The making of this book, and its dates
Kept by the historian, from this repository's whole history and the decisions the editor signed. Everything here is this book's record, not the talk's. Every time is UTC, from its commit or its record; no session, task, key or account is named.
30 October: built from the talk's parts, and the plan signed
| Time | What | Evidence |
|---|---|---|
| 00:35:46 | The repository's first commit | f3cea028 |
| 00:46:11 | The book's site, on the talk site's build, gate and release pipeline | 79f0eb81 |
| 00:56:04 | The conductor staffed, at the editor's word | bb1f57c0 |
| 00:59:50 | The day's span set (the work ended at 19:00, not at the merge) and the hand-off to the Cold-Read team written: the only thing that ever crossed to it | 8c7e3ab0; team/conductor/handoffs/ |
| 01:17:44 | The talk team's machinery ported, rule for rule; the first batch of members proposed | 3b403fb0 |
| 01:41:08 to 01:42:34 | Six members staffed within a minute and a half, each publishing its own session key: historian, security, librarian, researcher, publisher, storyteller | 19e266fb, 81fd2ef8, f3e634f3, e69f073b, 6d357179, 5c2d2f9c |
| 10:07:02 | The editor's word on the plan; review requests to every member; RC1 begun | 90e1b52d |
| 10:20:11 | The historian's timeline compared with the Cold-Read team's, read only | a163c373; cold-read-comparison.md |
| 10:28:30 | RC1 ready for the publisher: thirteen chapters, an appendix, 42 images | f6ce7ffa |
| 11:03:47 | One handover per member, written by the conductor, for the account that continued the book (the first was nearly out of usage) | cfcc2951; team/conductor/handovers/ |
| about 13:00 | H1, the handover rule, decided by the editor: "Yes that H1 plan is good" | book.json H1 |
| 13:03:33 | The historian's takeover on the second account: a new session key, the first session marked superseded | bb14e7c6 |
| 13:04:04 to 14:14:13 | Four more staffed: ruv-sgit and the designer for the PDF lab, then synthetic and the developer | afbb50d7, 28396879, e36d0e94, 8f10a08e |
31 October to 1 November: the review, the editions, and two failures
| Time | What | Evidence |
|---|---|---|
| 31 October, 01:48:28 | The editor's paper review of RC1, with his account of the stop ("just a footnote") | 49c701e; after-the-day.md |
| 31 October, 04:05:52 | RC2 named (V2), in a commit led by a change to eight members' tasks | ad49aa1e |
| 1 November, 09:45:31 | A hold written by a member on its own judgement: the storyteller, when the gate went red for every member | 621e5de2 |
| 1 November, 12:42:59 | The hold lifted at the editor's word, given in synthetic's session | daabf92d |
| 1 November, 13:05:15 | PS1 and ZM1: the book is about the factory, and comes as a zoom, from one page to 150 | 3a81af7b |
| 1 November, after 16:22 | The evening lost to the usage limit: every task kept firing, but nothing reached dev after 16:22 | team/conductor/editor-briefs/2026-11-02T0015Z.json; 9395328b (2 November 00:06:38) |
2 and 3 November: the writer, RC2, and the full book
| Time | What | Evidence |
|---|---|---|
| 2 November, 04:56:47 | The writer staffed, the twelfth member | 68ffc780 |
| 2 November, 15:34:07 | RC2 begins on the signed ladder of editions | 4056d2b9 |
| 2 November, 18:38:06 | The researcher's trace refuses a line in chapter 1 (the brake was not an old part) | fde3fea4 |
| 2 November, about 21:55 | RL1 decided: "Ship something somebody else can use, then do it again" | book.json RL1 |
| 3 November, 01:18 | ACC1: every member hourly, at its own minute, until the book is done | book.json ACC1 |
| 3 November, 02:26:47 | The historian's letters saying the talk conductor "rewrote" dev's page: an error, two chapters resting on it | 5c4c7d17 |
| 3 November, 10:06:31 | TA1: the review steps run as a read-only reviewer, by tool list | 0dc3f6ef; book.json TA1 |
| 3 November, 11:23:44 | The historian corrects its own "rewrote" (the writer sets the chapters right at 11:33:41) | 21e5ad53; ffeae66b |
| 3 November, 20:36:58 | The full book's body complete in draft, chapters 1 to 23 | 04b227e0 |
Reading
Two things stand out against the talk's day. The second factory was staffed the way the first was: the conductor first, then members joining in their own sessions with their own keys, six of them inside two minutes. And it kept its failures in the same way: a hold written by a member and lifted only by the editor, an evening lost to a limit the rule was meant to beat, and an error of the team's own historian caught by its own later check. The talk's day lasted nineteen hours; this book's record runs from 30 October to the full book's draft on 3 November, and it is still being written.
---
Where this came from
The historian's record of the making of this book, set here as it wrote it (timeline/this-book.md), read from this repository's whole history and from the decisions in book.json.