proposal and provides ten recommendations grouped in accordance with these these four themes. Over the next 18 months, FAIRsFAIR will be developing guidance resources in line with these recommendations to support further adoption of FAIR data standards and practices by research communities.
An overview of the ten recommendations detailed in the report appears below.
Theme A: Develop and implement data sharing and interoperability frameworks
A1: Describe research outputs using agreed terminologies and metadata standards to make data FAIR
A2: Build a culture of data citation
Theme B: Ensure data management is supported by data management plans (DMPs)
B1: Formalise and support appropriate data management plans (DMPs) for FAIR data
B2: Develop roadmaps, guidance and workflows for machine-actionable data management plans (DMP) to inform FAIR data stewardship
Theme C: Develop professional support for FAIR data
C1: Define and manage FAIR support costs and resources
C2: Develop and implement models for coordinating and supporting data stewards and research software engineers
C3: Develop and implement terminology for competence centres to annotate and retrieve training materials on enabling FAIR
C4: Develop and implement a self-assessment framework for Research Infrastructures, institutions, and other FAIR competence centres
Theme D: Ensure trusted curation of data
D1: Develop and implement guidance and support for selection of appropriate trusted digital repositories (TDRs)
D2: Develop and implement guidance and support for making sensitive data FAIR for reuse