Author notes — full detail, auditor-facing

Paper 3 is the Cipher paper (working title: *Alchemical Geometry: A Geometric Cipher for Predicting Crystal Structure From Atomic Number Alone*). It is the framework's most consequential paper for materials science and the basis for the framework's patent applications.

Status: HELD. Not published. Awaiting one specific result before release.

Why it's held

The cipher achieves 91/107 with 0 outright misses on the perspective read (cipher v12 — see the version-progression audit). But one piece of the derivation chain is still incomplete: the spiral coordinate's self-derivation from r=0.5 + void geometry.

The cipher reads electrons seated on a spiral coordinate (the C_potential's internal geometry). v11/v12 use a Thomson/Fibonacci hybrid for the seating positions — empirically this gives 98.1% A+P results. But the *spiral itself* is not yet self-derived from the framework's three axioms. It works; it just isn't proven to be unique.

The blocker: If the spiral were *fit* to give the right answers, the cipher would be a sophisticated curve-fitting exercise. The framework's intellectual-honesty discipline demands the spiral be *derived*, not fit. The two-curve cone cascade simulation (SIM-003 v6c) is the current attempt to derive the spiral from r=0.5 + cone geometry.

Pre-registered release criteria. Paper 3 ships when: 1. The spiral coordinate is derived from first principles (likely via SIM-003 cone cascade or via the framework's bulge/floor approach — see project_spiral_bulge_floor_derivation). 2. The derivation is independent of any cipher results (so the cipher becomes a *test* of the derivation, not the derivation's source). 3. The benchmark scoreboard reflects the canonical v12 perspective read with attribution to the derivation source.

What's already in the draft and stable

• The cycle-aware terrain (cipher v12 architecture).
• The 107-element benchmark methodology.
• The eigenvalue / coordination / topology read.
• The bicone simultaneous-derivation chain.
• The snap/tunneling pre-/post-snap mechanism.
• The non-Euclidean measurement finding (2–3° offsets = true curvature).

What's pending

• Spiral coordinate self-derivation (the blocker).
• Final benchmark table at v12 canonical.
• Cross-references to the magic-numbers note (cipher's predictive

reach extends to nuclear shells, not just crystal structures).

• IP-aware editing pass (some material has been moved to the patent

applications and shouldn't double-publish).

Why this matters for the framework

Paper 3 is what turns the framework from "interesting theoretical construct" into "predictive engineering tool." A cipher that predicts every element's crystal structure from atomic number alone is an empirically falsifiable, instantly testable claim. Publishing without the spiral self-derivation would mean publishing the most consequential paper with a known *fit-shaped hole* in its derivation chain. That violates the framework's discipline. So it's held.

Estimated release window. Conditional on SIM-003 v6c converging to r=0.5 ± 0.003 over the next 1–2 weeks: Paper 3 ships within 2–3 weeks after that. If SIM-003 v6c stalls, Paper 3 release pushes until the bulge/floor derivation path produces a result, or until a different self-derivation route succeeds. No premature publication.

Patent ordering vs Paper 3

The framework's patents (TPU + Generator, plus prediction-method and manufacturing-method follow-ons) are already filed (2026-03-29 provisionals). Paper 3's release does not block IP protection. Patent 3 (prediction method) and Patent 4 (manufacturing method) are ready to file before Paper 3 publishes; that ordering ensures IP is locked before the public-disclosure step.