The Fab City Index is one surface of a larger instrument; this is the map of all of them. Each surface does one job:
one place explains the index, one shows the global view, one draws the scales, one lets operators populate it, and the
rest are the cities and communities that feed it. Follow the map below to the surface built for what you came to do.
6 surfaces · 1 shared data contract · 3 ways data enterssome surfaces planned — dashed · v0
↩ρ — observation becomes response. The surfaces send what they see back to the communities that act — and the loop the index measures is how fast that happens.
Boundaries — some surfaces are planned. The network map and city subdomains are drawn dashed: the federated
map of cities and labs, and per-city addresses, are designed but not yet built. The spine is drawn in iron
because it is the one thing every surface shares. Hand-drawn schematic · v0.
Each surface owns one job; the iron contract is the one thing they all share. Dashed = planned. Every reading carries a
live / partial / mock
label, end to end. Hand-drawn schematic · v0.
The process, three ways through
Every reading takes the same journey — observation, then ingestion, then the contract, then the surfaces, then back to action.
The instrument has one spine and one loop. Wherever a reading comes from, it enters the same shared contract — a small
document called fci-cells-v0 that holds one value per cell, each tagged with how trustworthy it is — and from there it
reaches every surface. The loop the index measures is ρ (rho): how fast an observation becomes a fitted response. Here is
the path for each of the three postures.
posture 01
If you observe
read index.fab.city → the atlas & scale map → the observatory → a city page
You never touch the data; you read what the instrument has made of it. Start here on
index.fab.city for what the index is and the argument behind it; open the
aggregation atlas to see how readings nest — a measurement at the
Community scale travels whole into City, City into Region, and stops there. For the live, global frame — feeds and
stories across the whole network — go to the PLANETAI observatory. To read one
place in depth, open a city page. Every number you see carries a pill —
live, partial or
mock — so you always know how much to trust it.
posture 02
If you maintain data
connect a source → the workbench → the contract → the index & observatory
You run a partner lab or lead a city's data, and your job is to turn raw sources into populated cells. Two of the three
routes are yours. The first is automatic: an adapter pulls from an open platform — Smart Citizen,
OpenAQ, a city's open-data portal, Metroverse — on a schedule, and writes straight into the contract. The second is by
hand: where no feed exists, an operator enters readings through the
operator workbench, which also reviews submissions, checks coverage, and
holds the sovereignty gates that decide what may be published. Either way the reading lands in the same shared
contract, fci-cells-v0 — see what each cell measures — and from there it surfaces on
the index, the city pages, and the observatory at once.
posture 03
If you run a campaign
a campaign creates the source → the contract → a city page → the index
This is the third route, and the only one that creates the data instead of connecting to something that
already measures it. It is the answer to the cells no open platform will ever cover: community sensing, local surveys,
participation, the matters of concern a neighbourhood decides to track.
Making Sense Bali is the live exemplar — sensor kits, a
moderated WhatsApp reporting bot, and surveys, all run by a local team. The campaign produces its own readings, those
readings pass a calibration and consent gate, and then they enter the same contract every other source uses, showing up
as that place's cells on its city page and in the index. A campaign is how a community puts
itself on the instrument on its own terms.
And then the loop closes. Once a reading is in the contract and visible on the surfaces, the point is not the
number — it is what happens next. ρ measures how fast an observation produces a fitted, human-approved response at the right
level of government, and feeds that back to the people who can act. Observation becomes response; response changes the place;
the next observation reads the change. That returning iron curve on the map is the whole reason the instrument exists.
Where to go next
One instrument, many doors.
If you only remember one thing: the surfaces are not separate products — they are different windows onto a single
instrument, and each one owns a single job. The index explains the measure;
the observatory shows it live and global; the
atlas draws how it nests; the workbench is where it gets populated; the
city pages are where places appear; and campaigns like
Making Sense Bali are how communities create what no platform
holds. Pick your door from the home page and follow it in.