Answers are grounded in this paper and cite their sources.
What if space is a crystal that cannot close? Start with one polynomial, x³ − x − 1, whose discriminant is −23. Build a network of dodecahedral cells with 23 channels per node. Dodecahedra cannot fill flat space: around every edge a gap of 10.305° is left over, exactly the deficit by which a discrete network carries curvature. The rest of this paper follows from that gap and one rule, that the network never tears. Matter is frustration tied in knots. The gap escapes as defect strands that knot, the trefoil first. The no-tearing rule conserves charge and baryon number topologically, so the proton is exactly stable. The proton by counting. A cell offers 20 × 23 = 460 slots, and the nucleon cable needs an odd number. That gives m_p/m_e = 4 × 459 = 1836 and r_p = ƛ_e/459 = 0.84131 fm, within 0.9σ of CODATA 2022, with no QCD fit and no free parameter. Gravity from a common clock. Mass is a mixing frequency per tick, and all knots read the same tick, so everything falls alike. γ = 1 follows because the cells do not tile; G is reproduced at −0.44σ. Every weak-field test passed. The network Lagrangian is written out and varied, and its equations of motion pass the full first-order battery: redshift, light deflection, Shapiro delay, Mercury, Nordtvedt, Gravity Probe B, preferred-frame bounds, MICROSCOPE, the double pulsar, and the speed and polarizations of gravitational waves. Where it parts from Einstein, with falsifiable predictions: gravitational waves carry energy but do not gravitate, so there is no nonlinear memory after mergers (testable with stacked LIGO events and LISA); there is no horizon: shadows are 4.6% larger and the innermost stable orbit has an orbital frequency 6.9% lower; gas falling into Sagittarius A* never reaches a surface within the age of the Galaxy; the proton never decays. Open problems and corrections to earlier work are stated explicitly. Every number in the paper is checked by code.
No takes yet. Share an insight, caveat, or question.
ALFREDO LUIS ACOSTA PADILLA (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: