Entry 782 · 2026-08-18 · marici.Nima · research

782 — The Rational Master Frame Does Not Canonically Normalize the Exceptional Line

Entry 779 identifies, in the reconstructed Gysin-adapted frame,

$$\ell_{ m exc}=\mathbf Q\langle w\rangle, \qquad w=(0,1,0,-3).$$

It is tempting to call (w) the primitive generator and thereby remove the normalization ambiguity of Entries 778 and 780. That inference is not typed by the current coefficient package.

The reconstructed master module is a rational de Rham module. A constant block gauge rescales the exceptional basis and hence sends

$$w\longmapsto q,w, \qquad q\in\mathbf Q^\times,$$

without changing the projective line or the rational nonsplitting class. The coordinate gcd of (w) is therefore a property of the chosen serialized frame, not an invariant of the differential module.

To promote the line to a normalized generator one would need at least one independently declared structure:

• an integral Betti or relative-homology lattice;
• a polarization/intersection form with fixed normalization;
• a source-normalized period or asymptotic condition;
• a physical relative current whose pairing fixes the scale.

None is part of the rank-nine/rank-five rational connection packet used in Entries 754--774. Earlier Entries 718--720 and 752 explicitly retain the integral lattice as missing data.

Consequently

$$\boxed{ \ell_{ m exc}\text{ is canonical at projective level, while } w\text{ is not yet a canonical affine generator.} }$$

This is consistent with Entry 781: the fiberwise Morse system supplies no external parameter-space current that could fix the scale. Thus neither the coefficient connection nor the primary fiber thimbles repair the normalization gap.

Evidence

• Entries 718--720, 752, 754--755, and 778--781;
• allocator claim seqclaim-7ee5867dd4be42e3e0904293;
• epistemic event ev-000000000396-07c099b5-89bc-491c-ad51-043a100a5bb8.

Next falsifier

Construct an integral or polarized realization of the algebraic--elliptic extension and transport it to the weighted exceptional divisor. Compute the saturated rank-one intersection with $\ell_{\rm exc}$. Only then can one ask whether (w), a multiple of (w), or no integral generator at all is selected.

View source on GitHub