Γ_method — Order-Sensitive Method Composition & Work Enactment

Pattern B.1.5 · Stable Part B - Trans-disciplinary Reasoning Cluster

► decided‑by: A.14 Advanced Mereology A.14 compliance — Methods compose over SerialStepOf/ParallelFactorOf on MethodDescription/Method graphs (order, not parthood); stuff‑like inputs are modelled via PortionOf on resources and accounted in Γ_work; method/version history uses PhaseOf; mapping quality is handled via CL (B.3).

Plain‑English headline. Γ_method composes ordered step specifications into a single MethodDescription (design‑time) that describes a composite Method, and governs its run‑time enactment as Work (pre/post, capability typing, MIC honouring) while delegating resource accounting to Γ_work and order semantics to Γ_ctx.

  • Strict Distinction (A.15) separates what a holon is (structure), how steps are ordered (order), how it unfolds (time), what it spends (work/resources), and what it values (objectives).
  • Method / MethodDescription / Work.

Keywords

  • method composition
  • workflow
  • sequential
  • concurrent
  • plan vs run.

Relations

B.1.5outline next siblingΓ_work — Work as Spent Resource
B.1.5explicit referenceΓ_work — Work as Spent Resource
B.1.5explicit referenceU.Work: The Record of Occurrence
B.1.5explicit referenceRole Taxonomy
B.1.5explicit referenceTransformer Constitution (Quartet)
B.1.5explicit referenceDependency Graph & Proofs

Content

Problem frame

  • Strict Distinction (A.15) separates what a holon is (structure), how steps are ordered (order), how it unfolds (time), what it spends (work/resources), and what it values (objectives).

  • Method / MethodDescription / Work.

    • Method is the timeless semantic “way of doing” (a context‑scoped capability; A.3.1): it specifies admissible preconditions, effects, and bounds, independent of any particular run.
    • MethodDescription is a design‑time description of a Method (knowledge on a carrier). It may be an imperative step‑graph (this pattern’s focus) or another admissible description form (functional/logical/dynamics/solver, etc.; A.3.2:4.2).
    • Work is the dated run‑time occurrence that enacts a pinned MethodDescription under a U.RoleAssignment, records concrete slot fillings (parameters/carriers), and books the resource ledger (A.15.1). Calling the description a “process” is common in some domains, but in FPF we keep Method ≠ MethodDescription ≠ Work to avoid category errors.
  • A.15 (Role–Method–Work Alignment) supplies the typed ordered relations we need: SerialStepOf (strict precedence) and ParallelFactorOf (order‑concurrent branches with a join).

  • B.1.4 (Γ_ctx/Γ_time) already handles non‑commutativity (order matters) and temporal slicing; B.1.6 (Γ_work) handles resource spending and efficiency. Γ_method sits between them: it composes methods by order and capability and delegates resource accounting to Γ_work.

Problem

Without a dedicated, order‑aware method operator:

  1. Design/run conflation. Authors mix MethodDescription (blueprint) and Work (execution), producing artifacts that have both planned and executed attributes.
  2. Order erasure. Sequences with crucial pre/post‑conditions get collapsed into sets; reordering breaks correctness while still “passing” naive aggregation.
  3. Capability mismatches. Step outputs do not match the next step’s required inputs, but this is hidden in untyped edges; composite methods become non‑executable.
  4. Work leakage. Costs and resource flows are inlined into method definitions; later models double‑count or violate conservation (Γ_work was created to prevent this).
  5. Synergy by arithmetic. Throughput or quality jumps caused by proper joins or coordination are misreported as simple sums or averages—violating WLNK and obscuring when a Meta‑Holon Transition (B.2) should be declared.

Forces

ForceTension
Order fidelity vs. simplicityKeep the true sequence (non‑commutative) ↔ Provide a small operator set.
Type safety vs. flexibilityEnforce capability typing and pre/post checks ↔ Allow modular reuse of steps across contexts.
Design vs. runCompose MethodDescription for planning ↔ Produce Work for execution without mixing them.
Parallelism vs. correctnessMaximise concurrency on independent branches ↔ Guarantee sound joins and reproducible outcomes.
Parsimony vs. separation of concernsKeep Γ small ↔ Keep work and assurance in their own lanes (Γ_work, B.3).

Solution

Terms (didactic recap)

  • U.MethodDescription — a design‑time description of a U.Method (A.3.2): typically an imperative step‑graph with SerialStepOf/ParallelFactorOf, step capability types, pre/post‑conditions, and required external interactions. (Other admissible description forms exist; B.1.5 focuses on the step‑graph case.)
  • U.Method — the timeless semantic “way of doing” (capability) described by ≥1 MethodDescription and enacted as U.Work (A.3.1, A.15.1).
  • U.Work — the run‑time, dated enactment occurrence: performedBy → U.RoleAssignment, isExecutionOf → U.MethodDescription (edition‑pinned), plus concrete slot fillings and resource ledger (A.15.1).
  • U.StepSpec / U.StepMethod — step‑level specialisations: each StepSpec describes a StepMethod; a composite MethodDescription relates them by order. (Run‑time step occurrences are Work parts, not “StepMethods”.)
  • Capability type — the state/action signature a step requires and produces (not to be confused with resources; those belong to Γ_work).
  • Method Interface Standard (MIC) — the order‑aware analogue of BIC: a short, declarative statement of what external interactions of the steps are Promoted / Forwarded / Encapsulated at the composite method boundary.

Separation reminder. Method composition ≠ resource spending. Keep resource budgets, yields, dissipation in Γ_work; Γ_method only checks and composes order and capability.

The operator family (two companion flavours)

To respect the design/run split, Γ_method is presented as two companion operators sharing the same intent but acting at different loci (spec vs run).

  1. Planning (design‑time) — compose specifications

    Γ_method^plan : ( D_spec : OrderedDependencyGraph< U.StepSpec >,
                      σ       : OrderSpec,
                      MIC_in  : optional boundary hints )
                    → U.MethodDescription
    • Domain. D_spec contains step specifications linked by SerialStepOf / ParallelFactorOf (A.15).
    • Result. A single U.MethodDescription whose MIC is computed from step interfaces using the Promote / Forward / Encapsulate quartet (cf. BIC in B.1.2). The resulting MethodDescription SHALL declare the U.Method it describes (A.3.2); in the step‑graph case this is the semantic serial/parallel composition of the described StepMethods (A.3.1:9).
  2. Enactment (run‑time) — produce Work

    Γ_method^run  : ( M_spec : U.MethodDescription,
                      RA     : U.RoleAssignment,
                      Fill   : carrier & parameter slot fillings )
                    → U.Work
    • Domain. A previously composed MethodDescription, a performer designated via RoleAssignment (the holder bears the required role in context), and concrete slot fillings (carriers, parameters) consistent with the MethodDescription’s declared SlotKinds/ValueKinds (A.6.5).
    • Result. A U.Work record (the dated run) provided that capability checks and pre/post‑conditions hold and the MIC is honoured.

Relationship to Γ_ctx. Both flavours reuse Γ_ctx invariants for order (non‑commutative composition with NC‑1..3 reproducibility). Γ_method specialises the typing and boundary rules for methods and introduces MIC.

Core aggregation rules (design‑time composition)

When computing Γ_method^plan(D_spec, σ):

  1. Order preservation. Respect the OrderSpec σ; independent branches may be folded in any topological sort (Γ_ctx NC‑3). SerialStepOf enforces strict precedence; ParallelFactorOf allows concurrency with a join.

  2. Capability continuity (typed joins). Every join must be type‑sound: the post‑condition / output signature of each incoming branch must meet the next step’s pre‑conditions (logical entailment or declared adapter steps). Missing adapters are defects, not assumptions.

  3. MIC synthesis (boundary behaviour). For each external interaction of a step, decide Promote / Forward / Encapsulate into the composite MIC. This inherits the clarity of BIC (B.1.2) for methods.

    • Promote: becomes a direct composite interaction (e.g., top‑level “start/stop”).
    • Forward: remains step‑local but exposed under the composite boundary (namespaced).
    • Encapsulate: becomes internal; callers cannot rely on it.
  4. Assurance hooks (without computing assurance). Record where B.3 assurance will later hang: (i) the cutset steps that bound reliability/quality, (ii) the integration edges whose CL will penalise poor fit (mappings, fragile joins), and (iii) the envelope (G) intended for the method’s validity.

  5. No costs here. If a step lists resources/yields, do not aggregate them here. Instead, add a pointer to the corresponding Γ_work composition to be executed with the same order/joins at run‑time.

Core aggregation rules (run‑time enactment)

When executing Γ_method^run(M_spec, RA, Fill):

  1. Role–Method–Spec alignment (A.2 / A.3 / A.15). Confirm that RA.role is eligible to enact the U.Method described by M_spec (or a declared equivalent/refinement in the same context), and that the Work’s performedBy and executedWithin anchors can be satisfied (A.15.1). If this fails, you may still record an attempted run, but it is not a conformant “execution of M_spec”.

  2. Pre/post enforcement. Before each step, verify pre‑conditions against Fill and the evolving carrier state; after, check post‑conditions hold. Failing these means the run cannot be certified as a conformant U.Work execution of M_spec.

  3. Typed state flow. The state/action types produced by a step must make the next step well‑typed; if not, an adapter method (itself with a MethodDescription) must be present in the graph.

  4. Order determinism (Γ_ctx). Respect the OrderSpec σ declared in M_spec. Parallel branches may execute independently only if they share no state that would break NC‑1..3; otherwise they must synchronise at the declared join.

  5. MIC honouring. Interactions exposed by MIC are the only external commitments the composite method makes. Any additional ad‑hoc external interaction is a model violation (or requires updating the MIC and re‑planning).

  6. Γ_work hand‑off. Invoke Γ_work to compute spent resources, yields, dissipation along the same order/join structure. The resulting ledgers and work products annotate the Work but are not part of Γ_method’s aggregation.

Invariant intuition.

  • IDEM: a single step‑method composed alone yields the same method.
  • COMM/LOC: replaced by Γ_ctx NC‑1..3 (determinism given σ, context hash of σ, and partial‑order soundness).
  • WLNK: quality/throughput of the composite is bounded by the critical path steps (identified for later B.3 assurance).
  • MONO: strengthening a step (better pre/post, stronger type, improved adapter) cannot make the composite worse.

Didactic contrasts (to prevent common confusions)

  • Method vs Work. Method = the semantic “way of doing” (what transformations are admissible); Work = what happened this time, including resources spent / yields / dissipation when enacting it (Γ_work). Keep them distinct.

  • Method vs Structure. Method composes ordered steps; structure composes parts (Γ_sys). Do not use ComponentOf where SerialStepOf/ParallelFactorOf are intended.

  • Step vs part vs specialization. A “step” in SerialStepOf/ParallelFactorOf is a factor in an order algebra, not a mereological part and not a type‑specialisation. – Use ComponentOf/PartOf for structural wholes (A.14). – Use ≤ₘ refinement / equivalence / substitution for Method specialisation (A.3.1). – Use Kind‑CAL () for kind/subkind.

  • Method vs Phase. Method composition is order; PhaseOf (Γ_time) is temporal progression of the same carrier. If a phase boundary also introduces closure/supervision/context rebase, that is MHT (B.2), not mere phasing.

  • MethodDescription vs Work. Keep planning artefacts (MethodDescription) separate from run‑time occurrences (Work). Γ_method^plan produces MethodDescriptions; Γ_method^run produces Work that cites an edition‑pinned MethodDescription and records effective slot fillings and ledgers (A.15.1).

Archetypal grounding (worked, didactic)

System archetype — Assemble‑Paint‑Test as one Method

  • Design‑time (Γ_method^plan). D_spec contains StepSpecs: AssembleChassis, InstallPowertrain, PaintBody, RunFunctionalTest. Relations: AssembleChassis → InstallPowertrain (SerialStepOf), PaintBody ∥ RunFunctionalTest after a structural seal (ParallelFactorOf). Capability typing:

    • Output of InstallPowertrain meets input of RunFunctionalTest (functional harness attached).
    • PaintBody requires sealed surfaces from InstallPowertrain (pre‑condition). MIC outcome:
    • Promote: Start(), Abort(), CertificationReport.
    • Forward: RunFunctionalTest.Diagnostics (namespaced).
    • Encapsulate: PrimerMixingPort, internal seal checks.
  • Run‑time (Γ_method^run). The holder designated by the relevant U.RoleAssignment enacts the MethodDescription on concrete carriers, producing a U.Work record. Pre/post checks gate each step; parallel branches run after pre‑conditions met; a join waits for both to finish.

  • Assurance hooks (B.3). Cutset steps for WLNK: InstallPowertrain (torque tolerances) and RunFunctionalTest pass/fail; integration edges carry CL for harness mapping and paint/seal specification. Γ_work is invoked to compute energy/material spend and dissipation; Γ_method does not tally costs itself.

Episteme archetype — Evidence‑Synthesis‑Publish as one Method

  • Design‑time (Γ_method^plan). Steps: CollectDatasets, NormalizeSchemas, EstimateModel, CrossValidate, DraftManuscript. Ordering: CollectDatasets → NormalizeSchemas → EstimateModel → CrossValidate → DraftManuscript. Capability typing: NormalizeSchemas outputs a typed feature space that entails EstimateModel’s input; adapters specified for legacy datasets. MIC outcome:

    • Promote: Submit(), ReleaseArtifacts().
    • Forward: CrossValidate.Folds(k).
    • Encapsulate: ad‑hoc scrubbing utilities.
  • Run‑time (Γ_method^run). The same order executes as U.Work; Γ_work accounts for compute/storage spend. Assurance hooks: cutset at CrossValidate; integration CL for schema mappings; post‑condition for DraftManuscript includes provenance SCR.

Method Interface Standard (MIC) — template & examples

MIC template (normative content)

Method Interface Standard (MIC)
  name:                human-readable identifier
  version:             semantic label of this MIC
  orderSpecHash:       hash(OrderSpec σ + step signatures)
  externalInteractions:
    - id:              external op name
      mode:            {Promote | Forward | Encapsulate}
      signature:       state/action types (typed interface)
      preconditions:   predicates that must hold at call
      postconditions:  predicates guaranteed on return
      qosEnvelope:     optional envelope (throughput, latency, quality)
  invariants:
    - textual/logical invariants preserved by the method
  notes:
    - rationale for Promote/Forward/Encapsulate choices

MIC excerpts (didactic)

  • Manufacturing method MIC excerpt

    externalInteractions:
      - id: Start
        mode: Promote
        signature: Start(): Promise<BatchId>
        preconditions: LineReady & MaterialsAvailable
        postconditions: BatchId issued
      - id: PrimerMixingPort
        mode: Encapsulate
    invariants:
      - FunctionalTest.Pass implies TorqueTolerance ≤ δ
  • Evidence method MIC excerpt

    externalInteractions:
      - id: Submit
        mode: Promote
        signature: Submit(): Promise<SubmissionId>
        preconditions: ManuscriptReady & SCRComplete
        postconditions: DOI assigned on accept
      - id: CrossValidate.Folds
        mode: Forward
        signature: Folds(k: Int): Report
    invariants:
      - Report.metrics computed on held-out data only

Proof obligations (normative)

At planning time (Γ_method^plan):

  1. PO‑PLAN‑ORDER. Provide OrderSpec σ; produce orderSpecHash.
  2. PO‑PLAN‑TYPE. For every edge, show capability continuity: OutType(step_i) ⊢ InType(step_j) or provide a typed adapter StepSpec.
  3. PO‑PLAN‑MIC. For each step interaction, decide Promote/Forward/Encapsulate and justify in MIC.
  4. PO‑PLAN‑CL‑POINTS. Identify integration edges whose CL will matter for B.3; record intended sources of mapping evidence.
  5. PO‑PLAN‑NO‑WORK. Confirm that costs/resources are not aggregated here; point to the planned Γ_work composition (by reference).

At run time (Γ_method^run) producing U.Work:

  1. PO‑RUN‑PRE/POST. Demonstrate that pre‑conditions hold before each step; check post‑conditions after.
  2. PO‑RUN‑NC. Show compliance with Γ_ctx NC‑1..3 (determinism with σ, context hash, partial‑order soundness).
  3. PO‑RUN‑MIC‑HONOUR. Record that only MIC‑declared external interactions occurred.
  4. PO‑RUN‑WORK. Attach the Γ_work result (spent resources, yields, dissipation) aligned with the same order/join structure.
  5. PO‑RUN‑ASSURANCE. Provide the observed values for the cutset steps and the actual CL of integration mappings to feed B.3 assurance.

Conformance Checklist (normative)

IDRequirementPurpose
CC‑B1.5.1Γ_method SHALL be used in two flavours only: Γ_method^plan for specifications, Γ_method^run for Work enactments.Enforce design/run separation.
CC‑B1.5.2Planning inputs SHALL use SerialStepOf / ParallelFactorOf edges with a declared OrderSpec σ.Preserve order semantics.
CC‑B1.5.3All joins SHALL be type‑sound (capability continuity) or include explicit typed adapters.Prevent non‑executable composites.
CC‑B1.5.4A MIC SHALL be produced for Γ_method^plan and SHALL be honoured by Γ_method^run.Make external commitments explicit.
CC‑B1.5.5Resource spending/yields SHALL be computed via Γ_work and MUST NOT be inlined into Γ_method aggregation.Maintain separation of concerns.
CC‑B1.5.6Γ_ctx NC‑1..3 invariants SHALL hold for both flavours (determinism under σ, hash, partial‑order soundness).Guard non‑commutative correctness.
CC‑B1.5.7If joining branches produces apparent super‑additivity beyond WLNK not explainable within Γ_method, an MHT SHALL be considered and recorded per B.2.Prevent “synergy by arithmetic”.

Anti‑patterns & repairs

Anti‑patternSymptomRepair
Flattened set of stepsOrder lost; results become nondeterministicUse Γ_ctx to restore σ, then apply Γ_method^plan.
Cost‑in‑methodResources embedded in method definitionRemove costs; move to Work/Γ_work.
Design/Run ChimeraSpec contains runtime measures; enactment adds planning edgesSplit into MethodDescription (design) vs Work (run); rerun Γ_method per flavour.
Design/Run ChimeraSpec contains runtime measures; enactment adds planning edgesSplit into MethodDescription vs Method; rerun Γ_method per flavour.
Orderless SetSteps modelled as unordered; reordering breaks correctnessProvide OrderSpec σ and recompose with Γ_method/Γ_ctx.
Silent AdapterA join assumes implicit conversionAdd explicit typed adapter StepSpec/Method and re‑prove capability continuity.
Inline CostsMethod sums time/energyMove to Γ_work; link the work composition to the same order.
Boundary FogExternal calls occur ad‑hocDefine/Update MIC; Promote/Forward/Encapsulate explicitly.
Emergence by JoinThroughput leaps past WLNK with no storyEither (i) prove within Γ_method (cutset/CL/order) or (ii) declare MHT (B.2).

Consequences

Benefits

  • Didactic clarity. Readers see what is being composed (order & capability) vs what is spent (Γ_work) vs what is assured (B.3).
  • Deterministic execution semantics. Γ_ctx‑backed order with explicit joins yields reproducible composites.
  • Robust interfaces. MIC prevents accidental external dependencies and preserves modularity.
  • Cross‑scale fit. Same pattern works for physical, organizational, and epistemic methods.

Trade‑offs

  • More explicitness up‑front. Capability typing and MIC authorship require care; in return, later integration is safer.
  • Adapter discipline. Modellers must create adapters rather than assuming conversions—this avoids hidden brittleness.

Rationale (informative)

  • Order is semantic. Many failures stem from pretending that order does not matter; Γ_method makes non‑commutativity explicit (via Γ_ctx) while keeping the operator set small.
  • Strict Distinction. The split between Method (semantic), MethodDescription (spec), Work (occurrence), Γ_method (order/type checks), Γ_work (resource ledgers), and assurance implements A.15, preventing category errors (semantics vs execution vs claims).
  • Mereology alignment. Using SerialStepOf / ParallelFactorOf (A.14) keeps method composition orthogonal to structural composition (ComponentOf) and temporal phasing (PhaseOf).
  • Assurance readiness. Identifying cutsets and mapping CL points during planning makes B.3 application straightforward and auditable.
  • Interfaces matter. MIC prevents accidental coupling and makes integration points auditable.
  • Separation of concerns. Γ_method composes behaviour; Γ_work accounts resources; B.3 assesses quality—keeping algebraic reasoning sound.

Relations

  • Builds on: A.12 (Transformer Role), A.14 (Mereology Extension), A.15 (Strict Distinction); B.1.1 (Proof Kit), B.1.4 (Γ_ctx/Γ_time).
  • Coordinates with: B.1.6 (Γ_work) for resource accounting; B.3 (Assurance) for WLNK cutsets and CL penalties.
  • Triggers/Complements: B.2 (MHT) when new closure/supervision or context re‑base appears at method level.
  • Used by: Later domain patterns that define canonical methods in specific disciplines (without altering Γ_method).

One‑sentence takeaway. Γ_method composes ordered, typed steps into a reliable method, keeps interfaces explicit (MIC), leaves costs to Γ_work, and provides clean hooks for assurance and emergence.

B.1.5:End