Naming And Articulation

Naming and Articulation

Principle expression

Primary: P06 Supporting: P07, P16, P15

Scope

Own one judgment: given a concrete object, audience, and blocked decision, what is the smallest name, operative definition, explanation, and source location that makes the relevant relation usable without mistaking the label for the object or creating needless taxonomy?

A name is a handle, not a claim to exhaust essence. A definition is a decision boundary, not a dictionary flourish. An explanation reconnects the boundary to evidence and a reader's next action. Do not use this Skill for branding, stylistic polish, ordinary variable naming, a one-off sentence, generic article writing, skill-expression engineering, or a change to Sequence meaning.

Principle source

Use a host Sequence and matching interpretations when present. Otherwise use this package's read-only fallback in references/sequence.md. Read only P06, P07, P16, and P15 for this skill's stable lineage.

Start

State the concrete naming problem before suggesting words:

Object and source evidence:
Actor, audience, and next decision:
Current terms and nearest confusing alternatives:
What must survive this session:
Strongest case for leaving it unnamed or local:

If one actor can finish the task without a shared distinction or later recovery, keep the answer local. Do not manufacture a term, definition, glossary entry, or documentation page.

Core method

1. Recover the relation before the word. Read the actual object, history, owner, and evidence. Name the relation or boundary that changes an action; do not infer it from an attractive label, directory, or metaphor.
2. Find the contrast. State what the proposed term includes, excludes, and would otherwise be confused with. If no practical contrast exists, stop: a synonym or private phrase does not warrant a shared name.
3. Generate few candidates from the relation. Prefer a name that is compact, pronounceable in the host register, and durable across incidental implementation changes. A provisional name is allowed when action needs a handle before the object is fully known; mark its uncertainty rather than presenting it as final doctrine.
4. Write only the needed layers. Use the articulation layers: a name; an operative definition; an explanation; and, only if necessary, a projection. Every layer must enable a distinct reader action or recovery.
5. Place the explanation with its authority. A P-ID reading belongs in its interpretation; an architectural boundary in design; an agent method in the owning Skill or direct reference; an implementation detail near the code. A README index, glossary, or generated map points back to those sources and cannot become a second authority.
6. Test action and stopping. Ask the intended actor to distinguish the term from its nearest alternative and make the blocked decision. Remove a layer or term that does not change that decision. Route a new semantic consequence to principle-cultivation, a carrier choice to form-guidance, skill-expression wording to skill-engineering, and artifact placement to artifact-organization.

Boundaries

• Do not make “the real name” a substitute for evidence. Names may be revised; their sources, status, and decision boundary must remain visible.
• Do not add a global glossary because several local terms exist. Create a rebuildable index only when cross-source discovery is the proven problem.
• Do not define a term at every layer. Put the canonical operative definition at its owning source; use shorter linked explanations elsewhere.
• Do not turn a metaphor, translation, marketing phrase, or generated summary into doctrine. It is an audience projection unless separately governed.
• Do not change the Sequence or interpret a P-ID beyond its source line. Route that question to principle-cultivation.

Completion standard

The result is ready when it states the object, audience, contrast, chosen name, operative boundary, source placement, and a disconfirming observation. It is supported only when an action probe shows the reader makes the intended distinction, a boundary probe declines a request with no shared decision, and a context/authority probe preserves source and projection boundaries.