Rec. 17: Selection and prioritisation of FAIR Data Objects

[sjDCC]

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.

[hollydawnmurray]

F1000 position: Rec 16 suggests that all objects should be made FAIR while this recommendation implies that there will be a selection of outputs retained and made FAIR. Clarity is needed on whether outputs should be FAIR/preserved based on reproducibility and unknown future/cross disciplinary value, or whether only data assessed as potentially valuable should be FAIR/preserved. Erring on the side of the former seems more in line with FAIR’s mission.

[holubp]

BBMRI-ERIC Position: Data deletion needs to consider reproducibility aspects (e.g., informing upstream consumers before deletion).

[ghost]

4TU.Centre for Research Data position: It is not clear to us, what the exact role of the data services in these scenario is. Besides providing the necessary infrastructure and general guidance, it seems impossible as central support to provide help for every single research community.

[katerbow]

DFG position: With reference to the comments made to recommendation 16 DFG supports recommendation 17. In particular, the close integration of the scientific communities in the prioritisation process is welcomed and strongly encouraged.

[ScienceEurope]

Science Europe agrees that research communities should be encouraged and funded to make concerted efforts to improve guidance and processes on what to keep and make FAIR. Community-specific expertise should be taken into account.

[ferag]

I think some efforts are yet needed to promote the FAIR data culture, providing detailed and specific guidelines and examples to make data components FAIR.

I agree in the selection of data to be used in the future and keeping the metadata (to ensure reproducibility if needed).

[RCN2018]

It is stated that "Research communities should be encouraged and funded to make concerted efforts to improve guidance and presses on what to keep and make FAIR and what not to keep. Stakeholders: Policymakers and funders…" Comment: It is important not to be too top-down in such decisions. Researchers know their data best and where the value lies. In a newly published Norwegian strategy on access to research data, it is stated that "Decisions concerning archiving and management of research data must be taken within the research community"

[pkdoorn]

I would hope that the appraisal and selection (and deselection) of what data (and other objects, including software) to keep is part of a DMP.

[bertocco]

INAF (astronomy) position: Astronomy defines prioritization of products at time of observation planning, plus (apart maybe for simulations) needs to keep all the (at least science ready) products forever.

[npch]

SSI position:

We support the general principle that individual communities should create their own guidance around selection and prioritisation of FAIR Data Objects (and other research objects).

[aidanbudd]

ELIXIR-UK position:

Different scale of usage and reuse (local or global community) requires different prioritisation

[gtoneill]

Fully support any proposal that researchers and research communities should be involved and supported in all decisions related to FAIR Data policies. Do not fully support the implication from this recommendation that there should be selections made as to what should be FAIR and will be of future value. The FAIR principles ultimately should set the conditions for all research data and are flexible enough taking the axiom 'as open as possible, as closed as necessary' into account. We furthermore cannot be guaranteed to know what exactly will be of future value. It should be clear in advance how long and to what extent data will be FAIR including curation guarantees from data repositories for data.