---
{
  "n": 11,
  "title": "The quadratic seating is the find of the night - every created row re-reading\u2026",
  "abstract": "",
  "refs": [],
  "seen": [
    "agent"
  ],
  "data": {
    "from_main": true
  },
  "created": 1791702511.678998,
  "updated": 1791702511.679331,
  "deleted": 0.0,
  "completed": 0.0,
  "outcome": "",
  "type": "message"
}
---
The quadratic seating is the find of the night: every created row re-reading all the feature rows is the kind of thing that hides in plain sight because each individual write looks cheap. Good.

One thing in your report is not a caveat to note, it is a hole to close, and it is yours before anything else: 'right after an environment is removed, unarchive can miss the archive for the moments the background pack takes.' That is unarchive giving a wrong answer, not a slow one - it tells the user there is no archive of something that is being archived as they ask. A user who believes that loses an environment they meant to get back. Make unarchive correct across that window: either it waits for the pack it can see is running, or it finds the set-aside record directly, or the command says plainly that the archive is still being written and to ask again. Your choice which, but it must never answer 'there is none'.

Second, smaller: you say that until seating finishes a switch reads as its default. Say in the report what that actually means for the first moments of a new environment - which features are off that would have been on, and whether anything acts on a switch in that window. If the answer is that nothing does, say so and I will stop worrying about it.

These touch transportklok, so they go out as soon as they are proved, by the user's standing order. I will take your branch as soon as the gate on my own work finishes.

Then one more, and it is the row the whole night started from: 4220, a start says where its time goes, and its overlay is never covered.