# Registration: the gravitational constant, realised — and the fork (KS-CCC.3)

**Freeze date:** 2026-09-04 (the paper locked 2026-08-11; provenance anchored by the verifier 2026-08-10/11)
**Author:** G · Studio G · the420code.org
**Scope:** AP44 (*The Snap*) §§3–6, §9, §12; AP28 (*The Constant*) §§2, 4–7
**Switches engaged:** KS-CCC.3 (existence, direction, and the migration fork); KS-CCC.4a/4b (the apparatus tests); KS-CCC.1 (forced structure); KS-R.7 (AP28's one-per-cent bar)
**Reproduction:** `verify_family.py` in this directory
**Status:** frozen for commit. The commit hash is the timestamp.

---

## 1. The registered value

> **G = α_em²¹ (1 + 1/π) / (1 + α_em) · ℏc / mₑ² = 6.6719 × 10⁻¹¹ N·m²/kg²**

Measured (CODATA 2018, unchanged 2022): 6.6743 × 10⁻¹¹. **Residual −0.036%.**

AP28's structural value — the provisioned coupling, α_em²¹(1 + 1/π) — is
6.7206 × 10⁻¹¹, +0.69%, outside every laboratory determination ever made. It
stands as the structural value. The commitment closes about ninety-five per cent
of that discrepancy and moves the prediction from outside the experimental
landscape to its low edge — within one standard uncertainty of the lowest
published determination, below others. Stated against individual experiments,
as a hostile reader will state it first: twenty-nine standard deviations below
HUST-T-18 and twenty-five below UWash-00 in those experiments' own
uncertainties, sixteen below the CODATA recommendation at its formal 22 ppm.
Exclusion, not agreement, and claimed as nothing more.

---

## 2. The structure, in one breath

Every Actualization State instant produces exactly one unpaired ε. At every
instant after a configuration exists, that instant's ε is shed at the boundary:
the leakage (AP06). At a constitution instant the ε cannot be shed — if it were,
nothing present would hold the first record open, and by Axiom R the crack
would heal — so it is held. The constitution that concerns a universal constant
is the arena's: one arena, one snap, one ε. Matter's commitments are already
booked as mass (AP29). The arena's commitment lives in the arena's coupling,
which is G. The realised coupling is the provisioned coupling divided by
(1 + ε), and the value of ε in the quotient is α_em — selected in AP44 §5,
derived on the closure note of 2 September 2026 (D-CCC.4 paid).

No new axiom, no new measured input, no new number. The landing is weak
evidence by the paper's own §10 and is reported as non-exclusion, not evidence.

---

## 3. The sign flip, stated plainly

Under AP28 alone the gap ran prediction-above-measurement, the direction
downward apparatus leakage can explain. Under the commitment the residual runs
prediction-below-measurement, and the same leakage cannot absorb it: leakage
lowers what laboratories read; it cannot raise it. Adopting AP44 trades a large
residual with a comfortable story for a small residual with no story yet. The
−0.036% is reported open and unattributed.

---

## 4. Kill and confirm conditions (binding)

1. **THE FORK — KS-CCC.3.** As apparatus systematics are resolved, the adjusted
   (consensus) value of G migrates downward by roughly 300 ppm from its current
   centre, toward 6.672 × 10⁻¹¹. Convergence *upward* toward 6.7206 kills the
   commitment and revives AP28's leakage candidate in full. Stability at 6.6743
   as systematics close supports the null. The switch tests existence and
   direction, not magnitude: the admissible window is seven and a half per cent
   of ε wide against a 22 ppm formal uncertainty, and convergence cannot
   distinguish 1×ε from nearby fractions. No claim of magnitude confirmation
   will be made from it.
2. **KILL — gap scale (KS-CCC.4a).** If the Appendix C meta-analysis finds a
   common apparatus leakage term of order 6,900 ppm, Candidate 1 is confirmed
   and the commitment withdrawn.
3. **KILL — the bar (KS-R.7).** AP28's pre-published one-per-cent bar governs
   both values; the residual is defended by the bar set before the comparison,
   never by "least-precisely-measured."
4. **THE FENCE — KS-CCC.1.** Two clean fractions the data volunteered after the
   fact — 20/21 of ε and 17/18 of ε — are named in AP44 §10 for one purpose: to
   be refused. Data may eliminate structural candidates; data may never
   generate them. The paper states 1×ε, states nothing finer, and arms this
   switch over every future refinement that cannot show its forcing argument
   was written before its number.
5. **NO RE-ANCHORING.** The quotient form and the magnitude 1×ε are frozen.

---

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