The Armistice Banner
Dead Letter's login banner changed on or about 12 July 1997 and did not change again while the board ran. It is fifteen lines. It names nobody, refers to nothing directly, and was understood immediately by everyone who saw it.
┌────────────────────────────────────────────────────┐
│ │
│ \ | / │
│ --+-- │
│ │
│ DEAD LETTER 1 line, still │
│ │
│ the roster is not checked here any more. │
│ │
│ if you sat it somewhere else, you sat it. │
│ if you sat it here, you sat it. │
│ nobody is going to ask you where. │
│ │
│ forty days is forty days on any clock and │
│ i have been wrong about the clock, not the │
│ forty days. │
│ │
│ login: │
└────────────────────────────────────────────────────┘Why it worked
The concession is in the last three lines and it is precisely bounded. Ossuary does not abandon the position that the Count admits exceptions; he abandons the position that his exceptions and Halberd's are different rites. The distinction is thin and it was sufficient, because the operational problem — two rosters, forty days each, one caller — was a question about recognition and not about doctrine. Stop checking, and the problem is gone whether or not either man was right.
Nailhouse never posted a reply and never needed to. Halberd's recorded response, secondhand, is four words: he did it properly. See lore/halberd-and-ossuary for the eleven years this ends and for the hand-off ten months earlier that probably explains it.
The lore/open-hand at the top is the largest attested instance and is the only one anybody has argued about, because a mark that means found open, left open on a banner announcing that a gate has stopped being checked is either the best use it was ever put to or a category error. Both readings are current.
The shibboleth
The banner became a test, and the mechanism is worth setting out because it is not the obvious one.
Dead Letter closed in 2001 and the text survives in dozens of copies, all
slightly different. Nobody has a canonical file. What circulated was not the
banner but the memory of it — people typed it out for each other, from having
seen it several hundred times at login, and the variants differ in exactly the
places you would expect memory to differ: box-drawing versus dashes, the
capitalisation of the board name, whether the 1 line, still sits on the
board-name row or its own.
What the variants do not differ on is the last three lines, which are word perfect in every copy. So the test that emerged is not can you produce the banner — anyone can find a copy — but do your errors sit in the right places. A version that is wrong about the box characters and exact about the clock is a version someone saw. A version that is perfect throughout was copied from a file five minutes ago, and a version that is wrong in the last three lines was reconstructed by someone who has read about it.
Nobody administers this. It is not a question in lore/long-count and it has never been proposed as one. It works entirely as a thing people notice.
The variant question
The version above is a reconstruction and follows the majority in every disputed place. It is not the banner. It is what fifteen or so surviving copies agree on, which is a different object and should be cited as one — the same distinction lore/sisken built his card format around, and the same one this wiki draws between what was observed and what was merely asserted in machinery/provenance.
Related
- lore/index — the collection
- lore/halberd-and-ossuary — the feud it ended
- lore/long-count — the forty days both boards counted
- lore/open-hand — the mark at the top
- lore/marrow-tape — the other object that survives by being copied badly