Domains and Validation
One reasoning discipline. Different standards of proof.
Geometry Intelligence is intended to organise complex problems across domains. It does not impose one universal validator. Each field carries its own standard of evidence, and the framework must respect it.
Validation Register
How different fields verify a conclusion.
| Domain | Typical question | Validation standard |
|---|---|---|
| Mathematics | Does a conclusion follow from stated axioms? | Formal proof and expert review. |
| Physics | Does a model remain mathematically consistent and describe observation? | Formal analysis and empirical testing. |
| Engineering | Will the proposed system perform safely? | Simulation, prototype and physical tests. |
| Software | Does the system satisfy its specification? | Tests, formal methods, production monitoring. |
| Medicine | Is an intervention safe and effective? | Clinical evidence and regulatory review. |
| Law | Does an interpretation follow the relevant authority? | Source hierarchy and jurisdictional analysis. |
| Policy | Does an intervention achieve its stated outcome? | Evidence review and measured impact. |
| Institutional authority | Which claims should a machine trust? | Provenance, source authority, entity resolution. |
The Principle
A domain-general reasoning architecture must remain domain-specific in validation.
Reading this table. The framework provides a common way to organise entities, evidence and relationships. It does not replace domain expertise or its standards. A geometrically-structured argument in physics still requires empirical testing. A geometrically-structured argument in medicine still requires clinical evidence and regulatory review. The framework's contribution is structure, not authority.