"""v428 -- OVERDET.WITNESS.RECLASS.01: the honest correction of v427. A pattern search (and the right objection) exposed that v427's "seven disjoint arithmetic grammars that MULTIPLY" overstates independence: by v236 (TOPO.BRIESKORN.01) the (2,3,5) Brieskorn singularity x^2+y^3+z^5 is the ONE generator, and its Milnor monodromy organises the order-30 Coxeter clock, the Galois (Z/30)^x and the Z/2 x Z/3 x Z/5 sheet/family/carrier factorisation. So six of v427's seven witnesses are different number-theoretic READINGS of that one object -- by v427's own definition that is COMPRESSION, not multiplication. This module applies v427's multiply-vs-compress test to v427 itself and re-states the honest accounting. Consistent with v305 (generator economy) and v313/v236. [E] 1. THE ONE GENERATOR (v236). The (2,3,5) atoms generate E8/Coxeter-30: Milnor mu=(2-1)(3-1)(5-1)=8=rank E8; 30=2*3*5=h(E8); phi(30)=8. One object, not many. [E] 2. WITNESS RE-CLASSIFICATION (compression). Each v427 witness maps to a facet of that object -- Gauss Z[i]->prime 2 (mu4/sheet), Eisenstein Z[w]->prime 3 (family), cyclotomy Q(z5) & Galois (Z/5)^x->prime 5 (carrier), Pascal 2^4->prime 2, Coxeter phi(30)->30, lattice det E8->E8. Every facet divides 30 or IS E8, so the seven collapse to ONE (2,3,5)/E8 generator: COMPRESSION (distinct underlying objects = 1), NOT seven independent multiplications. [E] 3. THE GENUINE MULTIPLICATION (corrected). What honestly multiplies is the INPUT forced by genuinely independent arguments: the "8" in c3=1/(8 pi) from rank(E8)=8, the carrier Coxeter h(D5)=2(5-1)=8, the totient phi(30)=8, and the Milnor number (2,3,5)=8 -- four DIFFERENT mathematical facts, all 8; plus the FOREIGN witness alpha^-1 ~ 137 (prime, not in the (2,3,5)-derived skeleton; v305). [C] 4. VERDICT. v427's framework (multiply vs compress) is right; its classification was wrong -- the arithmetic witnesses COMPRESS one (2,3,5)/E8 object, they do not multiply. The honest over-determination is the multiply-forced inputs (the "8" four independent ways) plus foreign witnesses (137). This SHARPENS the accounting (correct classification), consistent with v236/v305/v313; it refines, and partly walks back, v427. Python-only (sympy/numpy exact arithmetic). """ import sympy as sp from tfpt_constants import check, summary, reset, g_car, N_fam SKELETON = {1, 2, 3, 4, 5, 7, 8, 11, 13, 16, 30, 55, 248} def run(): reset() print("v428 OVERDET.WITNESS.RECLASS: v427's seven witnesses compress one " "(2,3,5)/E8 object") # ---- 1. the one generator (v236) ---- milnor = (2 - 1) * (3 - 1) * (5 - 1) # 8 rank_E8 = 8 h_E8 = 30 cox_factor = sorted(sp.factorint(h_E8).keys()) # [2,3,5] phi30 = int(sp.totient(30)) # 8 one_generator = (milnor == rank_E8 == 8 and cox_factor == [2, 3, 5] and phi30 == 8) check("THE ONE GENERATOR [E] (v236): the (2,3,5) Brieskorn atoms generate " "E8/Coxeter-30 -- Milnor mu=(2-1)(3-1)(5-1)=8=rank E8, 30=2*3*5=h(E8), " "phi(30)=8; one object, not seven", one_generator) # ---- 2. witness re-classification: every v427 witness is a facet of it ---- # (name, v427 value, facet of the (2,3,5)/E8 generator) witnesses = [ ("Gauss Z[i] N(3+2i)", 13, 2), # i order 4 = mu4/sheet (prime 2) ("Eisenstein Z[w] N(3+2w)", 7, 3), # w order 3 = family (prime 3) ("Cyclotomy Q(z5) N(3+2 z5)", 55, 5),# z5 = carrier (prime 5) ("Galois |(Z/5)^x|", 4, 5), # Gal(Q(z5)) (prime 5) ("Pascal 2^4 / C(4,<=2)", 16, 2), # mu4 exterior (prime 2) ("Coxeter phi(30)", 8, 30), # the whole 2*3*5 clock ("Lattice det Cartan(E8)", 1, "E8"), # E8 = the (2,3,5) resolution ] # a facet is "inside" the one object iff it is a prime factor of 30, or 30, or E8 def inside(f): return f in (2, 3, 5, 30) or f == "E8" all_inside = all(inside(f) for _, _, f in witnesses) distinct_objects = 1 if all_inside else len({f for _, _, f in witnesses}) check("WITNESS RE-CLASSIFICATION [E]: every one of v427's seven witnesses maps " "to a facet of the SAME object -- primes {2,3,5} of 30=2*3*5, the clock " "30 itself, or E8 (the (2,3,5) resolution); all 7 collapse to ONE " "generator (distinct underlying objects = %d) => COMPRESSION, not seven " "independent multiplications" % distinct_objects, all_inside and distinct_objects == 1 and len(witnesses) == 7) # ---- 3. the genuine multiplication: the input forced 4 independent ways ---- rank_e8 = 8 h_D5 = 2 * (5 - 1) # Coxeter number of D5 = 8 forced_8 = {"rank(E8)": rank_e8, "h(D5)": h_D5, "phi(30)": phi30, "Milnor(2,3,5)": milnor} eight_multiply = (all(v == 8 for v in forced_8.values()) and len(forced_8) == 4) alpha_inv = 137 # nearest integer; foreign foreign = alpha_inv not in SKELETON and sp.isprime(alpha_inv) check("THE GENUINE MULTIPLICATION [E]: what honestly multiplies is the INPUT " "forced by independent arguments -- the '8' in c3=1/(8 pi) from rank(E8), " "h(D5)=2(5-1), phi(30) and the Milnor number, four DIFFERENT facts all =8; " "plus the FOREIGN witness alpha^-1~137 (prime, not in the (2,3,5)-skeleton, " "v305)", eight_multiply and foreign) # ---- 4. verdict (typed [C]; refines/partly walks back v427) ---- check("VERDICT [C]: v427's multiply-vs-compress FRAMEWORK is right, its " "CLASSIFICATION was wrong -- the seven arithmetic witnesses are readings " "of ONE (2,3,5)/E8 object (compression, a la v305), they do not multiply. " "The honest over-determination is the multiply-forced inputs (the '8' four " "independent ways) plus foreign witnesses (137). Refines and partly walks " "back v427; consistent with v236/v305/v313", all_inside and distinct_objects == 1 and eight_multiply and g_car == 5 and N_fam == 3) return summary("v428 OVERDET.WITNESS.RECLASS (v427's seven witnesses COMPRESS " "one (2,3,5)/E8 object; genuine multiplication = forced inputs + " "foreign 137; refines v427)") if __name__ == "__main__": raise SystemExit(1 if run() else 0)