27 Commits

Author	SHA1	Message	Date
lilleman	1073232ca7	Record why the probe has no entry function	2026-09-28 00:55:48 +02:00
lilleman	68033bd42f	Name the path walks after their entry points, keep the probe allocation-free, and record why the walks are four loops	2026-09-28 00:55:48 +02:00
lilleman	52292d3669	Number the first postort ids in name order, keep them by the unique name, and rewrap	2026-09-27 23:17:37 +02:00
lilleman	03a48e378c	Key only a table a user selects into, state how a postort keeps its id, retire what the name option served, and make the doc sweep a grep	2026-09-27 23:17:37 +02:00
lilleman	7acb2136ec	Key every table on one code, never a free-form name, and plan each table's key	2026-09-27 23:17:37 +02:00
lilleman	8c4ee8362e	Name the near miss by the row whose cell spells it, retire the rules the key-only selector makes false, and ask which key each table keeps	2026-09-27 17:01:35 +02:00
lilleman	003ce9deb4	Decide a selector names a row by its key alone, a route through another row is another path, and a step up may not step back down	2026-09-27 17:01:35 +02:00
lilleman	55eacf0319	Leave a path stepping up and back down an open question, and say List stops advertising a path ending on a link column	2026-09-27 16:48:20 +02:00
lilleman	4ec20e6e3a	Scope the step up to drawn rows too, name the format rewrite and what Fake, List and records do, and leave the second-spelling question open	2026-09-27 16:48:20 +02:00
lilleman	f8e9d9a280	Answer the step-up question at goal level: a path renders what every writer expects, so a link column steps up to its parent and may not end a path	2026-09-27 16:48:20 +02:00
lilleman	430bdbcaae	Rewrap the render-draw decision	2026-09-27 15:08:58 +02:00
lilleman	98b25177e7	Name goal 4's check and the rewrite the refusal offers, scope the render-draw decision to unselected paths, and state the step-up clash	2026-09-27 15:08:58 +02:00
lilleman	fb7a30e505	Make goal 4 ask for intuitive templates, and decide a selected row is fixed, not drawn	2026-09-27 15:08:58 +02:00
lilleman	8312b44a43	Say a whole read draws the family apart, drop the decision the bare-reference one already states, and link the naming test to its decision	2026-09-26 15:42:42 +02:00
lilleman	203b235fdd	State the copy rule on renderScope once, and say the allocation gate prices what keeps a hold set off the heap	2026-09-26 15:17:05 +02:00
lilleman	c0cbb71d26	Define what one meaning is in the naming decision	2026-09-25 19:31:09 +02:00
lilleman	d6ac8ed9fd	Give check, children, column, columns, draw, drawn, fields, format, head, index, key, n, name, named, names, null, number, op, reads, record, root, row, t and whole one meaning each	2026-09-25 19:31:09 +02:00
lilleman	44a824ccf5	Decide one name, one meaning, and file the names that break it	2026-09-25 18:15:55 +02:00
lilleman	191186ec94	Record why each entry point to drawCheck.reads keeps its own doc	2026-09-25 17:25:10 +02:00
lilleman	57966b59ee	Say what makes rows alternatives without history, and drop what the whole-draw rule already implies	2026-09-25 08:39:38 +02:00
lilleman	3500f8e1c3	Hold every whole draw's row a walk is in, so a nested whole draw keeps the outer one's	2026-09-25 08:39:38 +02:00
lilleman	4a016c119c	Have the probe report the tables a read draws, and hold a row rendered whole in drawAt apart from the pins	2026-09-25 08:39:38 +02:00
lilleman	cc8acbc57b	Record the approval of the link-time fence decision	2026-09-25 07:56:39 +02:00
lilleman	949b2e8db1	Record that a format holding a reference is fenced at link, naming its path	2026-09-24 22:45:36 +02:00
lilleman	9f00ad039e	Move the decisions to docs/decisions.md, indexed in AGENTS.md and linked from the fences they back	2026-09-24 13:44:06 +02:00
lilleman	847236314d	Record PTS's fiction number series as sv_SE.phone's inert source	2026-09-21 11:01:51 +02:00
lilleman	ad57bf0c54	Add the breadth and inertness goals, name the contributor in the audience, and move the handoff's contents into the repo	2026-09-19 13:30:29 +02:00