A.6.F - Function and Functional Precision Restoration (RPR-FUNCTION)
Type: Architectural pattern Status: Stable Normativity: Normative unless explicitly marked informative
A.6.F:1 - Problem frame
Use this pattern when function, functional, functionality, effect, or similar wording leaves an action-changing object, claim, or applicable rule unresolved. The reading may concern architecture, work, method, capability, a system-role kind or assignment, participation, actual functioning, responsibility, quality, mathematics, module allocation, or an interface. These are possible meanings to distinguish, not one semantic kind.
First recover the object or claim and the rule that defines or tests it. Return the repaired wording and next useful action, including any necessary stop. Use the optional FunctionUseRepair note in §4.2 only when a receiving use needs inspectable detail.
Stop when the source cue, entity, value, claim, or claim-bearing episteme, subject pattern, and the next admissible use are clear. If the next step must test whether named participants stand in a relation, add its admitted direct predicate. If it must preserve an affirmative, negative, or modal claim about that predicate, identify the C.2.1 relational-assertion episteme and its claim. If it must track one particular obtaining instance, apply the subject pattern’s identity rule and add the separately individuated occurrence. Otherwise leave those three branches empty. Add reusable declaration, other selected assertion, specification, or view episteme, or representation correspondence only when the next step needs it.
Misreadings to repair: a function becomes a root kind; functional architecture becomes a peer ontology beside architecture; a capability becomes a function; a method or work occurrence becomes a function; a mathematical function becomes design ontology; a module allocation becomes functional truth; or a quality claim hides behind “functionality”.
What A.6.F buys in practice: the practitioner can keep useful engineering language while naming the object or claim and going straight to its subject pattern. Direct participation, reusable declaration, claim-bearing description, and representation remain separate instead of becoming one generic function record.
Keep already clear wording, including an FPF claim whose object, meaning and applicable rule are recoverable. For a general relation question, evaluative language, grounded architecture adequacy, or an architecture structural view, use A.6.P, C.16.Q, C.30, or C.30.ASV respectively.
E.10.ARCH subject-pattern relation. When E.10 encounters function-like wording whose entity, value, claim, claim-bearing episteme, direct relation, or subject pattern is hidden, E.10.ARCH may apply A.6.F until that object and the remaining action are clear or the wording is lowered to ordinary prose, quote-only wording, reduced-use cue, blocked use, or incomplete rewrite. A direct relation names its actual participants; a reusable RelationSignature and declaration-local SlotSpecs, selected assertion, specification, or view episteme, and representation correspondence are added only for a current receiving use. After recovery, apply the subject pattern; A.6.F alone establishes no architecture, mathematics, quality, work, evidence, assurance, gate, decision, or release claim.
A.6.F:2 - Problem
FPF texts repeatedly use function-like wording for different FPF kinds and relations:
- required transformation or effect in an architecture view;
- capability of a System;
- method wording;
- work occurrence or work result;
- a system-role kind or assignment, participation, actual functioning, or responsibility;
- mathematical function or relation;
- quality, fitness, or characteristic wording;
- module allocation or interface relation;
- functional architecture shorthand.
These uses are all legitimate in ordinary engineering speech. They are not the same FPF object or claim. If the text does not name the entity, value, claim, or claim-bearing episteme and its subject pattern, subsequent reasoning cannot tell whether the sentence is about architecture, behavior, work, a system-role kind or assignment, participation, actual functioning, responsibility, mathematics, module structure, quality, evidence, or decision. When a separate direct relation, reusable declaration, assertion or specification, selected view, or representation is current, identify it as that separate object.
A.6.F:3 - Forces
| Force | Tension |
|---|---|
| Familiar engineering speech vs object and claim precision | Engineers naturally say “function”, “functional”, and “functionality”; when the phrase carries an FPF claim, the entity, value, claim, or claim-bearing episteme and its subject pattern must be recoverable. |
| Functional architecture vs peer ontology | Functional architecture is useful, but it is the FunctionalStructure case of ArchitectureOf@Context, not a separate root architecture kind. |
| Capability or effect vs work or method | A function-like phrase may describe what a System can do, what a method prescribes, or what work has done; those are different values, claims, epistemes, and, where applicable, direct relations, each with its own subject pattern. |
| Mathematical function vs design relation | Keep an adequate mathematical definition or calculation. Use C.29 when a separate question about choosing a mathematical representation, transferring structure or relying on an inference remains. |
| Module allocation vs functional relation | Functional dependencies may be allocated to modules, but function and module-interface structure do not become one FPF kind. |
| Small repair vs unneeded evidence, quality, decision, or assurance apparatus | Most cases need the object or claim, its subject pattern, and a stop condition. Add direct-relation participants, reusable declaration, selected claim-bearing episteme, or representation correspondence only when the current use needs that object. |
A.6.F:4 - Solution
A.6.F is an A.6.P RPR specialization for function-like wording. It does not mint U.Function. It assigns the use under repair to an entity, value, claim, or claim-bearing episteme and its subject pattern, then stops unless another claim remains current. It does not package direct relations, declaration-local SlotSpecs, assertions, specifications, views, and representation elements as peer records.
A.6.F:4.1 - Trigger rule
When function-like wording remains ambiguous, distinguish the readings below. They are possible meanings to inspect, not a U.* kind, claim kind, relation kind, or admission result:
One phrase may combine several readings or transfer metonymically between them. Recover the claim it makes and retain every distinction that changes its use.
- architecture or functional architecture;
- capability, effect, externally promised behavior, or user-visible functionality;
- method wording, work occurrence, or work result;
- a system-role kind or assignment, participation, actual functioning, or responsibility;
- mathematical function, mapping, relation, loss, objective, or value functional;
- quality, fitness, characteristic, score, or proxy wording;
- module allocation, interface, signature, port, API, protocol, flow, or mechanism relation;
- another independently established claim named by value, such as evidence, assurance, gate, decision, or release.
If none of those readings carries a current FPF claim, the wording may remain ordinary Plain prose.
A.6.F:4.2 - FunctionUseRepair
FunctionUseRepair is an optional pattern-local repair note for a receiving use that needs inspectable detail. Its functionLikeReadingUnderRepair value only helps a reader recognize and dispatch a possible reading; neither that value nor the three scan groups below is a U.* kind, claim kind, relation kind, or admission result. The recovered result belongs in exactGovernedObjectOrClaim under its subject pattern. The note carries no project-publication, evidence, decision, or U.Function authority. FunctionalStructure is an ArchitectureStructureKindRef value under C.30.ASV, not a kernel Function kind.
FunctionUseRepair ::= {
phrase,
functionLikeReadingUnderRepair: {
directObjectOrValueReading?:
systemCapability |
methodDescription |
mechanismRealization |
workPlan |
workOccurrence |
workResult |
mathematicalFunction,
claimOrConditionReading?:
requiredTransformation |
requiredEffect |
inputCondition |
outputCondition |
systemRoleKindOrAssignmentCue |
participationOrFunctioningCue |
responsibilityCue |
qualityExpression |
characteristicExpression |
functionalArchitecture |
evidenceClaim |
assuranceClaim |
gateClaim |
decisionClaim |
publicationClaim,
relationParticipantOrLocusReading?:
functionalElementLocus |
transformerSideFiller |
candidateBearer |
functionalPort |
methodPosition |
mathematicalRelation |
moduleAllocation |
interfaceRelation |
signatureRelation,
otherDeclared?
},
exactGovernedObjectOrClaim: oneOrMoreOf {
exactEntityOrValueRef?,
exactClaimOrClaimContent?,
exactClaimBearingEpistemeRef?
},
directRelationPredicateUse?: {
admittedDirectRelationKindRef,
relationKindToken?,
semanticPredicate,
actualParticipantRefs,
directRelationPatternRef
},
relationalAssertionUse?: {
relationalAssertionEpistemeRef,
assertedClaimContent,
assertedSemanticPredicate,
polarityOrModality,
actualParticipantRefs,
directRelationPatternRef
},
obtainingRelationOccurrenceUse?: {
individuatedRelationOccurrenceRef,
obtainingSemanticPredicate,
actualParticipantRefs,
occurrenceIdentityRuleRef,
directRelationPatternRef
},
reusableDeclarationUse?: {
relationSignatureRef,
declarationLocalSlotSpecRefs
},
selectedClaimBearingEpistemeUse?: {
assertionSpecificationOrViewEpistemeRef,
selectedClaimOrDesignation
},
representationUse?: {
representationElementRefs,
interpretationOrCorrespondence,
representedObjectOrClaimRef
},
sourceCueText?,
subjectPatternApplicationRefs?,
blockedLocalOverreadRefs?,
admissibleUse,
nonAdmissibleUse?,
nextAdmissibleUse,
stopCondition
}
The repair is complete when a practitioner can name the object or claim, apply its subject pattern, and state the remaining action. When a note is needed, at least one entity or value, claim or claim content, or claim-bearing episteme is required in exactGovernedObjectOrClaim. A source cue stays in sourceCueText; it is not a recovered value.
When a direct relation is current, first name its admitted kind, semantic predicate, and actual participants in directRelationPredicateUse. Add relationalAssertionUse only when one C.2.1 episteme affirms, denies, or otherwise modalizes that predicate. Add obtainingRelationOccurrenceUse only when the receiving use needs one separately individuated obtaining occurrence under the subject pattern’s identity rule, applied through A.6.REL; a predicate or assertion never supplies occurrence identity.
Add a reusable RelationSignature and declaration-local SlotSpecs only for reusable typed use; add another selected assertion, specification, or view episteme only when it is a separate claim-bearing object; add a representation and its interpretation or correspondence only when that representation matters, using C.29 when a separate mathematical-representation, transfer or inference question remains. If the text still hides a function, capability, work, method, system-role kind or assignment, participation, functioning, responsibility, module, evidence, gate, or mathematical-function collapse, the repair is incomplete.
Preserve the subject’s necessary applicability and stop conditions in the repaired claim. Add blockedLocalOverreadRefs or nonAdmissibleUse as an explanatory guard only under F.19:4’s full independent-ground, plausible-reader, contribution, and smallest-clear-correction test. An unused guard may be omitted without an absence entry.
A.6.F:4.3 - Repair assignments
When a function-like phrase is claim-bearing, recover the object or claim under concern before lowering or rewriting the phrase. C.30.ASV does not define a world-side or view-local FunctionalElement individual. Its FunctionalStructureView is the same ArchitectureStructuralView episteme about one selected functional U.Structure; a FunctionalStructureViewUse may cite FunctionalElementClaim epistemes and other values needed by the use. Required or desired effects, actual transformations, candidate bearers, capabilities, ports, allocations, and correspondences remain separately established claims or values.
If a distinct functional-element individual is needed, first define its predicate and identity in a subject pattern; otherwise stop at the smaller requirement, behavior or effect claim, capability, participant, condition, port declaration, or other directly defined object. A field name or source cue is not a substitute for that object.
Method-description guard. A procedure, code file, solver model, recipe, protocol, or algorithm is only a clue. First identify the knowledge object and the method it is about. Then point to at least one claim that says how that method is done, such as its transformation or enactment concern, applicability, precondition, intended effect or preserved condition, bound, generic participant meaning, or internal method composition. Only then classify that already identified U.Episteme as U.MethodDescription under A.3.2.
A title, author, citation, approval, file form, or runnable artifact alone is a near-miss. If no admitted U.Method is its EntityOfConcern, or no way-of-doing claim is present, do not use U.MethodDescription: keep the actual plan, work, result, formal substrate, mechanism declaration, representation, publication occurrence or form, or carrier with its subject pattern. A different code, text, diagram, or publication form does not decide membership.
If claim content, method, or effective reference scheme changes, C.2.1 first identifies the resulting episteme; then apply A.3.2 to that individual.
| Function wording use | Recovered object or claim and subject pattern | Boundary |
|---|---|---|
| required or desired functional behavior, transformation, or effect | Keep the requirement or other claim about the required or desired behavior or effect with its requirement, architecture, capability-gap, functional-view, method, or other claim-bearing pattern. Use U.Transformation only for one independently grounded actual bounded change under A.3.4. Use TransformationFlowStructure only for an independently selected structure over explicitly named loci, not for the required effect itself. | Requirement wording establishes neither an occurrence nor a filled FunctionalElementClaim. Stop at the claim pattern unless the changed referent, boundary, conditions, actual before/during/after facts, and continuity or reidentification rule are grounded. A functional view may cite the required claim, selected structure, bearer candidate, capability, and allocation without saying that the change occurred. |
| functional element in a view | Under C.30.ASV, use one ArchitectureStructuralView episteme whose EntityOfConcern is one selected functional U.Structure. When needed, add a FunctionalStructureViewUse that cites C.2.1 FunctionalElementClaim epistemes and only separately established behavior or effect, bearer, capability, port, allocation, or correspondence values. | This is a claim-and-view interface, not a FunctionalElement individual, loose table row, or module. If no bearer or candidate allocation is current, keep the requirement, required-behavior claim, required-effect claim, capability gap, functional-behavior slot, or candidate-allocation question with its subject pattern. |
| transformer-side filler and candidate bearer | For a design-only candidate, keep the candidate transformer-side system locus or candidate U.System reference without asserting an assignment or performed Work. If the local kind TransformerSystemRole is current, name that kind; add a separate judgment classifying a System under it only when that judgment obtains. If an assignment is independently current, recover its directly declared species and its obtaining occurrence with actual participant values, holder, applicability, and extent under A.2.1. If performed Work is independently current, point to its basis: A.13 first, independent A.15.1 Work admission second, and F.6 afterward only for precise assignment-bound attribution. Coordinate the selected locus with A.3.4 TransformerRef?, A.7, A.15, A.15.1, and A.15.2 only for the claims that are current. | A FunctionalElementClaim may cite an independently established candidate-bearer locus, but that citation is not the whole transformer ontology. A kind, classification judgment, assignment species, assignment occurrence, and dated Work are independent facts; none manufactures another. |
| input condition, output condition, and functional ports | A.3.4 InputConditionRefs?, OutputConditionRefs?, and FunctionalPortRefs?; U.Signature discipline through A.6.0 and A.6.5 when accepted or produced states, media, flows, signals, information, work products, formal objects, or functional port signatures matter | A functional port is not automatically a module interface. Use A.6.M only when module-interface or substitution compatibility is the claim. |
| capability of a System | the qualified ability claim about an identified System under A.2.2 | Does not imply that a method, module, work occurrence, or successful transformation exists. |
| method or algorithm wording | U.Method only when the claim concerns a reusable semantic way of doing; U.MethodDescription only for an already identified U.Episteme that passes the A.3.2 guard above: one admitted U.Method is its EntityOfConcern and at least one claim says how that method is done | Procedure, code, solver, recipe, protocol, and algorithm forms are clues only; they establish neither membership, execution, nor evidence. |
| mechanism wording | U.Mechanism through A.6.1 and E.20 when a law-governed realization or operation structure is the claim | Does not become a method, Work occurrence, capability, selected functional structure, or functional-view claim by label. |
| work plan, work occurrence, or work result | Recover the U.WorkPlan under A.15.2, one dated W : U.Work on its A.13 basis with independent A.15.1 admission and F.6 only for precise assignment-bound attribution, or the separately identified result entity or episteme together with the predicate that relates it to the current Work or application. Use A.15.PROD when production, entity inception, or production completion is current, and A.6.RCD only when the needed direct result relation has no current governor. | A plan, Work occurrence, and result are different objects. None implies reusable function ontology or completed functioning, and a result is not part of Work identity. |
| system-role or responsibility wording | VP.AllocationResponsibility is a recognition cue. A positive responsibility claim needs an applicable domain predicate and supporting facts about its participants and conditions; add explicit occurrence identity only when the use needs it. An absent predicate leaves an A.6.RCD missing governor, missing case facts leave the claim unknown, and contrary facts refute it. Establish any assignment independently under A.2.1. For actual Work use A.13 first, independent A.15.1 admission second, and F.6 only for precise assignment-bound attribution. | A system-role kind, classification judgment, assignment, function phrase, commitment, position, authority label, or Work attribution establishes no responsibility relation by form. Responsibility may obtain with or without a commitment; each relation keeps its own predicate and identity. |
| participation or actual functioning | Name the domain predicate, participants, conditions and case facts supporting the claim. Add explicit occurrence identity when the receiving use needs it. If no rule governs the needed predicate, return that A.6.RCD missing governor; if only case facts are missing, leave the claim unknown. | Assignment, capability, allocation, nearby Work, or a function label does not make participation or actual functioning obtain. |
| mathematical function or relation | Use its mathematical definition and applicable rule. When the use needs a separate C.29 lens account, state domain, codomain or relation domain, preserved and lost structure, admissible use and stop condition. | Mathematical form alone establishes no architecture, evidence, causal, assurance or decision claim. |
| quality or fitness expression | C.25, C.16, C.16.Q, A.17, A.18, or an admitted characteristic or measurement subject pattern according to the claim being made | Does not let “functionality” carry a quality claim without bearer and subject pattern. |
| module allocation | Recover the allocation or correspondence claim or relation under its subject pattern. When a functional view cites it, use FunctionalStructureViewUse with its ArchitectureStructuralView and the claim or relation reference; use A.6.M when a module-interface claim is being made. | Does not make function and module one FPF kind. One selected functional structure may have allocation claims involving several modules, one module may occur in claims about several functional structures, and a module may have no current functional-view claim. |
| interface relation, module-interface relation, or signature relation | Use A.6.RSIR first when bare interface, API, port, protocol, or service-access wording could point to several direct EoCs; then use A.6.M for the module-interface boundary and A.6.0 and A.6.5 for signature discipline, with A.6.B, A.6.C, or A.6.P:4.11a only when that boundary, interface condition, API, protocol, service, promise, or duty claim is being made | Does not turn a functional link, port label, API name, or signature into implemented compatibility. |
| evidence, result, assurance, gate, decision, or publication claim | the direct evidence, result, assurance, gate, decision, publication, or source pattern named by value | Function wording can point to these claims, but it does not authorize or prove them by itself. |
| functional architecture | ArchitectureOf@Context whose structureKindRefs includes FunctionalStructure, one selected functional U.Structure, and the ArchitectureStructuralView episteme about that structure. Use FunctionalStructureViewUse only when citations to FunctionalElementClaim epistemes or other separately established values change action. | Not a peer architecture ontology, functional-element individual, selected transformation-flow structure, or mathematical graph description by itself. |
Required-versus-actual check. “The cooling loop shall reduce outlet temperature by 8 °C within 60 seconds” remains a requirement or functional-view claim; by itself it identifies no U.Transformation. If a later run changes the loop state, identify that cooling occurrence separately under A.3.4 from the changed loop, boundary and conditions, actual before/during/after facts, and continuity or reidentification rule. Requirement-only material is the countercase and stop: it cannot admit an actual transformation, observed functioning, or evidence of success.
A.6.F:4.4 - Functional architecture boundary
Functional architecture is the FunctionalStructure case of ArchitectureOf@Context: the declared organization used to relate one selected functional structure to independently established claims about required behavior or effects, capabilities, functional dependencies, and constraints that a holon is to realize, before or alongside allocation to modules, local system-role kinds or assignments, work, evidence, control relations, selected transformation-flow structures, or mathematical descriptions of those structures. Under C.30.ASV, the view is an ArchitectureStructuralView episteme whose EntityOfConcern is that selected structure; it does not define a functional-element individual. A FunctionalStructureViewUse may cite FunctionalElementClaim epistemes and other separately established values needed by the use.
Functional architecture shorthand:
open the `ArchitectureOf@Context` form in the current C.30 edition;
name the described holon and select one functional `U.Structure`;
use the `ArchitectureStructuralView` episteme whose `EntityOfConcern` is that structure and whose viewpoint conformance obtains;
add `FunctionalStructureViewUse` only when `FunctionalElementClaim` epistemes or other separately established values change action;
require an independent A.3.4 basis for every actual-transformation reference;
fill every other C.30 field required by this architecture use.
The view keeps requirement, required-behavior/effect, capability, dependency, and constraint claims with their subject patterns; their wording does not turn them into SelectedStructureReference values or actual transformations. An actual-transformation reference is admissible only after A.3.4 independently grounds the occurrence. A selected TransformationFlowStructure, path slice, crossing, flow valuation, or mathematical description may be related to functional structure through C.30.TFS-REL, E.18, or E.18.2, but it is neither the required effect nor the functional architecture itself unless the positive selected-structure co-reference check succeeds.
A.6.F:4.5 - Function-flow-module alignment note
Recover the local alignment when functional wording touches flow or module allocation but does not yet require a full structural view or A.6.M module-relation repair. Use this note only when the receiving use needs an inspectable alignment record; otherwise state the recovered alignment and boundary in the repaired wording.
FunctionFlowModuleAlignmentNote:
required function or effect:
flow path or dependency:
proposed module allocation:
separateNeighborClaims:
known mismatch:
subjectPatternApplicationRefs:
admissible use:
non-admissible use?:
The note records only the local function-flow-module alignment and boundary. Its explanatory non-admissible-use guard is optional under the full F.19:4 test and needs no absence entry when unused. Functional architecture, module relation, implemented-interface, evidence-sufficiency, and architecture-decision claims remain with their subject patterns.
A.6.F:4.6 - Common kind and relation separations
| Confusion | Repair |
|---|---|
| function = module | Keep VP.Functional and VP.ModuleInterface distinct; connect them through declared correspondence, allocation, retargeting, or A.6.M module-relation repair. |
| function = capability | Capability is the qualified ability of a System under A.2.2. Keep a required or desired behavior/effect as claim content under its requirement, architecture, capability-gap, functional-view, method, or other subject pattern; neither that claim nor the capability establishes an actual transformation. |
| function = work | One W : U.Work is a dated world-side occurrence. Its result or output is a separately identified entity or episteme. Connect it only through the subject-specific direct result or production relation that obtains, or use one local A.15.PROD claim for the current production-work, inception, or completion question. Use A.6.RCD only when a needed relation has no current governor. Function wording remains design-side or description-side content unless a work-facing claim is current. |
| function = method | Method is a reusable way of doing. A method claim may state an intended effect, but that effect is neither the method nor an actual U.Transformation; apply A.3.4 only when the actual change occurrence is independently grounded. |
| function = system-role kind, assignment, participation, functioning, or responsibility | Recover only the branch the sentence asserts. A system-role kind is a local ...SystemRole kind; an assignment is one occurrence and its declared U.SystemRoleAssignment species; actual Work requires A.13 and independent A.15.1 admission, with F.6 only for precise assignment-bound attribution; participation, functioning, and responsibility each need their domain predicate and supporting case facts; a missing rule, unknown fact and refuted claim have different outcomes. |
| mathematical function = holon purpose | Recover the intended connection to purpose. Use C.29 when it requires a mathematical representation or inference account, including preserved and lost structure and limits of use; an ordinary mathematical function needs no such account merely to remain a function. |
| functional diagram = evidence | Distinguish the diagram, the claim-bearing episteme it expresses and any publication. The content may support a named claim under that claim’s direct rule; diagram presence alone does not. Use A.10 when the source-to-use account is implicit or must remain replayable, and G.6 when the receiving use needs an addressable provenance path. |
| functionality = quality | Recover the quality bearer and subject pattern through C.25, C.16, or C.16.Q before using the wording as an adequacy claim. |
A.6.F:4.7 - Composability and compositionality
Composability and quality compositionality are separate claims. If the text says parts can be assembled, keep that as a structure or use claim. If it says a quality of the whole follows from parts, assign the quality-composition claim to C.25 and the direct characteristic, measurement, or quality pattern governing that claim:
Composability:
exactGovernedObjectOrClaim: the A.6.M `ModuleInterfaceClaim` content for "A and B can be assembled under interface X"
selectedClaimBearingEpistemeUse: the C.2.1 episteme carrying that claim content
directRelationPredicateUse?: only a domain predicate independently admitted for a direct module-allocation or module-interface relation, with its actual participants
relationalAssertionUse?: the C.2.1 episteme that affirms, denies, or modalizes that admitted predicate, only when such a predicate is current
obtainingRelationOccurrenceUse?: only when that admitted relation has a same-versus-new-occurrence rule and the receiving use needs explicit identity to distinguish, reidentify or refer to one obtaining occurrence, or to use it as a participant through A.6.REL
reusableDeclarationUse?: one compatible RelationSignature and its declaration-local SlotSpecs, only when the receiving typed use needs them
subjectPatternApplicationRefs: A.6.M for `ModuleInterfaceClaim`; C.2.1 for its claim-bearing episteme; A.6.RCD when a reusable direct relation is needed but absent; A.6.REL only after the direct relation is admitted and the receiving use needs explicit occurrence identity; A.6.0 and A.6.5 only for the reusable declaration
Quality compositionality:
exactGovernedObjectOrClaim: the affected Q-Bundle and the structural-characteristic, causal-hypothesis, or evidence-relation claim that this use relies on
directRelationPredicateUse?: the predicate supplied by the applicable subject pattern and its actual participants, only when that relation is current
relationalAssertionUse?: the C.2.1 episteme and its affirmative, negative, or modal quality or evidence claim when that assertion is current
subjectPatternApplicationRefs: C.25; A.17 and A.18 for Characteristic and Scale questions; C.16 for measurement; C.16.Q for unresolved quality wording; A.10 when the source-to-use account is implicit or must remain replayable
Non-admissible:
successful assembly is not quality propagation
Compositional formalisms may express composition structures and view or model relations. They do not make safety, latency, reliability, or another quality propagate automatically.
A.6.F defines no separate quality-composition record. Use the claim form supplied by the applicable subject pattern. A composite family uses the C.25 Q-Bundle with its bearer and declared measures. Include scope, a qualification window, mechanisms, status, or evidence when that value changes the quality claim or receiving action. A single Characteristic and its Scale follow A.17 and A.18; use C.16 for a measurement question and C.16.Q when quality wording still needs repair. Keep a causal inference with its causal pattern. Use A.10 when the source-to-use account is implicit or must remain replayable. If a required value cannot be established, keep the dependent quality-composition claim unresolved rather than filling an A.6.F-only schema.
A.6.F:4.8 - Worked slices
Function-like module claim; no direct relation. A release note says, “The brake-control functional package is in the vehicle-control system.” The receiving question asks what interface BrakeControllerPackage presents in Release-2026Q2 under BrakeControlInterfaceSpec-v5. That question selects the module-interface claim; the adjective functional alone selects neither that claim nor a new U.Function kind. Recover:
exactGovernedObjectOrClaim: A.6.MModuleInterfaceClaimcontent namingBrakeControllerPackage,VehicleControlSystem,Release-2026Q2,VP.ModuleInterface,BrakeControlBoundary, and aninterfaceSpecificationRefthat resolvesBrakeControlInterfaceSpec-v5; its direct-relation disposition isnoDirectRelationClaimed;selectedClaimBearingEpistemeUse:BrakeArchitectureNote_v3 : U.Epistemeunder C.2.1 carries that content and hasBrakeControllerPackageas itsEntityOfConcern;directRelationPredicateUse,relationalAssertionUse, andobtainingRelationOccurrenceUse:not used, because no domain rule has admitted a direct module relation predicate or an obtaining occurrence for this case;- remaining action: apply A.6.M to the declared interface and admissibility conditions; do not infer a function allocation, direct relation, or implemented compatibility from the source phrase.
Interrupted assignment; occurrence identity needed. A maintenance log says, “Robot-7 resumed its inspection function after a documented period with no inspection assignment.” Treat function as a cue. Recover InspectorSystemRole and one declared direct species CellInspectorAssignment <: U.SystemRoleAssignment; then test its direct predicate for Robot-7 : U.System and the species-specific cell and interval participants. Keep Robot7AssignmentLog_42 as the separate C.2.1 episteme that states the interval facts. A.2.1 says that the demonstrated non-assignment period ends the first assignment occurrence and the later resumption begins another. When the maintenance history must distinguish them, apply A.6.REL with that identity rule to keep InspectorAssignment_PreGap and InspectorAssignment_PostGap distinct. A taxonomy or scheme is not an assignment participant, and neither assignment implies inspection Work.
Neighbor claims that need their own relation.
TestArticle-7 participates passively in TestWork-9 during TestInterval-9is not established byTestArticleSystemRoleorTestArticleAssignment-7. Until a direct passive-test-participation predicate supplies participant order and identity, returnA.6.RCD missing-governor[direct passive-test-participation relation]; the tester’s Work and F.6 attribution remain usable.Motor-M1 drives PumpAssembly-A during PumpRun-17needs a direct motor-drive-functioning predicate. Assignment, torque capability, and pumping Work remain separate; without that predicate returnA.6.RCD missing-governor[direct motor-drive-functioning relation].Hammer-3 supports PaperStack-9 during Interval-Pneeds a direct artifact-support predicate. Do not replace that claim with an interchangeable list of use, load, support, or function kinds; without the predicate returnA.6.RCD missing-governor[direct artifact-support relation].
Functional architecture phrase. A team says, “the functional architecture is the user journey.” A.6.F does not let the phrase create a separate architecture kind. For a receiving use that needs inspectable detail, the repair can be recorded as:
FunctionUseRepair:
phrase: "functional architecture"
functionLikeReadingUnderRepair: functionalArchitecture
exactGovernedObjectOrClaim: the `ArchitectureOf@Context` claim record and its one selected functional `U.Structure`
selectedClaimBearingEpistemeUse: the `ArchitectureStructuralView` episteme whose `EntityOfConcern` is that structure, plus any C.2.1 `FunctionalElementClaim` epistemes cited by the use
subjectPatternApplicationRefs: C.30; C.30.ASV
blockedLocalOverreadRefs: the source claim "the functional architecture is the user journey"
nextAdmissibleUse: when the view changes action, use `FunctionalStructureViewUse` to cite the view, claim epistemes, and only separately established bearer, capability, port, allocation, or correspondence values
stopCondition: ordinary phrasing remains Plain when no architecture claim is made; requirement-only material remains a claim; an actual transformation appears only on an independent A.3.4 basis
Functionality as quality. A product note says, “new functionality improves adequacy.” The repair separates the added-capability or required-effect claim from the quality claim. Capability or effect wording may stay as recognition, but the adequacy claim goes to C.25, C.16, C.16.Q, or the admitted characteristic or measurement pattern that states its bearer and criterion. A.6.F stops once those claims and subject patterns are clear; it adds no reusable declaration, view, or representation apparatus unless the receiving use needs it.
Mathematical function or loss. A model note says, “the loss function explains the holon purpose.” The repair keeps the mathematical function under C.29 lens discipline: domain, codomain or relation domain, preserved and lost structure, lens-use admissibility value, and stop condition. The loss may inform a reasoning move; it does not become holon purpose, evidence sufficiency, causal proof, assurance, or project decision by itself.
Pump-station functional dependency. A maintenance note says, “the backup pump function is degraded.” A.6.F first separates the required effect, the qualified ability claim about the holder System, the physical module allocation, the performed maintenance work, the evidence relation, and the quality claim. The functional wording may open a FunctionalStructure view under C.30.ASV or go to the capability pattern; it does not by itself prove the pump was tested, authorize operation, or make the backup module compatible with the main line.
Product-platform allocation. A hardware team says, “thermal management functionality moved to the chassis.” The repair separates required heat-removal effect, module allocation, interface constraints, signature constraints, architecture structural view, and any evidence or gate claim. A.6.F keeps the function-like wording useful for architecture work while sending module-interface and evidence claims to their subject patterns.
A.6.F:5 - Archetypal Grounding
| Working case | Grounding |
|---|---|
| Recovery question | A practitioner reads “the function”, “functional architecture”, or “this functionality” and needs to know whether the sentence is about capability, effect, method, Work, a system-role kind or assignment, participation, actual functioning, responsibility, module allocation, mathematical relation, quality, or architecture. A.6.F asks for the object or claim and its subject pattern when the phrase leaves the meaning or applicable rule unresolved. |
U.System | A robot, software system, plant, product platform, or AI-agent system may have capabilities, required effects, control functions, module allocations, runtime flows, and user-visible functionality. Those are not one FPF object; A.6.F identifies the entity, value, claim, episteme, or direct relation current in each use and requires its subject pattern for it. |
U.Episteme — method-description case | An algorithm file is not automatically a method description. It passes the A.3.2 membership test only when one identified claim-bearing episteme has one admitted U.Method as its EntityOfConcern and says substantively how that method is done. File form, name, author, citation, approval, or executability alone is the near-miss; stop at the direct representation, publication, form, carrier, or other current pattern. |
U.Episteme — other function-like claims | A functional diagram, modeling-language view, architecture view or note, generated-code architecture note, benchmark report, or mathematical model may cue a claim-bearing episteme; a publication form can express its selected edition and a publication occurrence can make that edition available. Read the claim before the form: use C.30 for architecture and view claims, C.30.ASV for structural-view claims, and the pattern that defines any required representation correspondence; keep a benchmark report as an episteme and publication rather than evidence or a method description, and establish the grounds for relying on its content, using A.10 when the source-to-use account is implicit or must remain replayable and G.6 when an addressable provenance path is needed; use its subject pattern for a mathematical or formal claim and use C.29 when a mathematical-lens use is current. Neither the episteme, publication occurrence, form, carrier, nor correspondence becomes the function, capability, Work, evidence relation, architecture, or mathematical object by its form. |
A.6.F:6 - Bias-Annotation
Scope: unresolved function-like wording that changes an FPF claim or its use.
| Bias risk | Mitigation |
|---|---|
| Function-root bias | Recover the object or claim and its applicable rule; A.6.F supplies no universal U.Function kind. |
| Functional-architecture exception bias | Functional architecture is normalized as FunctionalStructure, not a peer ontology. |
| Module bias | Function-to-module allocation uses correspondence or A.6.M module-relation repair; function and module remain distinct. |
| Mathematical bias | Keep adequate mathematical wording; open C.29 only for a remaining mathematical representation, transfer or inference question. |
| Check-only bias | Every conformance item carries a repair action or subject-pattern application. |
Use the checklist to verify a repair when needed; it adds no required project form.
A.6.F:7 - Conformance Checklist
| ID | Requirement | Failed-check repair |
|---|---|---|
| CC-A6F-1 Object or claim. | The repaired phrase makes its object or claim and applicable rule recoverable. A separate reference is needed only when the receiving use depends on it. When a direct relation is current, the admitted kind and semantic predicate are distinguished from the C.2.1 relational-assertion episteme and from any separately individuated obtaining occurrence; each current branch names its actual participants, and neither a predicate nor an assertion supplies occurrence identity. A reusable declaration appears only when typed reuse is current; another selected assertion, specification, or view appears only when that claim-bearing episteme is current; representation correspondence appears only when representation is current. | Restore the missing distinction or leave the dependent claim unresolved; use FunctionUseRepair only when its inspectable detail is needed. |
CC-A6F-2 No U.Function. | The use does not mint or rely on U.Function as a new root kind. | Assign the use to a functional view, capability, method, Work, system-role kind or assignment, direct participation, functioning, or responsibility predicate, mathematical lens, quality or characteristic, module allocation, or another subject pattern. |
| CC-A6F-3 Functional architecture expansion. | Functional architecture expands to an ArchitectureOf@Context claim record with one selected functional U.Structure. When the view changes action, use the ArchitectureStructuralView episteme whose EntityOfConcern is that structure and a FunctionalStructureViewUse that cites FunctionalElementClaim epistemes and only separately established values. No label or @Context suffix creates a functional-element individual or fills a claim. | Add the claim record, selected structure, view episteme, and any needed claim references; otherwise keep the phrase as ordinary recognition wording. |
| CC-A6F-4 Function and capability split. | Capability claims and function or effect claims remain distinct. | State the qualified holder-ability claim under A.2.2, using the subject pattern only as a locator, and keep the required behavior or effect claim with its requirement, functional view, method, or other claim-bearing source. |
| CC-A6F-4A Required-effect and actual-change split. | A required or desired behavior/effect remains claim content; every U.Transformation reference has an independent A.3.4 occurrence basis. | If only requirement, architecture, method, desired-effect, diagram, or assertion material is available, stop before U.Transformation. If an actual change is current, identify its changed referent, boundary, conditions, actual facts, and continuity or reidentification rule separately. |
| CC-A6F-5 Function, work, and method-description split. | Method, U.MethodDescription membership, work occurrence, and work result claims do not hide inside function wording; source form does not establish membership. | For a reusable way-of-doing claim, recover the U.Method under A.3.1. For U.MethodDescription, first identify the C.2.1 episteme, its admitted U.Method as EntityOfConcern, and at least one substantive way-of-doing claim under A.3.2. Otherwise use the direct plan, work, result, representation, publication, or other pattern. |
| CC-A6F-6 Function and neighboring relations split. | Local system-role kind, System-classification judgment, assignment species, assignment occurrence, participation, actual functioning, responsibility, Work, and capability remain separate. VP.AllocationResponsibility is only a recognition cue. | Name each current fact independently. A positive assignment claim needs its directly declared species and obtaining occurrence under A.2.1; actual Work points to its basis: A.13 first, independent A.15.1 Work admission second, and F.6 afterward only for precise assignment-bound attribution. For participation, functioning or responsibility, distinguish a missing governing rule from unknown or contrary case facts; use A.6.RCD for the missing rule. |
| CC-A6F-7 Mathematical function boundary. | A mathematical function keeps its definition and applicable rule. A separate C.29 lens account is needed only when a representation, transfer or inference question remains. | Resolve that question with the necessary correspondence, preserved and lost structure, admissible use and stop condition. Keep an already adequate mathematical use. |
| CC-A6F-8 Quality and functionality boundary. | Quality, fitness, characteristic, score, or “functionality” wording recovers bearer and subject pattern. | Assign the claim to C.25, C.16, C.16.Q, A.17, A.18, or the characteristic named by value or measurement subject pattern according to the asserted quality, characteristic, measurement, or comparison claim. |
| CC-A6F-9 Module-interface boundary. | Functional relation, module allocation, interface, signature, port, API, protocol, flow, and mechanism wording remain separated. | Add A.6.RSIR interface-cue recovery, FunctionFlowModuleAlignmentNote, the A.6.M module-interface boundary, the A.6.0 and A.6.5 signature discipline, declared correspondence, declared allocation, or A.6.M module-relation repair. |
| CC-A6F-10 Useful action. | The repair leaves a remaining admissible use: apply the subject pattern to the object or claim; open a functional view; add an alignment note; add the admitted direct-relation predicate, relational assertion, separately individuated obtaining occurrence, reusable declaration, selected claim-bearing episteme, or representation correspondence only for a named use; assign the claim to C.29, C.30, C.30.ASV, A.15, C.25, C.16, A.10, B.3, A.20, A.21, or C.11; or stop. | Restore that use, or classify the phrase as reduced-use cue, quote-only wording, blocked transfer, or incomplete rewrite. |
A.6.F:8 - Common Anti-Patterns and How to Avoid Them
| Anti-pattern | Symptom | Repair |
|---|---|---|
| Root function kind | The text treats function as a new universal FPF kind. | Name the object or claim and apply its subject pattern; use the optional FunctionUseRepair when the receiving use needs that record. |
| Functional architecture exception | Functional architecture is treated as a peer architecture ontology. | Expand to FunctionalStructure under ArchitectureOf@Context and C.30.ASV. |
| Capability collapse | What the System can do is treated as a functional dependency or vice versa. | Split capability claim from functional relation or effect claim. |
| Work collapse | Work occurrence or result is conflated with the function it serves. | Establish deliberately asserted Work through A.13 and independent A.15.1 admission, using F.6 only for precise assignment-bound attribution; relate any result separately. Keep ordinary functional wording when the intended claim is clear. |
| Algorithm-form shortcut | Procedure, code, solver, recipe, protocol, or algorithm form is treated as proof of U.MethodDescription membership. | Recover the claim-bearing C.2.1 episteme, one admitted U.Method as its EntityOfConcern, and at least one substantive way-of-doing claim under A.3.2; otherwise keep the source form with its subject pattern. |
| Mathematical-function import | A mathematical function, loss, objective, or value functional becomes design ontology. | Use C.29 and state preserved and lost structure plus stop condition. |
| Module allocation shortcut | A function is considered implemented because a module is named. | Add correspondence, allocation, module-interface boundary, signature-discipline boundary, or A.6.M module-relation repair. |
| Functionality as quality proxy | “Functionality” carries adequacy or quality claim without bearer and subject pattern. | Recover bearer and subject pattern through C.25, C.16, C.16.Q, or an admitted characteristic or measurement subject pattern. |
| Sterile kind repair | The wording is typed but no useful move remains. | Restore the direct action on the object or claim: use the pattern that defines or tests it, open the selected view, add the needed alignment or correspondence, or stop the stronger reading. |
A.6.F:9 - Consequences
| Benefit | Cost or trade-off |
|---|---|
Function-like prose remains usable without minting U.Function. | The object or claim and applicable rule must be recoverable; additional relation, declaration, assertion, specification, view, or representation objects appear only for a named use. |
| Functional architecture becomes a normal architecture-by-structure-kind case. | C.30 or C.30.ASV may be needed when the phrase carries an architecture claim. |
| Capability, method, Work, system-role kind, assignment, participation, functioning, responsibility, mathematical, quality, module, and interface claims stay separable. | One familiar word may require several independently established objects or claims when several readings are current. |
C.29, C.25, C.16, A.15, C.30, and A.6.M govern only the claims that belong to them. | A conforming use stops once the object or claim, subject pattern, and remaining action are clear instead of opening all possible subject patterns. |
A.6.F:10 - Rationale
Preserve useful function-like wording and clarify unresolved meanings. The smallest useful repair is not a new ontology or a generic record. Name the entity, value, claim, or claim-bearing episteme, apply its subject pattern, and state the remaining use. Add an explanatory rejected reading only under the full F.19:4 test.
This follows A.6.P: recover the direct predicate and participants when a relation question is current; add a reusable RelationSignature and declaration-local SlotSpecs only when typed use needs them; distinguish assertion, specification and view epistemes; and interpret a representation under its applicable correspondence. Use C.29 only for a remaining mathematical representation, transfer or inference question. Under C.30, functional architecture concerns selected structure for a described holon. A selected transformation-flow structure or mathematical graph description requires its own basis.
The pattern keeps ordinary language usable, including an FPF claim whose object and governing rule are already clear. Repair ambiguity when it changes an ontological, evidence, causal, assurance, bridge, gate, work, decision or admissibility claim. A generic record does not supply the missing meaning.
A.6.F:11 - SoTA-Echoing
| Practice or source line | A.6.F adoption | Action consequence | Boundary |
|---|---|---|---|
| ISO/IEC/IEEE 42010:2022 architecture-description discipline | Adapt view and concern discipline to functional architecture as a structure-kind view over an architecture claim. | Functional architecture expands through C.30 and C.30.ASV rather than becoming a separate ontology. | ISO terminology does not mint U.Function or turn diagrams into architecture. |
| INCOSE systems-engineering and MBSE functional-analysis practice | Adopt the practical need to separate function, requirement, behavior, physical allocation, and verification claims. | A function-like phrase can guide architecture work only after capability or effect, allocation, evidence, and verification claims are separated. | Functional analysis practice is not evidence sufficiency, assurance, gate passage, or project decision by itself. |
| ISO/IEC 25010 quality-model practice | Treat functionality or functional suitability as quality wording when it evaluates a product or service. | Assign quality-like uses to C.25, C.16, or C.16.Q before they carry adequacy claims. | “Functionality” is not a free adequacy score and not an ArchitectureOf@Context claim or selected functional view. |
| C.29 mathematical-lens discipline | Use domain, codomain or relation domain, preserved and lost structure, admissible use and stop condition when a mathematical lens account is needed. | Resolve the remaining representation, transfer or inference question; retain an adequate mathematical use without adding a second account. | Mathematical functions, objectives and value functionals do not establish holon purpose, evidence, causal proof, assurance or a decision by themselves. |
| GonzoML neural-network architecture discussions | Adapt practitioner operation language involving blocks, activations, path-selection, memory, cache, loss functions, pruning, ablation, and architecture search as recognition material. | Function-like neural-network wording must resolve to the mathematical object, flow relation, module-interface claim, capability or effect, quality characteristic, decision, or evidence claim and its subject pattern. | Neural-network labels and benchmark results do not become FPF ontology, architecture decision, evidence sufficiency, gate passage, or assurance by themselves. |
A.6.F:12 - Relations
Builds on: A.6.P, A.6.RSIR, A.6.0, A.6.5, A.6.B, A.6.C, A.6.9, A.7, E.10, E.10.ARCH, C.2.P, F.18, and E.8.
Coordinates with: C.2.1 when the task must preserve an assertion about the direct predicate; A.6.REL only when the task needs explicit identity for an obtaining occurrence; C.30, C.30.ASV, C.30.TFS-REL, E.18, A.15, A.2, C.29, C.25, C.16, C.16.Q, A.17, A.18, A.10, G.6, B.3, A.20, A.21, C.11, A.6.RSIR, and A.6.M when a module or interface claim is being made.
Does not replace: C.30 grounded architecture and selected-structure adequacy, C.30.ASV architecture structural-view adequacy, E.18 selected transformation-flow structure, E.18.2 mathematical descriptions, C.29 mathematical-lens use, C.25 Q-Bundles, C.16 characterization, A.15 work and method discipline, A.10 or G.6 evidence, B.3 assurance, A.20 or A.21 gate or release records, or C.11 decisions.