# Registration: the neutron–proton mass difference, realised (KS-FLIP.1)

**Freeze date:** 2026-09-04 (the paper locked 2026-09-01; anchored by the verifier 2026-09-03)
**Author:** G · Studio G · the420code.org
**Scope:** AP47 (*The Flip*) §§2–8 and §10; AP30 (*The Resistance*) §§3–4, Lemma 3; AP45 (*The Blink*) §5
**Switches engaged:** KS-FLIP.1 (the frozen row), KS-FLIP.2 (forced structure), KS-FLIP.3 (the truncation), KS-FLIP.4 (the scope); KS-30.2 inherited; KS-NPP.1 FIRED and shown
**Reproduction:** `verify_family.py` in this directory
**Status:** frozen for commit. The commit hash is the timestamp.

---

## 1. The corpse first

The bare row of 2026-08-02 — the flip priced as free —

    Δ_bare = 3 (1 − 1/2π) + α (1 + 1/2π) = 2.53099393 mₑ

fired on 2 August 2026 at **7.24σ** against 2.530988574(74) mₑ (KS-NPP.1;
`ERRATA-2026-08-02.md` §E6a). It stays on the board. Nothing in this file
repairs it. This file registers a *new* row beside the corpse, entered as new,
never as repair — the fence the family set on 16 August 2026, before this row
existed: derived from structure alone, computed once, blind, entering as a new
frozen claim.

---

## 2. The registered value

> **Δ_real = Δ_bare / (1 + α²/8π) = 2.53098857035 mₑ**  (CODATA-2022 α)

against the measured line, stated decade-safe:

    Δ_meas = 2.530988574 mₑ,  u = 7.4 × 10⁻⁷ mₑ (0.29 ppm)   — CODATA 2022,
    the exact difference of m_n/m_e = 1838.68366200(74) and m_p/m_e = 1836.152673426(32)

The residual is **−0.005σ** where the bare row fired at 7.24σ. The claim is the
expression, frozen at the named α line; the kill test runs against the
expression, not the display. Digits beyond the seventh decimal exist for exact
reproducibility and carry no metrological weight.

---

## 3. The structure, and why it has no free parameter

1. **The scope is definitional.** The neutron is not the proton plus a thing:
   the same three couplings, one in the other orientation. What it holds is a
   *distinction* — a window held open against the wind that wants it shut. The
   cost is defined on flips: no flip, no cost. The proton is excluded by the
   definition, not by a check (applying the cost to the ground state moves the
   proton by 121,576σ and kills it instantly).
2. **The form is a quotient.** A held distinction discounts the booking of the
   difference itself; a new thing would add a term. The quotient and the
   subtractive form differ by a sixty-five-thousandth of the bar.
3. **The order is α².** The window's leak: the held grain (α-worth, AP45 §5)
   leaks at the substrate's rate (α, AP30 Lemma 3). Worth times rate.
4. **The divisor is 8π.** AP30's own construction at support two: a distinction's
   static count is its orientation pair through the four spacetime dimensions,
   2 × 4 = 8 dimensional expressions, each leaking under AP30's isotropic lemma —
   the leakage normalised by the very structure it leaks through, exactly as
   the proton's 1/(84π) is. The temporal expression is the holding.

The order on the record: the structure in words first; the pair described
before any count arithmetic; the elimination afterward, as elimination. The
alternatives, run after the structure was described — support 3, 6, 18, 21; a
three-dimensional count (6π); 9π — are shown dead in AP47 §5.

---

## 4. Kill and confirm conditions (binding)

1. **KILL — the row.** KS-FLIP.1 dies if an improved measurement of the
   neutron–proton mass difference departs from the expression beyond the bar —
   the audited 1σ of the source lines, 7.4 × 10⁻⁷ mₑ (0.29 ppm).
2. **KILL — forced structure.** KS-FLIP.2 fires if any element of the
   construction — the order, the divisor, the scope, the form, the process claim
   — is ever shown to have been selected by comparison with measurement rather
   than by the stated structural argument. Binding on author and desks.
3. **KILL — the truncation.** KS-FLIP.3: no α³ term exists. The absent
   third-order piece sits at 3.9 × 10⁻⁸ mₑ — 5.3% of the audited bar, 0.020 eV —
   so adjudication waits on roughly twentyfold better metrology on the deuteron
   binding energy (B_d ± 0.373 eV, CODATA 2022). The silence at third order is
   itself a prediction.
4. **KILL — the scope.** KS-FLIP.4 dies if a ground state is ever shown to
   require the cost, or a configuration inside the domain shown to escape it.
5. **INHERITED.** KS-30.2 (leakage isotropy) is live in its parent; shown
   anisotropic, 84π and 8π move together, and this row moves with AP30's.
6. **NO RE-ANCHORING.** The expression is frozen. Any change to the order, the
   divisor or the scope is a new registration referencing this file's hash.

---

## 5. What this does and does not establish

**Does:** freeze one row, derived from a structural sentence — a held distinction
pays to stay held — priced by the family's one mechanism at its fourth site,
landing at the seventh decimal place, beside the corpse it does not touch.

**Does not:** identify Δ_bare's two terms with the standard electromagnetic /
strong decomposition (the claim is the total, never the anatomy); claim a
mass-face observable for the bound neutron (declared null, AP47 §8); or repair
KS-NPP.1. The transfer claim at the paper's centre — AP30's maintenance
mechanics read at a strained hold — is STATED, located and switch-guarded, not
derived from anything deeper, and every grade rides it in the open.

---

*Copyleft 2026. Don't be a cunt. Be kind.*