Common Core Ontologies
Home/Blog/PLOS paper sets out a FAIR roadmap for competency frameworks
NEWS

PLOS Paper Sets Out FAIR Roadmap for Competency Frameworks

PLOS paper sets out a FAIR roadmap for competency frameworks

PLOS Computational Biology published "Ten simple rules for creating and maintaining FAIR competency frameworks" on November 16, proposing an eight-stage lifecycle for turning professional skills and training standards from static PDFs into machine-actionable digital objects. The paper, also indexed on PMC, targets a problem familiar to anyone who has tried to compare two certification programs or map a course catalog against a hiring requirement: competency frameworks (CFs) are usually written for humans to read once and file away, not for software to query, compare, or reuse.

What the rules actually require

The ten rules move from scoping and plain-language drafting toward concrete technical obligations. Rule 4 calls for publishing a framework's structure (competency areas, proficiency levels, the relationships between them) in "JSON, JSON-LD, RDF or a documented tabular structure" alongside the human-readable version, specifically so that crosswalks between frameworks can be computed rather than eyeballed. Rule 5 pushes persistent identifiers (DOI, PURL, ISBN) down to the level of individual framework elements, not just the framework as a whole, and points to registries like the EMBL-EBI Competency Hub and Zenodo as places to register them. The authors cite the AI Skills for Business framework and the ISCB bioinformatics competency framework as working examples that already publish structured files to let other tools compare course outcomes against the standard automatically.

A familiar shape to anyone who has modeled a domain

None of this is new to a CCO-style audience: defining components, giving each one a stable identity, and making relationships explicit so a machine can traverse them is ordinary mid-level ontology work, just applied to a domain (professional competency) that has mostly stayed in spreadsheets and PDFs. The gap the paper is really describing is an engineering one: a training provider or certifying body can usually write down what a competency framework should say, but turning that into versioned RDF or JSON-LD with resolvable identifiers, a crosswalk engine, and an API other systems can query is a different skill set and a real backlog item. Organizations sitting on that backlog tend to bring in outside help to build the structured-data layer rather than let a framework stay a PDF indefinitely; YuSMP's GenAI integration team is the kind of partner that can take a rules-based framework like this and wire it into AI-assisted crosswalk and skill-matching tooling once the underlying data model is sound.

What to watch next

The paper frames FAIR compliance as a lifecycle, not a one-time export, which means the more useful signal over the next few months is which registries (EMBL-EBI Competency Hub, Zenodo) start showing version history and crosswalk links for frameworks that cite these ten rules, rather than a single big-bang adoption announcement.