#Kind Intent, Membership Judgment, and Extension

C.3.2 · Cluster C.I - Core CALs / LOGs / CHRs · Stable · Local kind declaration and classification pattern · Normative unless a section is explicitly informative · Part C - Kernel Extension Specifications

#About this pattern

This is a generated FPF pattern page projected from the published FPF source. It is canonical FPF content for this ID; it is not a FPF Reference product feature page.

#How to use this pattern

Read the ID, status, type, and normativity first. Use the content for exact wording, the relations for adjacent concepts, and citations to keep active work grounded without pasting the whole specification.

Type: Local kind declaration and classification pattern Status: Stable Normativity: Normative unless a section is explicitly informative

Use this pattern when repeated typed reasoning needs one explicit local-kind criterion, when one exact entity or non-entity value must be judged against that criterion in one context slice, or when a named use needs a representation of all candidates currently judged true.

#Relations

#Content

#Use This When

Use this pattern when repeated typed reasoning needs one explicit local-kind criterion, when one exact entity or non-entity value must be judged against that criterion in one context slice, or when a named use needs a representation of all candidates currently judged true.

What goes wrong if missed. A kind is confused with the document that declares it, a measurement or schema label creates membership, missing information becomes false, a mathematical set becomes ontology, or a guard decision rewrites the classification.

What this buys. A practitioner can state an ordinary result, pin the declaration edition and slice when reliance requires it, distinguish true, false, and unknown, and materialize an extension only for a receiving query or review. A manager can separately plan higher declaration formality, stronger assurance for a relied-on classification assertion, or a different claim scope without treating those changes as one ladder.

Primary EntityOfConcern. One local-kind classification use: an exact candidate, local kind, KindSignature edition, context slice, and judgment value.

First useful move. Write the readable result first: Pump #14 counts as a cooling pump in this plant slice because it satisfies the declared cooling-pump criterion. Cite the measurement only when the receiving use relies on that support. Create a reusable KindSignature only for repeated use, and materialize KindExtension only for a named set-consuming use.

Not this pattern when. Use the direct subject pattern to establish the candidate and the qualities, relations, or construction that make the criterion hold; A.14 for membership in a collection or collective; C.3.3 for cross-context use; C.29 when a mathematical representation changes a claim-bearing use; and E.24.UK for durable public U-kind admission.

#Problem Frame

A local kind can support useful typed reasoning without becoming a public FPF U-kind. Its intent may need a reusable declaration; one candidate may need a current judgment; and a query may need a set representation of true candidates. Those are different objects. The candidate's governed world-side or value-side features settle the criterion when they are available for evaluation. Evidence can support a claim about those features, but a record, carrier, observation, or proof does not manufacture them. The pattern is concept-level and notation-neutral: it requires no particular ontology language, schema technology, rule engine, or programming type system.

#Problem

The shorthand MemberOf(e,k,slice) is unsafe for this problem because readers can take it as an A.14 collection relation, an ontic classification-relation occurrence, a three-valued evaluation, a database lookup, or a guard. Likewise, U.EntitySet(slice) makes a set representation look like an admitted entity kind. Deterministic-looking notation then carries the wrong ontology. C.3.2 restores an explicit declaration, a judgment, and an optional representation while leaving candidate identity and direct features with their actual governors.

#Forces

#Four Objects, Not One

Scope is not a fifth object attached to the kind. A KindSignature episteme may have its own U.ClaimScope; a separate classification assertion has the scope of that assertion; and U.ContextSlice remains an explicit evaluation input.

#KindSignature Declaration

Author a reusable KindSignature only when a named receiving use needs the criterion and assumptions to persist across more than one classification.

Its claim content declares:

• the exact local kind that is its EntityOfConcern;
• the candidate ValueKind: the direct kind or value interpretation admitted as candidate input;
• the membership criterion in terms of direct governed candidate qualities, relations, constructive grounding, or other features;
• the exact U.ContextSlice conditions under which the criterion can be evaluated;
• the effective U.ReferenceScheme;
• named assumptions, dependencies, standards, versions, units, and temporal policy;
• its U.Formality;
• an optional ExtentRule stating how repeated candidate evaluations feed an extension when a varying extension is current.

In A.6.0 terms, SubjectKind is the broad candidate kind and RangedValueKind is the finite judgment value kind {true, false, unknown}. ExtentRule is declaration content, not a new ontic relation. Formality characterizes the declaration episteme—not the local kind, candidate, candidate value, judgment truth, or extension. A claim that relies on the signature content evaluates that dependency on its own F–G–R support path; raising signature formality does not upgrade an unrelated claim.

A changed membership criterion, evaluation-domain declaration, EntityOfConcern referent, or effective reference scheme identifies another U.Signature episteme edition. C.3.1 separately decides whether the same local kind continues across that declaration change.

#One Candidate Judgment

For exposition, this pattern uses:

J(candidate, kind, signatureEdition, slice) ∈ {true, false, unknown}

This is local notation for an evaluation result, not a newly admitted U-kind, an A.14 MemberOf occurrence, a direct classification relation, or an evidence relation. Evaluation is reproducible: fixed four inputs and unchanged governed candidate facts yield the same result. The slice names concrete versions and an explicit temporal selector; unqualified latest or current is not an evaluation input.

1. Recover the candidate first. An entity candidate is already individuated under its direct pattern. A non-entity value keeps the identity, unit, scale, and interpretation supplied by the pattern governing that value.
2. Pin all four inputs. Name the candidate, local kind, exact KindSignature edition, and exact U.ContextSlice.
3. Evaluate direct governed features. A satisfied criterion gives true; a known failed criterion gives false.
4. Keep non-settlement visible. Missing evidence, an unavailable declared dependency, or a candidate outside the declared evaluation domain gives unknown, not false.
5. Separate support from satisfaction. An observation, measurement result, source episteme, schema row, or evidence relation may support a classification assertion. It does not substitute for the candidate or make the criterion true merely by existing.
6. Separate guard disposition. A receiving guard checks scope coverage and the classification result as separate predicates and may decline use when the result is unknown. That fail-closed use decision does not convert the judgment to false and does not change the candidate's world-side features.

When a separate claim-bearing classification assertion is current, it is a C.2.1 episteme. Its exact EntityOfConcern is the governed entity about which the classification matters, and its claim content designates the candidate entity or value, local kind, signature edition, context slice, judgment, and relied-on evidence. A value classification may remain inside another claim's content instead of fabricating a value-shaped EntityOfConcern. The assertion creates neither candidate nor kind.

A domain that genuinely needs a durable classification-relation occurrence as an object of later relations must supply a separate direct pattern with exact participants, obtaining predicate, occurrence identity, and relation to this judgment. C.3.2 does not mint that occurrence by default.

#Extension as Representation

Materialize KindExtension(k, slice) only when a named query, quantification, comparison, review, or publication needs the current true-candidate set.

• Pin the KindSignature edition used by the representation even though the compact name shows only k and slice.
• State the declared candidate domain without inventing U.EntitySet.
• Include exactly the candidate values whose pinned judgment is true; do not insert unknown candidates as false or silently omit their unresolved status when the receiving use needs it.
• Treat braces, rows, indexes, or database results as representations. They do not create a collection holon, an A.14 membership occurrence, a direct classification relation, or the candidate features.
• Use C.29 when the mathematical lens or represented set changes a claim-bearing use. Otherwise the extension may remain a local calculation.

A changed candidate state or later context slice can change KindExtension(k, slice) without changing the signature or local kind. A changed extension row cannot repair an inconsistent declaration or subkind link.

#Subkind Monotonicity and Change

For exact reference-scheme edition RS, monotonicity is a law over judgments whenever SubkindOfObtains(k1, k2; RS) holds. Use an identified R_sub : U.SubkindOf occurrence only when a receiving use needs occurrence identity; an assertion that the predicate holds is a separate C.2.1 episteme:

When both judgments are defined for the same candidate and context slice under the paired signature editions used by the comparison, J(candidate, k1, edition1, slice) = true implies J(candidate, k2, edition2, slice) = true.

A counterexample diagnoses an inconsistent subkind link, incompatible signature editions, or an undeclared context bridge. Repair that governing defect; do not silently edit the extension table. Cross-context classification goes through C.3.3. When a kind bridge is used, C.3.3 governs its CL^k and reliance/assurance consequence; the bridge does not by itself change signature formality, claim scope, or either local classification judgment.

Keep these changes distinct:

#Required Worked Cases

#Physical pump

Within bounded context Plant-7, CoolingPumpKind is the local kind identified by Plant-7's declared cooling-function distinction. Signature edition CPS-2 names effective scheme PS-7, declares pump candidates, and states a criterion in terms of directly governed flow, heat-transfer, and operating-state features for plant slice S-14.

Pump #14 is independently identified as the physical candidate. A calibrated measurement-result episteme supports the assertion that its flow and temperature-difference features meet the criterion; the measurement result is not Pump #14 and does not constitute its cooling performance. With those feature facts settled, J(Pump #14, CoolingPumpKind, CPS-2, S-14) = true. An extension used by a maintenance query may represent Pump #14, but the query row does not create its classification.

#Episteme and publication form

The exact maintenance-instruction episteme MI-22 is evaluated against local kind DiagnosticInstructionKind using its claim-bearing content and governed subject. For one bounded maintenance-reading use, the selected page arrangement and notation MI-22-PDF-Layout is the publication form that expresses the chosen MI-22 edition, while exact digital file MI-22-PDF-File-7 is the U.PresentationCarrier that bears that form. Separately, selected arrangement and notation MI-22-HTML-Layout is another publication form that expresses the same chosen edition for that bounded use, while exact digital file MI-22-HTML-File-8 is the U.PresentationCarrier that bears the HTML form. This case asserts no C.29 representation because it selects no elements with an explicit correspondence to independently recovered objects and changes no admitted modeling or reasoning operation.

Changing the arrangement, notation, presentation carrier, or file encoding does not by itself change candidate episteme MI-22, satisfy the DiagnosticInstructionKind criterion, create another kind, or rewrite the classification judgment. If the claim-bearing content, exact signature edition, context slice, and governed candidate facts remain unchanged, the same judgment remains current.

#Non-entity temperature value

The value 87 °C, interpreted under a declared measurement scale, unit, reference scheme, and time, is evaluated against local kind HighTemperatureValueKind whose criterion is a declared interval. The candidate remains that governed non-entity value; no value-shaped entity is fabricated. The classification may stay inside the measurement or diagnostic claim content. The unit and interval must be pinned before a true or false result is possible.

#Schema label

A database row carries schema label Customer, but the receiving claim asks whether account holder #441 is a contractual customer. The label is a cue or supporting source, not the contractual relation that makes the world-side criterion hold. The practitioner must recover the actual candidate and the direct contractual facts. If the candidate were instead the row itself and the local kind concerned row shapes, that different candidate and criterion would have to be stated explicitly.

#Unavailable measurement

At later slice S-15, the cooling-pump signature still requires a governed flow measurement, but the measurement dependency is unavailable. The current evaluation returns unknown. A safety guard declines reliance on Pump #14 as a cooling pump for that use. The guard does not return false, prove that the pump lacks cooling performance, or remove it from a historical extension for S-14.

#Additional Transfer Cases

#Work Boundary

Classification does not weaken the work ontology:

• U.Work is the admitted U-kind;
• W : U.Work is one independently grounded, world-side, dated 4D work occurrence under the direct work pattern;
• a plan, expected-work item, log, card, database row, field bundle, assertion, or description about W is a separate episteme;
• performer assignment, enacted method, temporal extent, containing system, affected referent, material binding, resource use, transformation, production, result, delivery, and acceptance remain separately governed.

A local kind may classify an already identified W. A formal kind symbol, work label, plan, or record never occupies W's individual position, and record existence does not make planned work actual.

#Authoring Rhythm

1. Start with one readable classification sentence and its practical use.
2. Recover the exact candidate and direct features before discussing evidence.
3. Reuse an existing signature edition when it truly governs the criterion, scheme, dependencies, and evaluation domain.
4. Author a new signature edition only when repeated use needs the changed declaration.
5. Return true, false, or unknown without folding in the guard decision.
6. Create an extension representation only for a named set-consuming use.
7. If a separate assertion is required, give the C.2.1 episteme its exact EntityOfConcern, content, scope, evidence use, and edition.

#Conformance Checklist

#Common Anti-Patterns and Remedies

#Consequences

Benefits. Classification becomes inspectable without ontology growth, evidence-created truth, or two-valued coercion. Repeated criteria can be reused, and set-consuming uses can receive a bounded representation.

Costs. Reliance-bearing uses must pin a signature edition and context slice and preserve unknown through the receiving decision.

Risks avoided. Kind/declaration collapse, record ontology, implicit time, false-for-unknown, mathematical-set overread, silent subkind repair, and kind/individual substitution are blocked.

#Rationale

The kind, the declaration used to evaluate it, one candidate judgment, and a set representation change for different reasons. Treating them as one object makes evidence, time, scope, and notation rewrite ontology. Their separation preserves ordinary reasoning while supporting exact review when a repeated or safety-relevant use needs it.

#SoTA-Echoing

Model theory and type systems distinguish intensional declarations, satisfaction judgments, and extensions; measurement and evidence disciplines distinguish the subject feature from its observation or support. C.3.2 combines those separations with FPF's episteme identity, context-slice, representation, and direct-object boundaries.

#Relations

• Builds on: C.3, C.3.1, A.6.0 declaration identity, C.2.1 episteme identity, A.2.6 context slices and claim scope, and direct patterns for candidate identity and features.
• Coordinates with: C.3.3 cross-context bridges, C.3.4 local adaptations, C.29 mathematical representations, C.2.3 formality, F-G-R evidence and assurance, A.14 collection membership, and E.24.UK durable U-kind admission.
• Does not replace: the direct subject pattern, evidence-use relation, collection membership, claim-scope governor, guard decision, public-kind admission, or a separately justified durable classification-relation pattern.

#C.3.2:End

---

Last Updated: 2026-08-04 — upstream FPF commit 8b727cba (github.com/ailev/FPF)