Candidate Structural Properties
The current formalization represents seven candidate structural properties:
- observability
- controllability
- traceability
- reconstructability
- independence
- integrity
- evidence coverage
These are candidate contributors to Structural Assurability rather than a complete taxonomy.
The theory does not assume that they are mutually independent, universally necessary, jointly sufficient, or universally monotone.
A candidate property contributes to the theory when a separating construction can demonstrate an assurance-relevant distinction associated with that property.
Dimension-Specific Separating Constructions
The formalization provides dimension-locked, profile-grounded separating constructions for four candidate properties:
- Observability: strict inclusion of observable-feature profiles.
- Evidence coverage: strict inclusion of covered-element profiles, with an additional controlled variant isolating one named element.
- Traceability: strict inclusion of trace-target profiles for a specified evidence item.
- Reconstructability: strict inclusion of reconstructable-target profiles.
Each difference relation can certify only its designated dimension.
The corresponding smart constructors also require independently established Structural Assurability non-equivalence.
A structural-property profile difference does not, by itself, establish a difference in evidentiary capabilities or claim resolution.
The general separating-construction interface remains available and permits arbitrary caller-supplied difference relations.