Proposal for prioritisation of the indicators

09 Aug 2019
Groups audience: 

Dear members of the FAIR data maturity model Working Group,
As a next step towards a common set of core assessment criteria, we started
to explore the prioritisation of the FAIRness indicators.
The idea is that the indicators that we have developed over the last months,
with the help of many members of the Working Group, do not all have the same
priority. Some may be more important than others in evaluating FAIRness.
Some may be considered essential, some should be met if possible and others
may be nice-to-have.
As a result, we have created an initial ranking of the indicators according
to three degrees of priority;
* Mandatory: indicator MUST be satisfied for FAIRness
* Recommended: indicator SHOULD be satisfied, if at all possible, to
increase FAIRness
* Optional: Indicator MAY be satisfied, but not necessarily so
(Keywords MUST, SHOULD and MAY as in RFC2119
We have created a Google spreadsheet

L4w8RRAw/edit#gid=0> with an initial proposal for assigning these
priorities to the current set of indicators. We have also created four
GitHub issues, one for each of the four FAIR areas, to discuss the
priorities of the indicators under that area.
We invite all members of the Working Group to comment on these proposals on
GitHub and to suggest changes in the proposed priorities.
In the next two weeks, we will gather all comments and then re-issue the
Google spreadsheet based on your contributions.
The set of indicators and the priorities will be on the agenda for the next
WG online meeting on 12 September for further discussion.
Also, don't hesitate to comment on the approach and any other aspect of the
work of the editorial team!
Kind regards,
Makx Dekkers and Christophe Bahim
The editorial team

  • Ge Peng's picture

    Author: Ge Peng

    Date: 11 Aug, 2019

    Dear Makx and Christophe,
    Excellent work by the editorial team. I have complied all the mandatory
    indicators and reviewed if they are sufficient for ensuring FAIR data and
    if there may be any issue with implementing them.
    Attached file in pdf contains a list of the mandatory indicators and my
    comments for consideration. Please feel free to correct me if I have missed
    or misrepresented anything.
    Here is a recap:
    Is it necessary to have two individual indicators from the
    machine-assessment perspective?
    If not, perhaps F1-01D and F1-02D can be combined as one: "Data is
    identified by a universally
    unique and persistent identifier"
    There is no mandatory indicator for interoperability. I do not see any
    mandatory indicator from
    this list for ensuring data to be interoperable, perhaps R1.3-01D but see
    my comment below.
    R1.3-01D is ambiguous. There are well-established community metadata
    standards. Are there
    any well-established community data standards? What community standards
    does it refer to for data to be
    compliant with: file naming conventions, file formats, file structures,
    standard variable names, …?
    Implementation of this indicator may be extremely arbitrary.
    Hope it helps.
    Ge Peng

submit a comment