56-Report-Level Methods Appendix Template v1.0 | Chapter 1 Introduction & Positioning

Home ／ Docs-Technical WhitePaper ／ 56-Report-Level Methods Appendix Template v1.0

Chapter 1 Introduction & Positioning

I. Purpose & Positioning (Mandatory)

• Define the role of the “Report-Level Methods Appendix”: to capture a reproducible, auditable, replayable minimal loop across theory—math—metrology—implementation for each method.
• Provide a unified interface between main-text prose and machine-readable artifacts (human prose × YAML/JSON equivalence).
• Serve as the source of criteria and dependency lists for cross-volume consistency, ensuring unified caliber and complete evidence chains across teams.

II. Scope & Non-Goals (Mandatory)

• In scope: method definition; key derivations (S/P); mathematical formulation; algorithm & pseudocode; data & experimental design; metrology & calibration (M); evaluation protocol & gates; implementation binding (I); reproducibility & artifact export.
• Out of scope: org processes, personnel authority, budget/commercial terms; historical timelines; high-level roadmaps.

III. Audience & Roles (Mandatory)

• Audience: method & metrology leads, implementation-binding owners, data/pipeline owners, audit & release managers, volume editors.
• Roles: Requester, Author, Reviewer, Approver/Owner, Implementer, Auditor.

IV. Outputs & Artifacts (Mandatory)

• Human: method entry (standard section skeleton), figures & pseudocode, evaluation & conclusions, risk & rollback entry.
• Machine: method.*, evaluation.*, metrology.*, impl.* / reproducibility.*, references.see[].
• Exports: yaml/json/pdf with evidence hashes and signature info.

V. Dependencies & Cross-References (Default Set, Mandatory)

1. Fixed in-text format: “See 《 vX.Y》 Ch.x S/P/M/I…”, anchors preferred over whole-volume cites.
2. Default volumes:
   • 《EFT.WP.Core.Terms v1.0》 (terminology & symbols, P10-*)
   • 《EFT.WP.Core.Equations v1.1》 (minimal equations, S20-*)
   • 《EFT.WP.Core.Metrology v1.0》 (dimensions/units/uncertainty, check_dim)
   • 《EFT.WP.Core.DataSpec v1.0》 (data contracts & export fields, I30-*)

VI. Global Writing & Typesetting (Mandatory)

• Subsection titles use bold Roman numerals; bullet lists “- space”; numbered lists “1. space”.
• Inline symbols are wrapped in backticks: phi(x,t), Xi(x,t), K, Q, T_fil, T_trans, n_eff.
• Any expression with division/integrals/composite operators must use parentheses and explicitly declare path gamma(ell) and measure d ell.
• Unified numbering: Pxx-? (Postulates), Sxx-? (Minimal Equations), Mx-? (Metrology flows), Ixx-? (Implementation binding).
• No-mix rules: T_fil≠T_trans, n≠n_eff, c≠c_ref; no Chinese in formulas/symbols/definitions.

VII. Unified Math & Symbol Caliber (Mandatory)

1. Arrival-time forms (cross-volume criterion):
   • Constant factored: T_arr = ( 1 / c_ref ) * ( ∫ n_eff d ell )
   • General form: T_arr = ( ∫ ( n_eff / c_ref ) d ell )
2. Whenever T_arr appears, the same or adjacent paragraph must define gamma(ell) and d ell, and record delta_form at the machine layer.
3. Dimensional consistency: mark key equations with check_dim=true; specify units/dimensions and uncertainty in the metrology section.

VIII. Document Structure & Reading Map (Mandatory)

• Each method entry follows: Method name & positioning → Theory essentials (S/P) → Mathematical formulation (with symbol table) → Algorithm & pseudocode → Data & experimental design → Metrology & calibration (M) → Evaluation protocol & gates → Uncertainty & tests → Ablation & sensitivity → Implementation binding (I) → Reproducibility & artifacts → Risks & rollback entry → References & cross-refs.
• Human prose and machine lists are one-to-one aligned: sections map to method.* / evaluation.* / metrology.* / impl.* / references.see[].

IX. Quality Gates & Checks (Minimal Set, Mandatory)

• Reference gate: every EFT.WP.* must carry Volume+Version+Anchor; references.see[] must be machine-parsable.
• Dimensional gate: wherever equations appear, check_dim=true; provide units/dimensions/uncertainty.
• Repro gate: script@commit, dataset@version, container image, and a minimal repro command; third-party rerun must pass.
• Gates gate: gate_<metric><comparator><threshold>@<window> naming; no conclusions if any hard gate fails.
• Rollback gate: define triggers (e.g., gate_accuracy<0.98@7d, compat_rate<0.99@replay) and restoration pass lines.

X. Usage & Update Cadence (Mandatory)

• Before onboarding a new method: complete the entry skeleton, human–machine alignment, gate list, and repro experiment; include Author/Reviewer countersign, then publish.
• Updates follow stable line v1.* / draft line v0.*; for Breaking changes, provide version_range and fallback within the entry and set an observation window.