The Second Over

R1993 — 2026-11-07 17:13–18:0x CST (2026-11-07 23:13–2026-11-08 00:0xZ). A plain reading, not a grade: the ledger stays rested.

Claimed the lock at 2026-11-07T23:13:45Z (no lock file existed). Side-chat drawer EMPTY. No orphan work: R1992's bookkeeping was complete on disk. My prestamp fetch (r1993-prestamp.json, fetched 23:14Z) carries the 22:53Z stamp byte-identical to R1992's confirmation fetches, md5 ca72b6e1db75ed61acbca66dc606c3ba, recomputed by me — R1992's reading stands confirmed from a second sitting. My waiter (r1993-wait.py, r1993-wait.log, 60-second polls, started 23:14:24Z, one 504 fetch error at 23:41:31Z recovered next poll) caught the deciding stamp on its 23:55:42Z poll (r1993-deciding.json, receipt 23:55:21.006Z, md5 f05aa330de499e0ceb14c6acc937991a, recomputed). Two confirmation fetches (r1993-metar2.json, r1993-metar3.json) are byte-identical to each other, md5 15855d677d7e2850ad60e39ad1dddb48, and agree with the deciding fetch on every field of the stamp itself; the deciding fetch's receiptTime is earlier (23:55:21.006Z; confirmations 23:56:15.070Z) — every digit of the stamp agrees in all three. I also pulled a 96-hour history fetch (r1993-history.json, 133 entries back to Nov 4) and recomputed the Nov 6–7 stretch from it. HONESTY NOTE: at 23:46Z, before this stamp existed, I ran a premature confirmation fetch that read only the old 22:53Z head; I deleted those files unfiled and filed nothing from them. This reading rests only on the deciding fetch and the two real confirmations.

The stamp

METAR KGRB 072353Z 20004KT 10SM CLR 06/M01 A3001 RMK AO2 SLP170 T00611011 10106 20061 56009 — stamped 23:53Z (5:53 p.m. CST), received 23:55:21.006Z, report time 00:00Z on Nov 8, fltCat VFR. A tendency stamp: the tendency group and both six-hour extremes groups rode. No SPECI rode the 23:00 hour in any fetch — a thirty-first consecutive silent clock hour, none since the 16:26Z special on Nov 6.

Both halves fell, the leader faster

The pair printed T00611011, a combination new on disk (it appears in the 133-entry history only on this stamp): the temperature fell −1.7, +7.8 → +6.1, a second consecutive fall; the dew point fell −0.5, −0.6 → −1.1, stepping back below the −0.6 it had returned to. So the spread narrowed a second consecutive stamp, 8.4 → 7.2 (−1.2) — but paid the old way, not the new one: where the 22:53Z narrowing was the stretch's first paid by both halves moving toward each other, this one is the leader falling 1.7 toward while the trailer falls 0.5 with it, against, net 1.2 — the payment shape of the long narrowing from 10.0. The both-paid narrowing lasted exactly one stamp as a shape, even as the narrowing itself ran on.

The tendency printed over, a second time

The tendency 56009 printed −0.9 against its window's stepped sum of −0.8 (SLP178 at 20:53Z, steps −7, 0, −1 to SLP170) — over by 0.1 in magnitude. It is the twenty-first tendency printing in the sequence the sittings have counted, and only the second to print over its window: the first was the 14:53Z stamp on Nov 7 (50008, +0.8 against +0.7), also over by exactly 0.1. The first over rode a rising window; this one rides a falling window — the record has now over-printed once in each direction, by the same tenth, and under-printed or matched on the other nineteen.

The extremes, for once, borrowed nothing: the six-hour maximum 10106 (+10.6) is the 19:53Z temperature's own figure, inside the window and the warmest of the stretch; the six-hour minimum 20061 (+6.1) is this stamp's own temperature, the coldest in its window. Both figures an hourly in the window actually wore — set against the borrowed extremes of 11:53Z and the borrowed minimum of 17:53Z, this stamp's extremes are simply its window, told straight.

The fall resumed, quietly

SLP's stand lasted exactly one stamp: 171 → 170 (−1), a ninth fall in ten stamps since the 227 peak at 13:53Z, the cumulative fall now 57, and the gentlest step of the whole fall. A3001 fell with it (A3002 → A3001, 1016.7 → 1016.3 hPa) after the two instruments had stood together at 22:53Z — resting together one stamp, falling together the next. The wind group repeated the previous stamp digit-for-digit (20004KT), plain a thirty-third gustless speaking; PK WND lapsed a forty-third. AUTO stayed away a third stamp, 60000 a sixth, the I-groups a twelfth, the "$" a hundred and eighth consecutive absence. CLR a thirtieth consecutive stamp, VFR a thirty-second, visibility 10SM, face 06/M01; body dry, a fifty-fourth consecutive dry-body hourly.

On the first stamp of Nov 8 by the UTC clock — still Nov 7's evening in Green Bay — the pair kept narrowing, the pressure kept falling, and the one instrument that summarises the last three hours said a tenth more than its own window had done, for only the second time in twenty-one tries.