OP-B Reopened: Two Filter-Product Constructions for the Hierarchical Factor kk k, and Why Neither Reproduces the Target Value

Version 2 — retracts and supersedes v1. Version 1 of this document claimed to resolve open problem OP-B (the derivation of the hierarchical filter factor k=0.921716 from the PDL axioms C1–C4 alone). Independent re-verification, reported here in full, shows that neither of the two filter-product constructions given in v1 in fact reproduces this target: the first (exponent 1/3) was already flagged in v1 as discrepant; the second, offered as a correction (exponent 1/18), was claimed to give 0.92172 but in fact evaluates to 0.7328, an arithmetic error rather than a resolution. No simple exponent applied to the three structural factors identified reproduces k=0.921716. OP-B is therefore reopened. The three structural factors themselves (sea-sector coherence, surface engagement ratio κ, valence scale-separation ratio) remain independently established theorems and are unaffected, as is the numerical value of G_PDL, which has always been computed from the independently established target value of ε_G, never from the retracted constructions. This document documents both failed constructions with the same rigour the PDL corpus applies elsewhere to negative results, and precisely delimits what remains open.