Library / First Principles Framework (FPF) - Core Conceptual Specification
Jump to passage
In this reading

Link to current text

Published source confirmed at last check

Source changed 2026-10-03 10:39:28 UTC · snapshot created 2026-10-03 10:40:04 UTC · last check 2026-10-03 10:50:05 UTC

A.3.1:4.7 - Semantic identity and variants

Two U.MethodDescription epistemes describe the same U.Method when their EntityOfConcern references resolve to the same A.3.1 identity. For the proposed comparison or reuse, compare their claims about the following method bases separately:

  • effective reference scheme and local senses, when a meaning difference matters;
  • generic participant meanings and declared applicability;
  • compatible preconditions;
  • compatible intended effects or preserved conditions;
  • compatible safety and other non-functional bounds;
  • accepted nondeterminism or search behavior; and
  • the same work-facing acceptance relation, when that relation is part of the comparison.

Different control flow, proof notation, programming paradigm, diagram notation, or prose does not by itself make a different method. The converse also holds: the same name, repository, supplier label, or diagram family does not prove identity.

Keep one method across parameter ranges, equipment envelopes, or representation variants only when its declared applicability and the bases used by the comparison admit that variation. A changed intended result, participant meaning, safety bound, semantic basis, or acceptance criterion requires a stated refinement, substitution, or distinct-method decision.

Same-name locality replay. An emergency-department Triage method applies to patient presentations awaiting clinical assessment; a clinician enacting it uses clinical signs to assign urgency, escalates unsafe cases, and stops when the evidence cannot support that assignment. A software-defect Triage method applies to defect reports awaiting product handling; a product team enacting it uses reproduction evidence, severity, and ownership to choose routing and release impact, and stops when the report cannot support that choice. The shared label identifies neither method. Their participants, applicability, local senses, intended results, and stops distinguish them without a generic context object.

No-extra-locality replay. EuclideanGCD over positive integers closes as one method when the integer meanings, division-with-remainder rule, positivity precondition, decreasing-remainder invariant, and greatest-common-divisor result are stated. If those facts answer the comparison, add no claim scope, context slice, model-use structure, or other locality object.