Protected-State Correction Theory v1.2.0

Protected-State Correction Theory v1.2.0 — Structural Recoverability Milestone

Release date: 2026-04-19

Main Result

The release-level technical center is the restricted anti-classifier theorem package (OCP-049 to OCP-053): on declared witness classes, exact recoverability cannot be decided by amount-only descriptors such as rank tuples or fixed budget summaries. Exactness is controlled by compatibility structure, demonstrated through executable same-rank/same-budget opposite-verdict witnesses and mismatch stress tests.

What Is New in This Release

• repository-wide hardening of the six-lane architecture (theorem core, restricted results, methods, validation, physics translation, governance),
• tighter promotion boundary between known backbone and restricted contributions,
• portable active documentation (removal of machine-specific absolute path references),
• reviewer-facing audit controls for live-vs-local and push-readiness,
• release package aligned to branch-level theorem credibility and evidence traceability.

Known Backbone (Explicitly Adopted)

The release does not claim novelty for:

• factorization and fiber-constancy logic as the exactness backbone,
• restricted-linear row-space and kernel criteria,
• standard observability/identifiability framing,
• established projection and damping foundations used in CFD/MHD/QEC-adjacent lanes.

These are treated as adopted structure and cited accordingly.

Scope Limits

• Claims are branch-limited and class-dependent.
• No universal scalar invariant is promoted.
• No universal ontology claims are made.
• Physics-translation lanes remain explicitly labeled by theorem status (PROVED, VALIDATED, CONDITIONAL, OPEN, ANALOGY ONLY).

Validation Snapshot

Validation passed in repository .venv context:

• source .venv/bin/activate && bash scripts/validate/run_all.sh
• python3 scripts/validate/check_links.py
• python3 scripts/validate/check_naming.py
• python3 scripts/validate/check_workbench_static.py
• git diff --check

Canonical Reading Set

• README.md
• docs/overview/repo-authority-map.md
• docs/theorem-core/theorem-spine-final.md
• docs/theorem-core/no-go-spine-final.md
• docs/restricted-results/strongest-paper-lane.md
• docs/validation-evidence/evidence-map.md

Full Diff

• Compare: v1.1.0...v1.2.0

Protected-State Correction Theory v1.1.0 — Expansion and Structural Consolidation

Protected-State Correction Theory v1.1.0 — Expansion and Structural Consolidation

Release date: 2026-04-18

Overview

This release reflects a large expansion of technical work together with a focused structural consolidation of the repository. The project now reads as a branch-based research program with a clearer separation between core theorem material, branch-limited extensions, and exploratory notes.

What Changed Since v1.0.0

Since v1.0.0, the repository has added substantial new material in constrained observation and recoverability, context-sensitive exactness, augmentation and invariant analysis, bounded-domain CFD structure, MHD closure/obstruction analysis, and scoped physics extensions. In parallel, the public-facing layout was reorganized around branch navigation so readers can follow research lanes directly instead of working through scattered file groups.

Structural Consolidation

The repository now uses a branch-first top layer under branches/, with each branch landing page mapping to its canonical theorem/no-go documents, papers, tests, and artifacts. Process-heavy internal pass logs were moved out of front-door navigation to archive locations, and top-level entry points were reduced to the core surfaces needed by outside readers.

This does not remove reproducibility material; it separates active scientific surfaces from internal workflow history.

Branch Updates in This Release

Constrained observation and recoverability

The constrained-observation lane was expanded and formalized around exactness criteria, context-sensitive local-vs-shared decoder structure, and augmentation requirements on supported classes.

Positive recoverability and design

A restricted positive package was developed with explicit class definitions, theorem candidates, failure boundaries, and constrained completion diagnostics.

Invariants and augmentation

The invariant program was expanded and pressure-tested. The stable core remains fiber/factorization and row-space/kernel structure, with additive branch-limited quantities (CID, delta_free, delta_C, DFMI, IDELB, CL) retained where they provide measurable classification value.

CFD bounded-domain structure

The CFD branch now has clearer periodic-vs-bounded separation, including narrow positive bounded-family results and explicit no-go boundaries for naive transfer.

MHD closure and obstruction

MHD materials were consolidated into a clearer exact-vs-asymptotic split, with divergence-cleaning and closure links presented in branch-limited theorem language.

Physics extension placement

Quantum and BH/cosmology materials were retained as scoped extension lanes with explicit status discipline (PROVED ON RESTRICTED CLASS, KNOWN / REFRAMED, VALIDATED / NUMERICAL ONLY, OPEN). They are not promoted as core theorem spine replacements.

Documentation and Navigation Improvements

The root README, branch READMEs, finalization spines, core CFD/MHD notes, workbench docs, and major master reports were normalized to a single research tone: direct prose, branch alignment, and explicit scope. A release-level style normalization report was added to document what was rewritten and what remains in legacy style.

Breaking Changes

There are no API-breaking code changes in this release.

There are documentation-path and navigation changes. Branch-first pages under branches/ are now the canonical entry route, and some process-oriented docs were moved to archive paths. Existing core theorem and paper paths are preserved.

Validation Snapshot

Release checks passed:

• python3 scripts/validate/check_links.py
• python3 scripts/validate/check_naming.py
• python3 scripts/validate/check_workbench_static.py

Known Limits

The strongest theorem support remains in exact and restricted-linear classes. Several extension lanes remain conditional or validated-only and are explicitly labeled as such. Parts of the long-tail documentation corpus still use older style and will continue to be normalized in later passes.

Upgrade Notes for Readers

If you are coming from v1.0.0, start from:

• README.md
• branches/README.md
• docs/overview/main-contributions.md
• docs/finalization/theorem-spine-final.md
• docs/finalization/no-go-spine-final.md

Full Diff

• Compare: v1.0.0...v1.1.0

Protected-State Correction Theory v1.0.0 — Initial Public Research Program Release

Protected-State Correction Theory v1.0.0 — Initial Public Research Program Release

This first public release presents the repository in its finished research-program form.

It exists to make a specific question precise: when can a system remove what should not be there without damaging what must be preserved, and what operator structure is actually required for exact recovery, asymptotic correction, or impossibility?

The release does not try to force one universal law across every correction architecture. Instead, it brings together the parts that survived formalization: a theorem spine, a strong no-go layer, explicit operator constructions, a disciplined physics extension, a static workbench, and a clear path for outside readers.

Why this release matters

• The repository now has a clear public identity and no longer reads like an internal build log.
• The strongest exact and asymptotic results are easy to find and are separated cleanly from conditional or rejected directions.
• The workbench makes the theory usable rather than leaving it only as documents and scripts.
• Physics connections are kept only where the operator structure survives honest testing.
• Reviewers and new readers now have a readable way into the project without having to reconstruct the argument from scattered notes.

Included in v1.0.0

1. A finished theorem-first core

• exact protected-subspace recovery
• exact sector recovery under pairwise orthogonal compatible embeddings
• exact correction rank lower bound
• invariant-split continuous generator results
• self-adjoint PSD decay corollary

2. A finished no-go layer

• overlap / indistinguishability no-go
• mixing no-go for linear flows
• no finite-time exact recovery in the smooth linear-flow branch
• sector-overlap detection failure
• bounded-domain projector transplant failure

3. A clear exact-versus-asymptotic architecture

• exact projector and exact sector branches
• exact periodic continuous projection anchor
• asymptotic damping / generator branch
• explicit separation between exact correction and continuous suppression

4. A real physics extension

Kept:

• quantum error correction as the exact sector anchor under standard assumptions
• periodic Helmholtz/Leray projection as the exact continuous anchor
• Maxwell / Coulomb-gauge projection as a projector-class extension
• GLM divergence cleaning as an asymptotic comparator
• numerical-relativity constraint damping as a conditional asymptotic extension
• continuous quantum error correction as a conditional bridge

Rejected or demoted:

• naive periodic-projector transplantation to bounded-domain exact correction
• generic constrained-Hamiltonian inclusion without extra reduction data
• optimizer or ML unification claims
• universal scalar-capacity language

5. A public-facing workbench

The repository now includes the Protected-State Correction Workbench, a static GitHub Pages tool with:

• Exact Projection Lab
• QEC Sector Lab
• MHD Projection Lab
• Gauge Projection Lab
• Continuous Generator Lab
• No-Go Explorer

It is designed as a real companion to the theory, not as a detached demo.

6. A reviewer-facing reading path

• reviewer-oriented proof-status documents
• novelty-and-limits framing
• paper-outline draft structure
• physics-scope-and-limits memo
• citable outside-literature map for expansion

What people can do with this repository

Use it as a formal theory repository:

• read the theorem and no-go spine to see what has genuinely been proved

Use it as a testing surface:

• run the workbench and the validation suite to inspect exact recovery, asymptotic suppression, and failure modes

Use it as a physics comparison layer:

• see where the framework genuinely fits existing systems and where it stops

Use it as a research starting point:

• follow the citable directions and open-problem catalog to build outward without reintroducing weak claims

Reader entry points

• Start here: docs/overview/start-here.md
• Final architecture: docs/finalization/architecture-final.md
• Final theorem spine: docs/finalization/theorem-spine-final.md
• Final no-go spine: docs/finalization/no-go-spine-final.md
• Workbench: https://rrg314.github.io/Protected-State-Correction-Theory/docs/workbench/
• Physics system matrix: docs/physics/physics-system-matrix.md
• Citable expansion directions: docs/references/citable-expansion-directions.md

Validation snapshot

• Python suite: 24 passed
• workbench / Node suite: 9 passed
• markdown link check: pass
• naming consistency check: pass
• static workbench asset check: pass

Honest scope

This release does not claim that protected-state correction is now a universal law of physics or engineering.

What it does claim:

• there is now a finished theorem-first correction framework here
• the strongest exact and asymptotic branches are clearly separated
• the no-go layer is part of the main story rather than a footnote
• the physics extension is real where the operator structure survives and explicitly rejected where it does not
• the workbench makes the framework inspectable by other readers

What remains open:

• boundary-sensitive continuous correction beyond the rejected naive projector transplant
• stronger branch-specific capacity theory beyond the current lower-bound layer
• sharper theorem-level treatment of conditional physics bridges such as continuous QEC and constraint damping