TFPT scientific source excerpt
Original repository path: experiments/theory-contracts/elliptic-moebius-source-bridge-20261026/INDEPENDENT-BOUNDARY-REVIEW.txt
Original file SHA-256: 7fac8804f89364125b2970eed9c937e7eb5dad2edf7253ffa8e9cc25f3d429fc
Inclusive original lines: 1-154
Line numbers below refer to the original source.
This bounded excerpt is source evidence; it is not a new proof/checker run.
Conversation metadata and personal execution details are omitted where indicated.

000001 | INDEPENDENT FINAL REVIEW OF THE BOUNDARY AND MODULAR SOURCE REPORTS
000002 | ==================================================================
000003 | 
000004 | 26 October 2026. Read only review of TRACE-METRIC-AND-DOMAIN.txt and
000005 | MODULAR-ORBIFOLD-SOURCE.txt. No Atlas or original source was changed.
000006 | 
000007 | Verdict
000008 | =======
000009 | 
000010 | Both reports are accepted in their stated scope. The trace report needs one
000011 | domain precision in its outer function argument, but its range theorem and
000012 | its negative state identification remain correct. The modular report has the
000013 | correct marked monodromies, passport, finite versus full modular distinction,
000014 | and elliptic Belyi data. It does not identify either construction with the
000015 | old quarter operator or state.
000016 | 
000017 | 1. Trace metric and domain
000018 | ==========================
000019 | 
000020 | The change of variables is correct:
000021 | 
000022 |   |dz|=|dx|/(|c||y|),
000023 | 
000024 | so including the half density Jacobian once makes T_r unitary from the two
000025 | lifted L2 traces to standard L2(Gamma_r,C^2;|dx|). The raw weighted measure
000026 | and the Jacobian description are equivalent presentations and must not be
000027 | used simultaneously. At a branch mark, |y| is asymptotic to
000028 | 2|s|^(1/2), giving the stated |s|^(-1/2) measure and |s|^(-1/4) half density
000029 | factor. Both are locally L2 compatible. The four point branch matching is
000030 | invisible after L2 closure but remains genuine H1 domain data.
000031 | 
000032 | On |x|=1 one has
000033 | 
000034 |   |g|^2=|tan(theta)|^(1/2).
000035 | 
000036 | The weights |g|^2 and |g|^(-2) are A2 weights. Thus the densely defined
000037 | similarities g P_H g^(-1) and g^(-1) P_H g extend to bounded idempotents.
000038 | The displayed Fourier coefficient is also correct:
000039 | 
000040 |   (1/(2 pi)) integral |tan(theta)|^(1/2) exp(-2i theta)dtheta
000041 |     =-1/sqrt(2),
000042 | 
000043 | with the opposite sign for the inverse weight. Hence the idempotents are not
000044 | selfadjoint in the induced standard trace metric and cannot themselves be
000045 | positive CAR covariances.
000046 | 
000047 | The only required precision concerns the notation
000048 | 
000049 |   closure[g H2]=H2.
000050 | 
000051 | At the ramified boundary M_g is unbounded on H2, so gH2 is not a naive image
000052 | of every H2 vector. Define instead
000053 | 
000054 |   D_g={h in H2: g h lies in H2}.
000055 | 
000056 | Then
000057 | 
000058 |   g D_g is contained in Ran(g P_H g^(-1)) and in H2
000059 | 
000060 | on the natural similarity core. For the bounded extension, approximation by
000061 | that core gives
000062 | 
000063 |   Ran(g P_H g^(-1)) is contained in closure(g D_g) inside H2.
000064 | 
000065 | Conversely the range contains g times every analytic polynomial, while the
000066 | range is closed because the extended operator is a bounded idempotent. Since
000067 | g is outer, those polynomial multiples are dense in H2. Therefore
000068 | 
000069 |   Ran(g P_H g^(-1))=H2.
000070 | 
000071 | The identical argument with g^(-1) proves the second channel. This supplies
000072 | the rigorous domain version of the report's equation (19) and preserves
000073 | 
000074 |   P_orth=P_H direct-sum P_H.
000075 | 
000076 | The report correctly keeps this orthogonal range projector separate from the
000077 | oblique Plemelj idempotent, the smooth annulus covariance
000078 | (1+exp(B))^(-1), and the old quarter Hardy state. No forbidden state or
000079 | operator identity is asserted.
000080 | 
000081 | 2. Modular pullback
000082 | ===================
000083 | 
000084 | With
000085 | 
000086 |   a=M^2,   b=M U^(-1),   c=U,
000087 | 
000088 | the exact matrices satisfy
000089 | 
000090 |   a^3=b^2=c^4=abc=I.
000091 | 
000092 | For Phi(t)=-(t-1)^2/(4t), the marked local monodromies are precisely
000093 | 
000094 |   u_0       =c=U,
000095 |   u_1       =a^2=M,
000096 |   u_(-1)    =b^2=I,
000097 |   u_infinity=b c b^(-1)=(U M)^(-1).
000098 | 
000099 | Their product in the displayed Hurwitz order is I. Writing only
000100 | d=(UM)^(-1), as the report does, is correct; the conjugation formula above
000101 | makes the path marking explicit.
000102 | 
000103 | The composition with t=x^4 has degree eight and passport
000104 | 
000105 |   [2,2,2,2], [2,2,2,2], [4,4].
000106 | 
000107 | Its ramification contribution is 14=2*8-2. The four surviving peripheral
000108 | operators are U^k M U^(-k), and their product is the v117 relation. Q:x->ix
000109 | is the source deck transformation. It is not thereby the modular S path or
000110 | the marked fiber operator U.
000111 | 
000112 | The finite image qualification is necessary and correctly stated. In
000113 | PSL(2,Z/4)=S4 the cusp image has order four, so the points over zero and
000114 | infinity become apparent after the fourth power. For the full modular local
000115 | system the cusp T has infinite order and T^4 remains at both points. Hence
000116 | the same map does not close on P1 minus mu4 for the full PSL(2,Z) system.
000117 | 
000118 | 3. Elliptic degree four source
000119 | ==============================
000120 | 
000121 | The Tate calculation checks out. For
000122 | 
000123 |   E_b: y^2+x y+b y=x^3+b x^2,
000124 | 
000125 | the discriminant is b^4(1-16b). The critical collision equation has factors
000126 | (16b-9)(16b-1). The branch b=1/16 is singular, while b=9/16 is smooth. At
000127 | the latter value the normalized function has passport [4],[3,1],[4] and
000128 | 
000129 |   j=2*47^3/3^8=207646/6561.
000130 | 
000131 | An independent enumeration gives 24 transitive ordered branch triples of
000132 | the stated cycle types. They form one simultaneous S4 conjugacy orbit of
000133 | size 24, confirming the unique ordered Nielsen class. The source therefore
000134 | is not the fixed j=1728 ellipse. The sign twist needed to turn the traceless
000135 | permutation representation into the v117 character remains an additional
000136 | marked representation datum, exactly as the report states.
000137 | 
000138 | Final scope
000139 | ===========
000140 | 
000141 | The reviewed results establish a canonical positive Hardy range for the
000142 | ramified trace and an exact finite modular origin for the v117 relations.
000143 | They do not supply a propagation unitary from the smooth annulus state to the
000144 | graph, choose the old quarter state, identify Q with modular S, or promote
000145 | the finite Level 4 result to the faithful full modular local system.
000146 | 
000147 | Independent checks performed
000148 | ============================
000149 | 
000150 |   * exact matrix evaluation of u_0,u_1,u_(-1),u_infinity and their product;
000151 |   * numerical high precision replay of the two weighted Fourier
000152 |     coefficients +/-1/sqrt(2);
000153 |   * exact symbolic replay of the Tate critical resultant and j invariant;
000154 |   * complete S4 enumeration of the ordered Nielsen class.