Skip to main content
Join this collection

FAIR principles and Semantics on the Web: where is the meeting point?

FAIR ontologies

Published on: 06/04/2021 Last update: 09/04/2021 News

The FAIR (Findable, Accessible, Interoperable, Reusable) principles have gained attention and importance since their proposal in 2016 being adopted by numerous private and public organisations worldwide, for example the European Open Science Cloud (EOSC) or the Research Data alliance (RDA).

Ontologies play a relevant role for supporting interoperability and are nowadays widely used for different purposes and in a diversity of contexts. When produced as a research output, ontologies should be treated as other research artefacts, such as data, software, methods, etc.; following the same principles used to make them findable, accessible, interoperable and reusable (FAIR) to others. However, no much attention has been paid to developed guides, indicators and recommendations to make  FAIR ontologies in comparison with the efforts developed for FAIR data. 

In our paper "Coming to Terms with FAIR Ontologies" presented during the "22nd International Conference on Knowledge Engineering and Knowledge Management" an analysis of the FAIR principles and their relation to semantic web best practices and guidelines was described. More precisely, the FAIR principles were compared or aligned with:

  • FAIRsFAIR deliverable "D2.2 FAIR Semantics: First recommendations". The FAIRsFAIR project, started in 2019, is a European effort aiming to provide practical solutions for the use of the FAIR data principles throughout the research data lifecycle. The D2.2 provides a list of 17 preliminary recommendations related to the application of FAIR principles to improve the global FAIRness of semantic artefacts. Each recommendation and best practice is related to one or more FAIR principles and links to existing recommendations and related stakeholders (e.g: practitioners, repositories or the SemanticWeb community).
  • Best Practices for Implementing FAIR Vocabularies and Ontologies on the Web (open pre-print) that provides specific practical guidelines to help users with design of accessible ontologies (design of ontology name and prefix, choose between hash or slash URIs, whether to select opaque URIs, versioning strategy and use of permanent URIs), generation of reusable ontology documentation (ontology metadata, HTML documentation and diagrams) and publication of ontologies (content negotiation for multiple formats and making the ontology findable on the Web).
  • 5-star vocabularies by Bernard Vatant
    • ⭐Publish your vocabulary on the Web at a stable URI with a open license.
    • ⭐⭐Provide human-readable documentation and basic metadata such as creator, publisher, date of creation, last modification, version number.
    • ⭐⭐⭐ Provide labels and descriptions, if possible in several languages, to make your vocabulary usable in multiple linguistic scopes.
    • ⭐⭐⭐⭐ Make your vocabulary available via its namespace URI, both as a formal file and human-readable documentation, using content negotiation.
    • ⭐⭐⭐⭐⭐Link to other vocabularies by re-using elements rather than re-inventing.
  • 5-star vocabularies by Janowicz, Krzysztof, et al.:
    • ⭐There is dereferenceable human-readable information about the used vocabulary.
    • ⭐⭐The information is available as machine-readable explicit axiomatization of the vocabulary.
    • ⭐⭐⭐ The vocabulary is linked to other vocabularies.
    • ⭐⭐⭐⭐ Metadata about the vocabulary is available (in a dereferenceable and machine-readable form).
    • ⭐⭐⭐⭐⭐ The vocabulary is linked to by other vocabularies

During the analysis it was observed that some of the FAIR principles could be aligned with ontology publication in a more or less direct way but others need to be discussed within the various communities developing and publishing ontologies and semantics on the Web. Finally, the alignments and open discussion could be summarised as:

  • To be Findable
    • F1: For using globally unique and persistent identifiers, the ontology engineering practices allow to define URIs for ontologies. However, it still necessary to agree if the Semantic Web community should establish mechanisms and authorities to coin persistent identifiers (PIDs) for semantic artefacts and if these PIDs should refer only to semantic artefacts as a whole or also to each of their components (e.g., specific concepts or properties, specific SKOS concepts).
    • F2: For describing data with rich metadata, the ontology engineering community should agree on a minimum set of metadata that semantic artefacts should have and provide more technical guidelines for its declaration.
    • F3: The principle of metadata clearly and explicitly includes the identifier of the data it describes is not applicable to ontologies because metadata is auto contained in the ontology code. However, the ontology engineering community should decide in which cases metadata should be provided as a separate object.
    • F4: For (meta)data are registered or indexed in a searchable resource, the ontology engineering community should define a federation model for ontologies that may be combined with standard definitions of SAODs as well as SAODs discovery approaches.
  • To be Accessible
    • A1, A1.1, A1.2 principles are aligned with the Ontology Engineering practices since ontologies, published following the best practices, use existing technologies and protocols (HTTP or HTTPS) suggested in these principles.
    • A2: For metadata accessible, even when the data are no longer available, having preservation policies (for example how long a semantic artefact will be preserved, what version will be retained, what serialization formats will be stored, etc.) for publishing resources may be a good practice to adopt by the Ontology Engineering Communities.
  • To be Interoperable
    • I1: For (meta)data use a formal, accessible, shared, and broadly applicable language for knowledge representation, semantic artefacts use knowledge representation languages proposed by W3C such as RDF(S), OWL, or SKOS.
    • I2: For (meta)data use vocabularies that follow FAIR principles, in the Ontology Engineering field must be needed to reuse FAIR semantic artefacts. However, it is necessary to define FAIR indicators for semantic artefacts.
    • I3: For (meta)data include qualified references to other (meta)data, ontologies already include qualified references to other ontologies by means of an URI reference, explicit OWL constructs, or different relations for SKOS concepts.
  • To be Reusable
    • R1: For meta(data) are richly described with a plurality of accurate and relevant attributes, the minimum set of metadata recommended to be defined in F2 should be enough for this principle. Despite these metadata are used to generate human readable documentation, it should be necessary to define best practices to document and communicate ontologies.
    • R1.1For (meta)data released with a clear and accessible data usage license, the semantic artefacts should provide enough information about the permissions and conditions included in their licenses. Also, such license descriptions should be linked from the resources and provided in RDF, for example, providing the RDF description of the license using the Creative Commons vocabulary or ODRL.
    • R1.2For (meta)data associated with detailed provenance, the W3C already provides the PROV-O ontology and standard specification that should be adopted for semantic artefacts.
    • R1.3For (meta)data meeting domain-relevant community standards, it refers to the use of RDF(S) and OWL to describe ontologies as already proposed in I1. However, standards may involve another aspects which will depend on the communities. Therefore, there is a need here for each community to agree on common standards and best practices to follow in regard to ontology engineering.

For more details we refer readers to the conference paper (open pre-print) or the video.  

 

Paola Espinoza Arias and María Poveda Villalón