Collatz Final Gate v8.9 (Main + Gate B Bottleneck v1.9) — Sealed Level-2B Demo Packet (L0=16 anchor, I4 mandatory)

# Overview This record releases **Collatz Final Gate v8.9** (main paper) together with the **Gate B Bottleneck v1.9** companion and a **sealed Level-2B demo packet** implementing an *audited proof-input interface*. **Primary goal.** The project is a *reduction + certification program* that moves the last arithmetic bottleneck (Gate B) from “prose assumptions” to a **finite, mechanically auditable verdict certificate**. The intended closure semantics is: - **Global statement proved here:** the **All-\(L\) promotion** mechanism (deterministic implication chain).- **Finite proof-input requirement:** a **single base-scale trigger instance** at \(L_0=16\) whose verdict certificate checks the required inequalities (including the Corr/TwGap item). This is **not** presented as an unconditional proof of the Collatz conjecture at v8.9; it is an explicit proof interface whose remaining inputs are certificate-facing and auditable. # Record contents - **Main paper:** `Collatz_Final_Gate_v8.9.pdf` (+ `Collatz_Final_Gate_v8.9.tex`)- **Companion:** `GateB_Bottleneck_v1.9.pdf` (+ `GateB_Bottleneck_v1.9.tex`)- **1-page checklist:** `L0_16_complete_certificate_checklist_v8.9.pdf` (+ `.tex`)- **Sealed Level-2B demo packet (ZIP):** `Collatz_Demo_L2B_v8.9.zip` # Summary (v8.9) ## What is proved (global, paper-level) The only all-scales conclusion in the certification program is the **All-\(L\) promotion**: If a single base-scale verdict certifies the trigger inequality at \(L_0\),\[ηₜₒₜₐₗᶜᵉʳᵗ(L_0) < κ_0ᶜᵉʳᵗ(L_0),\]then the deterministic implication chain upgrades it to **all \(L ≥ L_0\)**. ## What is required as a finite proof-input (certificate-facing) Gate B is treated as a **finite, auditable certificate interface** at a fixed base scale \(L_0\) (not as an \(L→∞\) limit). The verdict certificate must include: - a defect budget \(ηₜₒₜₐₗᶜᵉʳᵗ(L_0)\),- a contraction margin \(κ_0ᶜᵉʳᵗ(L_0)\),- a **certified-grid** Corr/TwGap item (I4) as a **mandatory checked entry**, and- hash bindings (including `binder.manifest_sha256`) that pin the verdict to immutable packet bytes. ## Scope clarification: global promotion vs certified-grid inputs - **Global conclusion (proved here):** All-\(L\) promotion.- **Certified-grid inputs (finite, audited):** the Corr/TwGap item is certificate-facing and interpreted on a declared validation grid \(Gᵥₐₗ\), denoted \[ Corrᶜᵉʳᵗ, TwGapᶜᵉʳᵗ. \] The purpose is to make Gate B a finite proof-input object (with complete witness files), not a hidden standing assumption. # What is new in v8.9 (reviewer-facing closure upgrade) - **Explicit closure semantics.** The papers now consistently state that a **single** base-scale verdict at \(L_0=16\) triggers the All-\(L\) implication chain.- **Certified-grid terminology.** Corr/TwGap is presented in certificate-facing form (\(Gᵥₐₗ\), \(Corrᶜᵉʳᵗ\), \(TwGapᶜᵉʳᵗ\)) to prevent “finite grid = global theorem” misreadings.- **Artifact naming is simplified.** Reader-facing names are short (see file list above), while integrity is enforced by packet-side identity (`packet_identity.json`) and hash bindings (`manifest.json`, `binder.manifest_sha256`). # Audit / reproducibility (quickstart) 1. Download and unzip `Collatz_Demo_L2B_v8.9.zip`.2. From the packet root directory, run the reference audit entrypoint: - `python -B scripts/audit_all.py --mode sealed_full` - and, for non-synthetic mode: `python -B scripts/audit_all.py --mode sealed_full_nonsynth`3. The audit checks: - `thresholds.json` enforcement (checked inequalities with margins), - completeness of witness/report files referenced by the verdict, - provenance seal constraints, and - the hash-binding closure via `binder.manifest_sha256`. **No separate “replication toolkit ZIP” is shipped.** Two-environment semantic replication (runA/runB snapshots + comparator + replication audit) is included *inside the demo packet* under `replication/` and `specs/repro_contract.json`. # Scope & non-toy status This record aims to align the project with modern standards of computational proof practice:a finite, auditable object (the verdict certificate) plus a reproducible verification pipeline with explicit hash bindings. The remaining work is explicitly tracked as certificate-facing Gate B targets (e.g., paper-level Proxy ⇒ Corr/TwGap strengthening) rather than being hidden as informal assumptions. # Program closure and targets - **Closed (annealed / unconditional):** finite-level conductance/spectral gap mechanisms and their deterministic implications *within the program’s annealed engine*.- **Open (certificate-facing / arithmetic bottleneck):** Gate B discharge at base scale with a uniform margin that is strong enough to be accepted as final proof-input for the All-\(L\) promotion chain. If the community accepts the program’s proof-input specification, the “last step” is the production (and public audit) of a **single non-synthetic \(L_0=16\) complete certificate** meeting the checklist in `L0_16_complete_certificate_checklist_v8.9.pdf`. ========================= Author: Lee Byoungwoo leeclinic@protonmail.com