A.11.OP:10 - Rationale

Operational parsimony is about relevance, not abstract minimization. The fewest-step method can be wrong when one additional action realizes the chosen result, changes a later policy, or preserves a relied-on condition. The longest method can also be wrong when its extra actions have no substantive receiver. Comparing keeping and removing one proposed requirement makes that difference visible without inventing a global cost function.

The three branches distinguish contributions that a step-count screen would confuse. Decision-changing result recognizes exploration and discrimination. Selected realization recognizes deterministic work. Assurance or recoverability preservation recognizes a named relied-on condition. None alone establishes obtainability, net value or an obligation. Existing selection or reliance can settle the need; otherwise the direct choice or demand method completes it. This preserves useful work while allowing a probe that could matter to lose against an available continuation.

The horizon must be substantive and bounded. A next-event horizon hides delayed information value; an indefinite horizon lets hypothetical future usefulness justify everything. The nearest named receiver is the smallest horizon that can carry the reason and the smallest reopen boundary when the use changes.

The rule coordinates existing decisions, transformations, results, evidence, assurance, and recovery uses. Keeping these objects under their direct patterns preserves FPF layering while giving practitioners one discoverable admission question.