tscg-tensor-to-structural-grammar-migration

name

tscg-tensor-to-structural-grammar-migration

description

Migration pipeline for the TSCG foundational formalism reform: replaces the algebraic tensor product interpretation of ⊗ (Hilbert spaces, SVD, coupling matrices, ket notation) with the Structural Grammar formalism (Lambek calculus, monoidal product, categorical functors). Covers 3 phases: M3 surgical edits, M2 property renaming (automated script), M1/M0 verification. Use this skill whenever Michel mentions the structural grammar reform, the migration of hasTensorFormula, the Lambek calculus foundation, removing Hilbert space references from M3, or renaming tensor properties in M2. Also use when asked to apply the Structural Grammar reframing to any ontology file in the TSCG framework. Prerequisite: read references/Structural_Grammar_Foundation.md before starting any phase.

TSCG — Tensor to Structural Grammar Migration

Migration from: Algebraic tensor product (Hilbert spaces, SVD, ket notation)
Migration to: Structural Grammar formalism (Lambek calculus, monoidal product)
Scope: M3 (manual surgery) → M2 (automated rename) → M1/M0 (verification)
Key invariant: Formula VALUES are unchanged — only mathematical interpretation and property NAMES change.
Status: v16.0.0 COMPLETED (2026-05-13)

⚠️ Read first: references/Structural_Grammar_Foundation.md This is the authoritative specification of the new formalism.

---

What Changes vs What Stays the Same

CHANGES                               STAYS THE SAME
──────────────────────────────────    ─────────────────────────────────────
m2:hasTensorFormula      (rename)     Formula values "D⊗I⊗F"   (unchanged)
m2:hasTensorFormulaTeX   (rename)     Symbol ⊗                  (unchanged)
m2:hasTensorFormulaASCII (rename)     Scores A/S/F/I/D/R/E/V/O/I (unchanged)
"hilbert_space"          (remove)     δ₁ and SpectralClasses    (unchanged)
"orthonormality"         (remove)     @graph, @context           (unchanged)
SVD / αᵢⱼ / ket notation (remove)     owl:imports, ontologyType  (unchanged)
"basis_properties"       (rename)     rdfs:label, rdfs:comment   (unchanged)
Any "Hilbert space" text (remove)     @base (factorised in M1)   (unchanged)

Zero tolerance: "Hilbert space" must NOT appear anywhere in active ontology files. Only changelog entries documenting the reform are exempt.

---

Phase Overview

Phase 1 — M3 SURGERY       (manual, ~2h)
  ├── M3_GrammarFoundation.jsonld  → NEW apex ontology (no imports)
  ├── M3_EagleEye.jsonld           → Territory Grammar Gt
  ├── M3_SphinxEye.jsonld          → Map Grammar Gm
  └── M3_GenesisGrammar.jsonld     → Bicephalous hub (ex-GenesisSpace)

Phase 2 — M2 RENAME         (script + manual cleanup, ~1h)
  ├── Run cli_tools/migrate_properties/migrate_properties.py
  └── Manual: remove SVD/Hilbert from Domain and KnowledgeField metaconcepts

Phase 3 — M1/M0 VERIFY     (script verification, ~30min)
  ├── Run cli_tools/verify_migration/verify_migration.py
  ├── Fix any remaining occurrences
  └── Delete legacy alignment files (e.g. M2_MetaConcepts_Alignment.jsonld)

---

Phase 1 — M3 Surgical Edits

Read each M3 file before editing. Apply changes precisely with str_replace.

1.0 — M3_GrammarFoundation.jsonld (NEW — create first)

This is the new apex ontology — no owl:imports, no dependencies. It defines: 6 indexed operators (⊗ᵗ ⊗ᵐ ⊗ᵗ⇒ ⊗ᵐ⇒ ^opᵗ ^opᵐ), abstract classes M3Grammar and M3Dimension, type system 𝕋₀/𝕋₁/𝕋₂/𝕄₀, m3gf:intersubjectiveBenchmark, m3gf:defeasibilityStatus.

See ontology/M3_GrammarFoundation_README.md for full specification.

---

1.1 — M3_EagleEye.jsonld

TARGET: basis_properties block and any Hilbert/orthonormality references.

REMOVE these fields wherever found:

"hilbert_space": "ℋ_μ"
"orthonormality": "verified"
"completeness": "partial (completes with REVOI from Sphinx Eye)"

REPLACE basis_properties with:

"grammar_properties": {
  "grammar_name":   "Territory Grammar Gt",
  "grammar_type":   "free_commutative_monoidal",
  "primitives":     ["A", "S", "F", "I", "D"],
  "product":        "⊗ᵗ  (structural co-activation, Territory)",
  "primitive_role": "generators of Territory type expressions (M2 formulas)",
  "functor_role":   "F_A, F_S, F_F, F_I, F_D : System → [0,1]  (M0 evaluation)",
  "perspective":    "ontological / empirical",
  "independence":   "verified  (no primitive derivable from others)"
}

UPDATE owl:imports: add M3_GrammarFoundation.jsonld

UPDATE coupling_with_REVOI → coupling_with_MapGrammar:

"coupling_with_MapGrammar": {
  "Phi": "Φ : Gt → Gm  (natural transformation — observation)",
  "Psi": "Ψ : Gm → Gt  (natural transformation — interpretation)",
  "epistemic_gap": "δ₁ = ||ASFID_mean − REVOI_mean|| / √2"
}

REPLACE orthogonality_matrix → independence_matrix:

"independence_matrix": {
  "description": "5 primitive types are analytically independent",
  "verified": "true",
  "note": "Independence here means conceptual non-redundancy, not metric orthogonality."
}

VERSION: bump minor. CHANGELOG: rolling 3 entries.

---

1.2 — M3_SphinxEye.jsonld

Same pattern as EagleEye with Map Grammar specifics:

• grammar_name: "Map Grammar Gm"
• primitives: ["R", "E", "V", "O", "I"]
• product: "⊗ᵐ (structural co-activation, Map)"
• coupling_with_ASFID → coupling_with_TerritoryGrammar

UPDATE owl:imports: add M3_GrammarFoundation.jsonld

---

1.3 — M3_GenesisGrammar.jsonld (RENAME from GenesisSpace)

RENAME FILE: M3_GenesisSpace.jsonld → M3_GenesisGrammar.jsonld

UPDATE owl:imports: add M3_GrammarFoundation.jsonld alongside EagleEye + SphinxEye

REPLACE mathematical_properties (Hilbert block) with:

"grammar_foundation": {
  "formalism":    "Lambek Calculus / Free Commutative Monoidal Categories",
  "architecture": "Bicephalous Structural Grammar: Gt (Territory) + Gm (Map)",
  "connection": {
    "Phi": "Φ : Gt → Gm  (natural transformation — observation)",
    "Psi": "Ψ : Gm → Gt  (natural transformation — interpretation)",
    "epistemic_gap": "δ₁ = ||ASFID_mean − REVOI_mean|| / √2"
  }
}

REMOVE entirely: SVD, αᵢⱼ, coupling matrix Σ, ket notation, ℂ²⁵, ℋ_ASFID, ℋ_REVOI

UPDATE all "Hilbert space" text references — zero tolerance, even in examples:

• "Quantum Mechanics: Hilbert space dimension" → "state space dimension"
• "log₁₀(dim(⊗ᵢ Hᵢ)) where Hᵢ are Hilbert spaces" → "log₁₀(number of distinct states)"

VERSION: bump major (e.g. 3.9.0 → 4.0.0). CHANGELOG: rolling 3 entries.

---

Phase 2 — M2 Property Renaming

2.1 — Run the Automated Script

cd <tscg_repo_root>

# Windows — force UTF-8 encoding
python cli_tools/migrate_properties/migrate_properties.py --dry-run --root ontology
python cli_tools/migrate_properties/migrate_properties.py --root ontology

Renames across ALL .jsonld files:

Old	New
m2:hasTensorFormula	m2:hasStructuralFormula
m2:hasTensorFormulaTeX	m2:hasStructuralFormulaTeX
m2:hasTensorFormulaASCII	m2:hasStructuralFormulaASCII

2.2 — Manual Cleanup in M2_GenericConcepts.jsonld

• Remove couplingMatrix, svdDecomposition blocks from Domain/KnowledgeField
• Remove ALL "Hilbert space" text — zero tolerance
• Update rdfs:label of m2:hasStructuralFormula:

  "rdfs:comment": "Structural formula in the TSCG Structural Grammar.
    ⊗ is a commutative monoidal product — NOT an algebraic tensor product."
  

2.3 — M1 Files: @base Factorisation

All M1 .jsonld files must have @base present and factorised:

"@context": {
  "@base": "https://raw.githubusercontent.com/Echopraxium/tscg/main/ontology/",
  "m1core": "M1_CoreConcepts.jsonld#",
  "m2": "M2_GenericConcepts.jsonld#"
}

Namespace prefixes must NOT repeat the base URL. Use relative paths.

---

Phase 3 — M1 / M0 Verification

3.1 — Run the Verification Script

# Windows — UTF-8 encoding handled automatically by the script
python cli_tools/verify_migration/verify_migration.py --root ontology --strict

# Redirect output to file
python cli_tools/verify_migration/verify_migration.py --root ontology --strict > verify_output.txt 2>&1

Expected: 0 violations.

Excluded automatically: any directory whose name contains backup, _archive, Ref, docs, sparql, tools, _protos, reboot-kit. These are archives/references and do not need migration.

3.2 — Manual Fix if Needed

Violation type	Fix
hasTensorFormula not renamed	Run migrate_properties.py on that file
hilbert_space / orthonormality	Remove manually — zero tolerance
Hilbert space in text	Replace with grammar/categorical equivalent
couplingMatrix / svdDecomposition	Remove blocks entirely
Legacy alignment file (e.g. M2_MetaConcepts_Alignment.jsonld)	Delete if no longer needed
Backup dir scanned	Already excluded — no action needed

3.3 — Final Checklist

□  M3_GrammarFoundation.jsonld  — apex, no owl:imports, operators defined
□  M3_EagleEye.jsonld           — grammar_properties Gt, hilbert_space absent
□  M3_SphinxEye.jsonld          — grammar_properties Gm, hilbert_space absent
□  M3_GenesisGrammar.jsonld     — grammar_foundation, SVD/Hilbert removed
□  M2_GenericConcepts.jsonld    — hasStructuralFormula, no Hilbert text
□  All M1 files                 — @base present and factorised
□  verify_migration.py          — 0 violations
□  JSON validation              — all files parse without error
□  Version numbers              — all modified files bumped
□  Changelogs                   — 3-entry rolling maximum respected
□  git commit                   — with descriptive message

---

Critical Rules (Never Violate)

1. Formula values are SACRED — "D⊗I⊗F" strings must never change.
2. ⊗ symbol stays — only the mathematical interpretation changes.
3. Zero Hilbert tolerance — no "Hilbert space" in active files, even in examples.
4. JSON-LD structure untouched — @graph, @context, @base, owl:imports.
5. Encoding: always encoding='utf-8' in Python. Use -X utf8 on Windows.
6. Changelog max 3 entries — remove oldest before adding new.
7. str_replace for surgical edits — never rewrite entire files.
8. @base factorised — M1 namespace prefixes must not repeat the base URL.

---

References

• references/Structural_Grammar_Foundation.md — full theoretical foundation
• cli_tools/migrate_properties/migrate_properties.py — automated property renaming
• cli_tools/verify_migration/verify_migration.py — post-migration validation
• ontology/StructuralGrammar/ — documentation folder (5 READMEs)
• Lambek, J. (1958). The mathematics of sentence structure. American Mathematical Monthly.
• MacLane, S. (1963). Natural associativity and commutativity. Rice University Studies.