TFPT scientific source excerpt
Original repository path: experiments/theory-contracts/tfpt-unified-source-derivation-20261029/aps_source_review/F5-HARDY-SEAM-OPERATOR-REVIEW.txt
Original file SHA-256: 6968c825edaf6454138c228949441a2d67ceae41a034037930f02aeebca8c017
Inclusive original lines: 1-169
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 | F5, THE ORIGINAL SEAM CLUTCH AND THE ALREADY SOLVED SPIN HARDY SHIFT
000002 | Independent source/operator continuation, 2026-10-29.
000003 | No registered source edits, model scan, suite or Atlas mutation.
000004 | 
000005 | 1. The old one-copy norm result is retained as SOLVED
000006 | Atlas finding:same-spin-unit-shift-gives-native-module-class-with-odd-determinant-line.
000007 | Read completely:
000008 |   elliptic-moebius-source-bridge-20261026/boundary_origin_20261027/
000009 |   uniqueness/SAME-SPIN-REFERENCE-SIMPLIFICATION.txt:1-191;
000010 |   originals/ACTUAL-REFERENCE-SPIN-TEST.txt:1-259.
000011 | The same-space reference is M_1, not a newly invented scalar reference.
000012 | Its primitive winding has index -1; the actual three-sheet spin cable has
000013 | index -3 and relative line N_q^-1 tensor B_O^-1, with its retained odd parity.
000014 | Neither the exponent nor its finite native module class needs another search.
000015 | 
000016 | 2. The exact original gluing operation
000017 | Directly read:
000018 | documents/archive--paper-latex--tfpt-theory-fullv21/sections/
000019 | 001-section-introduction-the-problem-of-parameters.tex:220-258.
000020 | Lines225-228 cut the orientable double into FOUR boundary pieces
000021 | (C1,C2,Gamma_+,Gamma_-) and explicitly prescribe a unitary gluing U_Gamma
000022 | between the two seam pieces. Lines230-247 give i partial_theta,
000023 | U_Gamma=e^(i theta), and unit winding.
000024 | 
000025 | Thus relative boundary matching is an original operation.
000026 | In a GIVEN collar with Gamma_+ assigned to face0 and Gamma_- to faceell,
000027 | it is represented by v1=U v0. That assignment is an actual needed map;
000028 | two F5 faces cannot be identified with two of the four original pieces
000029 | just by counting them. Reversing face order or clutch convention changes
000030 | the index sign. The computations below retain the existing quarter convention
000031 | B e_n=(n+alpha)e_n, U e_n=e_(n+1), alpha=1/4 or3/4.
000032 | The original i partial_theta formula also requires its basis orientation
000033 | when compared with this -i partial_theta convention.
000034 | 
000035 | 3. Reuse of the whole F5 Cauchy graph and its I/O functor
000036 | Read:
000037 |   pfaff_source_review/NORMAL-KINETIC-PAIRING.txt:13-108;
000038 |   graded_boundary_functor/FINITE-COUPLED.txt:28-69;
000039 |   seam-collar-multifield-20261016/BOUNDARY-OPERATION-PROOF.txt:24-144;
000040 |   seam-collar-multifield-20261016/PROOF.txt:78-123.
000041 | Atlas finding:two-face-noncommuting-geometric-response-calculus is reused.
000042 | The raw-mask null-quotient failure, compressed incoming response
000043 | A^(1/2) M_f A^(1/2), and the oriented exterior ray Gamma(R) are ALREADY known.
000044 | They are not reannounced as a new derivation.
000045 | 
000046 | For the given charged h, let R=e^(-ell h) on its maximal spectral domain.
000047 | Its normalized graph map is
000048 |   V_h=(C,R C), C=(I+R^2)^(-1/2).
000049 | It is unitary from H to Ran G. This is the existing full graph, including
000050 | all negative modes and both labeled faces.
000051 | The BOUNDED relative matching map on these actual Cauchy data is
000052 |   F_U=pi_1-U pi_0 : Ran G -> H,
000053 |   F_U V_h=R C-U C.
000054 | Equivalently, using the OLD A=(I+R^2)^-1 and OLD normalized V_h,
000055 |   F_U V_h=[-U,I]V_h=sqrt(I-A)-U sqrt(A).
000056 | Everything in this mismatch formula is bounded; no unbounded product is
000057 | needed to define the boundary consumer.
000058 | This is an actual mismatch between two boundary values, not an operator
000059 | postulated on the minimal Gram quotient and not G diag(U,U) G.
000060 | It retains the original field/adjoint completion through the existing T_faces.
000061 | 
000062 | 4. Why full diagonal compression cannot supply the Hardy defect
000063 | In the existing scalar block, put r_n=e^(-ell(n+alpha)) and
000064 |   v_n=(1,r_n)/sqrt(1+r_n^2).
000065 | Under the normalized graph chart, compression of U on BOTH endpoints is
000066 | the bilateral weighted shift
000067 |   e_n -> w_n e_(n+1),
000068 |   w_n=(1+e^-ell r_n^2)/
000069 |       sqrt((1+r_n^2)(1+e^(-2ell)r_n^2)).
000070 | For every n,
000071 |   1 >= w_n >= sech(ell/2)>0.
000072 | Hence it has a bounded inverse and index0.
000073 | The one-sided Hardy shift is unilateral, index -1, or -3 on three cables.
000074 | There is no exact invertible operator comparison between those operations.
000075 | This does not restrict the actual oriented I/O sewing to that compression:
000076 | the original gluing operation is the RELATIVE matching map in section3.
000077 | 
000078 | [OMITTED: non-scientific conversation or execution metadata]
000079 | [OMITTED: non-scientific conversation or execution metadata]
000080 | [OMITTED: non-scientific conversation or execution metadata]
000081 | [OMITTED: non-scientific conversation or execution metadata]
000082 | [OMITTED: non-scientific conversation or execution metadata]
000083 | [OMITTED: non-scientific conversation or execution metadata]
000084 | [OMITTED: non-scientific conversation or execution metadata]
000085 | [OMITTED: non-scientific conversation or execution metadata]
000086 | [OMITTED: non-scientific conversation or execution metadata]
000087 | [OMITTED: non-scientific conversation or execution metadata]
000088 | [OMITTED: non-scientific conversation or execution metadata]
000089 | 
000090 | 5. Constructive same-graph Fredholm defect, all modes and closed domain
000091 | For that same scalar source define
000092 |   T=R-U : Dom R -> H.
000093 | It is closed, as a bounded perturbation of the closed selfadjoint R.
000094 | C:H->Dom R is an isometric isomorphism for the R graph norm, since
000095 |   ||C x||^2+||R C x||^2=||x||^2.
000096 | The exact normalized map is
000097 |   T C=diag(r_n/sqrt(1+r_n^2))
000098 |           -U diag(1/sqrt(1+r_n^2)).
000099 | Let P_+ project onto n>=0 and P_-=I-P_+.
000100 | The two diagonal coefficients differ from P_- and P_+, respectively,
000101 | by exponentially decaying sequences. Therefore T C is a TRACE-CLASS
000102 | perturbation of
000103 |   T_inf=P_- -U P_+.
000104 | T_inf maps all negative basis vectors identically and all nonnegative
000105 | ones to minus e_(n+1); its only missing output is e0. Thus
000106 |   index T=index(T C)=index T_inf=-1.
000107 | 
000108 | There is no kernel: T x=0 forces
000109 |   x_n=x_0 exp(ell[n(n+1)/2+alpha n]),
000110 | which cannot be in l2 unless x_0=0.
000111 | The adjoint has Dom T*=Dom R and a one-dimensional kernel:
000112 |   y_(n+1)=r_n y_n,
000113 |   y_n=y_0 exp(-ell[n(n-1)/2+alpha n]).
000114 | It decays quadratically at both ends, and R y has the same l2 norm as
000115 | the shifted sequence, so the adjoint domain condition is satisfied.
000116 | This proves a closed Fredholm operator with exactly ONE cokernel mode.
000117 | Three actual copies give three such modes and index -3.
000118 | No finite cutoff, new covariance, new vacuum or new Hardy projector
000119 | is needed for this source-relative Fredholm statement.
000120 | 
000121 | The TC=T_inf+trace-class relation is NOT an exact operator intertwiner.
000122 | It supplies the compatible Fredholm class and a concrete Gaussian cokernel.
000123 | It does not yet identify that cokernel with the actual q spin fibers.
000124 | 
000125 | 6. The concrete remaining domain map
000126 | The spin Hardy construction already supplies an explicit positive local map.
000127 | On each lifted open z disc, write a section as f(z)sqrt(dz).
000128 | At its normalized boundary its coefficient is
000129 |   exp(i theta/2) f(e^(i theta)),
000130 | giving the ordinary NS basis n+1/2 and the exact multiplier z=e^(i theta).
000131 | This is the existing half-form norm and closed Fourier-domain dictionary.
000132 | The old boundary source further gives B_cov=2(A_1/4+A_3/4) as a DIRECT SUM
000133 | and U_cov=[[0,U_base],[I,0]] in those quarter blocks.
000134 | A single quarter multiplier U_base is not silently the primitive U_cov.
000135 | 
000136 | To make this the original F5 seam defect, the actual trace map must now
000137 | identify the seam pair, its three-sheet pullback and spin half-form domain.
000138 | At the original boundary ramification points z-z0=a u^3, the L2 measure-zero
000139 | argument suffices for the multiplier defect, but not for an H1 operator
000140 | domain: pulling back d/dtheta has coefficient i z/(3a u^2) d/du.
000141 | The actual transmission/normalization and half-form connection must be
000142 | transported, not assumed from a smooth boundary point count.
000143 | This does not rule out the map; it names its first decisive local test.
000144 | 
000145 | For the FULL actual v252 coupling use section3 with
000146 |   h=A gamma_F-i gamma_F Phi
000147 | and its given H1 domain. The scalar calculation in section5 does not
000148 | replace that full source with one selected channel or assign it three
000149 | physical copies. Its actual intertwiner must transport Phi, finite
000150 | grading, charges and the original seam action as well.
000151 | 
000152 | For the scalar block, the given circle group exp(i t B) preserves G but
000153 | conjugates U to U_t=exp(i t)U. It therefore transports the clutch domain
000154 | v1=U v0 to v1=U_t v0, rather than preserving the fixed U domain. Exactly
000155 |   F_(U_t) diag(exp(i t B),exp(i t B))=exp(i t B) F_U
000156 | on the actual graph. This is a family covariance with a MOVING seam phase,
000157 | not a fixed-defect physical-time identification. The original rule must
000158 | supply that moving phase/gauge if it is to define the same physical time.
000159 | 
000160 | 7. Outcome
000161 | The original relative seam operation has an explicit, compatible F5
000162 | Fredholm consumer, with the primitive one-copy defect proved directly
000163 | on the existing complete two-face graph.
000164 | The missing comparison is the original closed trace/cabling map to S0
000165 | Hardy data, with calibrated faces, multiplier and domains. Equality of
000166 | indices, norm characters or normalized operator rays does not supply it.
000167 | The once-derived norm result remains solved. No Seed state/time or
000168 | physical P1 source is chosen anew by this continuation.
000169 |