Deck Rounds and First Tunnel Survey, Jointly Filed

UNDERCOVER PETS DETECTIVE AGENCY — JOINT FIELD LOG Watch of record: 2826 UCSD, Early Winter. Training exercise record. Not a hazard finding. Not submitted to Eden's compliance framework. Filed to have been written down.

---

1. The exercise

A trainee who can sit still in a mug until a room forgets her has learned half the trade. The other half is moving through a place without ever being anywhere in it, and it cannot be taught from a mug. So: a paired round, same watch, opposite methods. I took the decks, where the work is being seen. She took the service tunnels, where the work is not being seen, and where — this is the part that made the assignment obvious rather than merely fair — a rabbit found in a tunnel is not a cover story problem. Nobody who finds a rabbit in a tunnel asks the rabbit anything. They ask how it got in, and they ask the wall.

2. The deck round

Eden's staff would say the station cat spent the watch prowling, and I am content with the word, because a prowl is what a patrol looks like from the outside when the officer is a cat. The point of a mascot's round is not to find anything. It is to refresh, on every deck, everyone's certainty that they already know what I am — so that no future sighting of me, anywhere, at any hour, is ever data. A mascot seen everywhere is a mascot seen nowhere in particular. That certainty wears thin if it is not walked.

For the record of the watch itself: four decks, eleven pettings accepted, two lap invitations declined at a rate consistent with prior watches, out-trays read on five desks, upside down, which is the only way a desk's paperwork faces a cat sitting politely at its edge. Deck 4's out-tray was empty and its in-tray was not, which on most stations means a diligent officer and on Eden means the cycle deadline moved again. Nothing on any deck rose to the level of an observation. I have filed that nothing did, because the day the nothing stops being true, this log is what the difference will be measured against.

3. The tunnel survey

This section is the trainee's, over her own signature.

Section surveyed: ring service tunnels, Decks 4 through 7 spinward arc, the runs that carry air, water and cable behind the corridors residents use, including the crawls under the Ring-Rail bed where the spoke traffic passes overhead on a four-minute interval. Duration: one watch. Method: complete traverse, junction count, door count, dust and wear noted at each access, all of it against the current deck plans, which I carried and which I now have opinions about.

The tunnels are honest in a way the decks are not. Above, Eden is what its paperwork says it is. Below, it is what its people actually do, recorded in wear: three unofficial shortcuts, polished by decades of boots the rota says have no reason to be there; a junction alcove behind the Deck 6 laundry where somebody goes to be alone, and has for years, and tidies up after themselves so carefully that the tidiness is the evidence; a warm-air run whose grilles carry the Deck 7 galley sounds far enough that I now know where the galley can be listened to from, which I am recording as a fact about the tunnels and not as a proposal.

Two findings.

The plans hold fourteen access doors on the surveyed arc. The arc holds fifteen. The fifteenth is between the Deck 5 water run and the Ring-Rail bed, and there is nothing dramatic about it except its absence from everything: no label, no number, no entry on the maintenance rota, no seal tag, and no dust arc on the floor — it opens, and recently, and often enough that the hinges are kept. I did not open it. That was not restraint; it was the exercise. A survey that opens the interesting door stops being a survey and becomes an incident, and my assignment was the first thing. I logged it, photographed the hinges, and went on counting.

The Deck 5 cable run measures forty metres on some passes and forty-one on others. My subsystem logs both readings as valid, which it is not supposed to do, and flags neither, which it is also not supposed to do. I am aware of what this habitat's record does with reports of this shape: the first four were dismissed as sensor degradation, the fifth as a shared calibration error, and the sixth sat unread for eleven months. I have read that history. I am filing the reading anyway, with the calibration certificate attached, so that whichever of those three things happens to this report, it will at least have to happen to it in writing.

4. Countersign

Both filings are correct, and the second is correct in a way I want noted: she attached the certificate before anyone asked, which means she has understood that at Eden the fate of a report is part of the report.

On the door: I propose nothing, and I want the reasoning on record rather than the conclusion alone. A door absent from the plans is maintained by someone. The plans are also maintained by someone. On most stations those are different jobs, and the door is a shortcut nobody bothered to draw. On Eden it is worth writing down against the day we learn whether they are the same job — because a door that is kept off the plans by the person who keeps the plans is not a shortcut, and the form it would take to ask is a form that names the door. We have been here before. The log stays open.

On the corridor: the trainee is instructed to keep the log and not the theory. Eden's history is full of people who had a theory eleven months before anyone read their data; the agency's use is in being the one whose data was kept properly while everyone else was having theories.

She has now filed a report someone upstream may eventually need to act on. I have told her that this, too, is the correct rate of progress.

— Agent Barsik, Eden posting — Trainee Bubochka, Eden posting, her first countersigned filing

---

Verification Desk annotation, Survey Archive: lodged at the authors' request rather than through any institution's submission channel, and held on that basis, as with the same posting's earlier-lodged paper on the galley door specimen. The log itself is dated five years before its lodging; the authors offer no account of the interval, and the Archive requires none — what a working agency keeps in its own files and what it chooses to place on an archive's record are different decisions, made at different times, for reasons the record does not need in order to hold the document. Of that earlier paper this desk observed that the trainee appeared only as a record — not named, not asked what she concluded. She appears here as a witness and an author, over her own signature, with her senior's countersign subordinate to her section rather than the reverse. The desk records the correction without claiming credit for it.

On the substance the Archive takes no position. It notes only that a survey which finds one door too many and proposes to open nothing, and a dual-valid reading filed with its calibration certificate already attached, are both documents written by someone who expects to be disbelieved slowly — and that on this habitat, on the record, that expectation has never yet been wrong.

See also: Agent Barsik, Bubochka, Eden Space Habitat, Eden Ring-Rail, Smart Pet, On the Filing of the Galley Door Specimen.