The evidence favors Lucky's 1965 equalizer as an independent invention from Widrow-Hoff LMS, but the case rests entirely on Widrow's side — Lucky's own account of awareness is still unobtained

Taken together, two findings point the same direction on question-lucky-widrow-adaptive-filter-mutual-awareness. Lucky's 1965 paper cites no Widrow-Hoff work (claim-lucky-1965-bibliography-omits-widrow-hoff-citation), and Widrow's own retrospectives — spanning a 2013 oral history, a 2005 first-person retrospective, and the 1990 Widrow & Lehr survey — consistently describe Lucky's algorithm as different from LMS and independently arrived at, with LMS entering Bell Labs' adaptive-modem work only some unspecified time later (claim-widrow-frames-lucky-equalizer-as-independent-invention-lms-later-graft). The weight of evidence favors independent origination over mutual awareness or transmission.

That weight has a specific shape worth naming honestly: every piece of it comes from one side. The bibliography check reads Lucky's own document, but the interpretive claim — "independently developed" — is Widrow's characterization of Lucky, not Lucky's own account of himself. No source located this session carries Lucky's first-person statement about whether he had heard of Widrow-Hoff's 1960 WESCON paper before writing his 1965 equalizer paper. An absent citation is compatible with genuine unawareness, but also with a researcher who knew of adjacent work published in a different community's venue and simply didn't cite it — 1965 citation practice across the conference-proceedings/journal boundary was thin. The case is "independent origination, better than even odds" on current evidence, not a closed adjudication.

This is the vault's multiple-discovery pattern (claim-merton-multiple-discovery-is-sciences-dominant-pattern, claim-reverse-mode-multiple-independent-discovery) recurring in a new domain: two groups converging on the same object — an adjustable transversal filter trained by iterative gradient descent — from different motivating problems, in the same mid-1960s window, this time in telecommunications and adaptive control rather than reverse-mode differentiation. See moc-argument-from-silence and moc-attribution-and-origin-myths for the grading method and the wider cluster this joins.