Ontology is a structured concept-and-relationship model used to organise, connect and reuse language about skills, occupations or other related entities.
In this project, a skill ontology is best understood as a reusable language layer for search, matching, suggestion, interoperability and inference. It is not, by itself, a competence-assessment framework.
Ontology-based relatedness is the degree to which two skills, roles or concepts are considered close because of semantic similarity, graph structure, overlapping demand, co-occurrence or transition patterns.
Adjacency can be used as an informal market label for this idea, but it is not Greenbeam's preferred term because it is imprecise and can imply more than the evidence supports.
Greenbeam should prefer language such as:
- ontology-based relatedness
- skill similarity
- association-based inference
- graph proximity
These describe proximity or overlap. They do not, on their own, establish why a person will succeed in developing or performing a capability.
Commensurability is the property that competence judgments for different people, and for people versus capability requirements, can be interpreted on a calibrated common metric or common capability scale strongly enough to support consistent comparison and controlled aggregation.
In Greenbeam's model, competence is a latent construct inferred from evidence, not something directly observed like a physical quantity. So the measurement only becomes meaningful when it is anchored to a well-defined capability construct and calibrated strongly enough to support:
- like-for-like comparison with respect to a defined capability and level
- gap reasoning against a defined requirement
- controlled aggregation into measures such as coverage or severity
- optimisation logic that tries to reduce the mismatch between available competence and required capability
Decision use therefore depends on:
- validity
- reliability
- calibration
- evidence quality
This is one of the clearest ways to distinguish development-grade from decision-grade competence data.
Recourse is the property that a competence judgment can be traced, reviewed, challenged, corrected, and learned from because the capability standard, level, evidence, assessors, and rationale are documented clearly enough to support accountability.
In Greenbeam's model, recourse matters because a stronger measurement system should not only produce a better judgment. It should also make it possible to determine:
- who made the judgment
- what capability and level were being judged
- what evidence was relied upon
- what rationale connected the evidence to the judgment
- whether the decision should stand, be corrected, or trigger a process change
This is the accounting-like part of the argument.
Accounting is not superior simply because it deals with numbers. It is stronger because it uses documentation, review, control, and correction practices that make judgments challengeable and improve measurement quality over time.
For Greenbeam, decision-grade competence data depends on both:
Commensurability makes judgments usable on a common capability frame.
Recourse makes those judgments governable, challengeable, and improvable.
This is also one of Greenbeam's strongest contrasts with weaker skills-cloud approaches.
If the underlying skill definitions are loose and the proficiency descriptors are generic across skills, then disagreement becomes hard to adjudicate. The judgement may still be recordable, but it has weak recourse because there is too little standard, traceability, or capability-specific evidence to determine whether the person, the assessor, or the process got the decision wrong.
