Modeling: review scope of informationSensitivityClassification

[jgeluk]

Tracks part of #98. Consider raising domain from dcat:Dataset to dcat:Resource and/or add distribution-specific sensitivity handling. Refs #98

[jgeluk]

Proposal: review scope of dprod:informationSensitivityClassification

Current modeling (ontology):

dprod:informationSensitivityClassification
  a rdf:Property, owl:ObjectProperty ;
  rdfs:isDefinedBy dprod: ;
  rdfs:domain dcat:Dataset ;
  rdfs:range dprod:InformationSensitivityClassification ;
  rdfs:label "information sensitivity classification" ;
.

Current SHACL shape:

dcat:Dataset-informationSensitivityClassification
  a sh:PropertyShape;
  rdfs:isDefinedBy dprod-shapes:;
  sh:path dprod:informationSensitivityClassification;
  sh:class dprod:InformationSensitivityClassification;
  rdfs:label "dataset information sensitivity classification shape";
.

What this issue asks:

• Consider whether sensitivity should be expressible on resources other than datasets (e.g., distributions).

Options to consider:

• A: Broaden domain to dcat:Resource (attach to datasets, distributions, data services, etc.); add/update SHACL property shapes accordingly.
• B: Keep domain at dcat:Dataset and add a separate distribution-level property (e.g., dprod:distributionInformationSensitivityClassification) with corresponding SHACL.
• C: Keep as-is with rationale documented if per-distribution differences aren’t needed in DPROD.

Proposed acceptance criteria:

• Decide A/B/C and document the decision.
• If A: change domain and add SHACL for at least dcat:Distribution; update examples/docs.
• If B: introduce a distribution-level property and SHACL; update examples/docs.
• Rebuild and verify published spec reflects the change.

Refs #98

[srirambles]

Raised as OMG issue https://issues.omg.org/browse/DPROD-32