back to table · edit · history · where entries came from · files · download
2638 bytes, as of the version from 2026-11-15 23:03 (current). Recorded here, not run.
# Argument selection for T718
The natural family is one real Li coefficient lambda_n(D) of the completed
primitive quadratic Dirichlet L-function, with parameter order (D,n).
D is a nontrivial fundamental discriminant; n is a positive integer.
The completion and factor n are fixed in the definition. There are no free
continuous couplings, evaluation-point choices, branches, or extra quantities.
D is **arithmetic_structure**. Goldmakher, Theorem 1.1 and its following
remarks (sources/goldmakher.pdf), identify all primitive quadratic characters
with fundamental discriminants and give their conductors |D|. The enumeration
includes every eligible D with |D| <= 40, rather than only prime conductors or
famous characters. It covers 12 positive and 14 negative discriminants, both
parities, and prime/composite and odd/even conductors. This broad conductor
prefix supports comparison of character families without sparse cherry-picking.
The cutoff 40 is a noncanonical practical bound: all character sums have at most
40 residue classes and every entry can receive a separate 100-digit numerical
check. It is not a claim that larger conductors are less meaningful.
n is **discrete_index**. It is the coefficient order in Li's named sequence,
not an evaluation-point mesh. Lagarias, equations (1.3)-(1.5) and Lemma 2.2
(sources/lagarias.pdf), fix its derivative/generating-function meaning and
normalization. Every n=1,...,20 is selected for each D. Consecutive orders give
a usable initial series and allow triangular coefficient transforms without
holes. The cutoff 20 is a noncanonical budget choice, with independently checked
cancellation and ample precision; it makes no asymptotic or positivity claim.
Breadth across conductors precedes greater depth in a few exceptional cases.
Exact enumeration in increasing |D|, positive before negative at a tie:
-3,-4,5,-7,8,-8,-11,12,13,-15,17,-19,-20,21,-23,24,-24,28,29,-31,
33,-35,37,-39,40,-40, each paired with all n=1,...,20 (520 entries).
The elementary enumeration in theta_check.py and the Sage squarefree tests
in generate.py agree on this whole set. Programs demonstrates the same quantity;
the document's complete-note gives exactly these bounds, and complete is no.
No reference_sample or mixed exemption is used. The separately proposed central
series is not part of this table and is not claimed as existing coverage.
Instruction evidence: ../20261115T193637Z-build-skills/manifest.json, matching
local/live bundle SHA-256 d6661c2f21613524889fa3a57d3c95f8330344ca62e5a856211c08a282b7dbab.
An independent argument-selection verdict is still required from a fresh critic.