| Table of Contents | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Level | Nickname | Marketplace metaphor | Features | Picture |
3 | "Pretty FAIR" | "Specialized Local Markets | Transition to good and best practice |
Table level 3 summary
Level 3 - Capabilities
Findability is largely achieved in the context the organization, access protocols and controls are in place. Data can be re-used at least at departmental levels. Machine interpretation is possible at the local (e.g. department) level. Processes for FAIR are formalized, including training, documentation, integration into workflow. The organization can support those processes financially and operationally. Data that increasingly comply to FAIR data principles can be generated from the onset and the organization can begin to reduce efforts required by retrospective “FAIRification”
Level 3 - Business value
Department-level data reuse and insight generation; increased effectiveness and efficiency of data resources available in the organization. Reuse of data reduces procurement costs and the need for data generation.
Level 3 - Questions to ask
Are the identifiers for data and metadata persistent? Does the metadata contain “structured” elements?
Does the data and metadata identifier resolution support authentication and authorisation for access?
Is there a policy for persistence of metadata?
Does the metadata provide a pointer to disclose the owner or license to the data which is readable by machine?
How is FAIR data currently managed within different organizational departments, and across divisions, or functions?
How does leadership ensure that financing for FAIR implementation actions is included in the budget of data acquisition and related projects budgets?
How are FAIR strategies and guidelines embodied in organizational documents, and how are they communicated?
What key and emerging roles are identified for FAIR data management within the organization? Which roles are emerging?
How are metadata implementation and other FAIR practices standardized and integrated into existing processes?
What initiatives are underway to curate FAIR implementation tools and facilitate cross-functional FAIR initiatives?
How is FAIR data knowledge integrated into the organization's training curriculum and shared among business units?
What IT infrastructure mechanisms ensures that data is made FAIR prospectively rather than retrospectively in the organization?
What budget and capacity are allocated for delivering FAIR data at scale, and how are vendors selected for tooling and infrastructure implementation?
Is a knowledge representation language being used that has ontological machine-resolvable formats?
Level 3 - FAIR data
One can find “FAIR” data in local environments such as a organization’s departments, divisions or functions.
Data embeds local master and reference data. Metadata is identifiable and resolvable (at a local level), but data is still being determined (also at a local level).
Data is “conformed” using a domain-level data model. It is integrated, processed and audited to support specific consumption patterns.
While we achieve “Findability” and there is evidence of re-use at the local level, the "Interoperability" and "Reusability" principles are still a challenge in the broader scope of the organization.
There are controls in place on Access at the data level.
Creation of analytics-ready 'data-marts' enabling self-service analytics.
Machine interpretation is possible at the department level.
Initial FAIR data sets are in production and proving value. There is a company-wide supported plan to make all (relevant) data FAIR. More data sets are bornFAIR while the company's budget and competencies grow. There is evidence of local reusability and, to some extent, interoperability.
(Back to the FAIR Matrix).
Level 3 - FAIR leadership
Leadership (C level / or relevant department level) expectation is that any data project (or product) budgets include a component for FAIR implementation.
There are FAIR metrics for the use of organizational governance.
Leadership creates objectives, accountability and necessary training for FAIR data implementation.
Leadership has a vision and plans to make the entire company aware of FAIR data, FAIR practices and their value.
There is a clear leadership understanding of why making data FAIR adds value.
There is company-wide support for FAIR implementation (Human Resources and Financial Resources are made available).
Budget/resources are available to grow FAIR examples and the "Planned FAIR" attitude.
For example, leadership starts engaging outside the organization's boundaries to enable convergence standards.
(Back to the FAIR Matrix).
Level 3 - FAIR strategy
A refined version of the vision, strategy and plans for implementation exist.
There is (at least one) a high-level "FAIR strategy" embodied in a set of guidance documents (e.g. strategy, guidelines, policy, "how to" guides).
There is a clear understanding of why making data FAIR adds value; this may extend beyond a given corporate function or domain. This value-adding process begins to be tracked or monitored. Strategies begin to encompass enterprise-level dimensions.
There is a company-wide supported vision and plan supported by the C-suite and with enough budget and capacity building.
Budget/resources available to grow FAIR examples but also a "planned for FAIR" attitude.
"FAIR implementation" plans to enact the strategy are in place. These may still be at a functional or department level (not enterprise-wide).
The impact, organization-wide system architecture and decisions of which systems (including external sources) will be part of the FAIR implementation are evaluated and are part of the strategic documentation.
Some legal requirements (for example, research security and Sunshine ACT in the US) are (or begin to be) interpreted in terms of FAIR implementation strategy.
The data strategy accommodates federated domain-specific FAIR data, and there is an operating model, set of procedures, and defined models to harmonize, align, and connect functional domain data to expand and scale FAIR data beyond single or simple domain-specific use cases.
(Back to the FAIR Matrix).
Level 3 - FAIR roles
Key Roles are data standard expert, data curator, semantic web expert, knowledge engineer, data steward, and data strategist.
Emerging roles are Data architects, Citizen Data Scientists, Data product owners, and Domain owners.
The organization has started laying out a framework (e.g. a matrix or a triple store?) of FAIR roles and how they interact—some experienced employees in FAIR roles.
FAIR data has started to be part of some people's training in some departments.
Formal roles and training are being laid out and approved. This cultural change role is pivotal in promoting and embedding a data-driven culture aligned with the organization's data management goals and practices.
Teams with complementary roles emerge.
For example, IT and business functions collaborate in creating Knowledge Graphs, scientific domain experts and data architects for data mesh
(Back to the FAIR Matrix).
Level 3 - FAIR processes
The organization initiates formal training for FAIR roles while a majority have embraced FAIR principles for major resources, backed by well-trained personnel capable of evaluating and implementing FAIR standards. At least one business unit actively employs FAIR processes in production, supported by documentation and training, with other FAIR initiatives underway and coordinated efforts recognized.
The FAIR data training is part of employee onboarding and continuous education development. The organization tailors specific FAIR processes to suit its needs and incorporates guidelines for providing FAIR data into standard IT procedures. Efforts to make metadata implementation accessible in a standardized manner without changing existing processes. The organization begins to integrate FAIR practices into workflows and configure systems to facilitate FAIR practices, demonstrating a commitment to normalising FAIR standards within organizational operations.
Additional processes include retrospective data FAIRification, the definition of requirements for procurement of FAIR data, curation of FAIR implementation tools, and cross-functional (e.g., business- IT- R&D) FAIR initiatives.
A "FAIR implementation plan" to enact the strategy is in place.
(Back to the FAIR Matrix).
Level 3 - FAIR knowledge
FAIR data is part of the training curriculum, with some materials and experts available company-wide.
Business Units begin the formalisation of FAIR knowledge sharing.
Domain knowledge, providing a shared understanding and interpretation of data context within each domain, starts to be established.
The Community of Practice drives domain knowledge through the common understanding and interpretation of data and its context.
A community of knowledgeable FAIR practitioners have met and have started sharing experiences in making data FAIR, at least at the department level.
(Back to the FAIR Matrix).
Level 3 - FAIR tools and infrastructures
Established processes for producing and managing FAIR data, with tools, either internally developed or available by vendors and other organizations, could be limited in terms of entities and the level of sophistication.
Tools that help create FAIR Data and manage metadata collection templates and ontologies are available.
Ways of working ensure data is made FAIR from the onset, working prospectively rather than retrospectively (i.e., compared to doing FAIRification work after data creation).
The process can start with external resources (people or tools) but needs to shift to internal resources (employees), rolling the tooling and infrastructure out, including training and support.
An "anchor/reference model", or "enterprise-wide data model", is being built (cf. refer to the Strategy dimension). This model can also be a semantic model external to the organization but selected based on business needs and suitability.
There is a conceptual definition of Interaction mechanisms with FAIR data, although they are not yet machine-actionable, indicating a step towards making data more accessible and interoperable.
The approval of Budget and capacity for delivery at scale, along with the launch of RFPs or similar and vendors are selected to put tooling and infrastructure in place for company-wide use.
It becomes easier to make data FAIR company-wide, using the well-supported off-the-shelf tools than in the previous maturity level.
(Back to the FAIR Matrix).
Findability:
Tool(s) or infrastructure component which contributes, enhances or enriches Findability:
Accessibility:
Tool(s) or infrastructure component(s) which contributes, enhances or enriches Accessibility.
Interoperability:
Tool(s) or infrastructure component(s) which contributes, enhances or enriches Interoperability.
Reusability:
Tool(s) or infrastructure component(s) which contributes, enhances or enriches Reusability.