> For the complete documentation index, see [llms.txt](https://fairsharing.gitbook.io/fairsharing/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://fairsharing.gitbook.io/fairsharing/fair-assistance/fair-support-and-definitions/fair-metrics-co-creation.md). # FAIR Metrics co-creation {% hint style="warning" %} This page is in active development. {% endhint %} This page contains information on the metrics that our team has co-authored and whose homepages we host here. We encourage the re-use of these metrics whenever they are applicable to other communities. For a full list of co-authored metrics, take a look at the [metrics in FAIRassist that have the Oxford e-Research Centre as a related organisation](https://fairassist.org/registry?search=\(principle=The+FAIR+Principles%26organisations=oxford_e-research_centre,_department_of_engineering_science,_university_of_oxford,_oxford,_uk%26recordType=metric_ids\)). ## FAIR Metric – F2 - Metadata – tagging to aid discovery This dimension evaluates the metadata for the presence of at least one keyword or tag of any kind (free-text or controlled). The metric expects the identifier to point to structured metadata (e.g., schema.org, DataCite, or DC) and verifies that the keyword/subject property is not empty. The metric passes if at least one tag is identified and fails if the keyword attribute is missing or null. Keywords and tags are essential for enabling filtered discovery and faceted search. While a title or abstract provides narrative context, tags provide discrete hooks for search engines and repository interfaces to categorize the data. In a generalist institutional repository, where datasets span many disciplines, these tags often serve as the primary mechanism for users to narrow down search results to relevant topics. Measuring this ensures that researchers are providing the basic "labels" required for their data to be categorized. ## FAIR Metric – F2 - Metadata – has publisher information This metric evaluates whether the metadata includes explicit information regarding the organisation responsible for publishing the metadata record. It looks for a structured "publisher" field within the record. In the context of an institutional repository, this is typically the institution itself, or an external repository (like Zenodo) if the record is registering an object hosted elsewhere. The metric will fail if this value is not present. Publisher information is a discovery attribute that allows users and machines to filter resources by their "home" or source of authority, and to group resources based on their point of origin. ## FAIR Metric - F3 – Metadata - uses common formats to reference data identifier(s) F3 requires that metadata include the identifier of the research object it describes. To align with FAIR, if a metadata record is separated from its associated research object (which is common in harvesting and indexing), the connection between them must be explicitly preserved. This metric is concerned with the explicit linkage aspect of FAIR; F3 ensures that a machine-actionable relationship exists, allowing harvesters to unambiguously navigate from a descriptive metadata record back to the research object itself. Because many research objects cannot be arbitrarily extended to include references to their metadata, in many cases the only means to discover the metadata related to a research object will be to search based on the identifier of the research object itself. While Principle F1 focuses on the qualities of the identifier schema being used (globally unique, persistent and resolvable), Principle F3 relates to the linkage aspects of the research object identifier. It does not mandate that the identifier of the research object within the metadata record be immediately resolvable; rather, it requires that the identifier is present, and formally structured so that the relationship can be maintained and indexed by automated services. The discovery of a research object should be possible from its metadata. For this to happen, the metadata must explicitly contain the identifier for the digital resource it describes, and this should be present in the form of a qualified reference. This metric evaluates whether the identifier provided as input resolves to metadata that contains a research object identifier. Further, that research object identifier must be distinguished from the numerous other fields and values that will be present in the metadata via one of the following formats: [schema.org](http://schema.org) JSON-LD format, DataCite XML, DC XML. ## FAIR Metric – I1 – Metadata - Database-level knowledge representation languages (syntactic) #### What is being measured? This metric evaluates whether there is a declared relationship (related to, implements or outputs) to a model/format from the hosting database for the provided identifier in the FAIRsharing registry, and that the associated model/format has a machine-readable structure. #### Why should we measure it? I1 requires that metadata use a formal, accessible, shared and broadly applicable language for knowledge representation. The purpose of this principle is to ensure that metadata is expressed in a standardised format that can be interpreted and exchanged across systems. In this metric, the definition of syntactic and semantic formats follows the [FAIRsharing guidance](/fairsharing/fair-assistance/fair-support-and-definitions/syntactic-vs-semantic-knowledge-representation-languages.md). This metric evaluates whether there is a declared metadata model or format that ultimately resolves to a basic representation format that is considered syntactic by the FAIRsharing registry. Note that all formats that match the guidance on semantic formats also fulfil the requirements for being syntactic formats. Communities may wish to use this metric in conjunction with others that examine the precise format of the resolved metadata. Note that, in order for this metric to function, tests must know the FAIRsharing database record URL/DOI so that FAIRsharing metadata may be retrieved. This can be done in a variety of ways, e.g. using the FAIRsharing API to find possible matches between the identifier and registered databases, hard coding the FAIRsharing record identifier, or expecting it to be provided as a second argument to the test. ## FAIR Metric – I1 – Metadata - Database-level knowledge representation languages (semantic) #### What is being measured? This metric evaluates whether there is a declared metadata model or format that is based in a representation format that is considered grounded / linked data within the FAIRsharing registry (for more information see the [FAIRsharing documentation](/fairsharing/fair-assistance/fair-support-and-definitions/syntactic-vs-semantic-knowledge-representation-languages.md)). #### Why should we measure it? I1 requires that metadata use a formal, accessible, shared and broadly applicable language for knowledge representation. The purpose of this principle is to ensure that metadata is represented in a widely recognised and interoperable format. In this metric, the definition of syntactic and semantic formats follows the [FAIRsharing guidance](https://fairsharing.gitbook.io/fairsharing/about-our-records/fair-assistance/syntactic-vs-semantic-knowledge-representation-languages). This metric evaluates whether there is a declared metadata model or format that ultimately resolves to a basic representation format that is considered grounded / linked data by the FAIRsharing registry. Communities may wish to use this metric in conjunction with others that examine the precise format of the resolved metadata. Note that, in order for this metric to function, tests must know the FAIRsharing database record URL/DOI so that FAIRsharing metadata may be retrieved. This can be done in a variety of ways, e.g. using the FAIRsharing API to find possible matches between the identifier and registered databases, hard coding the FAIRsharing record identifier, or expecting it to be provided as a second argument to the test. ## FAIR Metric – I2 – Metadata – has FAST subject heading #### What is being measured? This metric evaluates whether there is at least one FAST subject vocabulary term used in the metadata that resolves to content in a linked data format (see [FAIRsharing guidance](/fairsharing/fair-assistance/fair-support-and-definitions/syntactic-vs-semantic-knowledge-representation-languages.md)). #### Why should we measure it? I2 requires that metadata use vocabularies that follow FAIR principles. The purpose of this principle is to ensure that descriptive terms are drawn from such vocabularies, enabling consistent interpretation across systems and communities. Under this semantic interpretation, the evaluation checks whether there is at least one FAST subject vocabulary term used in the metadata that resolves to content expressed in a linked data format. The resolved term (for example, ) must return structured content in a recognised linked data representation, with concepts and relationships defined. A resource satisfies this metric if at least one metadata element references a FAST subject vocabulary term that resolves to content in a linked data format (the definition of semantic formats follows [FAIRsharing guidance](/fairsharing/fair-assistance/fair-support-and-definitions/syntactic-vs-semantic-knowledge-representation-languages.md)). ## FAIR Metric – I2 – Metadata – has ISO language **What is being measured?** This metric evaluates whether there is at least one ISO 639-2 / 639:2023 language code term used in the metadata that resolves to content in a linked data format (see [FAIRsharing guidance](/fairsharing/fair-assistance/fair-support-and-definitions/syntactic-vs-semantic-knowledge-representation-languages.md)). **Why should we measure it?** I2 requires that metadata use vocabularies that follow FAIR principles. The purpose of this principle is to ensure that descriptive terms are drawn from such vocabularies, enabling consistent interpretation across systems and communities. Under this semantic interpretation, the evaluation checks whether there is at least one ISO 639-2 / 639:2023 language code term used in the metadata that resolves to content expressed in a linked data format. The resolved term (for example, ) must return structured content in a recognised linked data representation, with concepts and relationships defined. A resource satisfies this metric if at least one metadata element references an ISO 639-2 / 639:2023 language code term that resolves to content in a linked data format (the definition of semantic formats follows [FAIRsharing guidance](/fairsharing/fair-assistance/fair-support-and-definitions/syntactic-vs-semantic-knowledge-representation-languages.md)). ## FAIR Metric – R1.2 – Metadata - creator ORCiD #### What is being measured? This metric evaluates whether there is at least one dc:creator with an ORCID in the metadata retrieved upon resolution of the provided identifier, providing one facet of evaluation for this R1.2. #### Why should we measure it? R1.2 requires that metadata be associated with detailed provenance. The purpose of this principle is to ensure that the origins and responsible individuals associated with a digital object are clearly identified, supporting attribution and transparency. This metric evaluates whether there is at least one dc:creator with an ORCID in the metadata retrieved upon resolution of the provided identifier, providing one facet of evaluation for this R1.2. Under this interpretation, the evaluation checks whether the resolved metadata includes an ORCID identifier associated with a person in the defined dc:creator role.
## FAIR Metric – I3 – Metadata – Qualified References to Related Research Objects ## FAIR Metric – I3 – Metadata – Qualified References to Versioned Records ## FAIR Metric - R1.2 - Metadata - Creator ORCIDs ## FAIR Metric - R1.2 - Metadata - Linked Publications ## FAIR Metric - R1.2 - Metadata - Original Source ## FAIR Metric – R1.3 - Metadata – use of FAST subject vocabulary ## FAIR Metric – R1.3 - Metadata – use of ISO 639 language ## FAIR Metric – R1.3 - Metadata – use of recognised and structured generic metadata format --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://fairsharing.gitbook.io/fairsharing/fair-assistance/fair-support-and-definitions/fair-metrics-co-creation.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.