Open Problems and Next Steps

WIP-5 Fine structure constant loop corrections

The tree-level result \alpha = 1/135.1 (+1.45% from measured) is derived from C8 with zero new parameters. The gap is now understood to come entirely from a single missing vacuum-polarization correction — the modon self-energy — computed as a dispersion-relation integral (scripts/modon_self_energy.py) fed by the BdG doorway spectral function. A single \Pi(0) \approx 2.00 simultaneously closes \alpha (+1.47\% \to 0), (g-2)/2 (+1.62\% \to +0.15\%), and the Lamb shift (+7.55\% \to 0 — \alpha^5 amplification of the same fractional shift). The loop is QED-validated: with the exact one-fermion spectral function the integral reproduces the known leptonic \Delta\alpha(M_Z) = 0.031423 vs 0.031418 before any substrate input enters.

Source 2 is excluded. Matching \alpha via \sin^2\theta_W running alone needs \sin^2\theta_W = 0.2279 — below the M_Z value, i.e. energy above M_Z, the opposite of the low-energy bridge the running was supposed to supply. The modon self-energy must dominate; there is no double-counting.

The lineshape is pinned, and the gap reduces to one ratio. The weak-branch resonance the \alpha chain rides (\omega_0/\Gamma = 2.99, \delta_0 = 18.5°) was initially modeled as a Breit–Wigner; it is now computed exactly via finite-cell BdG diagonalization (Script 6, R_\text{cell}/L = 4.5, \mu_\text{bulk}=0), giving C_\text{sub} = -1.85 \pm 0.07 — robust across cell size and grid, between the pure-log (0) and the centered-Breit–Wigner (-2.6). Splitting \Pi(0) into its two pieces, \Pi(0) = \underbrace{\frac{R_\text{ch}}{3\pi}\ln\!\frac{E_F^2}{\omega_0^2}}_{\text{continuum plateau, }80\%} \;-\; \underbrace{\frac{R_\text{ch}}{3\pi}\,C_\text{sub}}_{\text{resonance, }20\%}, the resonance piece is cutoff-free (the exact dispersion weight saturates by E \sim 6\,\omega_0, well below E_F and the grid). The entire cutoff question is the plateau log, and the required band is E_F/\omega_0 \sim 42–244.

Candidate resolution: E_F/\omega_0 = 4/\alpha_{mf}^2. The owed ratio is not a new free input. The CdGM relation \omega_0 = \Delta^2/E_F (Fine Structure Constant § The Kramers Doublet) pins both endpoints from framework quantities: the gap \Delta = m_e c^2/2 (the WIP-12 zitterbewegung/pairing gap) and the UV cutoff E_F = m_\text{eff} c^2 = m_e c^2/\alpha_{mf} (the intrinsic effective mass from WIP-15 — not the TeV vortex-core scale E_\text{core} of the Weinberg-angle running, a distinct scale these two symbols were once conflated over). Then: \omega_0 = \frac{\Delta^2}{E_F} = \frac{\alpha_{mf}}{4}\,m_e c^2 \approx 38\ \text{keV},\qquad \frac{E_F}{\omega_0} = \left(\frac{2}{\alpha_{mf}}\right)^2 = 44.2, with E_F/\Delta = 2/\alpha_{mf} = k_F\xi, the CdGM rung count. The value 44.2 lands at the low edge of the required band — exactly where the exact lineshape puts it — giving \Pi(0) = 2.00 and 1/\alpha = 137.06, closing \sim 101\% of the gap (scripts/wip5_cross_scale_ratio.py). It is not circular: \alpha_{mf} = 0.30078 is computed from BdG geometry, not from \alpha. The cutoff log has a closed form, \ln(E_F/\omega_0) = 2\ln(2/\alpha_{mf}) = 3.79 — twice the log of the CdGM rung count. This also merges WIP-5’s cross-scale unknown with WIP-12’s standing m_e-vs-m_\text{eff} visibility residual into a single unknown: E_F/\Delta = 2/\alpha_{mf} is the visibility factor.

What remains: two identifications, one closed. The candidate rests on two identifications: (i) that the loop’s UV cutoff is the chemical potential E_F = m_\text{eff} c^2 (spectral weight exhausted at the Fermi scale, not the far-higher dc1 healing scale); and (ii) that the CdGM bulk gap \Delta is the WIP-12 inter-band gap m_e/2. Identification (ii) is automatic if the vortex order parameter is that gap — as in any BdG vortex — so it is closed in principle. Identification (i) is the genuine open piece. The \mu_\text{bulk}-restored finite-cell run (scripts/wip5_mu_bulk_termination.py) cannot settle it: \mu_\text{bulk} is a band-bottom (gap) knob, not a band-top (bandwidth) knob, so the single-cell toy has no Fermi sea and its continuum stays pinned at the grid regardless of \mu_\text{bulk}. The real tests are (a) a finite-density Lindhard pair susceptibility filled to E_F, where Pauli blocking terminates the particle–hole phase space; or (b) the periodic Bloch BdG (Script 8) showing the anomalous doublet band tops at E_F = m_\text{eff} rather than at the grid. WIP-15 closes the IR side — its self-consistent 3D gap equation derives the gapless nodal continuum the doorway hybridizes into, retiring the \mu_\text{bulk}=0 idealization and firming up C_\text{sub}=-1.85 — but WIP-15 is an IR (band-bottom) result; the UV band-top question is structurally separate and still owed.

WIP-10 Bridge equation

Bridge equation. All factors now have identified physical origins:

Factor	Value	Origin	Status
4\pi	12.566	Gauss’s law solid-angle factor; enters through BLV induced gravity self-consistency (\nabla^2\Phi = 4\pi G\rho). NOT from Tkachenko speed (8\pi).	✅ Step A
1/K	0.0638	Bessel matching: K = j_{11}^2 + 1 = 15.682 from Larichev-Reznik modon boundary condition.	✅ Established
1/\sqrt{2}	0.7071	GP kinetic energy: the factor of 2 in \hbar^2/(2m). Healing length \xi_\text{GP} = \xi_V/\sqrt{2}.	✅ Step B
\eta = 1	1.000	No 3D stacking correction: lattice is straight parallel lines (fiber bundle). Five-pillar argument from Saffman.	✅ Step D

Step C: algebraic verification (0.20%). Step E: constrained equilibrium derivation — three conditions (GP energy balance, SC2 gravitational self-consistency, modon matching) acting on one medium uniquely fix \xi with no remaining variational freedom (the shared n_1\omega_0 cancels, so \omega_0 drops out — it is gravity-fixed, not fixed here; see WIP-15).

Remaining formal work (not blocking):

• Step A: explicit Seeley-DeWitt computation for BEC+lattice to verify the exact 4\pi coefficient. The BLV decoupling condition (their eq. 22) — the deepest open theoretical question — is physically motivated (strong-coupling universality, Volovik self-tuning) but not proven. This same Seeley-DeWitt 4\pi also derives the Higgs VEV’s geometric prefactor 8\pi = 2\times 4\pi (the extra factor of 2 is the radiation-EOS gravitational weight of the massless chirality Goldstone sector): closing Step A simultaneously closes the VEV — see WIP-15 §5. The 2026-05-28 reframe (worked out in that section) sharpens what “closing Step A” means: the 4\pi is the Einstein-Hilbert normalization, automatic once the induced action is exactly Einstein-Hilbert, so the heat kernel never produces it directly. Step A reduces to the proposition that the substrate’s emergent Lorentz invariance is exact (covariance then forces the EH form at two-derivative order, hence the Poisson 4\pi) — an existing framework pillar, with close-packing’s single Planck scale (E_{\text{Pl}1}=E_{\text{Pl}2}=m_1c^2) the candidate mechanism that removes the non-covariant contamination.
• Domain size: the five-pillar argument (Step D) establishes domains of size L_\text{domain} \gg \xi but doesn’t compute L_\text{domain} from first principles. Even an order-of-magnitude estimate (horizon size at formation? Jeans length at condensation?) would strengthen the argument. Detailed determination requires the Phase 4 cosmological calculation.
• Step F (the 2D/stacking decomposition of f): depends on WIP-15. The Blatter mapping now confirms why the \eta = 1 result holds — the in-plane lattice is strictly 2D at the scale where the bridge equation operates (R < \Lambda = \xi).

WIP-11 dag retired — the lattice is self-pinned

Status (2026-06-05): resolved by retirement. The framework previously carried a heavy, sparse second dark-matter species, “dag” (M_d, n_d), whose stated job was to pin the vortex lattice against cosmological drift, and whose mass and number density were free parameters. The open question was which role it played — potential well that nucleates the effective quantum (n_d = n_1/\nu \approx 800 m^{-3}), or former of the vortex-lattice cores.

That question is now moot: dag is retired. Once dc1’s self-interaction is taken to be logarithmic (a Zloshchastiev superfluid-vacuum equation of state, the absorption worked out across sessions/svt-1…svt-12), every job dag was invented to do the substrate does intrinsically. Its length scale is the coupling |b| = m_1 c^2 — a fixed energy, not a density — so the cell scale \xi cannot drift as the universe expands, the condensate self-binds at the healing length, and it self-organizes to close-packing (n_1\xi^3\approx1) with no external anchor. The decisive numerical check is the bridge equation: the headline 0.20\% match is the f_d = n_d M_d/\rho_{DM} = 0 optimum, and any nonzero dag fraction only worsens it, monotonically (machine-verified, scripts/svt_dag_removal.py; full argument in Substrate Particles § Why the Scaffold Needs No Second Species and sessions/svt-10-dag-removal.md). Retiring dag removes a free parameter (M_d, n_d), deletes a \le3\% downside, and changes no reported number.

Residual caveat (carried, non-blocking). The log’s maximum-packing result is proven for a scalar droplet (Avdeenkov–Zloshchastiev 2011); extending “maximum density \Rightarrow vortex-lattice spacing” to a rotating vortex array is a natural step but not yet a theorem. It does not affect the f_d = 0 numbers, which use only n_1 m_1 = \rho_{DM}. The broader job of sweeping the remaining “dc1 substrate” naming out of the downstream chapters is the synthesis pass; this entry records the physics decision.

WIP-12 Photon energy and the Compton breath

Status (2026-05-29): largely resolved. The two questions below turn out to be one phenomenon at two scales, once a causal-consistency check is applied to the breathing picture. The waveform is pinned to a zero-parameter form, now understood as the BdG spinor’s zitterbewegung — there is no material boundary potential; the harmonic V_b is its effective shadow. The lone residual is the framework’s standing m_e-vs-m_\text{eff} visibility question, now appearing in the breathing gap (end of entry).

Minimum photon energy — the quantum of the lattice breath. E_\text{min} = 2\pi m_1 c^2 \approx 13 meV (\lambda \sim 100\;\mum, f \sim 3 THz). Below this, modons cannot form — energy transport crosses over to lattice phonons (gravitational waves). The clean reading: this is one full anti-phase breath of a single lattice cell. The dc1 cell breathes at its own Compton clock \omega_1 = m_1 c^2/\hbar = 3.1\times10^{12} rad/s — the lattice “breathes in pairs” mode — with energy quantum \hbar\omega_1 = m_1 c^2, and the smallest modon carries one complete cycle of it: E_\text{min} = 2\pi\,m_1 c^2, the 2\pi being the full 0\to2\pi breathing phase. Tested from below (2026-06-10): if sub-floor light rode the collective branch, it would arrive early with a +\nu^2 signature ($$4,500 yr at 600 MHz from z=0.5 — sign-flipped and shape-inverted relative to the plasma \nu^{-2} delay). A refit of 896 one-off bursts from the second CHIME/FRB catalog’s full-resolution dynamic spectra bounds the participation at \varepsilon < 6.5\times10^{-16} (95%): sub-floor EM propagates at c like the modon, so the floor marks a change of quantization character, not of speed (Photon as Modon; sessions/frb-nu2-advance-1.md). Tested at the crossing, by the CMB (2026-06-10): because the floor is a fixed local frequency and photons blueshift into the past, every CMB photon below 3 THz crossed the floor in flight, at z_\text{cross} = 3\,\text{THz}/\nu_\text{obs} - 1 — so the COBE/FIRAS frequency axis is a crossing-epoch map (600 GHz \leftrightarrow z\approx4, 60 GHz \leftrightarrow z\approx50). The soliton-to-collective conversion there is adiabatic by \mathcal{A} = \nu_\text{floor}/H(z_\text{cross}) \sim 10^{28}, so any spectral scar is \sim 10^{-28} (shape \propto\nu^{-3/2}, steepest in FIRAS’s cleanest channel) — 23 orders below FIRAS’s |\mu|,|y| limits, i.e. its 50-ppm blackbody confirms the band edge is non-dissipative for redshifting light (scripts/firas_crossover_adiabaticity.py, sessions/firas-crossover-1.md). The floor is now transparent to EM from two orthogonal observables — below it (FRB) and at the crossing (FIRAS); the deeper question was why, and it now has a derived answer (2026-06-11). A single sub-floor photon (h\nu < E_\text{min}) cannot be a localized modon, so it is the modon’s conserved circulation (winding) quantum delocalized over \lambda \gg \xi — a “stretched photon” — whose speed c = \hbar/m_1\xi contains no soliton size and so has no leading dispersion. Because circulation cannot mix into the circulation-free sound mode, the residual deviation is not power-law but exponential, \delta_g \sim \exp(-\nu_\text{floor}/\nu) (the substrate’s BCS/Mattis–Bardeen sub-gap analogue — a photon below the gap cannot reach the core-reconnection that would unwind it). An \exp(-\nu_\text{floor}/\nu) has no \nu^2 term at all, so the FRB null does more than confirm “c”: it bounds any (k\xi)^2 = \nu^2 term at \alpha_6 < 2.4\times10^{-16}, excluding a generic emergent gauge boson (which would allow it at order unity) and selecting topological protection (Photon as Modon § What Carries Light Below the Floor; sessions/below-floor-dispersion-1.md). The falsifiable residue: the exponential turns on only in the 0.1–3 THz band, where a long-baseline propagation test would separate the protected reading (essentially nothing until a sharp floor rise) from the sound reading (a smoothly growing \nu^2 advance). Deferred to theory — now computed at barrier level (2026-07-05): the modon-core reconnection action is a Gross–Pitaevskii phase-slip barrier (the winding-change transition state is a black soliton / vortex nucleation), giving \alpha \equiv S_\text{rec}/E_\text{min} \approx 0.2–0.6, an O(1) that confirms the mechanism and places the exponential’s turn-on in the lower part of the 0.1–3 THz band (scripts/reconnection_barrier_from_above.py, sessions/reconnection-barrier-phase1.md). The one place an in-band detection (not just a bound) could live is the laboratory bench — THz vacuum spectroscopy and the crystal-optics band-edge test — which sees the sharp 3 THz edge that the cosmological redshift smears into FIRAS’s \nu^{-3/2} continuum.

Compton oscillation dynamics — two breaths, not one. The electron’s energy shuttles between a contracted, all-kinetic phase and an expanded, all-boundary phase at \omega_C = m_e c^2/\hbar = 7.76 \times 10^{20} rad/s. The original question — what fraction of the cycle has what radius? — first needs a correction. An earlier reading had the electron breathing from r_\text{eff} = 150 fm all the way out to \xi \approx 100\;\mum every Compton cycle. That is causally impossible: in one period the fastest signal travels c\,T_C = \lambda_C = 2.43 pm, so reaching \xi (a further \sim10^5\,\lambda_C) would take \sim4\times10^7 cycles even at the speed of light — a radial speed of \sim10^8\,c.

What survives is a two-mode picture, each mode running at the substrate speed limit (amplitude \times frequency = c exactly):

Breath	Inner ↔︎ outer turning point	Frequency	What it is
Heartbeat (Zitterbewegung)	r_\text{eff}=150 fm ↔︎ \bar{\lambda}_C = 386 fm	\omega_C = 7.76\times10^{20} rad/s	electron Compton breath
Coherence dress	\xi_\text{GP}=\xi/\sqrt2 ↔︎ \xi\approx100\;\mum	\omega_1 = m_1 c^2/\hbar = 3.1\times10^{12} rad/s	dc1 / lattice breath

The per-cycle heartbeat amplitude is bounded by causality at \sim c/\omega_C = \bar{\lambda}_C = 386 fm — exactly the Zitterbewegung amplitude the framework already invokes, and exactly r_\text{eff}/0.388. The \xi\approx100\;\mum envelope is not a per-cycle breathing extent; it is the static coherence / pilot-wave dress, established over the coherence time \sim\xi/c \sim 3\times10^{-13} s and thereafter quasi-frozen on the heartbeat timescale (the dress is m_e/m_1 = \alpha_{mf}\nu \approx 2.5\times10^8 times slower than the heartbeat). The two scales are simply the two Compton wavelengths in the medium — \bar{\lambda}_C = \hbar/m_e c for the electron, \xi = \hbar/m_1 c for dc1 — and their ratio is

\frac{\xi}{\bar{\lambda}_C}=\frac{\omega_C}{\omega_1}=\frac{m_e}{m_1}=\alpha_{mf}\,\nu = 2.5\times10^8.

Why the breath is a complete (lossless) exchange. The breath is energy-complete because it is an anti-phase pair oscillation (the Cooper-pair / ³He-A pairing of WIP-15): when the electron vortex contracts, its counter-rotating partner (the inter-sheet intermediate layer) expands, and the energy is handed across the shared seam rather than lost. This is why mass is the clean time-average, and why the two C4 terms (kinetic, boundary) are extrema of one oscillation rather than independent reservoirs — answering WIP-14. Modeling the breath as radial motion at fixed L=\hbar in V_\text{eff}(r)=\hbar^2/2m_\text{eff}r^2 + V_b(r), the radial breathing runs at twice the orbital frequency for an isotropic restoring dress — the same factor as the \omega_\text{internal}=2K_r/I_\text{eff}=\omega_C identity (Electron § The Dual-Spin Gyroscope) and Dagan–Bush’s “source at 2\omega_C,” and the temporal face of the WIP-15 pairing-two.

The waveform — resolved, conditional on the boundary potential. With the heartbeat bounded between r_\text{eff} and \sim\bar{\lambda}_C, “what fraction of the cycle at what radius” becomes a closed problem once V_b(r) is named. The framework already pins V_b at two points (V_b(r_\text{eff})\approx0, rising to \sim m_e c^2 at the outer turn) and fixes the frequency (\omega_C, with the radial mode at twice the orbital rate — the \omega_\text{internal}=2K_r/I_\text{eff}=\omega_C identity). The framework’s own boundary-energy budget, E_\text{boundary}=\tfrac12\rho_\text{cr}(\Delta v)^2 A\,\delta \propto area, makes the boundary a stretched counter-rotating membrane: V_b(r)=\tfrac12 k r^2 (the competing \Delta v=v_\text{rot}\propto1/r reading gives E_b\approx const, which cannot confine the breath).

Fixing k=m_\text{eff}\omega_C^2/4 from the frequency leaves no free parameter, and the model then predicts the whole waveform:

Quantity	Prediction
Equilibrium radius	r_0 = 2\,r_\text{eff} = 300 fm = 0.78\,\bar{\lambda}_C
Inner / outer turning point	r_- = r_\text{eff} = 150 fm; r_+ = 4\,r_\text{eff} = 600 fm = 1.55\,\bar{\lambda}_C
Trajectory	ellipse 150\times600 fm; orbit at \omega_C/2, breath at \omega_C
Peak radial / total speed	0.58c / 0.70c — causal
Energy	rotational \leftrightarrow boundary swap, sum constant (the existing quadrature figure)

The exact solution is the 2D isotropic oscillator, r^2(t)=\tfrac12(r_+^2+r_-^2)+\tfrac12(r_+^2-r_-^2)\cos\omega_C t — an arcsine distribution in r^2: the electron spends \sim30\% of the cycle contracted (r<r_0) and \sim70\% expanded, lingering near the outer turning point (\sim30\% of the time within 10% of r_+, vs. \sim7.5\% near r_-). The time-averaged size is \langle r\rangle = 409 fm \approx 1.06\,\bar{\lambda}_C, and \pi\langle r^2\rangle\approx6\times10^3 barn \sim\pi\bar{\lambda}_C^2 — the Compton/Thomson length, with the measured \sigma_T=\tfrac83\alpha^2\,\pi\bar{\lambda}_C^2 = 0.665 barn supplying the QED \alpha^2 (two photon vertices).

Three results fall out for free: (i) the 150\times600 fm ellipse is the transverse cross-section of the Zitterbewegung helix (Electron § The Zitterbewegung Connection); (ii) breath-at-twice-orbit is the 2:1 gear ratio behind the 720° / spin-½ property; (iii) r_+=1.55\,\bar{\lambda}_C sits right at the causal ceiling c\,T_C/4=1.57\,\bar{\lambda}_C — the breath is as large as causality permits, which is why its amplitude is the Compton wavelength rather than something smaller.

Is V_b(r) a real potential? No — the breath is zitterbewegung. Trying to pin V_b from first principles turned up a cleaner answer: there is no material boundary potential. Three routes were tested:

• GP kinetic + interaction (rigorous). Under the scaling ansatz the breathing scale b feels U(b)=A/b^2+B/b^d — both terms decrease, so a free GP blob spreads, it does not breathe. Confinement would need an added increasing term.
• Surface tension (dead end). V_b=\sigma\,4\pi r^2 gives the right r^2 shape but the wrong size: a GP surface tension overshoots the stiffness by \mathcal{O}(10), swings 5\times with the assumed density profile, and is geometrically impossible — the bulk dc1 healing length \xi_\text{GP}=68\,\mum is 10^9\times too thick to wrap a 150 fm electron.
• Vortex pairing gap (dead end). A chiral p-wave vortex’s order parameter vanishes linearly at its core, so |\Delta(\vec r)|^2\propto r^2 is harmonic — but only for r\ll\xi_\text{eff}\approx82 fm. The actual breath lives at 150–600 fm, where the gap is saturated (flat, |\Delta|^2\to\Delta_0^2): the gap-as-spring provides no restoring force over the breathing range.

What works is intrinsic, not a potential. The effective quantum is a BdG/Nambu spinor (particle u, hole v), and a relativistic spinor trembles — zitterbewegung — from particle–hole interference, at the inter-band frequency 2\Delta/\hbar with amplitude \hbar c/2\Delta. Setting the particle–hole gap to the visible breathing energy, 2\Delta=m_ec^2 (Dirac mass m_e/2 — the pairing-two again):

\omega_\text{zitter}=\frac{2\Delta}{\hbar}=\omega_C,\qquad r_\text{zitter}=\frac{\hbar c}{2\Delta}=\bar{\lambda}_C,

both exactly, from a single input and with no confining potential. This unifies the two relations that fixed the heartbeat — \hbar\omega_C=m_ec^2 (energy) and \omega_C\bar{\lambda}_C=c (causality) — into one: the breath is the spinor’s zitterbewegung, and its amplitude is the Compton wavelength because that is what \hbar c/\text{gap} is. The harmonic V_b above is a faithful effective shadow (its turning point sits at the causal ceiling precisely because the real motion is relativistic), but it is not a fundamental potential — and the precise turning points (r_0, r_+) are shadow-model artifacts; the robust statement is that the amplitude is of order \bar{\lambda}_C.

The two interfering components close a loop: the particle u and hole v of the zitterbewegung are the anti-phase Cooper pair of WIP-16 — the contracting vortex and its counter-rotating partner. The lossless anti-phase exchange and the zitterbewegung are the same motion. Residual: why the relevant gap is the visible m_ec^2 rather than the intrinsic m_\text{eff}c^2 is the framework’s standing m_e-vs-m_\text{eff} visibility question (the same \alpha_{mf} that makes the observed mass m_e=\alpha_{mf}m_\text{eff}), now appearing in the breathing gap — the one piece still owed. This is the same unknown as WIP-5’s cross-scale ratio (2026-06-17). There the doorway loop closes \alpha iff E_F/\omega_0 = 4/\alpha_{mf}^2, which follows from the CdGM minigap \omega_0 = \Delta^2/E_F with this gap \Delta = m_ec^2/2 (visible) and the Fermi scale E_F = m_\text{eff}c^2 (intrinsic): the ratio E_F/\Delta = 2/\alpha_{mf} that sets it is the visibility factor. So the breathing-gap residual and WIP-5’s cross-scale residual are one and the same \alpha_{mf} = m_e/m_\text{eff} — closing either closes both.

WIP-13: Observable consequences of Tkachenko modes

The substrate’s vortex lattice supports Tkachenko (shear) modes at c_T \approx 9 km/s \approx 3 \times 10^{-5}\,c (from Baym’s stiff-limit formula c_T = \sqrt{\hbar\Omega/(4m_1)}; deeply incompressible). These are not photons or gravitational waves — they are slow lattice oscillations at f_T \approx 3{,}700 Hz.

Mode	Speed	Frequency scale	Origin
Sound / modons / GWs	c	c/\xi \sim 3 \times 10^{12} Hz	BEC quasiparticle spectrum
Tkachenko (lattice shear)	\sim 9 km/s	\sim 3{,}700 Hz	Vortex lattice elasticity
Outer rotation (\omega_0\xi)	\sim 800 km/s	—	Lattice-scale vorticity

Possible signatures: modulation of dark matter density at kHz frequencies; second-order coupling to baryon-photon plasma (tiny CMB imprint); laboratory detection via precision interferometry at \sim 100\;\mum scales. Whether any are detectable is open, but c_T is a zero-parameter prediction.

WIP-14: C4 may have zero effective degrees of freedom

The effective-quantum form of C4 gives \tfrac{1}{2}m_\text{eff}\,v_\text{rot,inner}^2 = \tfrac{1}{2}(m_e/\alpha_{mf})(2\alpha_{mf}\,c^2) = m_e c^2 — an algebraic identity. The contracted-phase kinetic energy equals the full electron rest energy with no remainder. If the Compton cycle is a complete energy exchange (all-kinetic at peak contraction, all-boundary at peak expansion), then C4 reduces to a consistency check automatically satisfied by Subsystem A. It adds zero new constraints and zero new freedom.

The electron pumps its rest energy between the contracted phase (all kinetic, near r_\text{eff} \approx 150 fm) and the expanded phase (all boundary) every Compton cycle (T_c = 8.1 \times 10^{-21} s). The time-averaged energy is m_e c^2 at every moment, but the spatial distribution oscillates. (The per-cycle breath is bounded at \bar{\lambda}_C \approx 386 fm by causality; the \xi \approx 100\;\mum coherence dress is a separate, much slower mode — see WIP-12.)

Needs verification: Is the Compton oscillation truly a complete exchange, or is there a residual that stays localized? If partial, C4’s two terms retain some independence. Also: the identity \tfrac{1}{2}m_\text{eff} v^2 = m_e c^2 implies \gamma = 2 for the effective quantum — is there an independent physical reason for this?

Resolved (2026-05-29): Both sub-questions are answered by the two-breath analysis in WIP-12. The exchange is complete (lossless) because it is an anti-phase pair oscillation — the contracting vortex hands its energy to its counter-rotating partner (the inter-sheet layer), so nothing stays orphaned and C4’s two terms are genuine extrema of one mode, not independent reservoirs. And the “\gamma=2” is not a Lorentz factor (the actual \gamma\approx1.58 at 0.776c, and the BEC dispersion E^2=\mu^2+c^2p^2 does not map onto relativistic kinetic energy): the two is the breathing/pairing factor — the radial breath runs at twice the orbital frequency for an isotropic restoring dress — the temporal face of the WIP-15 pairing-two (\xi^2 = 2\xi_\text{GP}^2).

WIP-15: Chirality-sheet stacking — dimensional repair, the inter-sheet scale, and the Higgs VEV

Status: the framework’s most productive thread, now migrated into the main chapters. WIP-15 began as a narrow bookkeeping worry — the bridge equation’s \xi_\text{SC2}^3 recipe and the 3D forms of SC2 and old C1 did not balance dimensionally — and grew into the chain that produced the inter-sheet spacing d_\text{GJO}, the full decomposition of the cell occupancy, and a near-derived Higgs VEV. Those are results, and they now live in the chapters (map below). This entry keeps only what remains useful: the map, the dead ends worth not re-treading, and the handful of genuinely open problems — the framework’s deepest unfinished business. The full blow-by-blow derivation history has been retired to git.

Where the results now live

Result	Lives in
Why 2D math works in a 3D medium (Blatter \Lambda=\xi)	Higgs Field § The open problem — and recent progress
C1/SC2 are inner-scale identities (Volovik speed, Compton clock); no \omega_0	Bridge Equation § Route 2 note · § How the constraints fix everything
Inter-sheet spacing d_\text{GJO}=\xi\sqrt{\ln(\xi/\xi_\text{GP})/4\pi}	Substrate Particles § The Vertical Scale
Packing-fraction decomposition f=(n_v^{(2D)}\xi^2)(\xi/d_\text{GJO})\,\varepsilon_\text{chirality}	Bridge Equation § Derivation Status
The \sqrt2 as pairing / Majorana (S_M=\tfrac12\ln2)	Substrate Particles § The Lattice Breathes in Pairs
Higgs VEV v=\sqrt{8\pi\,m_\text{eff}^2 c^4\,\nu}=246.1 GeV	Higgs Field § Near-derived: the Higgs VEV
8\pi=2\times 4\pi_\text{SC2}; the 2 = radiation-EOS weight	Spacetime § Pressure as a gravitational source
Cooper-pair anti-phase breathing = the Nambu pairing-2	Conductors § The pairing mechanism
The 4\pi = Einstein-Hilbert normalization = exact LI = Step A	Bridge Equation § What the 4π rests on · WIP-10

The settled chain in one breath. The dimensional “mismatch” that started it all dissolved: written honestly, old C1 is the Volovik speed c=\hbar/(m_1\xi) and SC2 is the Compton clock \Omega_v=2m_\text{eff}c^2/\hbar — two inner-scale identities, each balanced, neither containing \omega_0; the [\text{m}^3] junk and the \sim10^4 “projection factor” earlier drafts chased were both artifacts of inserting the outer rotation \omega_0 where it does not belong (\omega_0 is gravity-fixed, not fixed here). The inter-sheet spacing closed in closed form, d_\text{GJO}=\xi\sqrt{\ln(\xi/\xi_\text{GP})/4\pi}\approx16–19\;\mum, carved by the Glaberson–Johnson–Ostermeier axial-flow instability (three classical-fluid results, \kappa_q cancelling, no fitted coefficient). The cell occupancy decomposed completely, f=(n_v^{(2D)}\xi^2)(\xi/d_\text{GJO})\,\varepsilon_\text{chirality}=4\pi/(K\sqrt2) with \varepsilon_\text{chirality}=\sqrt{\pi\ln2}/K — every factor a substrate identity. Half-integer winding forces a paired ³He-A-class condensate (a scalar BEC admits only integer vortices), and the recurring factor of two is the Nambu/Cooper pairing: the line-tension logarithm is the Majorana zero-mode entropy, \ln(\xi/\xi_\text{GP})=\tfrac12\ln2=S_M. The marginal point ties the knot — the 2D chiral-p-wave BdG mode is massless only at \mu=0, so “the substrate emits light at c” forces close-packing ⟺ marginal ⟺ exact isotropic LI ⟺ massless sector as one condition. From these the Higgs VEV falls out, v=\sqrt{8\pi\,m_\text{eff}^2c^4\nu}=246.1 GeV (measured 246.22, 0.06%), the chain never referencing v and selecting the electroweak route over cosmology; the 8\pi=2\times4\pi is entirely gravitational (Gauss factor \times radiation-EOS weight). Of the two sub-items that gated exact LI, close-packing ⟺ \mu=0 is closed (dividing the two emergent-light relations gives n_1\xi^3=1 with no free numbers), while the in-plane-vs-vertical light-speed equality is sharpened, not closed — that residue is item 1 below. Finally, the vertical mass-motion (Saffman local-induction) branch is consistent with every gravitational and gravitational-wave observation from sub-millimetre benches to galactic scales, with a falsifiable window only near 0.5–1 mm.

Dead ends worth remembering

Readings the chain tried and discarded; each is kept because the way it failed sharpened the right answer.

• The projection factor g\sim10^4 and the \omega_0 “ambiguity” (mutually inconsistent values 4.7\times10^9…8.3\times10^{15} rad/s) — both artifacts of forcing \omega_0 into inner-scale identities that do not contain it. Nothing to derive; they evaporate together, leaving the single gravity-fixed \omega_0.
• The Lawrence–Doniach 7\;\mum saddle — the two-term energy has an extremum there, but it is an energy maximum, not a well. The GJO instability sets d directly; 7\;\mum survives only as an upper bound.
• The QCD strong-coupling VEV (v=2m_\text{eff}c^2\sqrt\nu with m_\text{eff}=m_e/\alpha_s) — no electroweak justification. The electroweak m_e/\alpha_{mf} works once the prefactor “5” is recognized as \sqrt{8\pi}=5.013; no QCD enters.
• The Tkachenko shear-modulus 8\pi — numerically excluded (the physical c_{66} uses m_1 at \Omega_F, the VEV’s uses m_\text{eff} at \omega_0, \sim10^3 apart); the c_{66} that reproduces v is SC2 in disguise, confirming the 4\pi is the Gauss factor, not lattice elasticity.
• Scenario B (boson-induced metric) and Frolov–Fursaev as a prerequisite — the half-integer-winding argument forces the fermion-induced (³He-A / Scenario-F) metric instead, and the 4\pi is automatic for any induced G, so the induced-gravity sum rules are demoted to a falsifiable bonus, never load-bearing.

Genuinely open, in order of reach

Four problems remain. They are tightly coupled: items 1, 3, and 4 are three views of one question — is the substrate’s emergent Lorentz invariance exact — and item 2 is the independent number the gravity sector still owes. Each is stated below as what is open, where it stands, and which approaches remain live.

1. The vertical-cone isotropy — is v_z = c_\perp exactly?

What is open. Exact LI requires the fermion’s Bogoliubov–Weyl cone to be isotropic — the emergent light speed along the stacking (z) axis must equal the in-plane speed. In closed form this is the inter-sheet-coupling identity t_\perp=\hbar c/2d_\text{GJO}; the from-scratch continuum BdG restates it as a single ratio, \frac{v_z}{c_\perp}=\frac{2E_F}{\Delta_0},\qquad\text{isotropic exactly at }E_F/\Delta_0=\tfrac12 .

Where it stands. Two things are now settled. (a) The naive anisotropy was an artifact. The old “factor of \sim12” (v_z\sim12\,c) came from pinning the gapless fermion to a Bogoliubov–Weyl node at a Planck-scale lattice momentum — but the substrate is one continuous condensate with a sub-core ripple (d_\text{GJO}<\xi_\text{GP}, cores overlap \sim4\times), not a Josephson stack, so the cone must be read off the continuum 3D chiral-p-wave problem, where it is the O(1) ratio above. (b) O(1) isotropy is automatic. Solving the coupled bulk gap + number equations, close-packing at the single Planck scale (k_F=k_0) lands the isotropy point at an on-shell gap \Delta_\text{node}=0.62\,\epsilon_0 — squarely single-Planck (gap \sim Fermi \sim cutoff), the same regime the in-plane tensor sits at, versus ³He-A’s \sim10^3 hierarchy that would give v_z/c\sim2000. So the framework’s own inputs — close-packing and the single Planck scale, which coincide at k_F=k_0 — make the vertical cone O(1)-isotropic by construction.

What remains. Whether close-packing pins E_F/\Delta_0 to exactly \tfrac12 (full 4D isotropy) or only to O(1). This is a finite-\mu core-marginality question, not the bulk \mu\to0 light condition (strict bulk marginality sends the node to the origin and the vertical dispersion goes quadratic, v_z\to0). Crucially it is the same Bogoliubov–Weyl node whose width fixes the Weinberg angle: the cone aspect (v_z/c_\perp=2E_F/\Delta_0, isotropic at \tfrac12) and the doorway width (\omega_0\tau=2.99\Rightarrow\sin^2\theta_W=0.2312) are two marginal observables of one node — the self-consistent bulk is gapless exactly where the Weinberg chain needs it — so closing one closes both. The one owed number is the doorway width \Gamma=0.67\,\Delta_\text{node}, which the vortex-core run does return at the close-packing cell (a/L\approx3.7–5, matching the independent Bloch-lattice constant, with the marginal doorway orbital’s 90\%-containment radius r_{90}=1.94\,L giving geometric close-packing at a_v=3.9\,L inside that window). The residue is the \sim15\% between geometric core-touch and the dynamical crossing, plus the core compactness itself — owed by item 3.

Approaches. The density-ripple route is closed negatively: the stacked-array GP profile is the cnoidal soliton lattice, and even at full depth (node-touching, a rigorous upper bound) it is \sim2000\times too shallow to set v_z=c (v_z\approx16\,c), because the sub-healing-length period a=d_\text{GJO}/2\ll\xi_\text{GP} makes the recoil energy dwarf the interaction. The sole surviving route is shared-condensate self-consistency (item 3) — the isotropic in-plane phase stiffness dragging the fermion cone to v_z=c with no deep ripple.

2. f_\text{cross} and the outer velocity v_L=\omega_0\xi (the 750 km/s).

What is open. f_\text{cross} is the boundary-transit probability that fixes the outer rotation \omega_0 in the gravity sector; v_L=\omega_0\xi\approx750 km/s is currently back-solved from G, not predicted. Deriving either would turn the galactic critical velocity into a prediction.

Where it stands. G=f_\text{cross}\,v_L/4\pi slaves f_\text{cross} to v_L, so \omega_0, v_L, v_L/c\approx2.5\times10^{-3} and f_\text{cross}\approx1.1\times10^{-15} are one number read four ways. The reduction is now sharp on both faces of the underlying vortex-friction function: the dissipative face \tfrac12\sin2\delta_0 gives the electron’s \alpha_{mf} at the inner rim, and the reactive face \mathcal C=\sin^2\delta_0 read at the outer scale is f_\text{cross} — so gravity’s leak and the electron’s mass are two coefficients of one function, with a built-in mechanism for gravity’s weakness (in the clean limit \omega_0\tau\to\infty spectral flow is blocked, \mathcal C\to(\omega_0\tau)^{-2}\to0). This turns the owed velocity into one owed scattering phase, \delta_0^\text{outer}\approx2.5\times10^{-3} rad, computable by the same BdG/Sonin machinery that fixed the inner 18.5°. Full development in The Outer Rim Onset.

What remains. The phase is reachable but not selected. The outer rim is the clean end \omega_0\tau\approx c/v_L\approx400 of the response, but every probe — the BdG width sweep, the GJO fold-stack, and a direct no-go — finds no feature there: the vortex sector is monotone in \omega_0 and cannot self-select it. The selector must therefore be external, and the one external relation, gravity, is cubic: v_L=(4\pi\,\nu\,c^2 G)^{1/3}, which predicts v_L from G and the condensation number \nu, owing only the standing 2D→3D unit projection (residual \approx6.7) already on the books for f_\text{cross}.

The projection is now pinned, and it collapses two owed items into one. The residual factors identically as R=\frac{\mathcal C/\nu}{4\pi G/v_L}=\frac{v_L^{\,3}}{4\pi G c^2\nu}=\frac{c}{G\nu^2}\cdot Q,\qquad Q\equiv\frac{\nu}{4\pi(c/v_L)^3}\approx1.03, so (i) the projection factor is P=G\nu^2/c (units \text{m}^2\text{kg}^{-1}\text{s}^{-1}) — the “6.7” is the SI reading of a dimensionful quantity, the direct source of the cubic’s broken units — and (ii) the second factor Q is the previously-quarantined “coincidence” \nu\approx4\pi(c/v_L)^3 (to \approx3–4\%). Its opposite exponent sign, once read as disqualifying, is exactly what makes the two relations multiply to isolate P: the coincidence and the projection debt are one fact. Both then reduce to a single clean-units law, \boxed{\,v_L=c\,(4\pi/\nu)^{1/3}\,}\quad\Longleftrightarrow\quad(\omega_0/\omega_1)^3\,\nu=4\pi, which already gives v_L\approx740 km/s (-1.3\%) from c and \nu alone — no G, no projection — and reads as the outer-rotation phase-space volume equals the solid angle 4\pi per condensed dc1 quantum. Deriving this one relation fixes P=G\nu^2/c, makes the gravity cubic exact, and demotes G to a consequence. The same clean-units law is also the outer-rim dual of the bridge equation’s Route 2 scaffold: feeding \nu=4\pi(c/v_L)^3 into the definitional identity \xi=\nu\,\hbar/(m_\text{eff}c) gives a length form \xi_\text{SC2}=4\pi\hbar c^2/(m_\text{eff}v_L^3) (Outer Reach § One count, two costumes) — but this is the same coincidence times a length, not an independent repair of the broken cube-root: the \nu appears cubed in the velocity relation and linearly in the length relation, one duality read two ways. Caveats: R is \nu^2-sensitive (4.9 at \nu=9.6\times10^8, 6.7 at \nu=8.3\times10^8), so “6.7” is not sharp; and P=G\nu^2/c still contains G, so it is a diagnostic identity, not yet a forward closure — the owed step is a mechanism for the solid-angle-per-condensate reading. (Verification: scripts/route_a_selection_cubic.py §X.)

The mechanism reading is sharpened, and the areal-vs-volumetric knot is resolved (2026-07-03; scripts/route_a_landau_rim.py). Two conflated debts — the owed Landau mechanism and the areal/volumetric ambiguity — turn out to be one insight, and the target narrows to a single O(1) coefficient. (a) The mechanism is not the textbook vortex-ring Landau velocity. The effective quantum has radius R_\text{eff}=\nu^{1/3}\xi (close-packing volume V_\text{eff}=\nu\xi^3); the ring speed \tfrac{\kappa}{2\pi R}[\ln(8R/\xi)-2] at R_\text{eff} overshoots \sim3\times (\approx2200 km/s) and carries a log — the wrong object. What is clean and log-free is the bare one-quantum circulation speed at the quantum’s rim, v_\text{circ}(R_\text{eff})=\hbar/(m_1R_\text{eff})=\kappa/(2\pi R_\text{eff})=c/\nu^{1/3}\approx319 km/s, and the whole clean-units law is v_L=(4\pi)^{1/3}\,v_\text{circ}(R_\text{eff}) with (4\pi)^{1/3}=2.325: the outer rim is the effective quantum’s rim circulation speed, dressed by a bare solid angle. The owed piece drops from “a Landau spectrum” to one solid-angle coefficient on the rim circulation speed — and that coefficient is not a Kopnin–Kravtsov scattering output: it is the Gauss/Einstein–Hilbert normalization of WIP-10 Step A, where the lesson already banked — this 4\pi is “automatic once the induced action is exactly Einstein–Hilbert, so the heat kernel never produces it directly” — governs the outer-rim twin verbatim (a vortex-core solid-angle integral is the heat kernel’s analog here). So Route A owns the -1/3 exponent; Step A owns the 4\pi coefficient, and chasing (4\pi)^{1/3} inside the scattering integral re-fights a battle already settled at the level of covariance. (b) The count is volumetric; the fold-stack is its 2D projection. Neither 400 nor 20 is the fundamental linear scale — R_\text{eff}=\nu^{1/3}\xi\approx940\,\xi is; “400” =c/v_L=\omega_0\tau=R_\text{eff}/[(4\pi)^{1/3}\xi] is the radial spectral-flow rung count and “20” =\sqrt{400} is the fold-stack’s in-plane side. The clean-units law’s cube is the fundamental (3D) count; the outer-rim chapter’s areal \omega_0\tau=20^2 is the same object’s boundary cross-section. The geometric form V_\text{eff}=4\pi\ell_L^3 (\ell_L\equiv c/\omega_0=(c/v_L)\xi, the outer causal length) makes the two faces one identity — the effective quantum fills the outer-rotation causal volume — and exhibits the 4\pi as a bare solid angle arising geometrically, without G (the same Gauss 4\pi of WIP-10 Step A). So items 3 and 4 of the outer-rim breadcrumbs are one insight: the volumetric reading fixes the -1/3 exponent, the solid angle is the coefficient owed to Step A, and closing it is the same 4\pi as the bridge equation — reinforcing that the clean-units law can stand G-free (the item-6 endgame of demoting G to a consequence). Two checks make the identification sharp: (i) the coefficient is 4\pi, exactly 3\times the ball volume \tfrac43\pi — the “fills its causal ball” reading gives c/v_L=(3\nu/4\pi)^{1/3}\approx583 (wrong) while “4\pi steradians \times radial cell \ell_L^3” gives \approx400 (right), so it is Gauss over a cubic cell, not close-packed spheres; and (ii) the residual Q\equiv\nu/[4\pi(c/v_L)^3]\approx1.03 must converge to exactly 1 as \nu and v_L tighten — “4\pi exactly” versus “\approx12.6” is the decisive falsifier and the live near-term check.

Two routes feed this: Route A (compute \delta_0^\text{outer} from vortex-core scattering, equivalently derive the clean-units law) and Route B (fix \omega_0 from the Hubble bias H_0, making a_0\sim cH_0 one cosmological bias). Route B currently carries no predictive content — \omega_0/H_0 is algebraically forced (“consequence, not cause”) — so closing the clean-units law v_L=c(4\pi/\nu)^{1/3} is the live path.

3. The from-scratch BdG → GP self-consistency.

What is open. Solve the Bogoliubov–de Gennes quasiparticle problem and the Gross–Pitaevskii order-parameter profile together in one shared condensate, tracking the particle–hole 2 through the pair-field healing length. This does double duty: it makes the triply-sourced \sqrt2 airtight, and — now that the density-ripple route is closed — it is the sole surviving route to item 1’s isotropy (the in-plane-pinned phase stiffness dragging the fermion cone to v_z=c).

Where it stands. The bulk is done. The uniform 3D chiral-p-wave gap + number equations are well-conditioned (the gap vanishes only on an integrable polar line node) and solved to 4 digits, giving the full crossover — weak coupling reproduces the ³He-A hierarchy (E_F/\Delta_0\approx26), strong coupling drives \mu down through isotropy (E_F/\Delta_0=\tfrac12) into the BEC regime — and confirming that O(1)-isotropy is automatic at the single Planck scale.

What remains. The inhomogeneous vortex-core gap equation — a non-local kernel with non-convergent iteration — which is what would derive E_F/\Delta_0 (and settle the exact \tfrac12) rather than read it off the bulk. Plus the \sqrt2 / particle–hole-2 GP-side healing-length bookkeeping. This is the critical-path computation for items 1 and 4 and the VEV.

4. Exact emergent Lorentz invariance (= bridge Step A).

What is open. That the substrate’s induced gravity is exactly Einstein–Hilbert, \int\sqrt{-g}\,R — the deepest thread, and the same problem behind both the bridge equation’s 4\pi and the Higgs VEV’s 8\pi. It splits cleanly into two layers.

Layer (i): one shared light cone / no birefringence. Reduced to item 1 — the fermion Weyl cone matching the homogenization-protected dc1 phonon cone. The in-plane Seeley–DeWitt tensor is now complete: the four Nambu channels exhaust the two-derivative content — isotropic g^{ij}, a time–space g^{0i}=v_s that is just the superflow (zero in the homogenized bulk), and a single non-covariant scalar \lambda_V^2k^4 — so all that survives of layer (i) is the lone vertical z-leg velocity, which is item 1.

Layer (ii): Einstein, not merely Lorentzian. The one-loop-dominance / Barceló–Liberati–Visser decoupling condition — the non-covariant induced terms must stay subdominant to \int\sqrt{-g}\,R. Its controlling factor is computed: at the marginal point the non-covariant/covariant ratio is O(\lambda_V^2)=O(1/4), set by close-packing’s single Planck scale (\lambda_V=O(1/2)), versus ³He-A’s \sim10^6 hierarchy that makes its induced gravity non-Einstein. What remains is this beyond the leading coefficient — and it is a shared problem with all of induced/analog gravity, not unique to the substrate. Supported empirically by GW170817 (c_\text{GW}=c to 10^{-15}).

The punchline of the whole chain stands: the bridge equation’s exact 4\pi and the Higgs VEV’s 8\pi are one problem — exact emergent Lorentz invariance — so the framework’s two hardest open threads are a single thread, now reduced to the finite-\mu core-marginality of one Bogoliubov–Weyl node.

Sources: Blatter et al. (1994), Rev. Mod. Phys. 66, 1125; Clem (1991), Phys. Rev. B 43, 7837; Lawrence & Doniach (1971); Donnelly (1991), Quantized Vortices in Helium II; Sonin (2016), Dynamics of Quantized Vortices in Superfluids, Chs. 3, 6, 8, 9; Kopnin & Kravtsov (1976); Barceló, Liberati & Visser (2001), gr-qc/0111025; Jacobson (1995), Phys. Rev. Lett. 75, 1260 and (1994), gr-qc/9404039; Read & Green (2000), Phys. Rev. B 61, 10267; Ivanov (2001), Phys. Rev. Lett. 86, 268; Volovik & Mineev (1976), JETP Lett. 24, 561; Volovik, The Universe in a Helium Droplet. Key scripts: bdg_vertical_cone_gap_selfconsistent.py, bdg_marginality_bridge.py, bdg_node_to_vortex_width.py, bdg_corepacking_closure.py, bdg_vortex_magnetic_bloch.py, gp_stacked_array_profile.py, route_a_selection_cubic.py.

WIP-16: Derive z_b from relaxation dynamics

The bulk deficit onset redshift z_b = 2.20 in the DESI dark energy profile should not be a free parameter — it should follow from the condition H(z_b) \cdot \tau_\text{relax} \sim 1, where \tau_\text{relax} is the substrate’s cosmological relaxation timescale. The key question is which relaxation mode governs the response.

Three candidate modes exist. Diffusive relaxation (\tau \sim L^2/D with D = \hbar/(2m_e) from C2) gives \tau \sim 10^{30} yr for horizon-scale distances — far too slow. Propagative relaxation at the Tkachenko speed (c_T \approx 9 km/s from WIP-13) gives \tau \sim 10^{15} yr — still too slow. The third possibility is a collective hydrodynamic mode: a global adjustment of the vortex lattice parameters (\omega_0, \xi) in response to the changing expansion rate, with a timescale set by the lattice’s elastic response rather than by signal propagation across the horizon.

An alternative interpretation avoids relaxation dynamics entirely: z_b may simply track the matter-to-\Lambda transition epoch. The value 2.20 falls near where \rho_m/\rho_\text{total} begins declining significantly. If the disequilibrium tracks \Omega_\Lambda/\Omega_m, the Lorentzian step form follows naturally with z_b set by the Friedmann dynamics alone.

The physics is captured by a relaxation ODE: d(\delta T)/dt = -\delta T/\tau_\text{relax} + \alpha H(t). In steady state, \delta T_\text{ss} \propto H\tau_\text{relax}, so \rho_\Lambda \propto (\delta T)^2 \propto H^2. But H^2 also determines the expansion, creating a self-consistent feedback loop whose fixed point determines f(z). The Friedmann equation with \rho_\Lambda(z) = \rho_\Lambda(0) \cdot f(z) must be solved simultaneously with this ODE — and z_b emerges from the coupled system.

If z_b is derivable, the dark energy background becomes a zero-parameter prediction: C = 1 from Volovik self-tuning, z_b from relaxation dynamics. Only the crust parameters (B, z_s) remain free — and those describe the previous cycle’s boundary, inherently unpredictable from within this cycle.

Sources: desi-dark-energy-crust-equations.qmd DE5, gravity-equations.qmd G4-G5, agent-constraint-system.qmd C7

WIP-17: Crust energy budget consistency

The crust amplitude B = 1.88 implies a specific total energy absorbed from the previous cycle’s remnant. At its peak (z = z_s), the crust energy density is \Delta\rho \approx 0.44 \times \rho_\Lambda(0) \approx 2.6 \times 10^{-27} kg/m^3 — of order \rho_\text{DM}. This is not a coincidence to ignore.

Three consistency checks are needed. First, the total crust energy (integrated over the crust shell’s comoving volume) should match the nucleation barrier energy from the dc1 free energy landscape (breadcrumb 1 in universe-that-boils.qmd). Second, the crust energy should scale as the surface area of the previous cycle’s bubble — it is a boundary energy, not a bulk energy. Third, the exponential decay (e^{-z/z_s}) should be derivable from the crust’s physical thickness and the bubble wall velocity; the decay scale z_s encodes both.

Update (2026-05-31): the crust budget and the cosmological-constant residual are one inherited number. The peak crust density \Delta\rho \sim \rho_\text{DM} is not a loose coincidence — it is the substrate’s single energy-density scale appearing again. The same scale fixes the dark-energy background: \rho_\Lambda^{1/4} = 2.24 meV \approx m_1 c^2, i.e. \rho_\Lambda \approx \rho_\text{DM} = (m_1c^2)^4/(\hbar c)^3 by close-packing (see Gravity § The residual). Read in the substrate’s own units the present disequilibrium is order unity, \delta T/T_c|_\text{substrate} = \sqrt{\rho_\Lambda/\rho_\text{DM}} \approx 1.6 — and that order-unity value is measured directly as the moraine wake we still sit in (f(0) = 1.25, z_\text{harm} = -0.25). So the crust amplitude B, the bulk-deficit onset z_b (WIP-16), and the cosmological-constant residual are not three independent fit parameters but three faces of one inherited quantity: how completely \mathcal{B}^{-1} relaxed before our bubble nucleated. The nucleation-barrier calculation (breadcrumb 1) that check #1 calls for would, if it sets that relaxation depth, set all three at once.

Sources: desi-dark-energy-crust-equations.qmd DE8, universe-that-boils.qmd breadcrumb 1, gravity.qmd § The residual

WIP-18: Full Friedmann self-consistency with f(z)

The current DESI fit uses the standard \LambdaCDM H(z) for the reference curves. This is a bootstrapping approximation — the model modifies \rho_\Lambda(z), which modifies H(z), which modifies the BAO distance scales against which the model is tested. A fully self-consistent analysis would solve the Friedmann equations with \rho_\Lambda(z) = \rho_\Lambda(0) \cdot f(z) and verify that the resulting BAO distance scales still match DESI measurements with the same parameters. This is a well-posed numerical calculation: iterate the Friedmann solver with the two-component f(z) until the parameters converge.

Status (2026-06-17): the calculation is done — and it relocates the crust’s evidence. The well-posed iteration above is implemented and converges (scripts/.../friedmann_self_consistency.py, ~13 fixed-point steps). Three findings, in order of importance:

• The distance side was already self-consistent; the leaks were in the model definition. The substrate distance integrals already fold f(z) into H(z) (E2_substrate =\Omega_m(1+z)^3+\Omega_\Lambda f(z), with D_M,D_H integrating it), so the “\LambdaCDM reference curve” worry was only half-true. What actually leaked \LambdaCDM in were three model-definition pieces, now closed: (L1) flatness — the expansion law must read E^2(z)=\Omega_m(1+z)^3+\Omega_r(1+z)^4+(1-\Omega_m-\Omega_r)\,f(z)/f(0); the code’s missing /f(0) meant that whenever the observer sits on a crust crest (z_\text{harm}<0\Rightarrow f(0)=1.28 at the headline parameters) it ran at H(0)/H_0=1.09, a 9% violation of H(0)\equiv H_0. The fix is \Omega_{\Lambda,\text{base}}=(1-\Omega_m-\Omega_r)/f(0) — the Volovik base density is f(0)\times below today’s observed density. (L2) DE-weighting — the crust’s [\Omega_\Lambda(z)/\Omega_\Lambda(z_\text{crit})]^\gamma factor now uses the model’s own \Omega_\Lambda(z)=\Omega_{\Lambda,\text{base}}f(z)/E^2(z), not \LambdaCDM’s. (L3) z_\text{crit} — re-solved from M(z)=H(z)d_\text{proper}(z)/c=1 with the modified H(z).
• Good news — z_\text{crit} is robust. The transcritical crossing moves only 1.588\to1.59 (z_\text{harm}=-0.25) or 1.66 (z_\text{harm}=0) under full self-consistency — a \le5\% shift. The zero-parameter M=1 prediction survives; it was not an artifact of the \LambdaCDM reference.
• Sharp news — DESI BAO alone, treated self-consistently, no longer needs the crust. Re-fitting with free (H_0,\Omega_m,B) under the corrected expansion law drives the crust amplitude B\to0: the fit lands essentially on \LambdaCDM (\chi^2\approx11 at H_0\approx69, \Omega_m\approx0.31; the \LambdaCDM re-fit gives \chi^2\approx10.3 on the 13 DR2 observables). A B-scan confirms \chi^2 rises monotonically with B for both z_\text{harm} values. The bootstrap’s apparent crust (B=0.28) was partly the unnormalized f(0)=1.28 masquerading as an effective local-H_0 offset; once flatness is enforced and H_0 is free, the data prefers a slightly higher H_0 and a flat profile. This is consistent with the wider literature: DESI BAO alone is compatible with \LambdaCDM — the evidence for evolving dark energy (and hence for any crust) lives in the joint fit, once CMB and SNe pin \Omega_m h^2, r_d, and the low-z H_0(z) descent. So the self-consistent treatment relocates the crust’s support from the BAO distances onto the Jia H_0(z) reconstruction and Pantheon+ — which must now be re-run through the same self-consistent solver (the headline \chi^2=10.21, H_0=71.8 result is DESI + Jia, not BAO alone). The model-independent cosmic-chronometer H(z) set is too noisy to decide: the crust at H_0=71.8 fits it marginally better than Planck-\LambdaCDM (\chi^2\approx16.7 vs 18.1 on 35 points), within the scatter.

The joint self-consistent fit is now done — the crust survives, the headline improves, and the Hubble mechanism is reframed (2026-06-19). Items (i) and (ii) above are complete (scripts/.../friedmann_joint_selfconsistent.py, fit_freeform_selfconsistent.py, run logs alongside). The flatness-corrected expansion law was pushed through the joint DESI BAO + Jia H_0(z) likelihood, with a Planck \Omega_m h^2 prior, two ways:

• The chapter’s headline freeform fit survives self-consistency and improves. Closing the leaks and re-running the 15/22-knot freeform spline gives \chi^2_\text{total}\approx9.0 (fixed-Planck background) to 9.5 (free H_0 + CMB prior) — better than the bootstrap 10.21 — against \LambdaCDM’s 886 (Planck-fixed) / 68 (free H_0, the fair joint baseline). \Omega_m h^2 stays exactly Planck (0.143), and the crust holds at f(0)\approx1.15–1.25 (>1). So the chapter’s central result is robust to closing L1/L2/L3 — the freeform spline reproduces the descending Jia H_0(z) by sculpting the low-z crust shape, not by the flatness leak.
• The relocation thesis is confirmed; z_\text{crit} is robust. BAO-alone still drives B\to0 (the crust’s evidence lives in Jia); z_\text{crit} moves only to 1.59–1.66 under full self-consistency. The stiff 3–4-parameter physical DSW profile (friedmann_joint_selfconsistent.py) reaches only \chi^2\approx33–48 and is forced to H_0\approx72 — it cannot make the sharp low-z crest the freeform spline can, which is why the chapter pivoted to the spline.
• The modeling fork (iii) is decided — crust-artifact reading. The bootstrap’s “H_0(\text{local})=71.8 vs background 67.4” decoupling was the flatness leak (unnormalized f(0)=1.25 inflating E^2(0)\approx1.17). Self-consistently H_0(\text{local})=H_0(\text{background}) by construction, and the fit keeps the true H_0 Planck-consistent (\sim67–70): the descending Jia H_0(z) that a \LambdaCDM observer reads as a high local H_0 is reproduced as the crust’s reshaping of low-z distances. So f(0)>1 (a crust crest near today) makes the reconstructed low-z H_0(z) exceed the true H_0 — the Hubble tension as a crust artifact — while the true H_0 and \Omega_m h^2 stay at Planck. This is the reading carried into the DESI chapter and predictions.

Residual (non-blocking). Pantheon+ is not yet folded in (the SH0ES release is not in-repo; DESI+Jia is the headline pair). And the very-low-z crust crest the freeform sculpts to hit the tight Jia z=0.1 point is the least-constrained part of the shape — whether it maps to the distance-ladder (SH0ES) local rate, or stays a reconstruction feature, is what DESI DR3 binning at z=0.3–0.7 would decide.

Sources: data/desi/bao_data/friedmann_self_consistency.py, friedmann_joint_selfconsistent.py, fit_freeform_selfconsistent.py (+ run logs)

WIP-19: Shape predictions for Euclid/Roman

The two-component model predicts specific non-CPL features in w(z) that are invisible to current DESI precision but testable by upcoming surveys. Three signatures distinguish it from all monotonic dark energy models:

• A local maximum in w near z \approx 0.1-0.2, from the crust’s rising edge.
• A minimum in w near z \approx 1.5, from the bulk deficit’s steepest slope.
• An asymptotic approach w \to -1 from below at z > 3 — not from above.

These features are structural predictions of the two-component decomposition: no monotonic quintessence, phantom, or CPL parametrization can reproduce all three simultaneously. Computing the expected measurement precision of Euclid and the Nancy Grace Roman Space Telescope for resolving these features would establish whether the model is falsifiable on a 5-10 year timescale.

Sources: desi-dark-energy-crust-equations.qmd DE6, DE9

WIP-20: Nonlinear structure growth with MOND-modified Poisson equation — closed, S_8 = 0.816 stands

The S_8 prediction (C16) currently uses a linearized growth equation with a modified G_\text{eff}(z). This is a placeholder. The substrate’s gravity is governed by the MOND field equation (GD4):

\nabla\cdot\left[\mu\!\left(\frac{|\nabla\Phi|}{a_0}\right)\nabla\Phi\right] = 4\pi G\,\rho_b

which is nonlinear in \nabla\Phi. Slotting a modified G into the standard linear growth equation does not capture the full dynamics.

The linearized result gives S_8 = 0.816 with \eta_\text{crust} = 2\alpha_{mf}^2 = 0.181 — zero new parameters, landing just above the edge of weak lensing survey measurements (0.76-0.79). Four numerical passes at the nonlinear piece (below) sharpen what the leading correction does — and the later ones correct the first. The bore’s tidal field splits into an isotropic part and a traceless shear, and they pull in opposite directions: a separate-universe calculation shows the isotropic part enhances growth at second order (a convex, super-sample response), while only the anisotropic shear suppresses it; a 3D-geometry bound shows the shear cannot reach the band; and a direct impulsive-heating calculation shows the one channel where “the crust breaks up structure” is rigorous physics is real but quantitatively negligible (\approx0.2\% of binding energy). So 0.816 remains the headline linear value, and no parameter-free gravitational channel returns it to the WL band — the nonlinear correction is a genuine competition rather than a clean one-sided suppression, the parameter-free return-to-band the first pass reported was an artifact of treating the whole (mostly isotropic) tidal field as suppressive, and the one remaining lever — the deferred MOND sector — is now also computed (fifth pass below) and does not rescue it: a literal MOND-modified P(k) pushes S_8 up, not down, so the problem is now closed with S_8 = 0.816 standing as the framework’s honest prediction.

Background self-consistency is now folded in (2026-06-19) — and it lifts S_8 to 0.816. The growth ODE’s substrate E^2(z) in scripts/substrate_galactic.py (solve_self_consistent + compute_S8_sc) was switched from the bootstrap law to the flatness/DE-weight/z_\text{crit}-corrected one of WIP-18; G_\text{eff} (the coupling route) stays out of Friedmann and is untouched. The result moves S_8 = 0.7923 \to 0.816. The shift is almost entirely the flatness leak (L1 alone gives 0.814; L2+L3 add +0.001) — the same leak that faked the high local H_0. Flatness-normalizing the base density (\Omega_{\Lambda,\text{base}} = 0.548 vs 0.685) lowers the dark-energy friction over the growth epoch, so the background route raises \sigma_8 and partly undoes the coupling suppression: the two G_\text{eff} features reach 0.783 on a flat background, the self-consistent background lifts the net to 0.816. So S_8 = 0.816 sits just above the WL band — a more modest tension relaxation than the bootstrap 0.792, and the S_8 relaxation now trades off against the Hubble-tension dissolution through the shared f(0) crest (see predictions § 2c). This is complication (c) flagged in the chapter’s S_8 Status note, now resolved; the nonlinear MOND piece below is what remains.

First pass — the leading nonlinear channel, modelled as a sign-definite sink (2026-06-19; scripts/wip20_bore_disruption.py; partly corrected by the second pass below). The smooth-envelope linear calc misses a third channel, distinct from the two it carries (background friction + G_\text{eff} coupling). The undular bore is a frozen spatial \rho_\Lambda pattern — the moraine is debris at fixed comoving radii, not a quintessence field with c_s^2=1 that would smooth itself — so it sources a real gravitational tidal field. With w\approx-1 the dark energy gravitates through \rho+3p=-2\rho_\Lambda (repulsively), and Poisson gives a scale-independent tidal tensor of size \mathcal{T}/H^2 = 3\,\Omega_\Lambda(z)\,\delta_\Lambda(z), \delta_\Lambda \equiv f(z)-1 — parameter-free from the fit, and order unity at the S_8 epoch (peak \mathcal{T}/H^2 \approx 1.3 at the z\approx0.23 carrier crest). The bore’s density gradient is comparable to matter’s own self-gravity exactly where S_8 is measured.

At first order the bore is nearly invisible to growth: swapping the steep 15-knot spline in for the smooth DSW envelope moves the linear S_8 only -0.008 — the crest-friction/trough-anti-friction near-cancellation the chapter already notes. But the disruptive response is second order — it scales as \mathcal{T}^2, which never changes sign and so does not cancel: like a boat in choppy water, a forming structure loses binding energy whether each density wave pushes it out or lets it fall in. The first pass invoked two sub-mechanisms — static tidal truncation (a crest’s excess repulsion shrinks the tidal radius of an over-density, stripping its outskirts) and impulsive heating (the carrier crests cross in \tau_\text{crest}/t_\text{dyn}\sim0.06–0.11, deep in the Spitzer impulsive regime, so the rapidly changing field heats marginally-bound regions rather than being adiabatically absorbed) — and took both as suppressive. The second pass below revises this: the truncation piece is the isotropic (trace) response, which is actually a separate-universe gain; only the impulsive-shear piece is genuinely sign-definite suppressive.

The coefficient is pinned to an O(1) gravitational efficiency — not \alpha_{mf}-filtered. This is the key distinction from the G_\text{eff} channel. There (\eta_\text{crust}=2\alpha_{mf}^2) the bore energy must couple into the counter-rotating boundary and disrupt its current-phase relation, filtered twice through mutual friction. Here the bore’s resulting gravitational field acts directly on collapsing matter — no mutual-friction filter, so the efficiency is a gravitational O(1), not \alpha_{mf}-suppressed. Modelling the second-order sink as c_\text{imp}\,\mathcal{T}^2 on the growth source, the whole weak-lensing band 0.76–0.79 is spanned by c_\text{imp}\in[0.70,\,2.07], and the impulse-approximation value c_\text{imp}\sim1 gives S_8\approx0.78. This first-pass reading was too optimistic, and the separate-universe pass below corrects it: the sink c_\text{imp}\mathcal{T}^2 uses the full tidal field \mathcal{T}, but \mathcal{T} is dominated by its isotropic (trace) part — and the isotropic part is not sign-definite suppressive. So the apparent “natural O(1) lands in the band” rested on counting the isotropic tidal power as a loss when it is in fact (at second order) a gain.

Second pass — the separate-universe / peak-background split decomposes the sign, and it is a competition (2026-06-19; scripts/wip20_separate_universe.py). Rather than impose a sign-definite sink, this pass computes the response of small-scale growth to the bore as a local long-wavelength environment, and splits the tidal tensor into its two physically distinct pieces — the trace (isotropic) and the traceless shear:

• Channel A — isotropic (separate-universe). It enhances. A local dark-energy excess raises the local H, adding Hubble friction and shrinking the matter source — but the suppression saturates (growth freezes once DE dominates), so the response D(\delta_\Lambda) is convex. A convex response to a zero-mean, finite-variance modulation gives a net second-order gain — the same positive super-sample-covariance by which superclusters boost their embedded structure. Computed on the fit (curvature \tfrac12\,d^2\ln D/d\delta_\Lambda^2 = +0.038, growth-weighted \langle\delta_\Lambda^2\rangle = 0.28), this lifts S_8 by \approx+0.009. The trace of \mathcal{T} — the bulk of what the first pass fed into its suppressive sink — therefore works the wrong way.
• Channel B — anisotropic (traceless shear). It suppresses, but it is small here. Only the shear is sign-definite suppressive (Zel’dovich delay of collapse + impulsive heating). For the bore’s concentric-shell geometry the traceless shear \Sigma(\chi) = \mathcal{S}(\chi) - \langle\mathcal{S}\rangle_\text{enc}(\chi) carries only \approx13\% of the tidal power (\langle\Sigma^2\rangle = 0.0034 vs \langle\mathcal{S}^2\rangle = 0.026). To reach the WL band against the Channel-A lift now needs an impulse coefficient c_\text{tid}\sim12–19 — an order of magnitude above O(1), not the natural value the first pass implied. (Consistency: feeding the full trace power back in, ignoring the lift, reproduces the first pass’s c_\text{imp}\in[0.70,2.07] — the machinery agrees; the disagreement is one of physics bookkeeping, not numerics.)

The honest net: standard perturbation theory on the spherical bore does not deliver a parameter-free return to the WL band. S_8 = 0.816 stands as the linear value; the leading nonlinear correction is a competition whose isotropic part is a (computed, small) gain. Two caveats keep the suppressive channel alive and are the natural next targets: (i) the 13\% shear fraction is a lower bound — it is specific to a perfectly concentric moraine, the geometry that minimises shear; a realistic 3D bore (or an isotropic-random tidal field, where shear carries \gtrsim the trace power) would raise it substantially. (ii) The MOND enhancements deferred below (a_0(z), modified \mu) are not in this pass and act in the suppressive direction. So the suppression is physically real but its magnitude now rests on bore geometry and the MOND coupling, not on a generic O(1) — a weaker and more honest claim than the first pass.

Third pass — the 3D-geometry boost is real but bounded; shear alone cannot carry the band (2026-06-19; scripts/wip20_shear_geometry.py). Caveat (i) is now quantified. For a statistically isotropic potential field the tidal tensor obeys \langle s_{ij}s_{ij}\rangle = \tfrac23\langle(\nabla^2\Phi)^2\rangle — the traceless shear carries \tfrac23 of the trace variance, against only \approx13\% for the concentric shells. Interpolating the two with an anisotropy parameter \beta (radial-coherent \to isotropic) gives a geometry boost of up to g = \langle\mathcal{S}^2\rangle/\langle\Sigma^2\rangle \approx 7.7 in shear power. That is a large help — it pulls the required impulse coefficient down from c_\text{tid}\sim12–19 to \sim2–4 — but it is not enough. Even at full isotropy (\beta=1, a generous upper bound that assumes the transverse debris has the same order-unity contrast as the radial bore), c_\text{tid}=1 reaches only S_8\approx0.809; the band edge needs c_\text{tid}\approx2.3 and mid-band c_\text{tid}\approx3.7. Two things cap it: the geometry-independent +0.009 monopole lift (Channel A), and the \tfrac23 ceiling on the isotropic shear fraction. And the fitted bore is strongly radial (it is a radially-propagating shock — \beta is small), so the realistic shear sits well below this bound. Conclusion: tidal-shear geometry alone does not return S_8 to the WL band; the band requires the deferred MOND enhancements (a_0(z) extending the modified-gravity regime, the |\nabla\Phi|\nabla\Phi coupling), which now become the load-bearing piece of the remaining suppression rather than a refinement.

Fourth pass — the impulsive-heating channel the first three could not see, computed at last, and it is negligible (2026-06-19; scripts/wip20_impulsive_heating.py). All three passes above integrate the linear growth ODE, and the linear growing mode is adiabatic by construction — it cannot represent the energy a marginally-bound clump absorbs from a fast tidal shock. Pass 1 named this channel (Spitzer impulsive heating, \Delta E = \tfrac16\langle r^2\rangle\,\lVert\mathsf J\rVert^2 \ge 0, sign-definite) and even verified the regime is deeply impulsive (\tau_\text{crest}/t_\text{dyn}\sim0.06–0.11), but then implemented it as a sink on the linear mode — which assumes the sign rather than computing it. This pass computes the heating directly. As cosmic time advances, a fluid element sees the local DE density sweep up a crest then down a trough (the bore is frozen in comoving space, so \rho_\Lambda(t) rises and falls as the universe ages through the bore epoch — the temporal dual of the spatial-shell picture used for the shear). Locally, on the $$10 Mpc clump scale, the Gpc bore is spatially uniform, so the tidal tensor it feels is very nearly the pure isotropic trace — meaning the heating samples the full trace power (the \approx87\% pass 2 found enhancing the linear mode). This is the genuinely interesting regime point: the same trace tide enhances a linear mode that adiabatically tracks the slow background (Channel A), yet heats a bound clump that cannot adiabatically respond — same field, opposite sign, decided by regime, not contradiction. The “choppy water” bookkeeping is explicit: the net impulse over a full crest+trough nearly cancels (the first-order statement), but the heating is the sum of squared per-half-wave impulses \sum_n \mathsf J_n^2, which cannot cancel (here \sum_n\mathsf J_n^2 = 0.022 vs net-squared 0.016, a 1.4\times survival — modest, because one broad low-z half-wave dominates rather than many alternating ones).

The verdict is a clean negative, and it matters. Summing the squared impulses with the Gnedin–Ostriker adiabatic shield (\langle A\rangle\approx0.88, confirming the crossings are impulsive) over the realistic 15-knot spline, the energy injected into a marginally-bound (\Delta_c\approx5.5, turnaround) 8\,h^{-1}Mpc structure is \Delta E/|E| \approx 0.0016 — about 0.2% of its binding energy, parameter-free up to the heating\to\sigma_8 efficiency c_\text{heat}. Tighter structures heat even less (\Delta_c=18\to0.0004; virialized \to0). Mapping to the observed (non-linear) S_8 — note this acts on the weak-lensing observable, where the tension lives, not on linear \sigma_8, so the 0.816 linear value is untouched either way — even a generous c_\text{heat}\sim1 moves S_8 by only \approx-0.001; reaching the band edge would need c_\text{heat}\approx40, mid-band \approx70. No plausible efficiency rescues a 0.2% heating fraction. So the impulsive-heating channel — the one place the “crust breaks up structure” intuition is rigorous, sign-definite physics — is real but quantitatively negligible: the DE tidal field, though order-unity in \mathcal{T}/H^2, acts over only \sim0.1–0.2 e-folds at each late crest, so its velocity impulse is a few percent of a structure’s internal velocity and the heating \propto(\text{few }\%)^2 is three orders of magnitude too small. This closes the last gravitational lever: across the linear competition (pass 2), shear geometry (pass 3), and now non-linear impulsive heating (pass 4), no parameter-free gravitational channel returns S_8 to the WL band. Either the deferred MOND sector does it, or S_8 = 0.816 stands as an honest, mild relaxation — and the gravitational sector is now exhausted as a source of further suppression.

Fifth pass — the deferred MOND P(k,z), computed at last, and it reverses the sign the earlier passes assumed (2026-06-19; scripts/wip20_mond_pk.py). Passes 2 and 3 deferred the MOND sector as the load-bearing suppressive channel — “the band requires the deferred MOND enhancements.” Carrying out the scale-dependent linear growth with the MOND-modified Poisson (GD4), a_0(z) = a_0(0)(1+z)^{3/2}, and the bore as an external-field mode-coupling source — on an Eisenstein–Hu transfer function normalised so \sigma_8^{\Lambda\text{CDM}} = 0.811, the same self-consistent H(z), and the same growth integrator that reproduces the 0.816 baseline — overturns that assumption. MOND’s leading effect on 8\,h^{-1}Mpc growth is enhancement, not suppression, and it is large: those modes sit at g_N \approx 10^{-3} a_0, deep in the MOND regime, where the boost \nu = G_\text{eff}/G is a decreasing function of acceleration — stronger-than-Newtonian gravity grows structure faster. The literal AQUAL boost on linear modes diverges (\sigma_8 \sim 230, the known pathology of static-boost linear MOND that a relativistic completion — TeVeS/AeST — tames but the substrate paper does not carry), and even a bounded boost capped at \nu_\text{max} \in \{1.5, 2, 3\} drives \sigma_8 up, never toward the band. So the very effect passes 2–3 were counting on to deliver the suppression in fact pushes S_8 the wrong way: MOND, taken literally, worsens the tension. This is the same self-correction the earlier passes made — the first pass’s “natural O(1) lands in band” was a sign-bookkeeping artifact; here passes 2–3’s “MOND will supply the suppression” was a sign error in the channel itself. (The \nu_\text{max} cap used here to tame the divergence is not, in fact, ad-hoc: the physical regulator is the v_L coherence threshold, which switches MOND off in exactly the collapsing super-v_L walls and nodes that would otherwise carry the runaway — the substrate’s own version of the relativistic completion the linear pathology needs. See The Cosmic Web § the v_L threshold caps the MOND runaway.)

The one genuine MOND suppression is the bore external-field effect, and it rides on the enhancement. The MOND nonlinearity does supply a sign-definite suppression — exactly the mode coupling complication (2) named: the \simGpc bore’s tidal field g_\text{ext}(z) acts as an external field on a collapsing 8\,h^{-1}Mpc clump, adding in quadrature to its own g_N, raising the argument of \nu and cutting the MOND boost during the crust epoch (the bore partially “Newtonises” the modes — the first-principles form of pass 1’s hand-built \mathcal{T}^2 sink). At the z \approx 0.23 carrier crest the 15-knot spline bore’s external field is 2–5\times the internal field of an 8\,h^{-1}Mpc mode, halving the local boost (R_\text{EFE} \approx 0.44–0.66). But this is a reduction of the enhancement, not a suppression of the baseline: it claws back part of MOND’s over-growth and can never pull S_8 below the no-MOND 0.816. Pinning pass 1’s scanned c_\text{imp} this way gives it a definite physical content — a boost-reduction coefficient computed from the bore field, not a free knob — but with no suppressive counterpart once the sign of the MOND boost is included. (The three complications the deferred treatment carried are thereby all resolved: (1) a_0(z) deepens the MOND regime at higher z — and so amplifies the over-enhancement, the opposite of help; (2) the |\nabla\Phi|\nabla\Phi coupling is the EFE just computed, carried as the k-dependent R_\text{EFE}; (3) any crust modification of \mu during the crust epoch is subsumed in the same external-field argument.)

WIP-20 is closed (2026-06-19). The five passes exhaust the channels. The gravitational sector cannot reach the band (passes 1–4: linear competition is a near-cancellation with an isotropic gain, shear geometry is bounded below the band, impulsive heating is \approx0.2\% of binding energy). The MOND sector — far from rescuing it — either over-enhances (AQUAL taken literally) or, in any tamed prescription, leaves the scale-independent G_\text{eff} \to 0.816 as the operative linear result while its one suppressive sub-channel (the bore EFE) only trims its own over-growth. The honest verdict is unchanged in its headline and sharpened in its physics: S_8 = 0.816 stands as the framework’s prediction — a mild \sim 1\sigma-high relaxation, traded against the Hubble-tension dissolution through the shared f(0) crest (see predictions § 2c) — and no parameter-free channel, gravitational or MONDian, returns it to the 0.76–0.79 weak-lensing band. What remains is not a calculation but a falsifiable prediction: if weak lensing tightens on S_8 \le 0.79 with small error, the framework’s 0.816 is in genuine tension, because the suppression budget is now demonstrably spent. One scope caveat. This closure is over the crust’s smooth-field channels — the frozen \rho_\Lambda curve f(z) and its gravitational + MOND action on growth. It does not touch the distinct discrete-matter channel — a population of massive relics inherited from \mathcal{B}^{-1} acting kinetically through velocity dispersion and seeding — which is a separate physical mechanism with a two-sided sign, tracked in WIP-31 and still uncomputed. So “0.816 stands” is exact for the smooth crust; whether texture moves it is a genuinely open, separate question.

Sources: desi-dark-energy-crust-equations.qmd DE11-DE14, galactic-dynamics-equations.qmd GD4, GD10, open calc 1, agent-constraint-system.qmd C14, C16; scripts/wip20_bore_disruption.py (first pass), scripts/wip20_separate_universe.py (separate-universe sign decomposition), scripts/wip20_shear_geometry.py (3D shear-geometry bound), scripts/wip20_impulsive_heating.py (non-linear impulsive-heating estimate), scripts/wip20_mond_pk.py (MOND-modified P(k,z), fifth pass — closes WIP-20)

WIP-31: Crust texture — the discrete massive-remnant channel for S_8

WIP-20 closed the S_8 question over the smooth-field channels: the crust modeled as a frozen \rho_\Lambda density curve f(z), acting on growth through its background friction, its G_\text{eff} boundary coupling, its gravitational tidal field (trace + shear), and its MOND-modified P(k) — five passes, no parameter-free return to the weak-lensing band, S_8 = 0.816 declared falsifiable. But the crust was only ever modeled as the relaxed density curve — the “pure substrate” limit. A real moraine is not just a profile: it carries erratics. \mathcal{B}^{-1} need not have relaxed to a featureless fluid before we nucleated inside it — its own massive objects (collapsed cores, compact remnants, the shaved husks) would survive the relaxation as a discrete matter population caught up near the outer reaches of \mathcal{B}^0’s expanding wall. This is a kinetic / discrete-matter channel, physically distinct from everything WIP-20 computed, and it is the calculation the how-far chapter explicitly defers here rather than narrates. This entry is the statistical face of the erratics — the population’s effect on growth. Its individual face — whether one erratic could be caught and identified as a survivor of \mathcal{B}^{-1} by a composition or radiometric-age anomaly — is a separate, complementary program with its own smoking gun (an object dating older than \mathcal{B}^0), developed in Erratics: Catching a Piece of the Previous Cycle.

The sign is genuinely two-sided — and that is the whole content of the entry, because the paper is otherwise careful never to claim a suppression without its sign. A discrete relic population acts on 8\,h^{-1}Mpc growth through two competing sub-channels:

• Hot, ablated component — suppresses. The transcritical crossing is a stirring event: the same wash that shaves the husks (why an interstellar visitor like ʻOumuamua arrives ablated, as if through an energy bath empty space cannot supply — solar-system-boundaries) heats a fraction of the relics to high velocity dispersion. As a warm/hot admixture during the growth epoch this free-streams and suppresses small-scale power — lowering S_8, the direction the WL band needs. This is the “missing small-scale velocity dispersion” the how-far chapter names.
• Cold, surviving cores — enhances. The densest cores survive the wash cold and intact. As rare massive over-densities (“bowling pins”) they seed small-scale clustering (a discreteness/Poisson term), raising S_8 — and simultaneously boosting early structure, aligning with early-structure-formation’s JWST resolution rather than fighting it.

So texture does not automatically improve S_8: the seeding reading pulls the wrong way. The clean resolution is that the crossing sorts the two — hot wanderers carry the suppression and become the wandering shards, cold cores carry the enhancement and become the seeds for \mathcal{B}^0’s structure — so each observational hook (ablated ISOs; early massive galaxies) lands on the right dynamical component, and the net S_8 effect is whichever dominates on 8\,h^{-1}Mpc.

What would settle it. A magnitude estimate needs three inputs the framework can in principle supply: (i) the relic abundance and mass function surviving \mathcal{B}^{-1}’s relaxation — bounded by the same nucleation-barrier / relaxation-depth quantity that sets the crust amplitude B (WIP-17); (ii) the hot-fraction velocity dispersion imparted by the transcritical wash — a Grimshaw–Smyth energy-partition question; (iii) the free-streaming vs. Poisson-seeding balance on 8\,h^{-1}Mpc, a standard warm-dark-matter-plus-discreteness power-spectrum estimate. Until (i)–(iii) are done this stays a framed hypothesis with a two-sided sign, not a number — kept out of the how-far speculative section so that section cannot borrow its credibility, and out of the DESI chapter’s headline because WIP-20’s 0.816 stands until this is earned.

First steps (2026-07-02): the sign collapses, and the magnitude reduces to one velocity. A deliberately standalone order-of-magnitude estimator (scripts/crust_texture_s8.py) — kept out of the canonical substrate_galactic.py so texture cannot become a tuning knob that dials S_8 anywhere — evaluates sub-channels (iii) directly at the 8\,h^{-1}Mpc scale that defines S_8. Two results sharpen the entry from “two-sided, uncomputed” toward a single question:

• The enhancement branch cannot bite at the S_8 scale. The cold-core Poisson/discreteness term contributes \Delta\sigma_8^2 = f_\text{cold}\,(M_\text{core}/\bar\rho_m)\,I_8, with the top-hat white-noise integral I_8 = 1/V_\text{tophat} = 1/[\tfrac43\pi R_8^3] (verified numerically to 0.2\%). Because M_\text{core}/\bar\rho_m is a minuscule volume for stellar or even galactic masses, seeding is \Delta S_8 \lesssim 10^{-4} for M_\text{core}\le10^{12}M_\odot and reaches even +0.01 only for cluster-scale cores (M\gtrsim5\times10^{14}M_\odot at a percent mass fraction). So the “two-sided sign” WIP-31 was careful to carry collapses to essentially one-sided suppression at 8\,h^{-1}Mpc: unless \mathcal{B}^{-1} bequeathed cluster-mass monsters, the seeding (wrong-way) channel is off where S_8 lives, and only the hot free-streaming (right-way) channel remains. This does not contradict the cold cores’ role elsewhere — they can still seed early massive galaxies (early-structure-formation) at scales far below 8\,h^{-1}Mpc; it says only that they do not raise S_8.
• The magnitude reduces to one velocity. The hot/ablated component acts as a warm admixture: \Delta\sigma_8/\sigma_8 \approx -4 f_\text{hot} once its free-streaming length reaches 8\,h^{-1}Mpc — switched on by whether the dispersion the wash imparts is large enough. So the whole magnitude collapses onto a single number, the hot-relic velocity dispersion \sigma_v (input (ii)); everything else (f_\text{hot}, the profile) is either constrained or scales it linearly.

Step 2 (2026-07-02): the Grimshaw–Smyth partition, and a modest, scale-tilted answer. The dispersion \sigma_v is now computed from the GS forced-KdV framework itself (scripts/crust_texture_gs_partition.py; theory in papers/dsw-2016 Sec. 8, Grimshaw–Smyth 1986), anchored to quantities the framework already carries — no new free parameter. Three links:

• The amplitude is pinned, not fit. At the crossing the detuning \Delta = M-1 = 0 is fixed by the zero-parameter z_\text{crit}=1.588, so GS’s headline result applies: the far-field response depends only on \Delta and the forcing F_m, not on the moraine’s shape, with steady transition |A_\pm| = \sqrt{F_m/3}. The crust f(z) we already fit is that response, with fractional amplitude A_\text{resp}\approx 0.25 (present crest f(0)-1) to 0.44 (peak \Delta\rho_\Lambda/\rho_\Lambda, WIP-17); GS supplies the structure (upstream + downstream DSW) and the KdV lead-soliton factor of 2.
• The dimensional bridge is the crux — and it is not c. The DSW carrier rides the substrate signal speed c (that is what M=U/c measures), but a carrier-entrainment \sigma_v\sim A_\text{resp}c would be relativistic and is unphysical for massive relics. Discrete relics couple by direct ablation/drag — the very mechanism the paper already invokes for ʻOumuamua arriving “ablated” — whose velocity is capped by the substrate’s matter-flow ceiling v_L=\omega_0\xi\approx 775 km/s (the WIP-13 outer-rotation / GJO coherence speed, independently cross-checked by the galaxy/cluster split and the Ulysses fast solar wind, 751.5 km/s). So \sigma_v \approx \varepsilon_\text{couple}\,(1\text{ or }2)\,A_\text{resp}\,v_L, with \varepsilon_\text{couple}\!\in\!(0,1] the ablation efficiency — the lone remaining unknown. (The gravitational tidal channel imparts no dispersion to small relics: a \simGpc bore’s tidal field is uniform across a pc-scale core, so by the equivalence principle it moves the whole object together — which is why WIP-20’s tidal passes, correct for an 8\,h^{-1}Mpc clump, do not govern relic dispersion.)
• The number. This gives \sigma_v \sim 100–680 km/s across the coupling/amplitude range. Done properly — with the accumulated comoving free-streaming from injection at z_\text{crit}, not a static-Jeans scale — reaching a full 8\,h^{-1}Mpc needs \sigma_v\approx 1190 km/s, so the wash free-streams only \sim2–7 Mpc and the switch g at 8\,h^{-1}Mpc is partial (\sim0.03–0.25). Net \Delta S_8 \approx -0.001 (conservative: weak coupling, transition amplitude) to -0.041 (optimistic: efficient coupling, lead soliton, f_\text{hot}=5\%) — i.e. texture modestly relaxes S_8=0.816 toward \sim 0.79–0.81, touching the WL band’s upper edge only in the optimistic corner, not reaching its center.

Step 3 (2026-07-02): the scale tilt made a curve, and a data bound on \varepsilon_\text{couple} that beats the ISO reading. Step 2 ended on two loose threads — (a) pin \varepsilon_\text{couple} from how processed arriving ISOs are, or (b) turn the “stronger at few Mpc” remark into a real \Delta P(k)/P curve. scripts/crust_texture_pk_tilt.py does (b), and (b) turns out to do (a)’s job more honestly. Three results:

• The tilt is real, large, and now quantified. Propagating the step-2 \sigma_v range through the same warm-admixture switch used at 8\,h^{-1}Mpc gives the scale-dependent rms suppression \Delta\sigma(R)/\sigma(R)=-4f_\text{hot}\,g(1/R) across R. The falsifiable number is the tilt |\Delta\sigma(3\,\text{Mpc})|/|\Delta\sigma(8\,h^{-1}\text{Mpc})| \approx 3–14\times (optimistic \to conservative dispersion): texture predicts extra small-scale suppression riding above the S_8 offset, not a uniform shift — a scale-dependent signature no smooth-crust or \LambdaCDM channel makes, and one readable in the Lyman-\alpha forest / small-scale \sigma_8.
• (b) subsumes (a): the data already bound \varepsilon_\text{couple}. The hot channel is massive-neutrino-like (\Delta P/P\approx-8f_\text{hot} for k\gg k_\text{fs}), so it is already constrained by existing small-scale power — no need to identify \varepsilon_\text{couple} with a present-day, local ʻOumuamua ablation standing in for a z_\text{crit} crossing (the heuristic leap (a) required). Requiring |\Delta P/P|\lesssim4\% at \sim3 Mpc — the massive-neutrino-calibrated ceiling (\sum m_\nu<0.12 eV \Rightarrow f_\nu<0.0045 \Rightarrow |\Delta P/P|\lesssim3.6\% on cleared scales) — caps \varepsilon_\text{couple}: the optimistic corner (peak amplitude + lead soliton, f_\text{hot}=5\%) needs \varepsilon_\text{couple}\lesssim0.15–0.20, i.e. it is already in mild tension. That is a firmer, cosmological bound on the lone unknown than “how ablated ʻOumuamua looks.”
• The consequence for S_8: bounded from both sides. The strong-suppression corner that alone touched the WL band’s upper edge is the same corner the small-scale data disfavor, so texture’s realistic pull collapses to the modest end, \Delta S_8\approx-0.005 to -0.02: S_8=0.816 relaxes to \sim0.80–0.81, still above the WL-band center. Texture is now bracketed on both ends, and WIP-20’s 0.816 stands — mildly relaxed, not overturned.

Step 3b (2026-07-02): into the actual observable — the Lyman-\alpha 1D flux power — and the bound tightens. Step 3’s ceiling lived in the 3D matter power at a single peg scale. The forest does not measure P_\text{3D}(k); it measures the 1D line-of-sight flux power P_\text{1D}(k_\parallel)=\tfrac1{2\pi}\int_{k_\parallel}^{\infty}kP_\text{3D}(k)\,dk. A first check of data/desi settled a prerequisite: it holds Lyman-\alpha BAO distances (D_H/r_s,D_M/r_s at z=2.33) and full-shape f\sigma_8, not the flux power — so the matching dataset (eBOSS DR14, Chabanier et al. 2019; or DESI DR1, Ravoux et al. 2023) is the one external file worth grabbing. Meanwhile scripts/crust_texture_p1d.py projects the predicted suppression into that frame, and the result is sharper than step 3:

• The projection tightens the bound, in one direction only. Because P_\text{1D}(k_\parallel) mixes in all 3D modes k\ge k_\parallel, the few-Mpc suppression bleeds across the whole measured band — so the 1D constraint can only be equal or stronger than step 3’s single-scale estimate, never weaker. The ablation-efficiency bound falls from step 3’s \varepsilon_\text{couple}\lesssim0.15–0.20 to \varepsilon_\text{couple}\lesssim0.05–0.10 (against the eBOSS/DESI few-percent \LambdaCDM agreement).
• A few-percent hot fraction is excluded almost model-independently. f_\text{hot}\sim few-\% of \Omega_m as a warm/hot relic gives \Delta P/P\to-8f_\text{hot}\sim-40\% on small scales — \sim10\times the massive-neutrino ceiling (f_\nu<0.0045). The only escape is small free-streaming, which is the same \varepsilon-squeeze.
• The key squeeze (robust, cutoff-independent). Relaxing S_8 at 8\,h^{-1}Mpc requires free-streaming to reach \sim8 Mpc; that same free-streaming over-suppresses the 1–3 Mpc Lyman-\alpha band, which the data forbid. The S_8-helping and Lyman-\alpha-forbidden regimes coincide — so texture’s realistic S_8 pull is pinned small, |\Delta S_8|\lesssim0.005, tightening step 3’s bracket toward its floor. (Honesty: the 1D magnitudes are order-of-magnitude — a rigorous P_\text{1D} needs the forest’s flux bias, redshift-space, and thermal-history modeling, and depends on the small-scale window k_\text{cut}; the direction of the squeeze does not.)

So the honest verdict is between the two WIP-20 declared: texture is neither negligible nor a rescue. It relaxes S_8 by a data-bounded amount that the Lyman-\alpha frame now pins to |\Delta S_8|\lesssim0.005–0.01 without a new parameter (step 2’s up-to-0.04 optimistic corner is disfavored, and step 3b’s 1D projection tightens the squeeze toward the floor), and comes nowhere near carrying 0.816 into the middle of the weak-lensing band. What it does leave is a live, falsifiable prediction — the few-Mpc scale tilt — whose sign and shape are now the whole content: any texture strong enough to help S_8 must leave a Lyman-\alpha P_\text{1D} suppression that existing data already bound, and the framework’s own crossing does not appear to supply it. That last claim is no longer an assertion — it is now checked directly against the eBOSS measurement (step 3c).

Step 3c (2026-07-02): confronted with the eBOSS DR14 measurement — texture is bounded to sub-percent, and the tilt is the tell. The predicted suppression is now laid against the actual eBOSS DR14 Lyman-\alpha P_\text{1D} (Chabanier et al. 2019, papers/chabanier), whose linear-power constraints at the pivot k_p=0.009\,(\text{km/s})^{-1} (\approx0.69\,\text{Mpc}^{-1} comoving, z_p=3 — squarely in the tilt band) are \Delta^2_L=0.31\pm0.02 (amplitude, 6.5\%) and n_\text{eff}=-2.339\pm0.006 (log-slope, 0.3\%), both consistent with Planck \LambdaCDM and post-marginalization over the IGM thermal history (so a real suppression is not freely absorbed by T_0(z),\gamma(z)). Texture is tested two ways (compare_to_chabanier()):

• Amplitude test (shape-independent). Texture’s |\Delta P/P| at the pivot must fit inside the 6.5\% measurement: this alone caps f_\text{hot}\lesssim0.8–2.6\%.
• Slope test (the tell — sharper, mildly shape-dependent). Texture’s tilt is a slope shift, \Delta n_\text{eff}=d\ln(1+\Delta P/P)/d\ln k, and n_\text{eff} is measured to 0.3\%. A percent-level hot fraction shifts n_\text{eff} by 0.05–0.20 — an 8–33\sigma deviation. This forces f_\text{hot} to per-mille wherever the wash free-streams to the pivot.
• No corner evades both. Low \sigma_v (free-streaming cutoff near the pivot) trips the slope test; high \sigma_v (free-streaming past it) trips the amplitude test. The combined bound is f_\text{hot}\lesssim0.2–0.9\% across the entire dispersion range — sub-percent, whatever \sigma_v was.

This closes the squeeze with real data. The load-bearing leg is the shape-independent amplitude test: even its weakest corner (f_\text{hot}\lesssim2.6\%) is small, and it bites hardest precisely where the wash free-streams to the 8\,h^{-1}Mpc scale — so the same correlation that lets texture touch S_8 tightens the forest bound, and fed back through step 2’s partition it pins |\Delta S_8|\lesssim0.005–0.01 on the amplitude test alone. The slope test then sharpens the hot fraction to per-mille, but that per-mille number (and the 8–33\sigma) is the mildly model-dependent leg; the S_8 conclusion does not need it. Either way texture cannot both help S_8 and hide from the forest, and the eBOSS data say it hides. So WIP-20’s S_8=0.816 stands, now bounded on the discrete channel too, by measurement rather than by analogy. The one genuinely new, still-open prediction survives intact and is sign-checked: texture would show as extra small-scale suppression that steepens n_\text{eff} toward more negative at k\sim0.7–2\,\text{Mpc}^{-1} — the current data disfavor any such steepening at more than the per-mille-f_\text{hot} level, so a future higher-resolution P_\text{1D} (DESI, WEAVE-QSO) that detected a small-scale n_\text{eff} steepening beyond \LambdaCDM + known IGM would be the framework’s signature, and its absence tightens the bound further. (Caveats carried from step 3b: the suppression model is order-of-magnitude and the slope test’s steepness assumes the logistic transfer; the amplitude test and the |\Delta S_8| conclusion do not.) The only remaining internal step, if earned, is to fold \eta_\text{nuc}(z) into the growth ODE (below) — but with the channel now measured to be sub-percent, it would only confirm the small effect.

A note on the four crossing “realms.” The nucleation picture — (1) complete re-nucleation deep in the bore, (2) partial, stars only, shaved to cores, (3) partial, stars-and-planets, leaving shards, (4) no nucleation at the fringe, old clumpiness preserved and merging with new — is one monotonic nucleation-completeness function \eta_\text{nuc} read across the bore profile (high z / bore body \to low z / outer fringe), with surviving-relic fraction locally \propto (1-\eta_\text{nuc}). It fixes the ratio f_\text{hot}\!:\!f_\text{cold} — the max-stirring crossing (realms 2–3) mints the hot ablated wanderers, the cold fringe (realm 4) the intact cores — but the 8\,h^{-1}Mpc estimate above is agnostic to where along the bore each relic was minted, because \sigma_8 integrates the growth history, not the injection map. So \eta_\text{nuc}(z) is where a future step (folding texture into the growth ODE, if the partition earns it) would enter; it is not needed to bound the magnitude now.

Sources: scripts/crust_texture_s8.py (step 1, channel structure), scripts/crust_texture_gs_partition.py (step 2, GS partition), scripts/crust_texture_pk_tilt.py (step 3, scale tilt + data bound on \varepsilon_\text{couple}), scripts/crust_texture_p1d.py (steps 3b–3c, Lyman-\alpha 1D flux-power projection + eBOSS DR14 confrontation), papers/chabanier (Chabanier et al. 2019, eBOSS DR14 Lyman-\alpha P_\text{1D}; \Delta^2_L,n_\text{eff} constraints), papers/dsw-2016 (Grimshaw–Smyth forced-KdV theory, Sec. 8); WIP-20 (smooth-field channels, closed); WIP-17 (crust energy budget / relaxation depth); universe-that-boils.qmd (the moraine’s erratics); how-far-does-the-match-travel.qmd (the deferred claim); solar-system-boundaries.qmd (ablated interstellar visitors); early-structure-formation.qmd (cold-core seeding).

---

New Directions

WIP-21: Braid topology and the gauge group from substrate structure

The Bilson-Thompson helon model represents first-generation Standard Model fermions as braids of three ribbons, with electric charge arising from chirality of the twists. Recent work (Asselmeyer-Maluga et al., arXiv:2501.03260, Jan 2025) establishes the complete mapping between braid group \mathcal{B}_3 and the weight lattice of SU(3)_c \times U(1)_{em}: braids correspond to on-shell spinor states of the Lorentz group, twists denote charges, and the CPT-invariant elements of \mathcal{B}_3 reproduce exactly the known fermionic content — no spurious states.

The substrate framework provides the physical mechanism that makes this mapping work. Each ribbon in a helon braid has a front and a back — a co-rotating / counter-rotating pair with its phase relationship governed by SU(2). The braid group’s embedding in SL(2,\mathbb{C}) (the double cover of the restricted Lorentz group) is the combinatorial shadow of the substrate’s co-rotating + counter-rotating pairing at the level of discrete topology.

Where the two frameworks meet: The braid model captures SU(3)_c \times U(1)_{em} but not SU(2)_L. The preon authors note that SU(2)_L appears to require extending \mathcal{B}_3 to more strands. The substrate framework identifies the same gap from the opposite direction: the weak asymmetry is not a topological property of the particle — it is a strain on the particle’s outermost counter-rotating boundary as it moves through an already chirally ordered background. The Higgs VEV provides the chiral ordering; the strain couples to W and Z.

The three-generation problem as inter-sheet penetration. The helon model cannot represent higher-generation fermions with additional braid crossings (even crossings > 2 already express composites). The substrate framework suggests a different mechanism: each generation corresponds to a persistent crossover event between chirality-coherent sheets — a fold where one lattice layer threads through another, creating a stable dynamical center with higher mass. First generation: single-sheet orbital systems. Second generation: one inter-sheet fold. Third generation: two inter-sheet folds. The three-generation limit would then follow from the maximum number of stable folds the layered lattice can support — n = 4 folds being dynamically unstable because the boundary complexity exceeds what the elastic restoring force can maintain.

This converts the generation problem from a symmetry question into a stability question: how many times can you fold a vortex sheet through itself before the polar-jet righting moment can no longer restore equilibrium? The answer is constrained by the stiffness hierarchy (c_{44} \gg c_{11} \sim c_{66}) and the finite coupling strength (\alpha_{mf} < 1/2, the Kopnin maximum). If the n = 4 fold requires \alpha_{mf} > 1/2 to stabilize, the three-generation limit is automatic.

Partial resolution (charged leptons). The mass-ratio half of this program is now substantially answered for the charged leptons in Three Generations from One Turning Knot. Reading the three generations as the three cube-roots-of-unity phases (\mathbb{Z}_3) of one three-fold junction makes Koide’s relation Q=(\Sigma m)/(\Sigma\sqrt m)^2 = 2/3 automatic (the cube roots force \Sigma\cos\theta_k=0, \Sigma\cos^2\theta_k=3/2), and quantizing the generational deviation at the lattice’s pairing-\sqrt2 pins the amplitude, giving Q=\tfrac13+(\sqrt2)^2/6=\tfrac23 — a 9-ppm match with no free parameter. A single residual phase, read as \delta=2/9 rad, then lands m_\mu/m_e=206.77 and m_\tau/m_\mu=16.82 to 0.001–0.006\%. So “compute the mass ratios from fold geometry” is met phenomenologically; what stays open is deriving the phase \delta (and the amplitude \sqrt2) from the confined-junction energy functional rather than reading them off the data.

Progress on the two derivation threads (2026-06-29). Putting all four fermion triads into the same \mathbb{Z}_3+\sqrt2 gauge — solving \sqrt{m_k}=\bar M(1+A\cos\theta_k) in closed form for each, with Q=\tfrac13+A^2/6 — sharpens both open bullets, advances the second substantially, and corrects one claim the chapter made (scripts/koide_triads.py).

Triad	Q	A/\sqrt2	reading
neutrinos (floor)	0.462	0.62	floor-compressed, below the balance point
charged leptons	\mathbf{0.6667}	\mathbf{1.000}	exactly on the \sqrt2 balance point
down quarks	0.731	1.09	color-loaded three-fold, above
up quarks	0.845	1.24	color-loaded three-fold, above

The amplitude A/\sqrt2 is a clean monotone ladder 0.62 < 1.00 < 1.09 < 1.24 with the charged leptons alone sitting at 1 — the \sqrt2 balance is not generic, it is the leptons’ alone. Two specific results follow.

• The quark deviation is structural, not a running artifact — the chapter’s “undo the QCD running” reading is wrong. Koide’s Q is invariant under a flavor-universal rescaling m_i\to\lambda m_i, which is exactly what leading-order mass running is; so “running shifts Q off 2/3, drifting back as it is undone” cannot be right. Brought to a common scale (m_Z) the up-type Q moves further from 2/3 (0.845\to0.888), not toward it. The framework-correct reading is the one the chapter already half-states: the quark three-fold is loaded with color, so its \mathbb{Z}_3 is doing double duty (color and generation) and is not the bare generation clock — the leptons are clean because their colorless three-fold carries generation only. The deviation is a structural color-dressing of the amplitude (A>\sqrt2), not perturbative QCD running.

• The neutrino floor-compression is confirmed and directional. Raising the lightest neutrino mass from 0 to the framework floor (\approx2.04 meV) drags Q monotonically from 0.585 down to 0.462, toward the democratic 1/3 — so the floor that pins m_{\nu,1} quantitatively predicts the compressed Q\approx0.46, with no new input. The neutrino triad keeps the clean \mathbb{Z}_3 phase geometry (the 120^\circ fit closes); only the amplitude is throttled below \sqrt2.

On the first bullet (deriving \delta,\sqrt2) the progress is sharper-posing, not closure. The \sqrt2 is the same two-component BdG/Nambu pairing amplitude carried everywhere else (\xi^2=2\xi_\text{GP}^2, \kappa=1/\sqrt2) — an identification, not yet a from-scratch amplitude. The phase has a clean geometric reading — \delta=2/9 rad is the angular offset of the heaviest lepton from the maximal-mass point of the Koide circle, equivalently the electron sitting \approx2.3^\circ inside the massless edge \cos\theta=-1/\sqrt2 — and is sharply over-determined: among simple two-and-three rationals only 2/9 lands m_\mu/m_e (the neighbours 1/5,\,3/13,\,1/4 give 75,\,353,\,2710). But it is still read off, not derived from the junction energy functional. That remains the open computation, now well-posed.

Phase-sector progress — the \sqrt2 comes back, and the bet narrows (2026-06-29, second pass). Attacking the phase sector directly (scripts/koide_phase_sector.py) yields one new zero-parameter result, one cleaner statement of the bet, and three robustness checks.

• The pairing-\sqrt2 governs the phase sector too — the massless edge is derived. The product of the triad collapses, via the cube-root identities plus \prod_k\cos\theta_k=\tfrac14\cos3\delta, to a single cosine of the tripled phase: f(\delta)\equiv\prod_k(1+A\cos\theta_k)=1-\tfrac34A^2+\tfrac14A^3\cos3\delta\xrightarrow{A=\sqrt2}-\tfrac12+\tfrac1{\sqrt2}\cos3\delta. The lightest mass reaches zero (the massless edge, |\delta|<\pi/12 being the all-positive window) exactly when \cos3\delta=1/\sqrt2, i.e. \delta_\text{edge}=\pi/12=15^\circ. So the same 1/\sqrt2 that fixes the amplitude (A=\sqrt2\Rightarrow Q=2/3, the 45^\circ tilt) fixes the phase edge (\cos3\delta=\cos45^\circ). This edge is forced by A=\sqrt2 with no rational chosen — a genuine derivation of the landmark the “2.3^\circ inside the edge” reading rested on.
• The bet reduces to one cross-sector identity, 3\delta=Q. In the tripled variable \phi=3\delta that governs the product, the leptons sit at \phi=3\cdot\tfrac29=\tfrac23=Q — numerically the Koide ratio. So \delta=2/9 is exactly equivalent to 3\delta=Q, i.e. \delta=Q/3=\tfrac19+A^2/18=\tfrac19+\tfrac19 (democratic floor 1/3^2 plus a pairing piece). This is no longer “a rational from the two-and-three”: it is the tripled phase equal to the amplitude-fixed Koide ratio — one relation, carrying no freedom once A=\sqrt2 is granted. The electron then sits a fixed offset \phi_\text{edge}-Q=\pi/4-2/3=0.119 rad inside the \sqrt2-edge, the charged-lepton mirror of the neutrino floor: in both triads the lightest member is held a “floor” off its massless point (neutrino by m_{\nu,1}\approx m_1, electron by this phase offset). What the junction calculation must now output shrinks from all of \delta to just this offset — equivalently, just 3\delta=Q.
• Robustness (three checks). (i) The best-fit \delta is pinned to 2/9 to <10^{-5} rad across the entire PDG m_\tau error bar (1776.86\pm0.12 MeV) — at the central+ value it is 2/9 to seven figures. (ii) The up>down>lepton amplitude ladder is robust at \sim24\sigma under PDG quark-mass Monte Carlo (A_u/\sqrt2=1.239\pm0.001, A_d/\sqrt2=1.093\pm0.006) — the rungs are real, not artifacts of the loose light-quark masses. (iii) The neutrino compression is robust to mass ordering (Q=0.46 normal, 0.42 inverted — both well below 2/3, above 1/3). Negative result, recorded for honesty (later superseded — see third pass below): the quark color-loading appeared to have no clean closed form — (A^2-2) is 1.07 (up) vs 0.39 (down), a ratio 2.75 that sits between charge^{3/2} (2.83) and charge^2 (4.0) with no simple power landing the ratio. The color-dressing direction (more electric charge \to more loading) is clear; its functional form looked unpinned. This was an artifact of testing only the ratio: fixing the absolute normalization resolves it (below).

Amplitude-sector progress — the four-triad ladder closes on one functional (2026-06-29, third pass). The “no clean closed form” negative result above used only the up/down ratio of (A^2-2). Fixing the absolute normalization (scripts/koide_color_loading.py) turns it positive: all four amplitudes hang on a single functional, A^2 \;=\; 2\,\bigl(1 + C\,q^{3/2}\bigr), with C the color flag (0 colorless leptons/neutrinos, 1 colored quarks) and q the electric-charge magnitude (1 lepton, \tfrac23 up, \tfrac13 down). The prefactor is not free — it is the same pairing-two already in the lepton base A^2=2. Rungs: lepton 1.000 (exact, C=0), down 1.092 (measured 1.093, a 0.07\% hit), up 1.243 (measured 1.239, 0.3\%, inside quark-mass scheme systematics), and the neutrino is assigned the same colorless base \sqrt2 (C=0), then dragged to A/\sqrt2=0.62 by the floor alone. Robustness: a PDG quark-mass Monte-Carlo gives the loading power p=1.46\pm0.10 (3/2 within 0.4\sigma; integer neighbours q^1,q^2 miss the 2.76 ratio), and the per-quark normalization comes back c_\text{down}=1.01\pm0.07, c_\text{up}=0.98\pm0.01 — both the pairing-two. Most strikingly, fed through Q=\tfrac13+A^2/6 the functional becomes a single generalized Koide formula Q=(2+C\,q^{3/2})/3: the famous \tfrac23 is the C=0 value and the quark Koide ratios are charge-set, Q_\text{down}=0.731, Q_\text{up}=0.848 (measured 0.731,\,0.845) — three previously-unexplained Koide ratios collapse to one formula in the charge. This substantially closes open-work bullet 2: the quark A>\sqrt2 and the neutrino base are now one functional; what is not yet derived is only the 3/2 exponent (read geometrically as a self-similar color blob over the orbital sheet — area q thickened by its own radius \sqrt q, giving volume q^{3/2}; note the cone-of-ball reading would give the wrong power q^1) and the floor-throttle magnitude. See Three Generations § The four-triad ladder closes on one functional.

Amplitude/phase factorization — the residual debt is localized to two numbers (2026-06-29, fourth pass). Putting all four triads through the same exact \mathbb{Z}_3 fit (\sqrt{m_k}=\bar M(1+A\cos\theta_k); three masses in, (\bar M,A,\delta) out, masses reconstruct to nine figures — scripts/koide_phase_amplitude.py) splits the charged-fermion mass content into two sectors that color breaks differently. Amplitude (Q=\tfrac13+A^2/6, the spread): charge-set for all three by Q=(2+C\,q^{3/2})/3 (lepton 0.6667, down 0.731, up 0.845). Phase (\delta, which fixes the within-generation hierarchy): the lock 3\delta=Q holds only for the colorless lepton (3\delta=0.667=Q); the color-loaded quark phases are unlocked (3\delta=0.33,\,0.23\neq Q). The consequence is a sharp localization of the framework’s remaining debt: the lepton triad is fully locked (amplitude charge-set and phase =Q/3), which is why one anchor lands all three lepton masses to 0.001\%; the quark triads are half-locked (amplitude charge-set, phase free). So, setting aside the one borrowed overall scale per triad (the electron anchor for leptons, the separate absolute up/down scales for quarks — the standing Yukawa-scale program), the entire un-derived content in the ratio structure of the charged-fermion mass matrix — once thirteen Yukawas — is now just the two quark within-generation phases (\delta is how far inside its own massless edge the lightest member sits: 2.3^\circ lepton, fixed by 3\delta=Q; 4.0^\circ down, 0.4^\circ up — the near-edge up phase being m_t/m_u\sim7\times10^4), and nothing else. Color is what frees them: it loads the amplitude (q^{3/2}) and in the same stroke unlocks the phase the bare colorless clock pins. (Refined by the fifth pass below: the unlock is not color’s doing — the colorless neutrino is unlocked too. The lock tracks the balance point A=\sqrt2; color is one way off it, the neutrino floor the other.) See Three Generations § What color sets and what color frees.

The phase lock tracks the balance point, not color — corrected by the neutrino (2026-06-29, fifth pass). The fourth pass read the phase lock 3\delta=Q as something color frees (holds for the colorless lepton, fails for the colored quarks). Adding the neutrino — also colorless — to the same exact \mathbb{Z}_3 fit (scripts/koide_phase_lock.py) shows that reading is incomplete: the neutrino phase is unlocked too (3\delta=0.81 vs Q=0.46). What the lock actually tracks is sitting exactly on the pairing balance point A=\sqrt2 — and the charged lepton is the only triad there. Ordered by amplitude, 3\delta-Q is a clean sign change through the lock: neutrino (A/\sqrt2=0.62, floor-throttled below the point) overshoots, 3\delta-Q=+0.35; charged lepton (A/\sqrt2=1.00, on the point) locks, 3\delta-Q=0; down (1.09) and up (1.24, color-loaded above) undershoot, -0.40 and -0.61. So color is one way off the point (up, A>\sqrt2) and the floor is the other (down, A<\sqrt2); either departure unlocks the phase, in the corresponding direction — the lock is the crossing, the bare colorless clock at A=\sqrt2 alone. A tidy corollary: A<1 (the neutrino) has no massless edge at all (the edge condition \cos3\delta=(3A^2/4-1)/(A^3/4) falls outside [-1,1]), so the neutrino’s lightest member is held off zero by the amplitude throttle, not by a phase offset inside an edge — the same held-off-zero fact as the lepton’s “2.3^\circ inside the edge,” read in the two sectors the balance point separates. This does not change the localization of the debt (still the two quark phases) but it re-targets it: what the junction calculation must output is why the lepton, the one triad on A=\sqrt2, also lands 3\delta=Q — the lock is the balance point’s own signature, not color’s. See Three Generations § What color sets and what color frees.

The lock is geometrically inevitable — only its location is owed (2026-06-29, sixth pass). Two checks (scripts/koide_phase_lock.py) harden the fifth-pass reading from an observed four-point ordering into a structural statement. (i) The neutrino overshoot is ordering-independent. The fifth pass placed the neutrino below the balance point (A<\sqrt2) with the phase overshooting (3\delta>Q) using normal ordering only; in fact both facts hold under normal and inverted ordering, for every floor from 0 up — inverted ordering pushes it further below (A/\sqrt2\approx0.5) with a larger overshoot. So the neutrino is a structurally-forced “below-and-over” point, not an artifact of an assumed ordering, and a neutrino triad measured at A>\sqrt2 or 3\delta<Q would falsify the balance-point reading. (ii) The sign change is forced, not lucky — the lock is the crossing of two opposite monotonicities. The Koide spread Q(A)=\tfrac13+A^2/6 is exact, derivable, and increasing in A; the measured phase 3\delta(A) runs the other way, decreasing (0.81\to0.67\to0.33\to0.23 along the amplitude ladder), its ceiling the already-derived massless-edge envelope 3\delta_\text{edge}(A)=\arccos[(3A^2/4-1)/(A^3/4)], itself monotone falling. An increasing curve and a decreasing curve cross exactly once — so a sign change of 3\delta-Q must occur somewhere, and the whole bet collapses to a single un-derived number: that the crossing sits at A=\sqrt2 (where it reads 3\delta=Q=2/3). The junction calculation no longer owes “why a lock exists” — a crossing is inevitable — only “why at the balance point.” See Three Generations § What color sets and what color frees.

From a location to a balance condition — and an honesty pass (2026-06-29, seventh pass). The sixth pass left the target as a coordinate (“why the crossing at A=\sqrt2”). The seventh (scripts/koide_phase_lock.py) makes it a condition and audits what is actually derived. Reframing: when the lock holds, the phase decomposes \delta=Q/3=\tfrac19+A^2/18 — democratic floor \tfrac19=1/3^2 plus pairing piece A^2/18 — and A=\sqrt2 is the unique amplitude where the pairing piece equals the floor. So the balance point has a meaning intrinsic to the phase sector (not just the borrowed amplitude one), and the open target becomes “why the lightest member equalizes the two halves of its phase” — an equilibrium statement, the kind a stress-tensor minimization outputs, rather than a coordinate it must hit. Honesty audit of the sixth pass, three points. (i) The neutrino promotes the lock from a boundary extremum to an interior crossing: among the charged triads alone 3\delta-Q=\{0,-0.40,-0.61\}, so the lepton is the boundary maximum, not an interior zero — only the (forced) neutrino overshoot makes the zero interior, and a neutrino at 3\delta<Q would demote the lock to a boundary touch. (ii) The “two opposite monotonicities” is partly rhetorical: the two genuinely derivable curves, Q(A) and the edge envelope, cross at A/\sqrt2\approx1.03, not \sqrt2; the crossing at \sqrt2 is Q(A) against the measured phase, which sits inside the edge, so “decreasing” is derivable only for the phase’s ceiling. (iii) No single-parameter law 3\delta(A) fits all four phases (best \sim\!26\%), confirming the two quark phases are irreducibly free — the localized debt is exactly two numbers, no fewer. Net: what is structural is an increasing Q(A) + a derivably falling ceiling + a forced neutrino overshoot \Rightarrow the lepton is an interior zero; the location \sqrt2 stays the bet, now posed as a balance condition. See Three Generations § What color sets and what color frees.

Open work:

• Derive the n = 4 instability from the energy functional \mathcal{F}[d, \xi, \omega_0] (WIP-15).
• Map the braid group’s weight lattice coordinates to substrate boundary layer quantum numbers.
• Cross-check the amplitude ladder against the arm ledger. WIP-32 supplies a second, independent charge–mass relation on the same three-fold junction: each arm’s circulation either escapes as monopole charge or is retained as mass, giving m\propto(1-|q|) and m_d/m_u=2 (measured 2.18, PDG range [1.89,2.50]). This entry’s ladder A^2=2(1+C\,q^{3/2}) ties the generational amplitude to the same q. Two charge-indexed relations on one geometry must be mutually consistent — and the arm ledger has so far been tested only on the first generation, so whether m\propto(1-|q|) survives into the second and third triads (where the \mathbb{Z}_3 fit already fixes the within-triad structure) is a sharp, unrun internal check. If the two relations conflict, one of them is reading the junction wrong; if they agree, the 3/2 exponent gains an independent constraint.
• Derive the cross-sector identity 3\delta=Q (equivalently \delta=2/9 rad) from the confined Y-junction. The boundary calculation no longer owes all of \delta: the massless edge \delta=\pi/12 is now derived from A=\sqrt2 (the \cos3\delta=1/\sqrt2 condition), and the bet has collapsed to the single relation 3\delta=Q — the electron held one fixed phase-offset inside the \sqrt2-edge, the mirror of the neutrino floor. What is owed is the stress-tensor calculation that outputs that one offset, i.e. shows the lightest charged lepton is pinned a visibility-floor short of the massless edge exactly as 3\delta=Q requires (Three Generations § The phase sector has its own \sqrt2). The fifth through seventh passes re-target this: the lock is the balance-point signature (3\delta-Q changes sign through A=\sqrt2, the neutrino overshooting from below and the quarks undershooting from above), and because that sign change is the crossing of an increasing derivable Q(A) with a decreasing measured phase whose ceiling is the derivable edge envelope, the lepton is forced to be an interior zero (the neutrino overshoot supplies the interior) — so the calculation no longer owes “why a lock at all.” What it owes is best posed not as a coordinate but as a balance condition: in the locked decomposition \delta=\tfrac19+A^2/18, the point A=\sqrt2 is exactly where the phase’s pairing piece A^2/18 equals its democratic floor \tfrac19 — so the stress-tensor calculation must output that the lightest charged lepton equalizes the two halves of its phase, which simultaneously fixes the amplitude at \sqrt2 and the phase at 3\delta=Q. (Honest caveat: the two genuinely derivable curves Q(A) and the edge envelope cross at A/\sqrt2\approx1.03, not \sqrt2; the \sqrt2 crossing uses the measured phase, and no simple law 3\delta(A) fits all four triads, so the two quark phases stay irreducibly free.) Deriving the pairing amplitude \sqrt2 from scratch (not as the carried BdG/Nambu identification) remains the amplitude-sector half.
• (Substantially closed, third pass.) The four-triad amplitude ladder now hangs on one functional, A^2=2(1+C\,q^{3/2}) (C = color flag, q = electric charge), with the prefactor locked to the pairing-two: leptons/neutrinos share the colorless base \sqrt2, the quark loading is the charge to the three-halves (A_d/\sqrt2=1.092 vs measured 1.093; A_u/\sqrt2=1.243 vs 1.239), and the neutrino departs only via the floor throttle. Still owed: a stress-tensor derivation of the 3/2 exponent (currently read as color filling the 3-volume \Omega^{3/2} of the charge’s solid angle) and of the neutrino floor-throttle magnitude (the floor that pins m_{\nu,1} is derived; its conversion to A/\sqrt2=0.62 is descriptive).

Sources: Bilson-Thompson (2005), arXiv:hep-ph/0503213; Asselmeyer-Maluga et al. (2025), arXiv:2501.03260; Koide (1981), Lett. Nuovo Cimento 34, 201; higgs-field.qmd; fermion-generations.qmd.

WIP-22: The biological scale — why \xi is the size of a cell

The coherence length \xi \approx 100\;\mum, the inter-sheet period d_\text{GJO} \approx 16\;\mum, and its counter-rotating boundary half-period d_\text{GJO}/2 \approx 8\;\mum are not arbitrary numbers. They are the scales of biological organization (the spacing is now derived in closed form — see Substrate Particles § The Vertical Scale — and read into cell and organelle sizes in DNA and the Living Lattice):

Substrate scale	Biological analog	Size
\xi \approx 100\;\mum	Typical eukaryotic cell diameter	10-100 \mum
d_\text{GJO}/2 \approx 8\;\mum	Red blood cell diameter	6-8 \mum
d_\text{GJO}/2 \approx 8\;\mum	Mitochondrial length	1-10 \mum
\xi	Maximum capillary spacing in tissue	\sim 100\;\mum
E_\text{min} \approx 13 meV	Thermal energy at body temperature (k_BT at 310 K)	27 meV

The minimum photon energy E_\text{min} \approx 13 meV corresponds to \lambda \sim 100\;\mum — the same band where liquid water has anomalous absorption features and where biological tissue transitions from transparency to opacity. This is the scale where substrate-mediated energy transport (modons) and thermal energy transport (phonons) cross over.

The framework predicts that \xi is the boundary of coherent substrate response. A cell larger than \xi cannot maintain coherent internal substrate coupling. A structure much smaller than the boundary half-period d_\text{GJO}/2 falls below the substrate’s finest boundary spacing and couples weakly to the in-plane dynamics. The observed range of eukaryotic cell sizes (\sim 10-100\;\mum) sits exactly where substrate coupling is maximal — between d_\text{GJO}/2 and \xi — with cell-spanning organelles settling onto the \approx 8\;\mum boundary half-period.

Testable consequences:

• THz spectroscopy of living tissue should show anomalous absorption or dispersion near \lambda \sim \xi that does not arise from molecular resonances alone.
• Cell size distributions across taxa should show statistical clustering near \xi and d beyond what metabolic scaling laws predict.
• The “quantum biology” observations (photosynthetic energy transfer, avian magnetoreception, enzymatic tunneling) that lack thermal explanations may involve the substrate’s counter-rotating boundary layers providing low-decoherence channels with coherence lengths set by \xi.

Caution: This is the most speculative extension. The scale coincidence is striking but not yet connected by a derivation. Distinguishing coincidence from causation requires a substrate-specific biological signature that cannot be explained by chemistry alone.

WIP-23: Scale-invariant feedback topology in mesoscale systems

The feedback loop — rotating disk, polar axial jets, counter-rotating boundary absorption — is the lowest-energy stable configuration for organized rotational energy in an elastic medium. This topology recurs across 25+ orders of magnitude because it is dictated by physics, not by scale. The framework has documented it at four scales (substrate lattice, stellar accretion, AGN, galaxy formation) but has not applied it to intermediate-scale systems where the substrate’s influence may be more directly observable.

Atmospheric and oceanic circulation. The Gulf Stream is a coherent flow channel maintained by rotational energy from Earth’s spin (co-rotating flow), atmospheric jet streams (polar axial coupling), and counter-currents (Labrador Current, deep return flow — the counter-rotating boundary). The substrate framework predicts these are the same pattern operating in a boundary-layered medium. The testable question: does the substrate lattice contribute measurably to atmospheric energy transport? The lattice’s collective modes — Tkachenko waves at c_T \approx 9 km/s, f_T \approx 3{,}700 Hz — overlap with atmospheric gravity wave frequencies. A coupling between substrate lattice modes and atmospheric dynamics, however faint, would be extraordinary.

Aromatic chemistry. Benzene’s delocalized \pi electron cloud is a ring of co-rotating flow — six electrons sharing a circular orbital above and below the molecular plane. The substrate framework interprets this as a miniature vortex ring whose stability comes from boundary-matching quantization. The aromaticity rules (Hückel’s 4n+2 rule) would then be a manifestation of the substrate’s boundary-parity constraint at the molecular scale: 4n+2 electrons give even boundary parity (stable), while 4n give odd parity (antiaromatic, unstable). For large aromatic systems (porphyrins, graphene sheets), the substrate correction should become measurable as a deviation from standard DFT predictions.

Magnetohydrodynamic dynamos. The Earth’s magnetic field is generated by organized fluid flow with differential rotation, polar outflow, and boundary layer coupling at the core-mantle boundary. The substrate framework predicts the D'' layer’s anomalous seismic properties may reflect substrate-scale organization at the solid-liquid boundary.

WIP-24: Gravitational field topology and frame-dragging as substrate flow

Status (2026-06-18): the azimuthal profile v_\phi(r,\theta) is derived and reproduces Kerr/Lense–Thirring to linear order; the radial 1/r^3 falloff is now parameter-free; only the absolute coupling (inherited from SC1) and the full nonlinear metric remain. Written up in Gravity § Rotating bodies; calculation in scripts/kerr_frame_dragging.py.

Gravity is modeled as an ebbing current — dc1 flowing inward through boundaries to produce v_\text{ebb}(r) = \sqrt{2GM/r}, giving the Schwarzschild metric in Painlevé-Gullstrand form (SC1, exact). Real astrophysical objects rotate, and their gravitational fields carry angular momentum, which must appear as an azimuthal entrainment v_\phi added to the radial ebb.

Frame-dragging IS the azimuthal flow (rigorous). The acoustic metric’s cross term -2\,\vec v\cdot d\vec x\,dt carries the gravitomagnetic sector; its azimuthal piece gives g_{t\phi}=-v_\phi\,r\sin\theta, g_{\phi\phi}=(r\sin\theta)^2, so the dragging rate is \omega_\text{drag}=-g_{t\phi}/g_{\phi\phi}=v_\phi/(r\sin\theta) — no assumption beyond the same acoustic metric that gave SC1. The radial ebb sits in g_{tt}; v_\phi is the only new field the rotating sector needs.

The crux — the 1/r^3 falloff — is resolved by geometry (2026-06-18). A free draining vortex conserves circulation (v_\phi\propto 1/r, \omega_\text{drag}\propto 1/r^2) — the wrong law. The substrate around a spinning body is entrained, not freely draining: a force-free azimuthal flow v_\phi=f(r)\sin\theta obeys the \phi-vector-Laplacian f''+\tfrac2r f'-\tfrac{2}{r^2}f=0, an Euler equation with indicial roots n=+1 (rigid interior) and n=-2 (decaying exterior). The unique decaying mode is v_\phi=\tfrac{2GJ}{c^2}\tfrac{\sin\theta}{r^2}\Rightarrow\omega_\text{drag}=2GJ/(c^2 r^3) — exactly Lense–Thirring. This is the spin-dipole (\ell=1) analog of the mass-monopole Newtonian 1/r: the radial law is fixed by geometry alone (zero parameters), the same status as SC1’s radial profile. Gravity Probe B: geodetic \approx6604 mas/yr (vs GR 6606.1, measured 6601.8\pm18.3) and frame-dragging \approx41 mas/yr (simple circular average; full GP-B model 39.2; measured 37.2\pm7.2) — both within the data.

The ergosphere as a substrate phenomenon. Where v_\phi exceeds the local sound speed, the substrate goes supersonic in rotation — an acoustic ergosphere, the rotating analog of Unruh’s “dumb hole,” inside which no static substrate parcel exists.

What remains. Two pieces, both analogous to open items in the radial sector. (i) The amplitude’s coupling is inherited, not independently re-derived: the rotlet coefficient 2GJ/c^2 uses the same G (hence f_\text{cross}) as SC1, applied to the boundary’s angular-momentum current J rather than its mass M — the explicit boundary-layer entrainment calculation (the no-slip fraction \sim compactness GM/c^2 R) that would fix it from first principles is owed, exactly as f_\text{cross} is owed for the radial ebb (WIP-15 item 2). (ii) The full nonlinear Kerr acoustic metric — radial v_\text{ebb} and azimuthal v_\phi together, to the “exact, not linearized” standard the Schwarzschild sector already meets — plus the Penrose/ergosphere energetics.

WIP-25: The nuclear binding-energy curve from boundary topology

Status (2026-07-25): the surface-to-volume ratio is computed from geometry, the asymmetry coefficient a_\text{sym} is now closed, and the absolute scale \epsilon is the single remaining debt. Started from correcting the He-4 fusion estimate (Solar & Stellar Dynamics § Helium-4) — a dimensionally inconsistent surface-area form \sigma\,\Delta A replaced by the flux-tube-length form \sigma\,\Delta L — and grew into a substrate reading of the whole binding-energy curve, now written up in Proton Core § The Binding-Energy Curve.

New (2026-06-18): the ratio a_S/a_V from close-packing seam geometry, zero parameters (scripts/nuclear_seam_geometry.py). Because a_V and a_S are the same per-contact seam energy \epsilon counted two ways (interior contacts made vs. surface contacts missing), \epsilon cancels in their ratio, which is then pure packing geometry. Counting nearest-neighbour seam contacts on a close-packed (FCC) droplet — interior coordination 12, so a_V=6\epsilon — and reading the A^{2/3} surface deficit (lattice fit, cross-checked by the orientation-averaged smooth-sphere surface tension) gives a_S/a_V\approx1.36 (smooth sphere; faceted-lattice fits 1.37–1.6). The empirical Myers–Świątecki value is 1.18, so the geometric number is \sim15\% high — and informatively so: a sharp surface breaks the most contacts, so close-packing is an upper bracket, while BPS-Skyrme (no surface/gradient term, §below) is the lower bracket at 0; the true value sits between, set by surface diffuseness = the gradient/\mathcal{L}_2 energy both routes flag as open. Folding the geometric ratio into A_\text{peak}=2a_S/a_C (framework-derived a_C=0.717 MeV; empirical a_V only for the still-owed absolute scale) gives A_\text{peak}\approx59–63, the observed Fe/Ni region (56–62), tighter than the bare-coefficient 52. So the peak’s location, not just its existence, now follows from geometry; only the absolute \epsilon remains.

What is established (three independent frameworks agree). The strong force is a residual counter-rotating boundary seam between near-distinct color-singlet nucleons (the \sim 100\times hierarchy below the \sim 929 MeV confinement boundary; the weakness is the Ikeda threshold rule [R115]). Lattice QCD confirms single-baryon confinement is a Y-string, V_{3Q}=\sigma_{3Q}L_\text{min}+\dots ([R109], [R110]) — validating the Steiner-tree form of fusion-as-junction-merger. The semi-empirical mass formula maps term-by-term onto boundary mechanisms ([R117]): volume = saturated seam energy (saturation = boundary locality = string-tension constancy, independently the BPS-Skyrme incompressible-droplet bound E=E_0|B|, zero classical binding [R112], [R113]); Coulomb a_C=\tfrac35\alpha\hbar c/r_0 from the framework’s own \alpha; pairing = the anti-phase Cooper/lattice breath one tier down. The iron peak is A_\text{peak}\approx 2a_S/a_C\approx 52 (iron–nickel region), the same surface-vs-Coulomb ratio as Myers–Świątecki’s fissility parameter.

What remains. With the ratio now geometric, the one open quantity is the absolute per-contact seam energy \epsilon (which fixes a_V=6\epsilon and hence a_S individually). It reduces to \epsilon=\sigma\times\Delta L with \Delta L the overlap length of two fused confinement boundaries — and the empirical a_V=15.68 MeV needs \Delta L\approx0.003 fm \approx0.0025\,r_0, i.e. a fractional tail overlap of a few parts per thousand. That tiny number is the \sim100–300\times residual-strong-force suppression itself — the exponential tail of the closed color-singlet boundary, not the main flux tube — and it is the genuinely hard piece: it is why nucleons stay nearly distinct (the Ikeda rule), so it cannot come from a naive Steiner re-optimization of all 3A quarks (that would re-confine them at the \simGeV scale, giving binding \sim100\times too strong). The right object is the merged-boundary tail seam, not the merged Y-string. Concretely: compute the overlap energy of two adjacent confinement-boundary tails (the residual seam), or equivalently the constrained Steiner reduction that holds each nucleon’s color singlet intact. Resolving \epsilon would turn A_\text{peak} from a (now geometry-pinned) ratio result into a fully zero-parameter absolute prediction. Convergent check: the BPS-Skyrme soliton derivation [R118] (Adam–Naya et al. 2013) reaches the same boundary — it derives the volume (BPS, zero binding), Coulomb (Z^2/A^{1/3}), and an isorotational asymmetry term, but has no surface term (the gradient term \mathcal{L}_2 is dropped, hence light-nuclei overbinding). So both the substrate and the soliton route leave the surface tension — a gradient/boundary energy (a_S \leftrightarrow \mathcal{L}_2) — as the one open piece. (The asymmetry term a_\text{sym} — the other coefficient the BPS isorotation touches — is now closed, and by this entry’s own method rather than by the Y-junction solve it was originally assigned to; see below.)

New (2026-07-25): a_\text{sym} closes, by the same “\epsilon cancels in the ratio” move (scripts/nuclear_asymmetry_seam.py). WIP-32 originally listed a_\text{sym} as a fourth target on the intra-nucleon Y-junction solve. That was the wrong tier: a_\text{sym} lives here, at the internucleon seam, and — like a_S/a_V — it is a ratio of seam energies, so the absolute \epsilon cancels identically. Letting the seam energy depend on whether the two nucleons across it carry matched or mismatched arm-winding tallies splits it into \epsilon_{np} and \epsilon_\text{like}, and the same FCC contact count gives a_V=3(\epsilon_{np}+\epsilon_\text{like}), a_\text{sym}^\text{int}=3(\epsilon_{np}-\epsilon_\text{like}) — their ratio being exactly Myers–Świątecki’s \kappa. The winding ledger’s rule (a counter-rotating seam binds only what it can cancel) sets \epsilon_\text{like}=0, hence a_\text{sym}^\text{int}=a_V; adding the free-Fermi-gas exclusion half E_F/3 gives a_\text{sym}=a_V+E_F/3=27.96 MeV against 28.06 measured ($$2%, limited by k_F). Four corollaries come free: unbound dineutron/diproton, unbound pure neutron matter, the N\approx Z valley, and the pairing term. Note the pattern this entry has now hit twice — both nuclear coefficients that turned out to be reachable were ratios, and the one still open (\epsilon) is the one absolute scale. Of the five liquid-drop coefficients, a_C and a_\text{sym} are derived, a_S/a_V is geometric, and only \epsilon (setting a_V,a_S individually) and the pairing amplitude remain.

WIP-26: The substrate ladder — discrete scale invariance and the √2 rung

Status (2026-05-31): pattern identified; ratio identified and its period now derived (given a tower); the tower’s existence — the limit cycle itself — still open; comb test instrumented and run (a first real-EEG pass puts the EEG fine ratio at \varphi, not \sqrt2; only the octave structure stays on the comb). The paper repeatedly predicts that some measured quantity clusters at discrete substrate-preferred values rather than varying continuously — grid-cell modules, EEG bands, cochlear octaves, organelle and vesicle sizes, microtubule resonances, earthquake magnitudes, currency denominations. The Substrate Ladder collects these and proposes a single mechanism. This entry tracks what is owed to make it a derivation rather than a reading. It is the systematizing partner of WIP-22 (why \xi is a cell) and WIP-23 (scale-invariant feedback): WIP-22 fixes two absolute scales, this fixes the ratio between rungs.

The mechanism (identified). The recurring ladders are spaced by a constant ratio, not a constant step — they are geometric, not harmonic, so an ordinary standing-wave overtone series (1,2,3,\dots) is the wrong template. A geometric tower is the signature of discrete scale invariance: a scale-free (critical) system handed a short-distance cutoff has its continuous scaling symmetry broken to a discrete subgroup, leaving states at 1,\lambda,\lambda^2,\dots. The Efimov effect (\lambda_0=e^{\pi/s_0}\approx 22.7 from a renormalization-group limit cycle; Wilson 1971, Braaten–Hammer 2006) is the clean precedent; Sornette’s log-periodic rupture/seismicity is the classical-medium version. The substrate is built to sit at exactly such a point — the marginal \mu\to0 point of WIP-15 item 10, gapless and scale-free, the condition for emergent light — and is cut off at \xi_\text{GP}, d_\text{GJO}, and \xi.

The ratio (identified, not derived). The proposed fundamental rung is \sqrt{2} — the pairing factor \xi/\xi_\text{GP}=\sqrt2 (\xi^2=2\xi_\text{GP}^2, the lattice’s anti-phase breath), with the octave =2 as the pairing-two and the vertical d_\text{GJO}/2\to d_\text{GJO} spanning exactly one octave. So the ladder is the lattice breath replicated across scale. The owed computation is the substrate’s analog of Efimov’s s_0: a from-scratch demonstration that a critical paired condensate cut off at \xi_\text{GP} has a DSI tower of ratio exactly \sqrt2 — the same marginal-point Bogoliubov problem that gates Step A.

Sharpened (2026-05-31). The owed piece is not merely the value of the ratio but the existence of a tower at all: a scale-free point generically has continuous scale invariance (a smooth power law, no comb), and a discrete ladder requires the marginal point to be a genuine limit cycle — a complex scaling exponent, the analog of Efimov’s fall-to-the-centre. So the \sqrt2 claim is two claims (limit cycle exists; its period is \ln\sqrt2), and the comb test probes the first. Candidate route: a 2D condensate’s breathing mode carries an SO(2,1) scale symmetry anomalously broken by the short-distance cutoff, with the radial-breath-at-twice-orbital relation (WIP-12) the natural seed of the factor of two (\tfrac12\ln2=\ln\sqrt2). The conceptual bridge — breath-as-ladder (the Efimov self-similarity: one self-similar paired-breath replicated up the tower, the way one three-body state is replicated up Efimov’s), the keyboard-not-string refinement of the musical reading, and the rungs-as-lossless-channel “currency” reading — is now written into the chapter.

Derivation attempt — the period follows, the existence does not (2026-05-31). Pursuing the SO(2,1) route splits the two claims into very different difficulties (arithmetic spine verified in scripts/ladder_so21.py).

• The conformal point is real. A 2D condensate with a contact interaction carries an exact classical SO(2,1) (Pitaevskii–Rosch) conformal symmetry whose protected, undamped breathing mode runs at exactly 2\omega — and that 2\omega is the framework’s “radial breath at twice the orbital rate” (WIP-12). So the marginal \mu\to0 point is a genuine conformal point and the breathing-two is its fingerprint, not a coincidence.
• The period is then forced — given a tower. A conformal point handed a short-distance cutoff yields, by the standard de Alfaro–Fubini–Furlan / Efimov machinery, a discrete tower of length-ratio \lambda=e^{\pi/s_0} whenever the effective inverse-square coupling is supercritical (g<-\tfrac14, the complex scaling dimension of Kaplan–Lee–Son–Stephanov “conformality lost”). The tower’s rungs lie on the marginal BdG dispersion E_k=\sqrt{(\hbar^2k^2/2m_f)^2+(c\hbar k)^2}, and on its non-relativistic branch (E\sim k^2\sim 1/r^2, k>k^\ast=1/\xi — exactly where the pairing rung \xi_\text{GP}\to\xi sits) a length rescaling by \lambda is an energy rescaling by \lambda^2. The framework’s pairing/breathing octave (energy \times2 = the 2\omega breath = m_1=2m_f) is therefore the length half-octave \lambda=\sqrt2. This is Efimov’s own structure — sizes scale by \lambda_0, energies by \lambda_0^2 — so the factor of two between the breathing octave and the \sqrt2 rung is just the z=2 dispersion, not a fresh assumption.
• A clean restatement. \lambda=\sqrt2 is identical to \pi/s_0=\tfrac12\ln2=S_M: the limit-cycle log-period equals the per-vortex Majorana entropy the framework already carries three other ways (the chirality packing factor \varepsilon=\sqrt{2\pi S_M}/K, the GP pairing-two, the Majorana state count). It requires s_0=\pi/S_M\approx9.06, g\approx-82 — deeply supercritical, hence a dense comb, the opposite of Efimov’s shallow g\approx-1.26/sparse 22.7; dense-comb \Leftrightarrow large-s_0 \Leftrightarrow strongly-attractive coupling is internally consistent.
• What is still owed. The factor of two fixes the period but not the existence: nothing above proves g<-\tfrac14 — that the anti-phase pairing actually drives the inverse-square coupling supercritical so a limit cycle exists at all rather than a smooth power law. That is the irreducible residue, and it is the same marginal-point BdG\leftrightarrowGP self-consistency already gating Step A and the vertical cone. So the honest ledger moved one notch: period derived (conditional on a tower) and tied to S_M; tower-existence still owed.

Comb test on data — instrument built, verdict mixed (2026-05-31). scripts/comb_test.py turns the falsification signature into a reusable instrument: fold each measured ratio’s logarithm modulo \ln\sqrt2, then score (i) a Rayleigh test for clustering and (ii) an on-tooth alignment C=\langle\cos2\pi\phi\rangle (+1 = ratios sit on the \sqrt2 teeth, negative = clustered between them), plus a non-circular period scan (a log-space Schuster periodogram restricted to the rung band) and a random null. Run on the cleanest data the paper already cites:

• Grid modules (Stensola 2012, mean adjacent ratio 1.42): lands on a tooth — 1.42 vs \sqrt2=1.414, 0.4\% — a genuine hit, but effectively one well-measured ratio, since the modules form a single geometric progression.
• EEG band edges (Buzsáki): only the power-of-two edges (0.5,4,8 Hz) land on the comb; 12/30/80/200 Hz scatter (C=+0.29, not significant) — but the edges are the wrong quantity (human conventions, dynamically re-set by cortical state). The band centers are an independently established geometric ladder (Penttonen–Buzsáki 2003), and that literature’s own reconciliation \varphi^2\approx e (van Albada 2013 — the named band is two fine steps) matches the ladder’s octave-is-two-half-octaves exactly. The unsettled number is the fine step: substrate \sqrt2=1.414 vs resting-EEG golden ratio \varphi=1.618 (\sim\!14\%), a clean test that needs resolved spectral peaks, not band edges. Why the cortical column occupies the octave sub-lattice (integer-ratio nesting vs golden-ratio desync) is developed in cortical maps. (That test is now run — see Three-comb fold below; the fine ratio leans \varphi.)
• Microtubule cascade (Bandyopadhyay–Sahu 2020, explicitly titled “fractal, scale-free”): the analysis confirms a self-similar (DSI) cascade — but it is a “triplet of triplets” whose band centres are spaced by \sim2–3 decades and fold between the \sqrt2 teeth (C=-0.71), not onto them. Its log-period is the triplet (\times\!\sim\!40–1000), not \sqrt2. (Its discrete GHz peaks, n=2, are weakly on-tooth but underpowered.)

So the general DSI/log-periodic structure is supported broadly (grid, the MT cascade’s own fractal claim, Gutenberg–Richter), while the specific \sqrt2 period is supported by the one clean datum (grid) and is not yet decided: the richest cascade (MT) shows a different, much coarser period — exactly the two-families tension below, now with numbers. A decisive test needs raw, machine-measured size distributions (cryo-EM vesicle radii, per-cell grid scales, deconfounded MT peak lists), not published summaries and human-set band edges. The instrument is ready for them.

Three-comb fold + first real EEG run — the fine ratio leans \varphi, not \sqrt2 (2026-05-31). comb_test.py was extended from a single-comb test into a three-comb fold that decides among \sqrt2=1.414 (the substrate half-octave), \varphi=1.618 (the golden ratio — the resting-EEG fine ratio of Pletzer–Kerschbaum–Klimesch 2010), and the octave 2.0 (the integer-nesting rung) in one pass, the on-tooth C reported for each. It is validated on synthetic ladders (each of the three is fingerprinted correctly; a harmonic stack — peaks at 2f,3f,4f of one non-sinusoidal oscillator — folds onto no comb, so the instrument separates real rungs from harmonics), with a power budget: separating \sqrt2 from \varphi needs peak precision \sigma\lesssim 0.06 or n\gtrsim 40 ratios. The real pipeline (--eegbci) runs raw PhysioNet resting EEG \to MNE Welch PSD \to FOOOF/specparam peak extraction \to fold. First run: 109 subjects, eyes-closed, posterior channels, 5734 FOOOF peaks. Pooling all adjacent peaks is junk (dominated by FOOOF over-splitting — a pile at \times1.1–1.3 — not harmonics); the clean estimator takes one representative peak per canonical band (\theta/\alpha/\beta/\gamma) per subject \to 220 inter-band ratios, median 1.71. The full-set fold appears to favour \sqrt2 (p=0.006) — but that is an artifact: the octave 2.0=\sqrt2^{\,2} sits on the \sqrt2 comb, so octave-spaced band pairs fake a \sqrt2 win. Re-folding only the sub-octave fine structure (ratios <1.87, n=154) de-confounds it: \varphi wins decisively (C=+0.28, p<10^{-4}), \sqrt2 is rejected (C=-0.07, n.s.), the octave is rejected. Reading: the robust structure is the octave (\times2 band spacing — the nesting sub-lattice, consistent with the \sqrt2 family since 2=\sqrt2^2), while the genuine fine ratio is \varphi, not \sqrt2 — the long-standing Klimesch result. This is exactly the “if \varphi” branch the cortical-maps chapter spelled out: the rest-EEG ladder is the column’s own desync optimization (the most-irrational number), not a direct readout of the substrate’s pairing geometry — so \sqrt2 stays where the evidence is structural (grid 1.42), and EEG, always the loosest member, goes to \varphi. Caveats (why a lean, not a verdict): the per-band medians are shaped by the canonical band edges, so the sub-octave fold is partly band-binning-driven; it is eyes-closed occipital (alpha-dominated); the FOOOF-median representative is crude. The decisive version is a band-free peak assignment on a larger cohort — LEMON (OpenNeuro ds000221, 228 subjects); the pipeline is built and ready. Artifacts: scripts/eeg_peaks.csv, scripts/eegbci_109_run.log.

Cone mosaic — the anti-lock pole’s planar (hyperuniform) face (2026-05-31). The two-poles reading — locking structures on the comb’s teeth, anti-locking ones in its one gap — gained a second clean spatial witness, and it sharpened what “the gap” is. Phyllotaxis reaches the gap as the single golden ratio \varphi because its organs are added sequentially around a centre — the only sequence of rotations that dodges every rational lock is the most-irrational one. The retinal cone mosaic has no centre and no sequence: it tiles a plane all at once, where there is no single rotation to be irrational about, so its anti-lock optimum is not a ratio but a disordered hyperuniform “blue-noise” packing — the structure Yellott (1983) showed scatters aliasing into incoherent noise rather than coherent Moiré, and that Torquato’s group identified outright as “a disordered hyperuniform solution to a multiscale packing problem” in the avian retina (Jiao et al. 2014). scripts/cone_mosaic.py demonstrates it on three equal-density planar patterns scored by structure factor + normalized number variance \sigma^2/\langle N\rangle (=1 for a random gas, \to0 as long-wavelength fluctuations are crushed): the triangular lattice carries a Bragg comb with \sigma^2/\langle N\rangle\approx0.03 (locked, uniform); the Poisson gas has no ring and \sigma^2/\langle N\rangle\approx1 (disordered, clumpy); the Lloyd-relaxed blue-noise mosaic has no comb (one diffuse ring) yet \sigma^2/\langle N\rangle\approx0.10 — as uniform as a lattice, as disordered as a gas. So \varphi and blue noise are one pole in two geometries, circular vs planar, selected by whether the structure is built in sequence or all at once; the eye-chapter section ties the disordered-hyperuniform mosaic to the substrate’s own domain-structured (disordered-yet-uniform) texture, and adds an eccentricity-tuned falsifier. This is forward-prediction support for the sign-rule in a second spatial domain that shares no chemistry with the brain — not a derivation of the ladder; the \sqrt2 tower’s existence (above) is untouched. Artifacts: scripts/cone_mosaic.py, scripts/cone_mosaic_run.log.

Prime cycles — the anti-lock pole’s discrete-temporal face (2026-05-31). The gap gained a third witness and, with it, the generalization that turns “two geometries” into one principle. The two spatial faces (phyllotaxis \varphi, cone-mosaic blue noise) and the continuous-temporal one (resting-EEG \varphi) all read the gap in a continuous variable. Read it where the variable is a whole number of generations and there is no irrational to take: the most non-resonant integer is the one that shares no factor with any cycle it must dodge — a prime. The clean datum is the periodical cicada (Magicicada), whose 13- and 17-year cycles are both prime; the textbook explanation is exactly anti-lock (a prime period co-emerges with a threat of cycle q only every \mathrm{lcm}(P,q) years, minimizing both predation overlap and cross-brood hybridization — Williams & Simon 1995; Yoshimura 1997; Tanaka et al. 2009). scripts/prime_resonance.py is the discrete-time analog of comb_test.py/cone_mosaic.py: scoring each candidate period by its gcd-weighted resonance load against a field of threat cycles, the primes 11,13,17,19 sit exactly on the resonance floor (\rho=1.000) while every composite pokes above it, and inside the observed 12–18 window the two lowest-resonance periods are exactly 13 and 17 — the anti-lock pole’s “golden angle” cast in integers. It is substrate-level, not insect chemistry, in the same sense Levitov (1991) made \varphi-phyllotaxis physics: Goles, Schulz & Markus (2001) showed prime cycles emerge as the attractor of a generic, cicada-free predator–prey avoid-resonance dynamic (“an encounter of biology and number theory”), and Webb (2001) reached primes by a cicada-free renormalization argument. So the gap has one principle — maximal incommensurability — and three arithmetics: the reals mod one (continued fractions \to\varphi), the plane (Fourier/packing \to blue noise), the integers (gcd \to primes). The new content is the unification — that 13 and 17 belong on the same axis as the golden angle and the hyperuniform retina — and the sign-rule now reaching a third domain (insect ecology) sharing neither chemistry nor history with meristem or brain. Not a derivation of the \sqrt2 tower; the tower’s existence (above) is untouched. (Grace note, not mechanism: 13 is also the smallest cicada-window Fibonacci number — the discrete shadow of \varphi — so it sits at the meeting of the gap’s circular and discrete faces; the prime property is what the resonance pressure selects.) Artifacts: scripts/prime_resonance.py, scripts/prime_resonance_run.log.

The prediction engine at both poles — the sign-rule reaching the brain’s own computation (2026-06-01). The brain-as-prediction-engine chapter was brought onto the ladder, and the genuinely new yield is that the predictive-coding engine runs at both poles by operation, not only by state: it binds on the comb’s octave/\sqrt2 teeth (theta–gamma integer nesting, working-memory maintenance, the canonical loop’s coherence-match) and separates in the \varphi gap (the resting/default-mode desync already measured at \varphi; dentate-gyrus pattern separation, the anti-lock partner of CA3 pattern completion — the hippocampal separation\tocompletion dyad is the two poles in series). Free-energy minimization is one-sided as usually stated; the engine must simultaneously keep its hypotheses distinguishable, so “lock to bind, anti-lock to keep separable” is the substrate reading. Two clarifications fell out alongside: the cortical eigenmode basis is a log-spaced multiresolution (wavelet) decomposition, not the evenly-spaced Fourier “spectral” basis the chapter had reached for (the same harmonic-template slip corrected elsewhere), which unifies the chapter’s eigenmode-basis and multi-rate-integration passes into one architecture; and each column is a string (length \to overtones) while the ladder lives in the \sqrt2-spaced set of column lengths (the keyboard) — the keyboard-vs-string cut again. Like the codon code and the olfactory/trafficking address codes, hippocampal separation is the framework’s own biology meeting the gap by the labelling/decorrelation route, so it is an application of the sign-rule (new prediction #5: separation codes test more \varphi-spaced and more blue-noise/hyperuniform than binding codes), not a chemistry-free independent face of the gap. Not a derivation of the \sqrt2 tower; the tower’s existence (above) is untouched.

The body modon at both poles — HRV as the macroscopic readout (2026-06-02). The vagal-highway chapter was brought onto the ladder, extending the both-poles set from the brain to the organism modon. Heart-rate variability is its readout: healthy resting HRV sits at the anti-lock pole — broadband, 1/f, fractal, the heart refusing to lock to any single rate so it stays adaptable (Kobayashi & Musha 1982; Ivanov et al. 1999; Goldberger et al. 2002) — and slides to the lock pole when the visceral rhythms must bind (respiratory sinus arrhythmia, the baroreflex at the Mayer rung, HeartMath cardiac coherence’s sharp 0.1 Hz Lorentzian). This resolves the standing HRV paradox (high variability and coherent locking both read as healthy): they are the two poles, and health is the capacity to slide, not residence at either — so the two clinical failure modes are the two stuck poles, single-band low-power over-lock (the reduced-complexity mortality predictor; Lipsitz & Goldberger 1992; Costa, Goldberger & Peng 2002) and flattened decoupling, and the polyvagal triad sorts as pole-pathology exactly as the bilateral disorders do. The genuinely new physical content is the criticality tie: healthy HRV’s 1/f continuum carrying discrete respiratory/Mayer/ultradian rungs is the body-modon readout of the substrate’s own marginal, scale-free \mu\to0 point with the DSI ladder riding on it — the heart well-composed when it sits where the substrate does. Like the cone mosaic and the codon code, the empirical facts are textbook; the new content is the unification and the sign-rule, and it is not a derivation of the \sqrt2 tower (whose existence, above, is untouched). New prediction #6 + falsifier (f) in the chapter; scorecard row in predictions.

Markets and transformers at both poles — the sign-rule past biology (2026-06-02). Two chapters written before the ladder were brought onto it, extending the both-poles set beyond biology. Economics: a market binds on the teeth (one clearing price, settlement, entrained business cycles) and rests at the anti-lock gap (diversification — Markowitz 1952 — and the strategy diversity of Lo’s adaptive markets, where holdings must stay uncorrelated to spread risk); a systemic crash is the gap failing, cross-asset and cross-participant correlations climbing toward one (the network-finance “systemic risk = synchronization” reading, Haldane 2009) — the organism-of-organisms seizure — and the suppressed-recession claim gains a pole (a system pinned to lock loses its decorrelation capacity, Minsky’s stability-breeds-instability). Transformers: a both-poles system by representational geometry — coherence-match attention on the lock pole, and superposition at the anti-lock pole, where the residual stream’s nearly-orthogonal feature directions (Elhage et al. 2022) are a Thomson/Tammes packing on the d_\text{model} hypersphere, the spherical twin of the cone mosaic’s blue noise (avoid feature interference as the retina avoids aliasing and the code avoids codon confusion). New predictions added to each (economics #7 + falsifier g: pre-crash correlation rise / resilient-market decorrelation; transformers #6 + falsifier f: SAE feature directions anti-lock-spread vs aligned routing subspaces), a clause in the predictions two-poles narrative, and two stale-spot fixes — economics had called firm sizes, EEG bands, and currency “the same factor-of-10 rung” (they are different ratios; the economic ladders are the coarse family, EEG band-centers the resonant one), and the transformer Chinchilla tokens-per-parameter was disentangled from the width-to-depth aspect ratio. As with the cone mosaic, codon code, and HRV, these are applications of the sign-rule by behaviour/engineering rather than chemistry-free witnesses of the gap, and not a derivation of the \sqrt2 tower (whose existence, above, is untouched).

Galactic dynamics — the boundary as the breath, the sign-rule reaching gravity (2026-06-02). The galactic-dynamics chapter was brought onto the ladder — the first of the Cosmology section, which predated it entirely. The load-bearing clarification: the counter-rotating boundary whose parity symmetry forces the quadratic MOND current-phase relation is the framework’s own anti-phase breath (the Cooper/antiferromagnet pairing), so the boundary is parity-even because it is a pair — MOND is the breath intact (the paired, coherent, parity-even quadratic response) and Newtonian gravity is that breath un-paired by the Hubble/external-field DC bias, which unifies the external-field effect with the very existence of a Newtonian limit. The Josephson chain (N=r/\xi stacked boundaries) is the breath replicated across scale — a coarse-family ladder spaced by \xi, not a \sqrt2 comb, like the solar boundary stack; and the crust undular bore’s chirped, rank-ordered train is DSI made cosmologically visible (also coarse family, its period KdV-set). The two-poles sign rule now reaches gravity a second way: the Landau-velocity threshold (v_L\approx750 km/s — superfluid-MOND below, normal CDM-like above) keyed to velocity, joining the orbital-resonance trap/clear reading — honestly flagged as a coherence threshold (lock engaged vs disengaged), not the anti-lock \varphi gap, which by the sign rule should not appear where the job is to bind. New synthesizing section + a sharpened one-knob RAR falsifier (any RAR scatter not reducible to the single external-field knob breaks the universal-tooth claim); scorecard row in predictions. Not a derivation of the \sqrt2 tower (whose existence, above, is untouched).

Testable consequence. DSI predicts observables are power laws modulated by a log-periodic function: histograms of clustered quantities, plotted against \log x, should show peaks evenly spaced by \ln\sqrt2\approx 0.347 — a \sqrt2 comb of octaves and half-octaves. Reanalysis of existing organelle-size, vesicle-radius, grid-spacing, and MT-resonance datasets is the immediate test (scripts/comb_test.py is the harness); the grid-cell \sim\!1.4\times module ratio is the cleanest datum already on the comb.

Caution. Two ratio families are not yet unified: the resonant family (grid cells, octaves, EEG centers) sits at \sqrt2 and low powers — though a first real-data run puts the EEG fine ratio at \varphi, not \sqrt2 (only the octave structure, 2=\sqrt2^2, stays on the comb; see above); the coarse family (mechanoreceptor half-decades; the \times6–12 spatial-nesting steps; the economic ladders — currency denominations, firm sizes, business cycles; the transformer’s width-to-depth aspect ratio; \xi/d_\text{GJO}\approx 6.9, which need not be \sqrt2-commensurate since \xi and d_\text{GJO} come from different physics) is looser. Whether the coarse family is the same tower sampled every \simsixth rung or a second log-period is open. As with WIP-22, distinguishing a real substrate ladder from a chemistry-set coincidence requires the log-periodic signature to survive in data that chemistry alone does not organize.

WIP-27: The e-fold count from the bridge length

Status (settled to a scaling estimate; the bounce closes it). The number of inflationary e-folds is not tuned by an inflaton potential here — it is geometric, counting how far the nucleating bubble’s wall recedes before it exits the observer’s past light cone. The striking result is that it is already fixed, to ~15%, by the bridge equation’s own coherence length:

N_* \approx \ln\!\left(\frac{c}{H_0\,\xi}\right) \approx 69.4,

built from only c, H_0, and \xi \approx 97\;\mum — no new parameter. It overshoots the canonical \sim60 by ~15%, and the correction is negative and small (subluminal wall; light-cone exit at the inflationary, not the present, Hubble rate). The same thin-wall bounce gives S_E/\hbar \sim \mathcal{O}(1) — near-spinodal nucleation, the same scale-free \mu\to0 marginal point (WIP-15 item 10) that lets the substrate emit light at one speed and that WIP-26 reads as discrete scale invariance: emergent light, the ladder, and barrierless nucleation are three faces of one criticality.

Open work. The one calculation that closes it is the Gross–Pitaevskii bounce, which fixes both the critical-bubble prefactor R_c = \alpha\,\xi (order unity) and S_E/\hbar. A second, coupled, unknown is the wall velocity v_w(\alpha_{mf}) vs. the inflationary Hubble rate H_\text{inf} — the crust’s resolved structure (the 15-knot spline) constrains the product v_w \cdot t_\text{inf}, giving two equations in two unknowns. Full treatment in Why \sim60 E-folds; the spectral index n_s \approx 0.968 rides on N_* (spacetime-dynamics-inflation.qmd).

Sources: early-structure-formation.qmd § Why ~60 E-folds (Open Calculation 1: the GP bounce); spacetime-dynamics-inflation.qmd § The Number of E-Foldings Is Natural.

WIP-28: Why the substrate is invisible — stealthy hyperuniformity

Status (synthesis from existing commitments; the owed calculations now run). The framework’s single most-asked objection — if space is full of a dense superfluid, why don’t we see it? — now has a second answer alongside emergent Lorentz invariance, and it is sharper. Visibility is governed by the structure factor S(\mathbf q)=\tfrac1N|\langle\rho|e^{i\mathbf q\cdot\mathbf r}\rangle|^2 (single-scattering intensity \propto S), which is literally the coherence-match of the lattice density with the probe plane wave. Three textures can fill space; the sky rules out two. A periodic aether carries a Bragg comb — it diffracts starlight into an opal and its reciprocal vectors define a rest frame (a sharper Michelson–Morley failure than the timing argument). A random gas scatters at every \mathbf q — space would be fog. Only a disordered hyperuniform texture threads both needles: S(\mathbf q)\to0 as \mathbf q\to0 (Torquato–Stillinger), no Bragg comb, a single diffuse ring at |\mathbf q|\sim2\pi/\xi. This is not an added assumption: the framework already requires the lattice to be a domain glass of triangular Abrikosov crystallites (the bridge \eta=1 factor), locally crystalline and globally disordered-yet-uniform — which is the construction of a disordered hyperuniform solid. Stealth and isotropy are the same fact about the texture. Stealthy hyperuniform media are transparent at finite density (Leseur, Pierrat & Carminati 2016) and open isotropic photonic band gaps (Florescu, Torquato & Steinhardt 2009; Man et al. 2013); the rigorous transparency regime (\lambda\gtrsim\xi) coincides exactly with the framework’s collective-modon band, and the stealth-window edge (|\mathbf q|\sim2\pi/\xi, \sim3 THz, \sim13 meV) coincides with the modon-localization crossover E_\text{min}=2\pi m_1c^2 — one number, reached two ways. The vacuum is thus the deepest member of the paper’s anti-lock roster (cone mosaic, genetic code, phonemes, transformer features): name its job — carry signals without scattering them off your own texture — and the pole is fixed (anti-lock, S\to0). Full treatment in The Stealth Vacuum.

Open work — now computed (2026-06-11). The substrate analog of measuring the cone mosaic’s structure factor has been run, lifting the scripts/cone_mosaic.py number-variance instrument from the plane to the 3D lattice (scripts/stealth_structure_factor.py, sessions/stealth-structure-factor-1.md). The framework’s literal texture — a domain glass of 48 randomly-oriented triangular Abrikosov crystallites in a periodic 12\xi box, with S(q) read on the box’s own reciprocal grid (FFT, window-deconvolved, so the finite-box envelope cancels) — comes out disordered hyperuniform: S(\mathbf q\to0)=0.28 against a gas’s 0.99, normalized number variance 0.44 against the gas’s 0.97 (both \sim2\times below the Poisson floor), and a single diffuse powder ring at |\mathbf q|=1.12\,q_0 (q_0=2\pi/\xi, the cell scale) carrying no Bragg comb — the 48 orientations turn the single crystal’s sharp spots (directional contrast S_\text{max}/S_\text{mean}\approx1560) into an isotropic Debye–Scherrer ring (contrast \approx14, two orders lower), so no reciprocal lattice and no preferred frame survive. Seed-robust across {7, 42, 137, 2718}. The ring’s inner edge |\mathbf q|\simeq2\pi/\xi is the modon-localization crossover E_\text{min}=2\pi m_1c^2 — one number, two derivations.

The stealthy limit (2026-06-11). The Part-1 caveat — that the small 48-grain/12\xi box is only class-III hyperuniform (S\to0 suppressed but not a hard S=0 window) — is now resolved as a finite-grain artifact (scripts/stealth_limit.py, sessions/stealth-structure-factor-1.md §4). A polycrystal’s small-q fluctuations live on its grain boundaries, so S(\mathbf q\to0)\sim (boundary fraction) \sim1/D for grains of linear size D. Growing the crystallites drives the stealth-window S(\mathbf q\to0) down monotonically (0.16\to0.055 over D=2.6\to7.9 cells), and a fit S=S_\infty+c/D extrapolates to S_\infty\approx0.01 (statistically zero across four seeds) against the gas’s \approx1.0: the well-relaxed, large-grain domain glass is stealthy — S=0 across a finite window, not merely hyperuniform at \mathbf q=0. The physical substrate (\xi\sim100\,\mum cells, domains spanning many cells) sits at the large-D, near-stealthy end. (Two modelling notes: the crystallite is the net-density isotropic cell — the d_\text{GJO} sub-layer is the anti-phase partner whose net modulation cancels above one cell — and a single crystal in a cubic box carries an incommensurability seam, so the seam-free glass extrapolation is the floor.) So the stealth identification is now a computation reaching the stealthy limit — that result fixes the small-q amplitude; the small-q shape (the hyperuniformity class) is computed next. The short-wavelength regime (\lambda\ll\xi) is not covered by the stealth theorem — there hyperuniformity earns only “no Bragg comb, so no opal,” and the positive account of visible transparency stays with the crystal-optics off-resonant-boundary mechanism; the two partition the spectrum at \xi.

The hyperuniformity class, now measured from scratch (2026-06-17). The one residual the stealthy-limit result left open — the shape of the small-q suppression, i.e. which hyperuniformity class the as-constructed texture occupies, the piece WIP-22’s caution attached to — is now computed (scripts/stealth_class_exponent.py). A hyperuniform medium climbs out of the origin as S(\mathbf q)\sim|\mathbf q|^\alpha, and the exponent \alpha names the class (Torquato 2018): \alpha>1 class I, \alpha=1 class II, 0<\alpha<1 class III. Measured two independent ways on the isotropic-cell domain glass — a direct ensemble- and shell-averaged S(\mathbf q) fit below the ring, and the window-free number-variance exponent \sigma^2(R)\sim R^p read as \alpha=3-p — the from-scratch texture is class III with \alpha\approx0.5 (direct \alpha=0.51\pm0.06 pooled over grain sizes; number-variance \alpha=3-p\approx0.55, p\approx2.45), the two estimators agreeing, against a Poisson-gas null both routes read as \alpha\approx0. Seed-robust across {137, 42}. (The direct k-space slope is window-sensitive — \approx0.3 at the smallest accessible q, steepening toward the ring — because a finite box’s smallest |\mathbf q| still sits where the texture is turning over; the real-space number variance, free of any k-space fit window, is the robust anchor.) Exponent and amplitude are two faces of one suppression: the unrelaxed texture is class III (\alpha<1 — it climbs out of \mathbf q=0 as a sub-linear power), and growing the crystallites (the stealthy-limit result above) drives that climb’s amplitude to zero, flattening it into the hard S=0 window of a stealthy medium. WIP-22’s caution thus no longer hangs on an unmeasured exponent: the class is named.

Testable consequence — a far-IR opacity pinned to \xi (2026-06-11). Feeding the computed S(\mathbf q) into a Born scattering integral (scripts/stealth_farir_opacity.py) turns the ring into the one thing the sky can test. Back-scatter caps momentum transfer at q=4\pi/\lambda, so single scattering off the ring (q=2\pi/\xi) switches on only at \lambda\le2\xi: the vacuum is transparent for \lambda>2\xi (the stealth window) and the onset is the cell scale and nothing else — 2\xi=194–224\,\mum, \nu=c/2\xi=1.34–1.55 THz (Routes 2 and 1), rising to the ring/floor at \lambda=\xi (\nu\approx2.9 THz, E\approx13 meV). The opacity shape there is the measured S(\mathbf q); only the per-cell coupling (the cell polarizability) is free at this stage — and that amplitude is itself pinned below (it is the dispersion contrast squared, not an independent parameter). Because the edge is at a fixed local frequency, a photon observed at \lambda_\text{obs} scattered only for z>\lambda_\text{obs}/2\xi-1 (1 mm \leftrightarrow z>3.8, 500 \mum \leftrightarrow z>1.4, \lesssim200\,\mum \leftrightarrow whole path), so the far-IR sky reads the ring tomographically. Confrontation: the universe is optically thin in the far-IR/submm to z\sim4–6 (Herschel PACS/SPIRE resolve far-IR sources to the diffraction limit; ALMA detects dusty galaxies at z>4–6), which bounds \sigma_\text{cell}<3.8\times10^{-31}\xi^2 (\tau<1 to z=4). The contrapositive is the result: a random gas (\sigma_\text{cell}\sim\xi^2) would give \tau\sim2\times10^{30} — the far-IR universe opaque by \sim30 orders — so its transparency requires the \sim10^{30} per-cell suppression that only a hyperuniform texture supplies (the “poisson = fog” verdict, read on the sky). The signature, if ever resolved: a single diffuse ring around a dark S\to0 hole switching on in the 0.3–10 THz gap — neither a Bragg comb (periodic aether; breaks Michelson–Morley) nor flat speckle (random medium; fogs the sky) — and nothing below it.

The amplitude is not free — it is the dispersion contrast, squared (2026-06-18). The one piece the far-IR opacity left open — the per-cell coupling \sigma_\text{cell} (the vortex-cell “polarizability”), which the structure-factor program bounded (\sigma_\text{cell}<3.8\times10^{-31}\xi^2) but did not derive — turns out not to be an independent parameter (scripts/stealth_cell_polarizability.py). A photon crossing one cell sees a per-cell forward amplitude whose real part is the index contrast \delta n — the phase advance that is the speed deviation the FRB/floor dispersion program bounds — and whose imaginary part is, by the optical theorem (\sigma=\tfrac{4\pi}{k}\,\text{Im}\,f(0)), the extinction \sigma_\text{cell}. They are Kramers–Kronig partners of one response; in the Born/Rayleigh regime \sigma_\text{cell}\sim\tfrac{8\pi}{3}k^4\alpha^2 with \alpha=\tfrac{\xi^3}{3}\delta\varepsilon, so \sigma_\text{cell} is the dispersion contrast \delta n read through its square rather than linearly — it cannot be set independently of it. Why the contrast is tiny rather than order-unity is the same fact that kills the FRB \nu^2 term: the Volovik identity c=\hbar/(m_1\xi) fixes light’s speed from the circulation quantum and the healing length alone, not the local substrate density, so a denser/sparser vortex cell shifts c at no power-law order (\delta n\to0, residual only \exp(-\nu_\text{floor}/\nu)). The winding conservation that forbids the dispersion power law forbids order-unity elastic scattering of the same conserved winding — so the cell is invisible twice over: its neighbours are arranged so S(\mathbf q)\to0 (the spatial suppression, above) and each cell on its own presents \delta n\to0 (this per-cell topological suppression). The cross-check is the payoff: two utterly independent observables bound one microscopic number. The square root of the spatial bound, \sqrt{\sigma_\text{cell}/\xi^2}=\sqrt{3.8\times10^{-31}}=6.2\times10^{-16}, lands on the temporal (CHIME/FRB) per-cell contrast bound \varepsilon<6.5\times10^{-16} to better than a factor of two (ratio 0.95) — and inverting Rayleigh, both read the per-cell index contrast at \sim10^{-16}–10^{-17}, from Herschel/ALMA source counts at z\sim4 on one side and FRB timing at 600 MHz on the other. So the “free amplitude” collapses: the residual genuinely still owed is no longer \sigma_\text{cell} as a separate unknown but the single modon-core reconnection action that fixes the dispersion exponential’s coefficient and this scattering amplitude together (the “radio-photon microphysics” of below_floor_dispersion.py) — now computed at barrier level (2026-07-05) as a Gross–Pitaevskii phase slip, \alpha \approx 0.2–0.6 (scripts/reconnection_barrier_from_above.py), closing both at once. (Note this action fixes the sub-floor law and this amplitude only; it is not the coefficient of the speculative tier-below \delta — that is the tower coupling s_0^2, thirty orders too large to be a reconnection barrier; see The Tier Below.)

WIP-29: Neutrino flavor oscillation from boundary-strain mass eigenstates

Status (2026-06-17): qualitative neutrino narrative complete; oscillation genuinely open. Surfaced while assembling The Standard Model in the Substrate, which collects the framework’s neutrino reading into one place for the first time. The static picture is coherent and rests on results already derived: a neutrino is the one fermion with two of its three geometric handles removed — the braid word has only crossings, no twists (charge zero) and no three-fold junction (color zero), leaving boundary strain as its only coupling. That single fact delivers production (only via chirality-flipping weak processes, the pp-chain flip), near-non-interaction, the left-handed-only rule, and the sterile right-handed twin / seesaw (Higgs Field).

What is missing: flavor oscillation. Nothing in the framework yet explains why a neutrino born \nu_e can be detected \nu_\mu. The natural substrate reading is that the three neutrino mass states are the three radial harmonics of the twist-free, junction-free knot — the leptonic generation ladder of WIP-21 and the generation-as-harmonics picture, but with charge and color absent. A propagating neutrino is a superposition of these boundary-strain eigenstates, each with a slightly different internal energy; they beat against each other in flight, so the measured flavor cycles with distance. Mixing is then the misalignment between the weak-interaction basis (which strained-boundary configuration the W produced) and the mass basis (the harmonic eigenstates).

Open work:

• Compute the three neutrino mass eigenvalues as harmonic excitations of the twist-free knot — the same fold-energy calculation as the charged-lepton ladder (m_\mu/m_e, m_\tau/m_\mu in WIP-21), minus the junction terms — and check whether the absence of charge/color naturally compresses the splittings to the observed \Delta m^2 \sim 10^{-3}–10^{-5}\,\text{eV}^2.
• Derive the PMNS mixing angles as the overlap between the boundary-strain (weak) basis and the harmonic (mass) basis; ask why lepton mixing is large where quark (CKM) mixing is small — plausibly because the junction that rigidly aligns the two bases for quarks is absent for neutrinos.
• Pin the sterile right-handed mass scale from the seesaw and check consistency with the tiny left-handed masses and the neutrino’s position (\alpha \sim 10^{-9}) on the visibility spectrum.

Candidate direction (2026-07): the mixing asymmetry from the \mathbb Z_3 clock. The second bullet — why lepton mixing is large where quark mixing is small — now has a concrete mechanism that needs no new machinery (Three Generations § Mixing). Mixing between two triads is the relative rotation of their two \mathbb Z_3 clocks, and the chapter’s already-published 3\delta-Q lock table fixes the sign: up and down quarks are color-loaded to the same side of the lock (3\delta-Q=-0.61,-0.40), so their two mass bases co-rotate and nearly cancel — small, hierarchical CKM; the charged lepton sits on the lock (0) while the neutrino is dragged the opposite way by the visibility floor (+0.35), so the lepton bases straddle it — large PMNS. This is the framework’s own “the junction that aligns the two bases for quarks is absent for neutrinos,” made literal (color co-loads both quark sectors identically). One suggestive anchor rides along (flagged in-text as a coincidence the framework absorbs, not a forced number): the maximal atmospheric angle \theta_{23}\approx45^\circ echoes the pairing-\sqrt2 tilt \cos45^\circ=1/\sqrt2 the Koide sector carries — though the two \sqrt2’s live in different spaces (a \sqrt{\text{mass}}-space amplitude vs a flavor-basis rotation angle). Structurally, the \mathbb Z_3 singlet is the democratic column (1,1,1)/\sqrt3 — trimaximal mixing, and the same \mathbb Z_3\subset A_4/S_4 mainstream flavor models use for tri-bimaximal. CP (2026-07 addition): the same reading hands the CP phase a home — a \mathbb Z_3 clock is built from the complex cube roots of unity \omega=e^{2\pi i/3}, so its mixing unitary is intrinsically complex and a large Dirac CP phase is generic, not tuned; the two sectors share that near-maximal phase and differ only in how far color collapses their real angles (\delta_\text{CKM}\approx65^\circ and the near-maximal PMNS hint are both consistent; small J_\text{CKM} comes from the small quark angles, not a small phase). This is the one place PMNS and CKM might both be large, and it predicts a near-maximal \delta_\text{PMNS}. What remains open: the mechanism fixes the sign of the real-angle asymmetry, the one 45^\circ echo, and the genericness of large CP, but not the individual angles or the CP phase’s value — the raw 3\delta-Q gaps run the right direction (0.21 quark vs 0.35 lepton) yet understate how sharply CKM collapses. Turning the lock displacement into actual |V_{ij}| still owes the overlap integral between the harmonic and boundary-strain bases. (A 2026-07 numerical check asked whether the clock reproduces the Gatto–Sartori–Tonin relation \theta_C\approx\sqrt{m_d/m_s} without an imposed constant; verdict honest-but-partial, scripts/wip29_gst_cabibbo_check.py: the \sqrt{}-form is native — the clock’s variable is \sqrt m, so a ratio of adjacent clock amplitudes is \sqrt{m_\text{light}/m_\text{heavy}} identically — but the angle’s magnitude, the up/down pairing, and the relative sector phase stay empirical, so it does not derive the Cabibbo angle. Recorded, not claimed.)

This is the quantitative partner to the Yukawa program: both turn boundary architecture into mass numbers, and resolving the lepton harmonics would feed directly into both.

Sources: standard-model.qmd; higgs-field.qmd; mass-rotational-energy.qmd; proton-core.qmd; WIP-21.

WIP-30: The condensation number \nu — is the electroweak lift a derivable dimensionless number?

Status (2026-07-04): the two-route spread RESOLVED — it is the cell occupancy, and it closes to the SC2 value \nu=8.3\times10^8; the bottom-up derivation of \nu remains open (the DSI tower). Earlier (2026-07-01): target isolated and moved onto legal ground; numbers sharpened. This entry states, as precisely as the framework now allows, what a bottom-up “Route 2” to the lattice size would actually have to compute — and why the anti-phase breathing, rightly read, reframes that task rather than performing it. Sharpened (2026-07-01): \nu anchored at 9.6\times10^8 (\ln\nu=20.68) on the clean cosmology route, its spread identified as the bridge’s own two-route closure; the two candidate forms shown to be one statement (the BCS exponent fixes the tower’s rung count); the reduced/full Compton notation trap flagged and removed; the C-09 dilemma split into its two horns (length-horn dissolved via dimensional transmutation, circularity-horn made concrete at the tower); and the “one length + one scaffold + one open number” reframe carried into the bridge chapter and substrate-particles. It is the constructive companion to the bridge equation’s C-08/C-09 dilemma (the “broken cube root”): where that critique says what the electroweak route cannot honestly do, this says what is left to do and where it lives.

The whole residue is one pure number. Everything dimensionally honest in the bridge equation — the geometric 4\pi/(K\sqrt2), the GP-2, the effective quantum m_\text{eff}=m_e/\alpha_{mf} — leaves exactly one thing unexplained: the nine-decade lift from the electroweak reduced Compton length \bar\lambda_C(m_\text{eff})=\hbar/(m_\text{eff}c)\approx116 fm to the lattice cell \xi\approx111\;\mum (the cosmology anchor’s value; the round “\sim100\;\mum” used loosely below). That lift is the condensation number

\nu \;=\; \frac{m_\text{eff}}{m_1} \;=\; \frac{\xi}{\bar\lambda_C(m_\text{eff})} \;=\; \frac{2\pi\,\xi}{\lambda_C(m_\text{eff})} \;\approx\; 9.6\times10^{8}, \qquad \ln\nu \approx 20.68,

anchored on the clean cosmology route (m_1=\rho_\text{DM}^{1/4}=1.77 meV, the marginal-fluid identity \xi_\text{Compton}=\xi_\text{packing} with no broken cube root; scripts/wip30_condensation_number.py). The residual wobble to 8.3\times10^{8} (\ln\nu=20.54) is not a loose convention but the bridge equation’s own two-route closure: the same m_1 read off the SC2 particle-physics side (\xi=96.9\;\mum) rather than cosmology — the 13–15% the framework already carries, no new slack. (This anchor is superseded below: the spread is now resolved as the cell occupancy, and \nu=8.3\times10^8 — the SC2 value — is adopted as primary. The 9.6\times10^8 here is the f\to1 limit.) (One notation trap: the reduced length is \bar\lambda_C\approx116 fm — the “\approx120 fm” quoted loosely elsewhere — so the 2\pi form must be paired with the full \lambda_C=730 fm; the trap-free statement is \nu=\xi/\bar\lambda_C, no 2\pi.) Predicting \nu is predicting Route 2; nothing else in the electroweak side is unaccounted.

A third vote, from the outer rim (2026-07-04). The two-route \nu spread now has an independent tie-breaker that does not come from the bridge: the outer-rim clean-units law v_L=c(4\pi/\nu)^{1/3} (Outer Rim Onset, item 4). Fed the measured fast-solar-wind anchor (v_L\approx751.5 km/s, Ulysses), it reproduces v_L to -1.3\% on the SC2 leg (\nu=8.3\times10^8, \xi=96.9\,\mum) but only -6.0\% on the cosmology leg (\nu=9.6\times10^8), so the solar wind favors the SC2 anchor by a factor \sim4 in residual. This also collapses a would-be separate open item — the outer rim’s “Q\to1” convergence test, Q\equiv\nu/[4\pi(c/v_L)^3], is not an independent test but this same two-route closure: its error budget is \nu-route-dominated (16\% swing vs <1\% from the wind anchor; scripts/route_a_landau_rim.py), because \nu=\xi/\bar\lambda_C makes the \nu spread and the \xi spread one object. Net: closing WIP-30 to the SC2 value simultaneously drives Q\to1.03 and turns the residual 3\% into the outer rim’s clean coefficient/shape test for the Gauss 4\pi; the cosmology value would leave Q\approx1.19 and a 6\% outer-rim miss. The framework’s cheapest cross-check on the \nu route is therefore the fast solar wind.

Resolved (2026-07-04): the two-route spread is the cell occupancy, and it closes to SC2. scripts/wip30_packing_reconciliation.py. The “13–15% closure” is not two routes at all — it is one close-packing calculation quoted at two cell occupancys. Close-packing reads n_1\xi^3=f; with the Volovik identity \xi=\hbar/(m_1c) and n_1=\rho_\text{DM}/m_1 it becomes \rho_\text{DM}\,c\,\xi^4/\hbar=f, so \xi\propto f^{1/4}. Both legs use the same \rho_\text{DM}; they differ only in f. The cosmology quote (\xi\approx111.8\,\mum, \nu\approx9.6\times10^8) sets f=1 — the round “n_1\xi^3\approx1.” The SC2 leg uses the framework’s own derived cell occupancy f=4\pi/(K\sqrt2)=0.5666 (bridge equation, zero-parameter), giving \xi=111.8\,f^{1/4}=97.0\,\mum and \nu=9.6\times10^8\cdot f^{1/4}=8.3\times10^8 — the SC2 value to the quoted precision (the entry’s own dag check already records \rho_\text{DM}c\,\xi_\text{SC2}^4/\hbar=0.566=f). Since f is derived, not approximate, the honest close-packing length carries it: adopt \nu=8.3\times10^8, \xi=96.9\,\mum, m_1c^2=2.04 meV as primary. The cosmology 9.6\times10^8 is recovered exactly as the f\to1 limit, high by precisely the dropped f^{-1/4}=1.153 — a rounding of the close-packing condition, not a second measurement.

This is corroborated by over-determination: \nu is now fixed by three genuinely independent inputs that agree to \sim5\% — the derived packing f (8.33\times10^8), the measured electroweak VEV v=246.22 GeV via \nu=v^2/8\pi m_\text{eff}^2c^4 (8.36\times10^8), and the measured fast-solar-wind v_L=751.5 km/s via the clean-units law \nu=4\pi(c/v_L)^3 (7.98\times10^8). Only the f=1 cosmology quote sits outside the cluster (mean \nu\approx8.2\times10^8), high by exactly f^{-1/4}. The \rho_\text{DM}, v, and v_L inputs share no physics — cosmological density, collider electroweak, heliospheric flow — so their landing at one \nu is a real triangulation, not a restatement.

What this does not do is derive \nu bottom-up — that stays the DSI-tower question below (the C-09 circularity horn, still open). It resolves which anchor value is correct and why the two differed, retiring the “two-route spread” as an open uncertainty: the spread was the cell occupancy all along. (Bookkeeping still owed: the numeric sweep 9.6\to8.3\times10^8, 111\to97\,\mum through the boxed \nu equation above and the downstream chapters — bridge equation, substrate particles, gravity — carried through for consistency.)

The breathing reframes a forbidden length as a legal number. The naive picture — the electron vortex breathing from r_\text{eff}\approx150 fm out to \xi\approx100\;\mum each Compton cycle, physically carrying the scale up — is causally impossible and already retired (WIP-12): per cycle the breath is capped at \bar\lambda_C\approx386 fm, and the 100\;\mum envelope is a quasi-frozen coherence dress, not a dynamical stroke. What survives is more useful. The two-breath analysis shows the 10^9 is not a stretched length at all but a two-clock ratio — \xi/\bar\lambda_C=\omega_C/\omega_1=m_e/m_1=\alpha_{mf}\nu — i.e. a collective occupation number: how many dc1 quanta co-orbit to make one effective quantum (the three-tier hierarchy). This is the decisive move for C-08/C-09. C-08 first (the unit-dependence): the residue \nu=m_\text{eff}/m_1 is a ratio of two masses — manifestly unit-invariant, unlike the cell occupancy f whose whole pathology was that it comes out dimensionless only in metres. Restating the residue as a mass ratio rather than an f is the C-08 answer. C-09 next, which is a dilemma with two horns: independent ⇒ dimensionally broken (the illegal length) and honest ⇒ circular (the balanced form needs m_1). The breathing kills the first horn outright — C-09’s length-theorem forbids building the 100\;\mum length from \{\sin^2\theta_W, m_e, \hbar, c\} without importing an IR scale, but it says nothing against generating the dimensionless number \nu from dimensionless inputs, and generating a huge pure number from an O(1) coupling is exactly dimensional transmutation — the same legal mechanism behind \Lambda_\text{QCD} and the BCS gap. The breathing relocates the Route-2 question from illegal ground (a length) to legal ground (a pure number). What it does not do is dissolve the second horn: the anchor value \nu\approx9.6\times10^8 is itself computed from m_1=\rho_\text{DM}^{1/4}, so a genuinely bottom-up \nu must be sourced from inputs that do not smuggle m_1 back in. That circularity horn is not closed — it is relocated and made concrete (the tower, below). So C-08/C-09 do not close the door on a bottom-up route the way they closed it on the cube root; they forbid the length form of it and set a precise bar for the number form.

One factor of the lift is already derived, not imported. The lift splits as m_e/m_1 = \alpha_{mf}\cdot\nu. The \alpha_{mf}=0.3008 half — the m_e\!\leftrightarrow\!m_\text{eff} “visibility” — is computed from vortex-core BdG geometry, independent of \alpha (WIP-5, WIP-12 residual: E_F/\Delta=2/\alpha_{mf} is the visibility factor). The genuinely open pure number is therefore \nu alone.

Its natural home is the log-EOS discrete-scale-invariance tower, not the breath amplitude. A bottom-up \nu must be an exponentially large dimensionless number generated by an O(1) coupling. The framework contains exactly one mechanism built to do that — the logarithmic EOS driven to a marginal (\mu\to0) critical point, whose discrete scale invariance yields a geometric tower \nu=\lambda^N, i.e. \ln\nu = N\ln\lambda (WIP-26). This is also where the circularity horn now lives, made concrete: the tower earns a bottom-up \nu only if its marginal (\mu\to0) point is fixed by the substrate’s own criticality rather than by feeding m_1 back in — and whether the marginal-fluid condition can be stated without m_1 is itself open. This is the right kind of object; the breathing amplitude never was. There is one lead here, not two — the geometric tower and the BCS-shaped exponential are the same statement, wearing two hats:

• The tower and the exponential are one. With the pairing rung \lambda=\sqrt2 (the ladder’s own half-octave), \ln\nu=20.68 needs N=\ln\nu/\ln\sqrt2=59.7\approx60 rungs — i.e. \nu\approx2^{30}, thirty octaves. The closest single-coupling BCS form, \nu\sim\exp(2\pi/\alpha_{mf}), is not independent: writing \exp(2\pi/\alpha_{mf})=\sqrt2^{\,N} fixes the rung count at N=4\pi/(\alpha_{mf}\ln2)=60.3. So the exponential predicts the tower’s count, and the exponent has a clean reading — 2\pi/\alpha_{mf}=\pi\cdot(2/\alpha_{mf})=\pi\times6.65, i.e. \ln\nu\approx\pi\times the CdGM core-state count 2/\alpha_{mf}=E_F/\Delta the framework already carries. One object, two faces: a tower whose rung count is set by \alpha_{mf}.
• The honest miss. Against the anchor \ln\nu=20.68: the clean integer N=60 (\nu=2^{30}) lands +0.5% in \ln\nu (a factor 1.12 in \nu); the exponent 2\pi/\alpha_{mf}=20.89 lands +1.0% (factor 1.23). Both overshoots sit inside the bridge equation’s own 13–15% closure, so “60 rungs” is a clean count to the precision \nu is known — the earlier “un-clean count” worry was an artifact of quoting \ln\nu too loosely. But 0.5–1.0% in \ln\nu is not an identity: a dimensionless, electroweak-sourced exp-of-inverse-coupling with the right profile is filed as a lead to chase, not a result. The tower’s existence — a genuine limit cycle rather than a smooth power law — is precisely the WIP-26 residue, so \nu and the ladder stand or fall together.

What this means for the paper’s framing (now carried through). The site had spoken of “two paths to the lattice size.” The honest structure is one length + one scaffold + one open number: (1) a top-down determination, now doubly anchored — \xi=\rho_\text{DM}^{-1/4}(\dots) and the independent dark-energy length differing by the known (\Omega_\Lambda/\Omega_\text{DM})^{1/4}; (2) an electroweak/geometric scaffold that fixes everything about the cell except the pure number \nu; (3) \nu itself, dimensionless and open, its home the DSI tower. Under this framing there is no illegal length anywhere — C-08 and C-09’s length-horn dissolve, and C-09’s circularity-horn sharpens to one concrete task (source \nu from the tower without re-importing m_1) — the cosmology result is untouched, and “Route 2” is stated precisely as the single object it would have to compute. This reframe now lands in bridge-equation.qmd’s framing and the “two paths” language: both now read the bridge as one length + one scaffold + one open number.

Sources: WIP-12 (two breaths, causal ceiling); WIP-26 (DSI tower, \sqrt2 rung); WIP-5 (the \alpha_{mf} visibility factor); Bridge Equation Route 2; Gravity § The residual; critique C-08/C-09. Numerics: scripts/wip30_condensation_number.py.

WIP-32: The winding ledger — charge balance as conserved circulation, and the ledger’s two banked numbers

Status (2026-07-25): substantially advanced. The entry’s original framing — “a_\text{sym} is a fourth target on the one Y-junction solve” — was wrong in a productive way, and correcting it closed the target. a_\text{sym} is not a moment of the intra-nucleon junction flow; it lives one tier up at the internucleon seam, and it is a ratio of seam energies rather than an absolute one, so the framework’s hard open scale \epsilon cancels — exactly the move that already fixed a_S/a_V in WIP-25. Result: a_\text{sym}=a_V+E_F/3=27.96 MeV against 28.06 measured, zero free parameters (scripts/nuclear_asymmetry_seam.py). A second, independent consequence of the same ledger fixes the quark mass ratio, m_d/m_u=2. Surfaced while writing The Two Ledgers of the Boil, the charge-side companion to Why Matter Won.

What is established. Electric charge is a reading of vortex circulation — banked at the FQHE Laughlin quasiparticle (e/3 = winding fraction) and in the junction-flow charge derivation. Given that, three consequences are forced, not fitted:

• Exact neutrality. An irrotational vacuum can mint net circulation only in canceling \pm pairs (Kelvin/Helmholtz; quantized-circulation conservation), so every +1 baryon knot forces a -1 lepton into being — the electron. The universe is neutral to <10^{-20} per particle ([R135]) because charge is conserved winding, not by tuning.
• The knot/float selection rule. Fractional winding must Borromean-lock into a three-arm junction summing to an integer (→ the heavy, confined proton); integer winding floats free (→ the light electron). Recasts m_p/m_e=1836 as the confinement cost of hiding the same conserved charge two ways.
• \beta-decay is the ledger settling. A neutron re-winds one Type-B arm to Type-A (\Deltawinding =+1) and must bud off a -1 quantum — the electron — with the antineutrino carrying the chirality entry from the other ledger.

Correction to the original framing — the targets live at three tiers, not one. The entry previously tabulated four numbers owed by “one over-determined Y-junction solve.” Two of the four do not belong there:

Target	What it actually is	Tier	Status
Quark charges \pm\tfrac23,\pm\tfrac13	monopole of the co-rotating flow	intra-nucleon junction (~929 MeV)	open; now narrowed to “why thirds” (below)
Proton charge radius \approx0.84 fm	second moment of the same flow	intra-nucleon junction	open
Quark mass ratio m_d/m_u	retained fraction of the same flow	intra-nucleon junction	banked: =2 vs measured 2.18
Asymmetry energy a_\text{sym}	mismatch ratio of two seam energies	internucleon seam (~8 MeV, 100\times down)	banked: 27.96 vs 28.06 MeV
\eta_B per-interface bias \varepsilon_\text{chirality}	chiral free energy across the bubble wall	the boil	open (why-matter-won)

Merging these into one solve obscured which were reachable. Separating them showed that the two now banked are precisely the two that are ratios — needing no absolute scale — while everything still open needs one. That is the reusable lesson, and it is the same one WIP-25 learned with a_S/a_V.

Banked result 1 — a_\text{sym} from the vanishing like-nucleon seam. The internucleon seam is a counter-rotating boundary: it binds by cancelling circulation across the interface. An n–p contact presents opposite Type-A/Type-B arm excess and cancels; an n–n or p–p contact presents the same winding and has nothing to cancel. Hence the seam energy is two-valued, with \epsilon_\text{like}=0 — the seam binds only what it can cancel. Running WIP-25’s own close-packed contact count (z=12, six shared seams per interior nucleon, \delta=(N-Z)/A):

B = 6A\Big[\tfrac{1-\delta^2}{2}\epsilon_{np} + \tfrac{1+\delta^2}{2}\epsilon_\text{like}\Big] = \underbrace{3A(\epsilon_{np}{+}\epsilon_\text{like})}_{a_V} - \underbrace{3(\epsilon_{np}{-}\epsilon_\text{like})}_{a_\text{sym}^\text{int}}\frac{(N-Z)^2}{A},

so a_\text{sym}^\text{int}/a_V=(\epsilon_{np}-\epsilon_\text{like})/(\epsilon_{np}+\epsilon_\text{like}) — which is exactly Myers–Świątecki’s \kappa, with \epsilon cancelled. With \epsilon_\text{like}=0 this is 1, and adding the exclusion half (E_F/3, free-Fermi-gas, already the framework’s via spin-statistics) gives a_\text{sym}=a_V+E_F/3=15.68+12.28=27.96 MeV vs 28.06 measured. Read backwards, the kinetic-subtracted \kappa_\text{int}=1.007, i.e. \epsilon_\text{like}/\epsilon_{np}=-0.003.

Four further facts come free from the same rule: deuteron bound but dineutron/diproton unbound; pure neutron matter unbound (\delta\to1\Rightarrow B\to0); the N\approx Z valley; the pairing term. And because the symmetry cost rides on the same contact tally as the binding, it inherits the same surface deficit — which is why MS needed one \kappa on both a_V and a_S, and why the plain SEMF coefficient is conventionally \sim23 MeV while the volume symmetry coefficient is \sim28.

Honest sizing. k_F\in[1.29,1.37] fm^{-1} moves the prediction across 27.2–28.7 MeV, so this is a $$2% result. The kinetic/interaction split is textbook nuclear physics; the framework’s contribution is specifically why \epsilon_\text{like}=0, and that claim is load-bearing for the four corollaries as well.

Banked result 2 — the arm ledger fixes m_d/m_u. Each junction arm carries one unit of circulation with exactly two destinations: it escapes to the confinement boundary as monopole charge |q|, or is absorbed by the arm’s counter-rotating boundary as mass. So |q|+(1-|q|)=1 per arm, giving m\propto(1-|q|) and

\frac{m_d}{m_u}=\frac{1-\tfrac13}{1-\tfrac23}=2,

against a measured 2.18; PDG’s m_u/m_d=0.474^{+0.056}_{-0.074} gives [1.89,2.50], so 2 is inside and the proton chapter’s previous boundary-area estimate (3/2) is outside. Quark mass ratios are RG-invariant, so this is scheme-free. The ledger closes across the proton’s three arms (\tfrac53 escaped + \tfrac43 retained =3) and recovers m_n-m_p after Coulomb.

This also narrows target 1: |q_A|+|q_B|=1 is the ledger statement, and combined with three-fold quantization into thirds it forces \{\tfrac23,\tfrac13\} outright, with the mass ordering assigning them. So the junction solve no longer owes “produce 2/3 and 1/3” — it owes only “confirm the split is into thirds,” and the solid-angle argument is demoted from load-bearing derivation to illustration.

Caveat. Tested against one ratio, using the loosest masses in the PDG. The implied per-arm quantum M\approx6.8 MeV (numerically \approx4\,m_\text{eff}) is not vetted and should not be built on.

The selection rule is now argued, not asserted. The entry previously listed “prove fractional winding cannot be free” as open, while the chapter cited FQHE — where fractional charge does float — as its headline support. Resolving that tension supplies the argument: a fractional excitation can be free only inside a medium already carrying the complementary winding (the Laughlin condensate’s attached flux lends it). The dc1 background is irrotational — the same premise that gives exact neutrality — so it has nothing to lend, and fractional winding must lock. One premise, two consequences: neutrality and confinement. FQHE thereby stops being an analogy and becomes the control experiment. This is a physical argument, not yet a topological proof; the proof remains open and ties to WIP-21’s gauge-group program.

Open work.

• The absolute seam energy \epsilon remains the one scale everything nuclear still waits on — unchanged, and now more clearly isolated: with a_S/a_V and a_\text{sym}/a_V both geometric, \epsilon is the only un-derived quantity in the liquid-drop formula besides the pairing amplitude. It is WIP-25’s merged-boundary tail-overlap problem (\Delta L\approx0.003 fm), and it is what the \sim100–300\times residual-strong-force suppression is.

  It may be a function, not a number (2026-07-31). Note the pattern this entry has itself been remarking on: everything in which \epsilon cancels lands; everything requiring \epsilon itself stays open. The quadrupole-residual section offers a reason rather than a coincidence. Three-fold symmetry lets a sheath cancel the junction’s monopole exactly — that cancellation is color neutrality — and the dipole by the same symmetry, leaving \ell=2 standing. (Identically one tier up: the stealth vacuum has the lattice’s anti-phase breath cancelling through the dipole and bottling the quadrupole in the honeycomb hollow.) If the uncancelled residual is a quadrupole, the internucleon seam is a quadrupole–quadrupole contact and \epsilon = \epsilon(\Omega) is orientation-dependent — in which case contact-counting ratios survive orientation-averaging over \sim12 neighbours intact while the absolute scale needs the un-averaged function. That is exactly the observed success pattern, and it means the debt may have been misdescribed as a missing number.

  Three consequences the binding sections cannot currently produce: (i) a tensor force, which the framework has otherwise been entirely silent about — the deuteron’s Q_d = +0.2859 fm² and 4–6\% D-state become the direct measurement of the junction residual ([R142], [R143]); (ii) an explanation, not a noted convergence, for ^4He’s anomalous binding — four junctions on a cube satisfy all six pairwise orientations at once, which is the O_h symmetry of the B=4 Skyrmion ([R112]), and ^8Be is unbound because the packing does not extend; (iii) a clean split of the seam’s two independent cancellation conditions — winding (monopole: which pairs bind, the \epsilon_\text{like}=0 rule above) versus orientation (quadrupole: how strongly). Falsifier: a purely central substrate seam is ruled out by the deuteron as it stands. Not delivered: no \epsilon(\Omega), no angular form, no tensor-to-central ratio. Next calculation: the angular dependence of the interaction energy between two three-fold vortex junctions at \sim1 fm — the same Saffman-style problem already owed for the charge fractions and the charge radius, carried to a third observable.

• Test the arm ledger elsewhere. m\propto(1-|q|) was read off one triad. Does it survive into the second and third generations, where WIP-21’s \mathbb{Z}_3 amplitude ladder A^2=2(1+Cq^{3/2}) already ties amplitude to charge? The two are independent charge-mass relations on the same junction and must be mutually consistent — a sharp internal cross-check that has not been run.

• Formalize local co-production at the boil. Show net charge =0 within each causal patch of the bubble wall, forbidding primordial super-horizon charge separation.

• Prove the selection rule topologically, upgrading the irrotational-background argument above.

Predictions / falsifiers to track. Exact neutrality; no free fractional charge in vacuum (but expected inside a wound medium — FQHE); no primordial large-scale charge asymmetry; no bound pure-neutron system below the gravitational regime; m_u/m_d tightening to a value outside [1.9,2.1] would strain the arm ledger, near 1.5 would break it.

Honest status. Two numbers are now banked where there were none, and both are Tier 2a (known physics re-derived) on the scorecard — a_\text{sym} at $$2%, m_d/m_u inside a loose PDG interval. Exact neutrality remains a postdiction of an already-exact fact. The entry’s real value is the same as before but better earned: unification — neutrality, fractional-charge confinement, \beta-decay, the nuclear valley, the unbound dineutron, and now the quark mass ratio become one conservation law read at four scales — plus the methodological lesson that the framework’s reachable targets are its ratios. Its one remaining nuclear debt was described here as a single absolute scale; the quadrupole reading above suggests it is instead an orientation function \epsilon(\Omega) whose average is what the ratios have been quietly using — which would explain why the ratios were reachable and the scale was not.

Sources: two-ledgers-of-the-boil.qmd; why-matter-won.qmd; proton-core.qmd; scripts/nuclear_asymmetry_seam.py; WIP-21; WIP-25. Refs [R135] (Bressi et al. 2011, matter neutrality), [R117] (Myers–Świątecki).

---

WIP-33: The maximum density of dc1 — and whether a sub-horizon saturated lump can exist

Status (2026-08-06): opened. A number the framework has leaned on twice without ever computing, now with a second job that makes it observationally consequential.

The debt. The black-holes chapter retires the singularity by asserting that the dc1 condensate has a maximum density: the logarithmic EOS is a Zloshchastiev superfluid vacuum, and the Avdeenkov–Zloshchastiev result says a logarithmically self-binding condensate packs to a finite ceiling and stiffens without bound as it approaches it. That is the right shape of argument. What is missing is the value: the framework nowhere states \rho_\text{max}, and the phrase it uses for the ceiling — “close-packing” — is doing double duty in a way that needs untangling. At ambient conditions the framework already sets n_1\xi^3\approx1 with n_1\approx6.6\times10^{11} m^{-3}, i.e. \rho\approx2.4\times10^{-27} kg/m³ — which is \rho_\text{DM}, the cosmological mean. So the ambient substrate is already described as close-packed, while the black-hole core is described as close-packed at densities “far above the ambient.” Both cannot be the same ceiling. Either the core packs cells tighter than \xi (in which case \xi is not incompressible and the cell scale’s status as a coupling constant needs restating under compression), or the core’s excess density is carried some other way — more circulation per cell, higher core rotation, or a second packing tier. Naming which is the first task of this entry.

Why it now matters beyond the singularity. A second, independent question turns on the same number. The horizon floor says nothing below M_\text{min}=\xi c^2/2G\approx6.7\times10^{22} kg can be a black hole. That leaves a category the framework has never examined: a self-bound lump of saturated dc1 below the horizon mass — maximally packed, hugely dense, but with no trapped surface. If such objects are stable, the framework owns a compact-relic class in the 10^{14}–10^{19} kg range, and the erratics chapter’s supernova channel has a physical supplier for the white-dwarf-transit trigger. If they are not, that channel loses its positive half and the framework is left with only the negative claim (no PBHs) and no replacement.

What the calculation needs.

1. Fix \rho_\text{max} from the logarithmic EOS. With |b|=m_1c^2 fixed as an energy rather than a density, the Avdeenkov–Zloshchastiev ceiling should be computable directly, and its ratio to \rho_\text{DM} is the number the black-hole core has been using implicitly. This is a self-contained one-page calculation and should be done first — everything else here is downstream of it.
2. Resolve the close-packing collision. Whether compression is absorbed by cell count, cell size, or per-cell circulation, and what that implies for c\propto\rho^{1/3} and for \xi’s claimed pinning inside a black-hole core.
3. Stability of a sub-horizon saturated lump. Given \rho_\text{max}, is there a self-bound branch below M_\text{min}? A logarithmic condensate has a known Gausson soliton family; the question is whether the gravitating version has a stable branch at 10^{14}–10^{19} kg, what radius it carries (the transit mechanism needs \ll the white-dwarf pressure scale height), and whether it survives the transit intact.
4. A formation route, or the honest absence of one. \mathcal{B}^0 cannot make them — the boil’s Gaussianity forbids the density contrasts. So the only supply is inheritance through the moraine from \mathcal{B}^{-1}, which means the abundance is an inherited initial condition of the same character as the crust amplitude B (WIP-17) and not predictable from within. This is a real limitation: the framework can predict the shape of the supernova channel’s metallicity trend (an onset at z\approx2.2) but not its amplitude, so it can be falsified by the trend and never confirmed by the rate.

Predictions / falsifiers to track. \rho_\text{max} finite (else the no-singularity claim fails outright); a maximum black-hole compactness following from it, feeding the GW-echo prediction; no stable sub-horizon lump would remove the erratic reading of the SN Ia channel without touching the PBH exclusion.

Honest status. This is a genuinely open entry, not an advanced one — it is being logged because writing the horizon floor exposed that a number the paper has treated as available has never been produced, and because a second chapter now depends on it. Item 1 is cheap and should not have waited. Item 3 is the substantive physics and may well come out negative.

Sources: black-holes.qmd; erratics-of-the-previous-cycle.qmd; substrate-particles.qmd; WIP-11 (the Avdeenkov–Zloshchastiev maximum-density result, used once already to retire the dag). Refs [R14] (Zloshchastiev), [R144], [R145] (Leung et al., the transit-ignition channel).

WIP-34: The hourglass coefficient — the regulating speed and local density of compactor inflow

Status (2026-08-14): opened alongside the hourglass section. The form is fixed; the rate is not.

The debt. The hourglass section establishes that the fuel filling a compactor is the ambient dc1 — the CMB is retired by twenty-one orders of magnitude, baryons by the Eddington valve and a finite reservoir — and that medium capture takes the Bondi form \dot M = 4\pi G^2 M^2 \rho_\text{dc1}/v_\text{reg}^3, a finite-time runaway with t_* \propto 1/(M_0\rho). Everything qualitative downstream of that (guaranteed arrival at the nucleation barrier, mass-ordering of the pop queue, cascade loading) survives any coefficient. The timescale does not: the two natural readings of the regulating speed differ by (c/v_L)^3/4 \approx 1.6\times10^7. With v_\text{reg} = v_L = 751 km/s (the Landau critical speed — bulk flow past it shreds the lattice, which is a physical reason it should regulate the infall) Phoenix A’s t_* at cosmic mean density is \sim4\times10^4 Hubble times; with v_\text{reg} = c (a bare horizon-flux reading, \dot M = 4\pi r_s^2\rho c), \sim6\times10^{11}.

What the calculation needs.

1. The regulating speed, from the substrate’s own hydrodynamics. Classical Bondi uses the far-field sound speed; the substrate has two candidate speeds (c for excitations, v_L for bulk flow) and a transcritical transition between flow regimes when v_\text{ebb} crosses v_L — at r_L = 2GM/v_L^2 = (c/v_L)^2\,r_s \approx 1.6\times10^5\,r_s, far outside the horizon. Whether the accretion rate is set at that outer sonic-like surface (favoring v_L, the fast clock) or at the horizon (favoring c, the slow clock) is the crux, and it is the same class of transcritical problem the framework already solves for the moraine (Grimshaw–Smyth). Related: does the shredded, post-critical lattice inside r_L fall ballistically or re-form?
2. The local \rho_\text{dc1}. Cosmic mean density is a floor; galactic-center halo densities run 10^2–10^5\times higher, shortening every t_* accordingly and re-weighting the pop queue by environment (t_* \propto 1/M\rho, so a modest hole in a dense nucleus can outrank a giant in a void). The queue ordering claimed in the hourglass section is by M\rho, not M alone; making that concrete needs the framework’s own halo profiles (galactic-dynamics.qmd).
3. Back-reaction on the reservoir. The runaway formally diverges, but t_* at these scales exceeds the timescale on which the holes’ own consumption (plus expansion) depletes the ambient density. A self-consistent \rho(t) turns the divergence into a race between the drain and the dilution — this is where “does every hole actually arrive at the barrier, or only the early queue” gets decided.
4. Junction to the nucleation barrier. t_* is when the mass formally diverges; the pop happens earlier, when the core density hits the barrier from WIP-33 / boil breadcrumb 1. Given \rho_\text{max} and the barrier, the actual cycle period is t_\text{pop}(M_0, \rho) < t_* — the hourglass read against the barrier is the framework’s cycle clock, computed rather than asserted.

Predictions / falsifiers to track. None near-term, deliberately — the section claims an ordering and a floor, not a date. The falsifiable content is inherited: a demonstrated mechanism that halts dc1 accretion (an Eddington-analog for the dark channel) would break the “valveless drain” premise; the transparency check against quiet-majority prediction 3 is already banked.

Honest status. Open at the coefficient level, exactly like the horizon floor’s cell count. The M^2 law, the finite-time form, and the CMB retirement are robust; the seven decades between v_L and c readings are not a detail but the difference between a cycle clock of 10^4 and 10^{11} Hubble times, and item 1 is the calculation that collapses it.

Sources: black-holes.qmd § The hourglass; universe-that-boils.qmd (breadcrumbs 1, 2); quiet-majority.qmd; outer-rim-onset.qmd (v_L from the solar wind); WIP-33 (the barrier end of the same clock).

---

Open Theoretical Questions

• Bell test mechanism: verify the Kelvin wave channel calculation independently. Compute v_\text{ch} from substrate parameters to predict L_\text{max}. Design an experimental protocol for extreme-distance Bell tests that could detect degradation of correlations.

• Derive the nuclear coupling \alpha_{mf}^{(N)}\approx552 independently. The mass relation m_\text{eff}\,\alpha_{mf}=m with a shared effective quantum (m_\text{eff}\approx1.70 MeV/c^2) lets us write the proton as \alpha_{mf}^{(N)}=m_p/m_\text{eff}\approx552 effective quanta — but \alpha_{mf}^{(N)} is back-solved from the measured proton mass, so the identity \alpha_{mf}^{(N)}/\alpha_{mf}^{(e)}=m_p/m_e\approx1836 is algebraic, a re-parametrization rather than a prediction. It would become a genuine prediction of the proton-to-electron mass ratio if the count \sim552 could be fixed independently from nuclear-scale vortex packing (the three-fold Borromean junction), the way \tan^2\theta_W fixes the electronic \alpha_{mf}^{(e)}=0.3008. Related: WIP-25 and the Proton Core treatment.

  Explored and set aside (2026-07; the full twelve-step path and its interim write-ups are in git history, with numerics in scripts/nuclear_coupling_*.py). We tried to derive 552 — equivalently m_p/m_e=6\pi^5 — from vortex-junction geometry via a phase-space-measure route. Three results are worth keeping; the rest was scaffolding around a known coincidence. (1) 552 is a count, not a coupling. A single boundary obeys the Kopnin ceiling \alpha_{mf}\le\tfrac12, so 552\gg\tfrac12 must be a seam count N, categorically unlike the sub-maximal electronic leakage \alpha_{mf}^{(e)}=0.3008 — dissolving the “why is one O(1) and the other O(500)” puzzle. (2) 6\pi^5=3\cdot\mathrm{Vol}(S^3)\cdot\mathrm{Vol}(S^5) is an exact identity, and that volume is a genuine color-neutral phase-space (Berezin–Toeplitz) measure — the trace of a coherent-state resolution of identity over the junction’s \mathbb C^2\otimes\mathbb C^3 order parameter (nuclear_coupling_phase_space_measure.py). (3) The +18.8 ppm gap has the right sign for free. Reading 6\pi^5\leftrightarrow m_p/m_e as a finite-level correction, finite-level counts exceed the continuum volume and the data exceeds 6\pi^5 (nuclear_coupling_lenz_correction.py).

  Propagated to the main chapters (2026-07-31). Result (1) above had been banked here but never reached the body of the paper, which continued to read 552 as a coupling and to conclude that the proton “leaks nearly all of its energy” and “shows almost everything it stores” — a claim the Kopnin ceiling forbids, and one that also contradicted the same chapter’s EMC prediction, which needs a large reactive ledger inside the bound nucleon. The main chapters now carry the count reading: m = N\alpha_{mf}m_\text{eff} with \alpha_{mf}=0.3008 universal and N_p = m_p/m_e \approx 1836, so the proton stores \sim3.12 GeV, shows 938 MeV, and hides the same 70\% the electron does. Two consequences worth noting. The electron is the heaviest fermion a single boundary can contain — the ceiling \alpha_{mf}^\text{eff}=\tfrac12 falls in the gap between the electron (0.3008) and the up quark (\approx1.3), not through the middle of a family. And the string tension independently prefers the stored energy: L_Y = E/\sigma gives 1.15 fm per Y-arm from 3.11 GeV (sane against the 0.38 fm tube radius) versus 0.34 fm from 929 MeV (a tube shorter than it is wide). Unrun cheap check: whether the lattice constant C_{3Q} tracks the reactive share (1-\alpha_{mf})/\alpha_{mf} times the visible energy.

  A reframing of the target, not progress on it. The open problem is now “derive the seam count N\approx1836,” which is a different question from the one this entry set aside. A three-fold junction admits no closed stream surface — Type A and Type B orbital orientations are mutually perpendicular, so no plane and no symmetry axis is shared, and the three-body vortex flow destroys the invariant tori a single ring enjoys. The count is then “how many locally-tangent patches cover a surface with no global tangent structure.” That does not revive the phase-space-measure route below; it only says the unanswerable form of the question (“why is a coupling 500 when couplings cap at \tfrac12”) has been replaced by a geometry question. See Proton Core § What the sheath cannot cancel.

  Why it was set aside — the honest bottom line. None of this derives the number; it re-encodes the Lenz (1951) coincidence m_p/m_e\approx6\pi^5 as a geometric identity. The residual +18.8 ppm has no visible path to 10^{-5} precision: the correction level k\approx2.1\times10^{5} is not a computable framework quantity (it is defined by the match — the natural candidate \nu\approx8\times10^{8} misses by three to four orders), and the absolute scale gives no independent handle because m_p=m_\text{eff}\cdot552=m_e\cdot6\pi^5 is the same 6\pi^5 wearing a mass. This is the identical pattern to the charged-lepton Koide result: a known empirical relation (there Q=2/3, off by \sim6 ppm; here 6\pi^5, off by 19 ppm) re-encoded via the same three-fold \mathbb Z_3 junction, with the precision residual absorbed by an admittedly-underived fitted number (there the phase \delta=2/9; here the level k). Both are self-consistent structural re-encodings of pre-existing coincidences, not forward predictions — a genuine unifying picture, but not a derivation. Do not re-tread the twelve-piece path. The only moves that would change the verdict are an independent physical identification of a computable k (or \delta), or a genuinely new number the same machinery forecasts rather than reproduces — a focused search for the latter (2026-07) turned up none.

• Is N_\text{eff} really 1? We assumed one effective quantum per electron. If N_\text{eff} = 2, v_\text{rot} = c\sqrt{\alpha_{mf}} = 0.549\,c and r_\text{eff} = 0.212 pm. The angular momentum per quantum would still be \hbar, but total L = 2\hbar, requiring l=1 orbital state.

• DNA as a substrate antenna. The double helix has co-rotating and counter-rotating strands (two sugar-phosphate backbones wind in opposite senses), with energy transport through base-pair stacking (the helix’s polar axis). If the helix geometry is optimized for coupling to substrate modes, DNA is a lattice-scale antenna whose parameters should be derivable from (d, \xi, \alpha_{mf}) without invoking molecular evolution as the sole explanation.

• Phase transitions as substrate reorganization. Every condensed matter phase transition involves the substrate reorganizing its local vortex structure. The He-3 superfluid transition, which Volovik uses as the template, should be the calibration case — reproducing the He-3 A-phase and B-phase transition temperatures from substrate parameters would provide a direct laboratory verification path.

• The cosmological constant problem, resolved and sharpened. The substrate resolves the 10^{122} discrepancy by construction (Volovik self-tuning), and the residual now splits cleanly into three pieces (full treatment in Gravity § The residual). (i) The dark-energy scale is predicted: \rho_\Lambda^{1/4} = 2.24 meV \approx m_1 c^2, since close-packing makes \rho_\Lambda \approx \rho_\text{DM} — one substrate density, dissolving the coincidence problem. (ii) The apparent fine-tuning is not a separate small number: \delta T/T_c|_\text{Planck} = (m_1/M_\text{Pl})^2, so small \Lambda is the same fact as weak gravity, set by the same f_\text{cross}\approx10^{-15} — deriving f_\text{cross}/\omega_0 (WIP-15 item 2, WIP-16) predicts both, the same collapse found between the bridge 4\pi and the Higgs 8\pi. (iii) The genuine disequilibrium, read in the substrate’s own units, is order unity (\delta T/T_c|_\text{substrate}\approx1.6) — measured directly as the moraine wake we still sit in (f(0)=1.25). What stays open is the value of \Lambda itself — equivalently the de Sitter horizon entropy S_\text{dS}\approx2\times10^{122}, an inherited initial condition (the relaxation depth of \mathcal{B}^{-1}, same status as crust B in WIP-17). Self-tuning plus the horizon-fluctuation law \delta T/T_c = 1/\sqrt{S_\text{dS}} make \rho_\Lambda = \rho_\text{Pl}/S_\text{dS} a marginal self-consistency — it fixes the form, not the value, exactly as expected at the substrate’s critical (\mu\to0) point. If the nucleation barrier (WIP-17, breadcrumb 1) sets that relaxation depth, the value too becomes a prediction of the substrate’s phase diagram.