fpf.tools
Library / First Principles Framework (FPF) - Core Conceptual Specification
Jump to passage
In this reading
Source · lines 35987–36019

Link to current text

Published source confirmed at last check

Source changed 2026-10-03 05:29:54 UTC · snapshot created 2026-10-03 05:30:57 UTC · last check 2026-10-03 07:30:20 UTC

A.19.ULSAM:0 - At a glance (didactic, informative)

  • Suite stage: fold_Γ? (ordering lives only in A.19.CHR:suite_protocols; mechanisms[] membership is a set, not an order).
  • Input surface: an admitted MeasureSetSlot, CNSpecSlot, CGSpecSlot, and GammaFoldSlot, with the grouping or membership basis, fold and policy editions, claim scope and selected slices, qualification window, evidence basis, contributors, and intended result declared by those inputs; MinimalEvidenceSlot? may override the CG-Spec minimum.
  • Output surface: AggregatedMeasureSlot (+ optional ContributorSetSlot? as an explanation surface).
  • Non‑goals: no scoring, no comparison, no selection, no “method catalog”, no hidden defaults, no hidden thresholds.
  • P2W seam: an A.15.2 baseline selects the fold, evidence policy and their editions. A.15.3 typed filling applies only to independently declared receiving positions under the CHR P2W hook. Actual fold arguments bind under §4.1; a planned value does not establish that binding.
  • Failure mode: tri‑state guard GuardDecision := {pass|degrade|abstain}; unknown/insufficient evidence never coerces to “pass”.
  • Rule of thumb: if you are about to “average/sum/roll up”, you probably need an explicit ULSAM Fold_Γ stage (or a justified decision to not fold).

What this mechanism is. ULSAM is the CHR mechanism that makes aggregation explicit: it performs an explicit Γ‑fold over a set of admitted measures, producing an aggregated measure (and optionally a contributor surface) under declared admissibility.

What this mechanism is not.

  • It is not a scoring method (that is USCM).
  • It is not a comparison mechanism (that is CPM).
  • It is not a selection mechanism (that is SelectorMechanism).
  • It is not a “method catalog”: method specifics belong to SoTA packs and wiring (G.*:Ext.*), not here.
  • It is not a place to hide defaults (“implementation default fold”) or hidden thresholds.

When you need ULSAM.

  • You want to “roll up” multiple measures into one measure (e.g., an overall reliability/assurance coordinate, a single aggregated risk measure, an aggregate score coordinate).
  • You need the fold to be auditable (what contributed; what was excluded by evidence/admissibility).
  • You need the fold to be scale-lawful (no ordinal arithmetic; no illegal mixing of units).
  • You need the fold to be policy-bound and edition-stable (replayability and pin traceability).

Choosing the fold. If the law for the intended result or a required property is unresolved, use A.9 or reuse a sufficient domain result. When the result selects a fold, its law and conditions supply the basis for the explicit fold reference; ULSAM still governs admission of the measures, scale lawfulness and the actual mechanism operation.

Where it sits in CHR.

  • In the CHR suite protocol, ULSAM corresponds to the optional stage fold_Γ? (i.e., explicitly optional and never hidden inside score/compare/select).

60‑second script for engineer-managers.

“If you’re about to average, sum, or otherwise compress multiple measures into one, stop. Ask: (i) do we have a declared Γ‑fold policy and SCP admissibility, (ii) are the measures admissible and scale-compatible, (iii) what do we do if evidence is missing? If you cannot answer with explicit pins/refs, you are not folding – you are smuggling an assumption. Use ULSAM’s Fold_Γ, record the effective Γ‑fold and contributor set, and keep the fold as an explicit step.”