Follow the source behind the result.

The registered Python source is shown as text. Displaying it does not execute the code.

"""v29 -- machine certificates for the two research contracts (U_wall, G_metric).

Backs the cleanly-certifiable lemmas of the two research contracts
(tfpt_research_contracts), and records one HONEST correction.

C_U^(1)  WALL ENUMERATION (Contract 1, Lemma U1 -- corrected).
  Enumerate the 6^4 = 1296 weight matrices whose columns are permutations of the
  cusp triple (0,1,2).  The parabolic wall constraint (row-sum multiset {6,3,3},
  i.e. weights (2,1,1) lifting degrees (-2,-1,-1) to pardeg 0) leaves 144; with
  the O(-2) line distinguished (its row carries the max sum 6), 48.  HONEST
  RESULT: under the FULL natural symmetry D4(punctures) x S3(lines) x S3(values)
  these still form FIVE orbits, not one.  So W_wall is NOT singled out by
  symmetry alone -- the uniqueness must come from the TFPT SELECTOR (Lemma U4:
  det R=8, Spec(Q+)={1,2,3}), not from U1.  W_wall is one of the five.

G5  GAP-DOMINANCE (Contract 2, Lemma G5).
  The strong form 2||V_metric||_rel < Delta with ||V_metric||_rel <= 248 c3^2
  = 31/(8 pi^2) = 0.39262 and Delta = 6 log(3/2) = 2.43279 gives an effective
  transfer gap Delta_eff > 1.647 > 0.
"""
from itertools import permutations, product
import mpmath as mp
from tfpt_constants import check, summary, reset, c3

mp.mp.dps = 30


def _wall_orbits():
    perms = list(permutations([0, 1, 2]))
    mats = []
    for cols in product(perms, repeat=4):
        M = tuple(tuple(cols[c][r] for c in range(4)) for r in range(3))
        mats.append(M)
    def rowsums(M):
        return tuple(sum(M[r][c] for c in range(4)) for r in range(3))
    walls = [M for M in mats if sorted(rowsums(M), reverse=True) == [6, 3, 3]]
    # full symmetry group: D4 on columns, S3 on rows, S3 on values
    def rotate(p):
        return [p[(i + 1) % 4] for i in range(4)]
    D4 = []
    p = [0, 1, 2, 3]
    for _ in range(4):
        D4.append(tuple(p))
        D4.append((p[0], p[3], p[2], p[1]))
        p = rotate(p)
    D4 = list(set(D4))

    def canon(M):
        best = None
        for d in D4:
            Mc = tuple(tuple(M[r][d[c]] for c in range(4)) for r in range(3))
            for rp in permutations([0, 1, 2]):
                Mr = (Mc[rp[0]], Mc[rp[1]], Mc[rp[2]])
                for vp in permutations([0, 1, 2]):
                    X = tuple(tuple(vp[Mr[r][c]] for c in range(4)) for r in range(3))
                    if best is None or X < best:
                        best = X
        return best
    orbits = set(canon(M) for M in walls)
    Wwall = ((2, 1, 2, 1), (1, 0, 0, 2), (0, 2, 1, 0))
    return len(mats), len(walls), len(orbits), (canon(Wwall) in orbits)


def run():
    reset()
    print("v29  research-contract certificates (C_U^(1) wall enum, G5 gap)")

    # C_U^(1): wall enumeration (honest 5-orbit result)
    total, nwall, norb, has_W = _wall_orbits()
    check("C_U^(1): 6^4 = 1296 column-permutation weight matrices", total, 1296, exact=True)
    check("C_U^(1): 144 matrices satisfy the {6,3,3} wall constraint", nwall, 144, exact=True)
    check("C_U^(1) HONEST: 5 wall orbits under D4 x S3 x S3 (NOT 1) "
          "-> uniqueness needs the selector (U4), not symmetry (U1 corrected)", norb, 5, exact=True)
    check("C_U^(1): W_wall is one of the five wall types", has_W)

    # G5: gap-dominance strong form
    Vrel = 248 * c3**2
    Delta = 6 * mp.log(mp.mpf(3) / 2)
    check("G5: ||V_metric||_rel <= 248 c3^2 = 31/(8 pi^2) = 0.39262",
          Vrel, mp.mpf(31) / (8 * mp.pi**2), tol=mp.mpf('1e-20'))
    check("G5: strong form 2||V|| = 0.7852 < Delta = 6 log(3/2) = 2.4328", 2 * Vrel < Delta)
    Delta_eff = Delta - 2 * Vrel
    check("G5: effective transfer gap Delta_eff = Delta - 2||V|| > 1.647 (positive)",
          Delta_eff > mp.mpf('1.647'))
    return summary("v29 research-contract certificates")


if __name__ == "__main__":
    raise SystemExit(1 if run() else 0)

984bb2d73dfbb40a66020cdedda90ec79e801433cb096d4489765b295a3d2cc5

The public registry includes 116 document views, 6,693 section records and 985 physical LaTeX sources. Active publication documents, research material and archives retain their recorded roles. This is source navigation, not a new semantic review or proof run.