nx_acceptview.nx

nx_acceptview.nx -- THE OPERATOR'S REVIEW QUEUE, PUBLISHED. WHY IT EXISTS (operator, 2026-09-03): "for human acceptance you need to publish a queue and evidence to review like how we have rigview but you arent anywhere". That was exactly right and it named the missing half of the three-party doctrine. The ledger was already complete -- nx_accept list computes, for every subject, which of the three parties is missing, and knowledge/store/accept- already holds the operator's own verdicts and words. NONE OF IT WAS PUBLISHED ANYWHERE. Acceptance you cannot SEE is not acceptance, and asking an operator to attest to a queue that only exists as a CLI verb is asking them to sign blind. ★★★★★★A DECISION THAT REQUIRES A HUMAN AND IS NOT PUT IN FRONT OF THEM IS NOT A PENDING DECISION, IT IS A STALLED ONE -- and from the inside it looks identical to the human being slow. COMPOSES, NEVER RE-DERIVES. The verdict and the missing-party call come from nx_accept list (one classifier, nx_accept_lib, shared by every consumer); the operator's words come from the accept- plane through nx_store_put's own loader. This organ computes NO verdict of its own -- a second opinion rendered beside the first is the duplicate-ruler defect wearing a stylesheet. SORTS BY WHO IS BLOCKED. Subjects awaiting the OPERATOR come first, because this page exists for them; everything else is shown but marked as waiting on a seat or a referee. A queue that does not say whose turn it is makes the reader compute that themselves, every time. ESCAPES BY CONSTRUCTION. Operator words are free text written by a human and are rendered through av_esc -- an unescaped quote or angle bracket in a rejection note must never become markup. nx_acceptview [droot] default sites/nishifamily/compare (page: <droot>/acceptance/index.html) license_tier: ORIGINAL No hw writes (Rule 26). expect_exit: 0