Context

Documentation status: architecture — see Maturity and evidence.

Context is a semantic input to logical interpretation. It is not merely a request-local dictionary or an ambient application variable.

In the current engine, context is passed into core logical operations including subtype testing, implication, and comparability. Logical relations also carry a dedicated context position alongside their source, target, and polyadic role map.

Conceptually:

IsSubtypeOf(A, B, roleMap, context)
Implies(A, B, roleMap, context)
IsComparable(A, B, roleMap, context)

This means the answer to a structural/logical question can depend on the context under which the relation is interpreted.

Context in logical relations

A logical relation is structurally closer to:

logical relation
  PolyadicMap
  Source
  Target
  Context
  attributes: weight, order metadata, ...

The context is therefore part of the relation's semantics, not an after-the-fact annotation.

Modeling guidance

Use an explicit conceptual context when validity genuinely changes with a modeled scope: for example a policy context, interpretation regime, scenario, tenant-specific ontology, temporal reasoning frame, or another domain-defined semantic scope.

Do not automatically turn every request correlation id, UI state, or thread-local value into a conceptual Context. Operational execution context and conceptual logical context are different layers.

Reasoning consequence

If two statements appear structurally compatible but were established under different contexts, do not assume they can be compared or composed without passing the intended context through the reasoning operation.