Standing Orders
How this register works, what it claims, and what it does not.
I. What this is
A neutral, public catalogue of unusually large or otherwise anomalous buildings in
Capital Rift. It records
where a building is, how big its floors are, how many floors it has, who owns them, and roughly
what it appears to earn. It takes no position on how any building came to exist. It is maintained
by its readers.
II. Names are self-declared and unverified
Capital Rift publishes no API and offers no way for this site to confirm who you are. The only
known route into the game's data is replaying your own session cookie, which is a full account
credential — so this site will never ask you for one, and you should be wary of any site that does.
You therefore simply state the name you play under. It is signed into your browser session and
attached to whatever you file or edit. Anyone can type any name. Treat attribution
on this register as a claim, not a fact.
III. Right of removal
If the Capital Rift player named on an entry says they did not file it, the entry is
removed. No argument, no review. State that name in your session, open the entry, and use
Dispute this entry — removal is immediate.
The same applies if you are named as a building owner or floor owner on an entry and want it gone.
Removals are soft: the record is hidden from the register but kept in the history, so a bad-faith
takedown can be undone and seen for what it was. Every removal and restoration is logged publicly
with the name that performed it.
IV. Anyone may edit
This register is wiki-style. Any entry can be corrected by anyone, and every change is recorded in
that entry's history with the name and time. If something looks wrong, fix it.
V. On the rent figures
Every rent figure on this site marked est. is an estimate. It is not a reading from
the game and it should not be relied on.
The model was calibrated on six floor observations taken on 22 September 2026. Those six are worth
looking at honestly, because they do not say what you might expect:
Calibration observations
| Building | Floor area | Rent /min | Per 1,000 m² |
Floor area spans 2.40× across that set. Rent spans only 1.50× — and runs the wrong way: the
63,652 m² floor earns less than the 26,526 m² one. Linear, power-law, logarithmic and
saturating curves were all fitted; every one lands around R² ≈ 0.15, which is to say floor
area explains essentially none of the variation in rent above roughly 25,000 m².
So the model does not multiply by square metres. It reports the observed average with the observed
spread, and flattens above 25,000 m² exactly as the data does. Below that line there are no
observations at all, so any figure shown is an assumption and is labelled assumed.
Foot traffic is not in the model yet. It is recorded on every entry so that it can
be fitted once there is enough data. Until then the estimates ignore it — which is very likely why
two floors of nearly identical size can earn 48% differently.
If you can read a floor's actual rent in game, put it in the observed rent field. Observed
figures are shown as fact rather than estimate, and they are what will make a better model possible.
VI. Sabotage status
Sabotage, here, is not damage. It is a market undercut: small ordinary buildings bought up
around a bugged one and listed cheaper, so the renters walk and the big floors stop filling.
The building still stands and still says it is worth a fortune. It simply has nobody in it.
Every entry carries one of three states, with the name that reported it and when:
- Unchecked — nobody has reported either way. A dashed gold ring on the map.
- Untouched — renters still in place, still earning. A solid gold disc.
- Sabotaged — undercut, renters gone. A slate disc struck through.
This exists because the useful question about a building is nearly always has someone
already done this one?, and the answer saves a wasted trip. Anyone can set the state in
one click from the entry, filter the map down to any of the three, and sort the docket by it.
Every change lands in that entry's history like any other edit.
Sabotaged is the end of the story. The cheap buildings do not go away, so a
building that has been undercut stays undercut. Moving an entry off that state is not a
recovery — it is a retraction of a report that was wrong, and the history records it as one.
The register takes no position on sabotage. It does not mark a building as
deserving it, it does not rank who has done the most, and a sabotaged entry is not a scoreboard
line. The state is recorded the same way floor area is recorded: because it is true of the
building, and because knowing it is more useful than not knowing it.
It is still a claim like any other here, made under a name nobody has verified. If you find a
building earning happily that someone has marked sabotaged, correct it.
VII. Coordinates
Every entry carries its coordinates as two numbers separated by a single space — latitude first.
That string pastes straight into the search box in Capital Rift. The copy button is on every entry;
press C with an entry open to copy without reaching for the mouse.
VIII. Map data
Everything geographic here comes from OpenStreetMap — the same source the game world is
built from. Searching a building by name goes through
Nominatim,
OSM's own gazetteer, so a name that exists in the map data will be found whatever script it
is written in. Footprints and floor areas are read from the OSM API, falling back to the
Overpass API — areas are
computed from the building's actual polygon, not estimated.
Map data © OpenStreetMap
contributors, available under the Open Database License. Basemap tiles from OpenStreetMap,
darkened in the browser. Place search by OpenStreetMap Nominatim.
These are volunteer-run services shared by everyone who uses OSM. Searches are only sent when
you actually ask for one — never on a keystroke — and every result is cached for a day, so
this register stays a light guest on them.
Tallinn Co. Justice is an unofficial fan project. It is not affiliated with, endorsed by, or
connected to Capital Rift or its developers.