fair-data-eg / action-plan Goto Github PK

Funders should report annually on the outcomes of their investments in FAIR and track how the landscape matures. Specifically, how FAIR are the research objects that have been produced and to what extent are the funded infrastructures certified and supportive of FAIR data.

• Statistics should be published on the outcome of all investments to report on levels of FAIR data and certified services
  Stakeholders: Funders; Institutions.

• The results of monitoring processes should be used to inform and iterate data policy.
  Stakeholders: Policymakers; Funders; Institutions.

The Rules of Engagement for EOSC must be broadly-defined and open to enable all existing service providers to address the criteria and be part of the European network.

• The Rules of Engagement for EOSC must be consulted on widely, drawing in views from a broad range of stakeholder groups beyond the core European Research Infrastructures and E-Infrastructures to include research communities, institutions, publishers, commercial service providers and international perspectives.
  Stakeholders: Data services; Research communities; Institutions; Publishers.

• The resulting Rules must be fit-for-purpose to enable all existing data services and capacities developed by different communities to be exploited for best return on investment. The Rules should be reviewed regularly to ensure they remain viable.
  Stakeholders: Data services; Research communities; Policymakers.

Repositories need to be encouraged and supported to achieve CoreTrustSeal certification. The development of rival repository accreditation schemes, based solely on the FAIR principles, should be discouraged.

• A programme of activity is required to incentivise and assist existing domain repositories, institutional services and other valued community resources to achieve CoreTrustSeal certification.
  Stakeholders: Funders; Data services; Standards bodies.

• A transition period is needed to allow existing repositories without certifications to go through the steps needed to achieve trustworthy digital repository status. Concerted support is necessary to assist existing repositories in achieving certification.
  Stakeholders: Data services; Institutions; Data stewards.

• At an appropriate point, the language of the CoreTrustSeal requirements should be reviewed and adapted to reference the FAIR data principles more explicitly (e.g. in sections on levels of curation, discoverability, accessibility, standards and reuse).
  Stakeholders: Global coordination fora; Data services; Institutions.

• Repositories may need to adapt their services to enable and facilitate machine processing and to expose their holdings via standardised protocols.
  Stakeholders: Data services; Institutions.

• CoreTrustSeal should also be supported to achieve scalability to meet the needs of repository certification in the FAIR context.
  Stakeholders: Funders, Standards bodies.

• Mechanisms need to be developed to ensure that the repository ecosystem as a whole is fit for purpose, not just assessed on a per repository basis.
  Stakeholders: Global coordination fora; Research communities.

Agreed sets of metrics should be implemented and monitored to track changes in the FAIRness of datasets or data-related resources over time.

• Repositories should publish assessments of the FAIRness of datasets, where practical, based on community review and the judgement of data stewards. Methodologies for assessing FAIR data need to be piloted and developed into automated tools before they can be applied across the board by repositories.
  Stakeholders: Data services; Institutions; Publishers.

• Metrics for the assessment of research contributions, organisations and projects should take the past FAIRness of datasets and other related outputs into account. This can include citation metrics, but appropriate alternatives should also be found for the research / researchers / research outputs being assessed.
  Stakeholders: Funders; Institutions.

Data skills of various types, as well as data management, data science and data stewardship competencies, need to be developed and embedded at all stages and with all participants in the research endeavour.

• Data skills, including an appropriate foundational level of in data science and data stewardship, should be included in undergraduate and postgraduate training across disciplines, and in the provision of continuing professional development (CPD) credits for researchers.
  Stakeholders: Institutions; Data services; Research communities.

• More in-depth data science and data stewardship skills should be embedded in Master’s degree courses for Information Professionals, so future generations of librarians, archivists and information systems staff are equipped to deal with the increasing complexity of research outputs.
  Stakeholders: Institutions; Data services.

The realisation of FAIR data relies on, at minimum, the following essential components: policies, DMPs, identifiers, standards and repositories. There need to be registries cataloguing each component of the ecosystem and automated workflows between them.

• Registries need to be developed and implemented for all of the FAIR components and in such a way that they know of each other’s existence and interact. Work should begin by enhancing existing registries for policies, standards and repositories to make these comprehensive, and initiate registries for DMPs and identifiers.
  Stakeholders: Data services; Standards bodies; Global coordination fora.

• By default, the FAIR ecosystem as a whole and individual components should work for humans and for machines. Policies and DMPs should be machine-readable and actionable.
  Stakeholders: Data services; Global coordination fora; Policymakers.

• The infrastructure components that are essential in specific contexts and fields, or for particular parts of research activity, should be clearly defined.
  Stakeholders: Research communities; Data stewards; Global coordination fora.

• Testbeds need to be used to continually evaluate, evolve, and innovate the ecosystem.
  Stakeholders: Data services; Data stewards.

Data repositories and other components of the FAIR data ecosystem should be supported to explore business models for sustainability, to articulate their value proposition, and to trial a range of charging models and income streams.

• Examples of different business models should be shared, and data services given time and support to trial approaches to test the most viable sustainability paths.
  Stakeholders: Funders; Data services; Global coordination bodies.

Steps need to be taken to develop two cohorts of professionals to support FAIR data: data scientists embedded in those research projects which need them, and data stewards who will ensure the management and curation of FAIR data.

• Formal career pathways must be implemented to demonstrate the value of these roles and retain such professionalised roles in research teams.
  Stakeholders: Institutions; Global coordination fora.

• Key data roles need to be recognised and rewarded, in particular, the data scientists who will assist research design and data analysis, visualisation and modelling; and data stewards who will inform the process of data curation and take responsibility for data management.
  Stakeholders: Funders; Institutions; Publishers; Research communities.

• Professional bodies for these roles should be created and promoted. Accreditation should be developed for training and qualifications for these roles.
  Stakeholders: Institutions; Data services; Research communities.

Research funders should require data management costs to be considered and included in grant applications, where relevant. To support this, detailed guidelines and worked examples of eligible costs for FAIR data should be provided.

• Details on the costs of data management, curation and publication should be included in all DMP templates.
  Stakeholders: Funders, Institutions, Data services.

• Guidelines should be provided for researchers and reviewers to raise awareness of eligible costs and reinforce the view that data management, long term curation and data publication should be included in project proposals.
  Stakeholders: Funders; Institutions.

• Information from existing and completed projects should be used to retrospectively identify costs and develop examples and guidelines based on these.
  Stakeholders: Funders; Institutions; Data services.

Research communities must be supported to develop and maintain their disciplinary interoperability frameworks. These incorporate principles and practices for data management and sharing, community agreements, data formats, metadata standards, tools and data infrastructure.

• Enabling mechanisms must be funded and implemented to support research communities to develop and maintain their disciplinary interoperability frameworks.
  Stakeholders: Funders; Standards bodies; Data services; Global coordination fora.

• Disciplines and interdisciplinary research programmes should be encouraged to engage with international collaboration mechanisms to develop interoperability frameworks.
  Stakeholders: Funders; Policymakers; Institutions; Data stewards; Global coordination fora.

• Mechanisms that promote the exchange of good practices and lessons learned within and across disciplines should be facilitated.
  Stakeholders: Data services; Research communities; Global coordination fora.

Systems providing citation metrics for FAIR Data Objects and other research outputs should be provided. In parallel, next generation metrics that reinforce and enrich citation-centric metrics for evaluation should be developed.

• Citation of data and other research outputs needs to be encouraged and supported, for example by including sections in publishing templates that prompt researchers to reference materials, and providing citation guidelines when data, code or other outputs are accessed.
  Stakeholders: Publishers; Data services; Institutions.

• The Joint Data Citation Principles should be actively endorsed and implemented in the scholarly literature for attribution and in research assessment frameworks for recognition and career advancement.
  Stakeholders: Publishers, Institutions, Funders.

• A broader range of metrics should be developed to recognise contributions beyond publications and citation. These should recognise and reward Open and FAIR data practices.
  Stakeholders: Funders; Publishers; Institutions.

The components of the FAIR ecosystem need to be maintained at a professional service level with sustainable funding.

• Criteria for service acceptance and operation quality, including certification standards, need to be derived and applied with the aim to foster a systematic and systemic approach.
  Stakeholders: Research communities; Global coordination fora; Funders.

• Regular evaluation of the relevance and quality of all services needed to support FAIR should be performed.
  Stakeholders: Research communities; Data stewards.

• Sustainable funding and business models need to be developed for the provision of each of these components.
  Stakeholders: Data services; Funders.

While there is much existing infrastructure to build on, the further development and extension of FAIR components is required. These tools and services should fulfil the needs of data producers and users, and be easy to adopt.

• The development of FAIR compliant components needs to involve scientific communities, technical experts and other stakeholders. They should be provided with a forum for the exchange of views.
  Stakeholders: Data services; Research communities; Global coordination fora.

• Engagement of the necessary stakeholders and experts needs to be facilitated with appropriate funding, support, incentives and training.
  Stakeholders: Funders; Institutions.

• FAIR components will need regular iteration cycles and evaluation processes to ensure that they are fit for purpose and meet community needs.
  Stakeholders: Data services; Research communities.

Research facilities, in particular those of the ESFRI and national Roadmaps, should be incentivised to provide FAIR data by including it as a criteria in the initial and continuous evaluation process. Strategic research investments should consider service sustainability.

• The metrics and criteria by which research infrastructure are assessed should reference and build on the FAIR principles, incorporating language and concepts as appropriate, in order to align policy with implementation and to avoid confusion and dispersion of effort.
  Stakeholders: Funders, Data services.

• Investment in new tools, services and components of the FAIR data ecosystem must be made strategically in order to leverage existing investments and ensure services are sustainable.
  Stakeholders: Funders; Institutions.

• The RCN thinks that the FAIR Data Action Plan provides an impressive list of recommendations and it is obvious that a lot of work has been invested. An idea for developing the report further is to make a clearer distinction between the "low hanging fruits" and the more immature recommendations. It would be an improvement if the plan was divided into short-term and more long-term recommendations, as several of the long-term recommendations depend on the realisation of the short-term ones.
• The RCN thinks that making data interoperable and machine readable is a very demanding and costly exercise. It is important that different academic communities are involved and that research funders do not enforce "computer languages" that are not perceived as useful by the researchers. Interoperability standards take time to develop and this work must not be rushed.
• The RCN notes that several of the recommendations mention research funders as the responsible stakeholders. It is important to bear in mind that several of the decisions related to e.g. interoperability and what data should be preserved, must be taken by the research communities and not by research funders and authorities.
• The RCN thinks that the FAIR principles lack a time horizon; who will decide how long data can and should be taken care of and who will pay for this? In principle, we talk about keeping data indefinitely and this can be very expensive.

Research data should be made available by means of Trusted Digital Repositories, and where possible in those with a mission and expertise to support a specific discipline or interdisciplinary research community.

• Policy should require data deposit in certified repositories and specify support mechanisms (e.g. incentives, funding of deposit fees, and training) to enable compliance.
  Stakeholders: Policymakers; Funders; Publishers.

• Mechanisms need to be established to support research communities to determine the optimal data repositories and services for a given discipline or data type.
  Stakeholders: Data services; Institutions; Data stewards.

• Concrete steps need to be taken to ensure the development of domain repositories and data services for interdisciplinary research communities so the needs of all researchers are covered.
  Stakeholders: Data services; Funders; Institutions.

• Advocacy via scholarly societies, scientific unions and domain conferences is required so researchers in each field are aware of the relevant disciplinary repositories.
  Stakeholders: Data services.

A concerted effort should be made to coordinate, systematise and accelerate the pedagogy and availability of training for data skills, data science and data stewardship.

• Curriculum frameworks should be made available and be easily adaptable and reusable.
  Stakeholders: Institutions.

• Sharing and reuse of Open Educational Resources and reusable materials for data science and data stewardships programmes should be encouraged and facilitated.
  Stakeholders: Institutions; Global coordination fora; Data services.

• Train-the-Trainer programmes for data science and data stewardship roles should be developed, implemented and supported, so they can scale.
  Stakeholders: Institutions; Data services; Data stewards; Funders.

• A programme of certification and endorsement should be developed for organisations and programmes delivering Train-the-Trainer and/or data science and data stewardship training. As a first step, a lightweight peer-reviewed self-assessment would be a means of accelerating the development and implementation of quality training.
  Stakeholders: Institutions; Global coordination fora; Standards bodies.

ELIXIR-UK has the following comments about the documents:

• No notion of return on investment - this needs to be delegated to communities to prioritise. FAIR is a spectrum and not all Data is equal
• Under-emphasis of the lightweight glue need to connect institute-national and international repositories
  • “Dark indexing” potential gains
  • “On boarding” researchers on the first mile of FAIR
• FAIR is a spectrum. It is not binary.. A FAIR maturity model is needed, not binary criteria
• It is important to separate infrastructure funding from research investment

Interoperability frameworks should be articulated in common ways and adopt global standards where relevant to enable interdisciplinary research. Common standards, intelligent crosswalks, brokering mechanisms and machine-learning should all be explored to break down silos.

• Efforts should be made to identify information and practices that apply across research communities and articulate these in common standards that provide a baseline for FAIR.
  Stakeholders: Standards bodies; Research communities.

• Case studies for cross-disciplinary data sharing and reuse should be collected. Based on these case studies, mechanisms that facilitate the development of frameworks for interoperability and reuse should be developed.
  Stakeholders: Global coordination fora; Data stewards.

• The components of the FAIR ecosystem should adhere to common standards to support disciplinary frameworks and to promote interoperability and reuse of data across disciplines
  Stakeholders: Data services; Research communities; Global coordination fora.

The Open Data mandate for publicly funded research should be made explicit in all policy. It is important that the maxim ‘as Open as possible, as closed as necessary’ be applied proportionately with genuine best efforts to share.

• Steps should be taken to ensure coherence across data policy and issue collective statements of intent wherever possible.
  Stakeholders: Research funders; Policymakers.

• Policies should require an explicit and justified statement when data cannot be Open and a proportionate and discriminating course of action to ensure maximum appropriate data accessibility, rather than allowing a wholesale opt out from the mandate for Open Data.
  Stakeholders: Funders; Policymakers.

• Sustained work is needed to clarify in more detail the appropriate boundaries of Open, the proportionate exceptions to data sharing and robust processes for data that needs to be protected.
  Stakeholders: Research communities; Data services; Global coordination fora.

• Concrete and accessible guidance should be provided to researchers in relation to sharing sensitive and commercial data as openly as possible.
  Stakeholders: Data stewards; Data services; Institutions; Publishers.

Funders of research data services should consolidate and build on existing investments in infrastructure and tools, where they demonstrate impact and community adoption. Funding should be tied to certification schemes as they develop for each of the FAIR ecosystem components.

• Funding decisions for new and existing services should be tied to evidence, metrics and certification schemes validating service delivery.
  Stakeholders: Funders; Institutions; Research communities.

• Effective guidance and procedures need to be established and implemented for retiring services that are no longer required (ref. Principles for Open Scholarly infrastructures).
  Stakeholders: Data services; Data stewards.

Efforts should be made to align and consolidate FAIR data policy, reducing divergence, inconsistencies and contradictions.

• Concerted work is needed to update policies to incorporate and align with the FAIR principles to ensure that policy properly supports the FAIR data Action Plan.
  Stakeholders: Policymakers

• A funders’ forum at a European and global level should do concrete work to align policies, DMP requirements and principles governing recognition and rewards.
  Stakeholders: Funders.

• Information on practice in relation to exceptions should be captured and fed into a body of knowledge which can inform future policy guidance and practice.
  Stakeholders: Policymakers; Global coordination fora.

• Policies should be versioned, indexed and semantically annotated in a policy registry.
  Stakeholders: Policymakers; Data services; Global coordination fora.

Automated processing should be supported and facilitated by FAIR components. This means that machines should be able to interact with each other through the system, as well as with other components of the system, at multiple levels and across disciplines.

• Automated workflows between the various components of the FAIR data ecosystem should be developed by means of coordinated activities and testbeds.
  Stakeholders: Data services; Standards bodies.

• Metadata standards should be adopted and used consistently in order to enable machines to discover, assess and utilise data at scale.
  Stakeholders: Data services; Research communities.

• Structured discoverability and profile matching mechanisms need to be developed and tested to broker requests and mediate metadata, rights, usage licences and costs.
  Stakeholders: Data services.

DMPs hold valuable information on the data and related outputs, which should be structured in a way to enable reuse. Investment should be made in DMP tools that adopt common standards to enable information exchange across the FAIR data ecosystem.

• DMPs should be explicitly referenced in systems containing information about research projects and their outputs (CRIS). Relevant standards and metadata profiles, should consider adaptations to include DMPs as a specific project output entity (rather than inclusion in the general category of research products). The same should apply to FAIR Data Objects.
  Stakeholders: Standards bodies; Global coordination fora; Data services.

• A DMP standard should be developed that is extensible (e.g. like Dublin Core) by discipline (e.g. Darwin Core) or by the characteristics of the data (e.g. scale, sensitivity), or the data type (specific characteristics and requirements of the encoding).
  Stakeholders: Standards bodies; Global coordination fora; Data services.

• Work is necessary to make DMPs machine readable and actionable. This includes the development of concepts and tools to support the creation of useful and usable data management plans tied to the actual research workflows.
  Stakeholders: Funders; Data services; Data stewards.

• DMPs themselves should conform to FAIR principles and be Open where possible.
  Stakeholders: Data services; Research communities; Policymakers.

• Information gathered from the process of implementing and evaluating DMPs relating to conformity, challenges and good practices should be used to improve practice.
  Stakeholders: Data services; Funders; Research communities; Global coordination fora

FAIR is not limited to its four constituent elements: it must also comprise appropriate openness, the assessability of data, long-term stewardship, and other relevant features. To make FAIR data a reality, it is necessary to incorporate these concepts into the definition of FAIR.

• The FAIR principles should be consulted on and clarified to ensure they are understood to include appropriate openness, timeliness of sharing, assessability, data appraisal, long-term stewardship and legal interoperability.
  Stakeholders: Global coordination fora; Research communities; Data services.

• The term FAIR data is widely-used and effective so should not be extended with additional letters.
  Stakeholders: Research communities; Data services.

• The relationship between FAIR and Open should be clarified and well-articulated. FAIR depends on appropriate Openness which can be expressed as ‘as Open as possible, as closed as necessary’.
  Stakeholders: Research communities

Research communities and data stewards should better define which FAIR data objects are likely to have long-term value and implement processes to assist the appraisal and selection of outputs that will be retained in the long term and made FAIR.

• Research communities should be encouraged and funded to make concerted efforts to improve guidance and processes on what to keep and make FAIR and what not to keep.
  Stakeholders: Policymakers; Funders; Data services; Global coordination fora.

• The appraisal and selection of research outputs that are likely to have future research value and significance should reference current and past activities and emergent priorities.
  Stakeholders: Research communities; Data stewards; Data services.

• When data are to be deleted as part of selection and prioritisation efforts, metadata about the data and about the deletion decision should be kept.
  Stakeholders: Research communities; Data stewards; Data services.

Semantic technologies are essential for interoperability and need to be developed, expanded and applied both within and across disciplines.

• Programs need to be funded to make semantic interoperability more practical, including the further development of metadata standards, vocabularies and ontologies, along with appropriate validation infrastructure.
  Stakeholders: Funders; Standards bodies; Global coordination fora.

• To achieve interoperability between repositories and registries, common protocols should be developed that are independent of the data organisation and structure of various services.
  Stakeholders: Data services; Standards bodies.

Skills transfer schemes should be supported to equip researchers from various domains with information management skills or vice versa. Such individuals will play an important role as intermediaries to broker relations between research communities and infrastructure services.

• Investigate and learn from existing programmes that have demonstrated success in sharing skills across research scientist and information professional roles
  Stakeholders: Funders; Institutions; Research communities.

• Support programmes of activity that enable skills transfer across communities.
  Stakeholders: Funders; Institutions; Data services.

Three suggestions regarding the action plan based on my previous work on technical standards and specifications.

1. Adopt IEFT RFC 2119 (https://www.ietf.org/rfc/rfc2119.txt) as source of definitions for the terms "should", "need", etc. I think this will clarify recommendations and support developing metrics. (It may also make some of those conversations more drawn out.)

2. Try to describe the properties and attributes of FAIR objects in the present tense. So write "Data are assigned a unique and persistent identifier ...." rather than "Data should be assigned ...." Stronger sentence and fewer words.

3. There are a large number of recommendations. Can a phased action plan be suggested in each area? Resources are limited; what recommendations are to be implemented first?

Funders should incentivise data reuse by promoting this in funding calls and requiring research communities to seek and build on existing data wherever possible.

• Researchers should be required to demonstrate in DMPs that existing FAIR data resources have been consulted and used where possible before creating new data.
  Stakeholders: Policymakers; Research communities.

• Appropriate levels of funding should be dedicated to reusing existing FAIR data by schemes that incentivise this.
  Stakeholders: Funders; Institutions.

Implementing FAIR requires a model for FAIR Data Objects which by definition have a PID linked to different types of essential metadata, including provenance and licencing. The use of community standards and sharing of code is also fundamental for interoperability and reuse.

• Universal use of appropriate PIDs needs to be facilitated and implemented.
  Stakeholders: Data services; Institutions; Publishers; Funders.

• Educational programmes and tools are needed to raise awareness, understanding and use of relevant standards and routine capture of metadata during the research process.
  Stakeholders: Data stewards; Institutions; Data services.

• Systems must be put in place for automatic checks on the existence and accessibility of PIDs, metadata, a licence or waiver, and code, and to test the validity of the links between them.
  Stakeholders: Data services; Standards bodies.

A set of metrics for FAIR Data Objects should be developed and implemented, starting from the basic common core of descriptive metadata, PIDs and access. The design of these metrics needs to be mindful of unintended consequences, and they should be regularly reviewed and updated.

• A core set of metrics for FAIR Data Objects should be defined to apply globally across research domains. More specific metrics should be defined at the community level to reflect the needs and practices of different domains and what it means to be FAIR for that type of research.
  Stakeholders: Global coordination fora; Research communities.

• The European Commission should support a project to coordinate the activities of various groups defining FAIR metrics and ensure these are created in a standardised way to enable future monitoring.
  Stakeholders: Funders.

• The process of developing, approving and implementing FAIR metrics should follow a consultative methodology, including scenario planning, to minimise to the greatest extent possible any unintended consequences and counter-productive gaming that may result. Metrics need to be regularly reviewed and updated to ensure they remain fit-for-purpose.
  Stakeholders: Global coordination fora; Publishers; Data services.

There are large amounts of legacy data that is not FAIR but would have considerable value if it were. Mechanisms should be explored to include some legacy data in the FAIR ecosystem where required.

• Research communities and data owners should explore legacy data to identify indispensable collections with significant reuse potential that warrant effort to make them FAIR.
  Stakeholders: Research communities; Institutions; Data services.

• Funding should be provided to adapt legacy datasets that have been identified as particularly crucial in a given discipline.
  Stakeholders: Funders; Institutions; Research communities.

Certification schemes are needed to assess all components of the FAIR data ecosystem. Like CoreTrustSeal, these should address aspects of service management and sustainability, rather than being based solely on FAIR principles which are primarily articulated for data and objects.

• Building on the model of CoreTrustSeal, new certification schemes should be developed and refined by the community to assess and certify other core components needed in the FAIR data ecosystem, such as identifier services, standards and vocabularies.
  Stakeholders: Global coordination fora; Data services; Standards bodies.

• Formal registries of certified components are needed: these must be maintained primarily by the certifying organisation, but should also be communicated in community discovery registries such as Re3data and FAIRsharing.
  Stakeholders: Data services.

• Steps need to be taken to ensure that the organisations overseeing certification schemes are independent, trusted, sustainable and scalable.
  Stakeholders: Funders; Research communities.

FAIR data should be recognised as a core research output and included in the assessment of research contributions and career progression. The provision of infrastructure and services that enable FAIR data must also be recognised and rewarded accordingly.

• Policy guidelines should recognise the diversity of research contributions (including publications, datasets, online resources, teaching materials) at the level of biography and in templates for researchers’ applications and activity reports.
  Stakeholders: Funders; Publishers; Institutions.

• Credit should be given for all roles supporting FAIR data, including data analysis, annotation, management, curation and participation in the definition of disciplinary interoperability frameworks.
  Stakeholders: Funders; Publishers; Institutions.

• Evidence of past practice in support of FAIR data should be included in assessments of research contribution. Such evidence should be required in grant proposals (for both research and infrastructure investments), for career advancement, for publication and conference contributions, and other evaluation schemes.
  Stakeholders: Funders; Institutions; Publishers; Research communities.

• The contributions of organisations and collaborations to the development of certified and trusted infrastructures that support FAIR data should be recognised, rewarded and appropriately incentivised.
  Stakeholders: Funders; Institutions.

FAIR should be applied broadly to all objects (including metadata, identifiers, software and DMPs) that are essential to the practice of research, and should inform metrics relating directly to these objects.

• Policies must assert that the FAIR principles should be applied to research data, to metadata, to code, to DMPs and to other relevant digital objects.
  Stakeholders: Policymakers.

• The FAIR data principles and this Action Plan must be tailored for specific contexts and the precise application nuanced, while respecting the objective of maximising data accessibility and reuse.
  Stakeholders: Research communities; Data services; Policymakers.

• Guidelines for the implementation of FAIR in relation to research data, to metadata, to code, DMPs and other relevant digital objects should be developed and followed.
  Stakeholders: Data services; Data stewards; Research communities; Funders.

• Examples and case studies of implementation should be collated so that other organisations can learn from good practice.
  Stakeholders: Global coordination fora; Research communities.

Any research project should include data management as a core element necessary for the delivery of its scientific objectives, addressing this in a Data Management Plan. The DMP should be regularly updated to provide a hub of information on the FAIR data objects.

• Research communities should be required and supported to consider data management and sharing as part of all research activities.
  Stakeholders: Funders; Institutions; Data stewards; Publishers; Research communities.

• Data Management Plans should be living documents that are implemented throughout the project. A lightweight data management and curation statement should be assessed at project proposal stage, including information on costs and the track record in FAIR. A sufficiently detailed DMP should be developed at project inception. Project end reports should include reporting against the DMP.
  Stakeholders: Funders; Institutions; Data stewards; Research communities.

• Research institutions and research projects need to take data management seriously and provide sufficient resources to implement the actions required in DMPs.
  Stakeholders: Institutions; Data stewards; Research communities.

• Research communities should be inspired and empowered to provide input to the disciplinary aspects of DMPs and thereby to agree model approaches, exemplars and rubrics that help to embed FAIR data practices in different settings.
  Stakeholders: Data services; data stewards; Research communities.