#KD‑CAL

C.2 · Cluster C.I - Core CALs / LOGs / CHRs · Stable · 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-memory 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.

Scope & exports. A substrate‑neutral calculus for composing epistemic holons (U.Episteme) and reasoning about their motion and equivalence. Exports: (i) three point‑characteristics—Formality F, ClaimScope G, Reliability R—that locate a single episteme; (ii) a pairwise ladder of Congruence Levels (CL 0…3); (iii) four Δ‑moves (Formalise, Generalise/Specialise, Calibrate/Validate, Congrue); (iv) composition rules (Γ_epist) for aggregates; (v) propagation laws for CL through mappings and notation bridges. KD‑CAL is typed by U.EpistemeSlotGraph and never confuses ClaimGraph, DescribedEntitySlot, GroundingHolonSlot, Viewpoint, View, ReferenceScheme, notation, publication form, or carrier. All F–G–R computations are context‑local; Cross‑context traversals require an explicit Bridge with CL and apply the B.3 congruence penalty Φ(CL) to R. // Contexts ≡ U.BoundedContext; substitution is plane‑preserving only.

Formality F is the rigor characteristic defined normatively in C.2.3. All KD‑CAL computations and guards SHALL use U.Formality (F0…F9) as specified there; no parallel “mode” ladders are allowed.

FPF fixes two archetypal sub‑holons: U.System (physical/operational) and U.Episteme (knowledge holon). KD‑CAL is the primary composition pattern for U.Episteme, giving engineers a compact, testable way to say (a) how strictly an episteme is written (F), (b) how much structure it manages (G), (c) how well it is warranted by evidence or severe tests (R), and (d) how closely two epistemes coincide (CL). KD‑CAL is built atop C.2.1 U.Episteme — Epistemes and their slot graph, which reifies every episteme through U.EpistemeSlotGraph: ClaimGraph, DescribedEntitySlot, GroundingHolonSlot, Viewpoint, View, and ReferenceScheme. Notation, publication forms, carriers, and work occurrences remain outside episteme content and are linked by their own FPF relations.

#Keywords

• knowledge
• epistemic
• evidence
• trust
• assurance
• F-G-R
• Formality
• ClaimScope
• Reliability
• provenance.

#Relations

#Content

#Problem Frame

FPF fixes two archetypal sub‑holons: U.System (physical/operational) and U.Episteme (knowledge holon). KD‑CAL is the primary composition pattern for U.Episteme, giving engineers a compact, testable way to say (a) how strictly an episteme is written (F), (b) how much structure it manages (G), (c) how well it is warranted by evidence or severe tests (R), and (d) how closely two epistemes coincide (CL). KD‑CAL is built atop C.2.1 U.Episteme — Epistemes and their slot graph, which reifies every episteme through U.EpistemeSlotGraph: ClaimGraph, DescribedEntitySlot, GroundingHolonSlot, Viewpoint, View, and ReferenceScheme. Notation, publication forms, carriers, and work occurrences remain outside episteme content and are linked by their own FPF relations.

#Problem

Teams routinely entangle programs, specifications, proofs, and datasets; a “proof” is treated as a tested routine, a “program” is cited as if it entailed a theorem. Trust decays because justification and evidence freshness are not explicit. Epistemes are anthropomorphised as actors (“the standard enforces…”), producing category errors at execution. Without a shared composition and equivalence calculus, aggregates hide weakest links and analogies harden into overclaims. KD‑CAL must stop these failure modes with a single constitution and scale‑set.

#Forces

• Universality vs domain idioms. One calculus must cover physics theories, legal codes, safety specs, algorithms, and formal proofs without flattening their differences.
• Meaning vs materiality. Meaning must be independent of carrier, yet accountable to it historically.
• Deductive vs empirical. Axiomatic certainty and empirical trust have different evidence-continuity profiles; both must compose.
• Abstraction vs enactment. Epistemes constrain action; systems act. The calculus must keep the roles distinct.

#Solution

#Coordinates and the episteme slot graph

KD‑CAL characteristics (single‑episteme, point‑values).

• Formality F. From free prose to machine‑checkable proof/specification. Litmus: would a machine reject it if wrong?
• Claim scope (G), a set‑valued applicability over U.ContextSlice, with ∩/SpanUnion/translate algebra; CL penalties apply to R, not to F/G. Litmus: how wide is the declared scope, and under what minimal assumptions does the claim hold?
• Reliability R. From untested idea to continuously validated claim. Litmus: where is the last successful severe test? R‑claims MUST bind to evidence and declare relevance windows; stale bindings degrade R or require waiver per ESG policy.

Congruence Level (CL), pairwise ladder. CL‑0 Opposed/Disjoint (contrastive; no substitution); CL‑1 Comparable / Naming‑only (label similarity; no substitution); CL‑2 Translatable / RoleAssignment‑eligible (structure‑preserving mapping in a declared fragment with stated loss; theorems may transport); CL‑3 Near‑identity / Type‑structure‑safe (invariants match; type‑structure substitution allowed). CL is a characteristic of a relation between two epistemes; it is not a fourth member of the F–G–R assurance tuple and it is not a characteristic space of its own. Norm: substitution is permitted only if plane‑preserving and CL ≥ 2; substituting type‑structure requires CL = 3.

Slot-graph link. The assurance components are stated over U.EpistemeSlotGraph: F by the internal ClaimGraph and formal substrate, G by the ClaimScope attached to the described entity and assumptions, and R by evaluation templates and evidence bindings. Notation belongs to representation and reference-scheme structure; carriers remain outside the episteme and link through SCR/RSCR or other exact carrier relations. Multiple notations are allowed only when their relation is explicit; authors SHOULD register NotationBridge(n₁,n₂) with an associated CL to make conversion loss explicit.

#Four Δ‑moves (epistemic motion)

• ΔF — Formalise. Rewrite for stricter calculi/grammars; raise proof obligations.
• ΔG — Generalise / Specialise. Widen or narrow the claim scope (assumptions & scope). Changes to decomposition granularity are an orthogonal view and do not change G unless they alter the envelope.
• ΔR — Calibrate / Validate. Strengthen severe tests or add live monitoring; update evidence bindings.
• ΔCL — Congrue. Establish and record the sameness relation between two epistemes (ladder 0→3). Moves compose into paths; CL along a path is the minimum of its links.

#Composition (Γ_epist) and propagation

Let Γ_epist combine epistemes {Eᵢ} into a composite episteme Γ that makes a joint claim (AND‑style) or exposes an interface (series composition). KD‑CAL imposes safe defaults:

• R (Reliability). Along any justification path P, compute R_eff(P) = max(0, min_i R_i − Φ(CL_min(P))) (weakest‑link with congruence penalty). For series composition (claims needed conjunctively), the path‑wise weakest‑link applies; for parallel support (independent lines to the same claim), use R(Γ) = max_P R_eff(P) (annotate independence); never exceed the best attested line. Cross‑context steps and NotationBridge traversals contribute to CL_min(P).

• F (Formality). F(Γ) = minᵢ F(Eᵢ) (monotone non‑increasing along used paths). To raise F, apply ΔF to the weakest parts.

• G (ClaimScope). On any dependency path, take the intersection of claim scopes (the narrowest overlapping scope). Across independent support paths to the same claim, set G(Γ) = SpanUnion({G_path}) constrained by support (drop unsupported regions). Widening/narrowing the scope is an explicit ΔG± operation.

• CL (Congruence). For a chain of mappings E₀ ~ E₁ ~ … ~ Eₖ, the path congruence is min CL(Eⱼ,Eⱼ₊₁). Passing through a NotationBridge sets CL to the bridge’s declared level; the Φ(CL) penalty is applied in the R fold for any path that traverses it.

These rules keep Γ aligned with the holonic kernel: Γ is only defined on holons and respects identity/boundary discipline from the core.

#What must not be conflated (normative guards)

• Representation structure ≠ carrier. Files, PDFs, or repositories are carriers outside the episteme; they never count as parts of U.Episteme (see C.2.1 EP‑1; CC‑EPI‑2/3).
• Epistemes do not act. Only systems perform work; epistemes carry constraints and evaluation criteria through their ClaimGraph, described entity, grounding holon, scope, and evidence bindings (per Core A.15 / CC‑EPI‑3).
• CL is not a score. It is a qualitative ladder of preservation classes; do not average it.

#✱ Archetypal Grounding (Tell–Show–Show)

Universal rule (tell). Compose knowledge by Γ_epist with weakest‑link R, monotone F, and explicit CL on every bridge; keep ClaimGraph, described entity, grounding holon, viewpoint, view, reference scheme, notation, publication form, and carrier in their exact FPF relations.

System (show, Sys‑CAL lens). Consider a battery‑pack thermal subsystem integrating a physics model of heat flow and an operating envelope for fast‑charge. As a system, it composes pumps, sensors, and controllers by physical Γ with conservation constraints (Sys‑CAL). The assurance story depends on epistemes about the model and envelope; the system acts, epistemes constrain. (Archetypes and boundary discipline per core.)

Episteme (show, KD‑CAL lens). Consider a CMIP‑class climate projection episteme (post‑2015 generation): its ClaimGraph covers PDEs and parameterisations; its described entity and grounding holon identify what projection claim is about and how it is grounded; its ClaimScope names historical forcings, resolution, and assumptions; its representation may include domain equations and a tabular schema linked by a NotationBridge with an explicit CL. Compose sub‑epistemes for radiation, clouds, and ocean mixing: R = min across the critical path; an independent hindcast line can raise R only up to its own level; F is bounded by the least‑formal sub‑claim unless the composition adds formal invariants.

#Bias‑Annotation

• Metric worship. Treating [F,G,R] as ends rather than means; mitigation: require evidence bindings and narrative of limits in the claim scope and grounding envelope.
• Category slip. Equating a notation or carrier with ClaimGraph, described entity, or grounding holon; mitigation: slot-graph and carrier separation under C.2.1.
• Analogy inflation. Presenting CL‑0/1 as identity; mitigation: always name the CL rung for cross‑mappings.

#Conformance Checklist

1. C2‑1 (Slot graph). Every U.Episteme MUST satisfy C.2.1 slot discipline for ClaimGraph, DescribedEntitySlot, GroundingHolonSlot, Viewpoint, View, and ReferenceScheme; carriers link through SCR/RSCR or other exact carrier relations and are never parts of the episteme.
2. C2‑2 (Coordinates). Each episteme SHALL declare [F,G,R] with a brief rationale; F is U.Formality ∈ {F0…F9} per C.2.3, exactly one episteme‑level F computed as the min over essential parts. CL is declared for pairs only. Sub‑anchors: ** Contexts MAY mint named sub‑anchors (e.g., F4[OCL], F7[HOL]), which MUST preserve the global order and map to their parent anchor from C.2.3.
3. C2‑3 (Composition). Authors SHALL choose Γ_mode (series vs parallel). For any justification path use R_eff(P) = max(0, min_i R_i − Φ(CL_min(P))); for parallel independent lines to the same claim, take R(Γ) = max_P R_eff(P) (never exceeding the highest-R support line). Compute F(Γ) = min along the used paths. For G, use path‑wise intersections and then SpanUnion({G_path}) constrained by support. Cross‑context traversals MUST use a Bridge with CL and apply Φ(CL) to R.
4. C2‑4 (NotationBridge). Multi‑notation representation components SHOULD register NotationBridge edges with CL and loss note; any cross‑notation reasoning MUST cite the bridge’s CL.
5. C2‑5 (No action). Epistemes MUST NOT be assigned actions; work is executed by systems in role.

#Consequences

Benefits. A single, compact map for all knowledge epistemes or publications; fast detection of weakest‑link R in aggregates; disciplined reuse across domains with explicit CL; consistent separation of meaning from material carriers. Trade‑offs. Authors must learn to declare Γ‑mode and CL explicitly; multi‑notation work requires bridge bookkeeping; mitigation: the episteme slot graph and CL scale keep the discipline brief and repeatable.

#Rationale

KD‑CAL turns the coarse legacy semiotic picture into holonic composition over U.EpistemeSlotGraph, where formal structure and claim scope (F,G), evidence (R), and cross‑mapping congruence (CL) are visible and composable. The explicit C.2.1 slot graph prevents carrier confusion; the characteristics provide a manager‑readable yet formalisation‑ready scale (with G grounded in scope/envelope, not part‑count); the CL scale replaces overloaded “alignment” with a typed sameness relation.

#Relations

• Depends on: U.Episteme — Epistemes and their slot graph (C.2.1): identity invariants, slot definitions, carrier separation, and evidence bindings.
• Peers: Sys‑CAL (C.1), which composes systems; KD‑CAL composes epistemes and feeds assurance lenses in Part B.
• Constrained by authoring: Architectural patterns must include Tell–Show–Show with Archetypal Grounding (this section).

#Worked mini‑examples (post‑2015 flavours)

• Formal lift (ΔF). Recasting a 2019 variational free‑energy narrative into a typed calculus raises F, clarifies scope, and enables CL‑2 bridges between biological and ML formulations—without claiming empirical gain (R unchanged).
• Parallel evidence (R, max). Two independent hindcast lines (circa CMIP6, 2019) supporting the same forecast allow R(Γ)=max(R₁,R₂); if one line drifts, the composite is bounded by the higher-R support line until series constraints apply.
• Notation bridge (CL drop). A 2021 type‑theoretic specification rendered in a semi‑formal DSL requires a NotationBridge with a CL<3 note; any theorem transported across must respect the bridge’s declared preservation.

(No tooling is implied; these are conceptual moves within the calculus.)

#C.2:End

---

Last Updated: 2026-05-15 — this section last modified in upstream FPF commit 37a19061 (github.com/ailev/FPF)