Methodology and Provenance
How Flock Clocker builds its map, where the data comes from, and how to check our work.
What this app is, plainly
Flock Clocker is a civic-records transparency tool. We map where government surveillance cameras, including automated license plate readers, are deployed across the United States, and we map the public contracts and government spending that pay for them. Everything we publish is drawn from records that are already public.
We map infrastructure and public spending, not people. We do not track anyone, we do not show real-time positions, and we do not provide any way to route around anything. The public map shows generalized coverage areas, never exact pinpoint locations.
Why this app is permissible
This is the same kind of work as data journalism and public-records reporting: taking documents that are already public, organizing them, and putting them where the public can see them. A resident has a right to know that their city council voted to fund a surveillance network, what it costs, and roughly where it operates. We surface that, and we tie every record back to the document it came from.
We have deliberately designed against the things that make this category risky:
- We do not publish exact camera positions to the public. Coverage is shown as generalized area zones built from aggregated map cells. There is no public path to a precise camera location.
- We provide no real-time tracking, no alerts to dodge anything, and no routing or evasion feature of any kind.
- We keep evidence tiers separate and labeled, so an unconfirmed community sighting is never presented as an established fact.
- Every record is disputable and traces to a public source, and anyone, including a named operator or business, can ask us to correct or remove a record (see Corrections).
Where the data comes from
Flock Clocker combines a few kinds of public information, and we keep them clearly distinct because they carry different weight.
1. Government procurement and public meeting records. Our core layer is mined from lawful public records: government procurement filings, city and county council agendas and minutes, and cooperative-purchasing records. These tell us who contracted for surveillance cameras, in which jurisdiction, for how much, and on what schedule. This layer carries counts, agencies, contract details, and deployment status (deployed, under contract, proposed, or vote scheduled). It does not contain exact camera coordinates, because the source documents do not.
2. Community sightings from OpenStreetMap. Some location detail comes from OpenStreetMap, the public, openly licensed map project, including surveillance-camera nodes contributed by the DeFlock community mapping effort. We treat these as community sightings, unverified. They are contributed by members of the public, not confirmed by us, and they are labeled that way everywhere they appear. We never present a community sighting as a confirmed or official camera installation. On the public map, community-sourced detail is generalized into area coverage, never shown as an exact pin.
3. Agency transparency portals (such as Eyes on Flock). Where a government agency publishes its own surveillance transparency record, we cite the agency's published portal as a source. We link to it; we do not relicense or republish it as bulk data.
How the map shows it (generalized areas, not pins)
The public map shows coverage as generalized area bubbles, aggregated map cells that say "there is documented surveillance-camera coverage in this general area," along with a count and the operating agency. It does not show exact camera positions, it is not a live feed, and it does not move with anyone. A momentary visual flourish when you tap an area is a one-time animation, not a live signal or a tracking ping.
We do this on purpose. The civic value, that a network exists, who runs it, and who pays for it, is fully delivered at the area level. Exact pole-level positions are not necessary for accountability, and publishing them to everyone would invite the wrong uses. So we do not.
Evidence tiers and confidence
Different sources carry different certainty, and we never blur them together. Every record shows which tier it belongs to and links to its source:
- Confirmed / deployed: a camera documented as installed (from a sighting or press tier source).
- Contracted / planned: a procurement record showing cameras are under contract or funded, not yet confirmed on the ground.
- Proposed / vote scheduled: an item on a public agenda, not yet approved.
- Community sighting (OpenStreetMap), unverified: publicly contributed map data we have not independently confirmed.
For attribution to a specific operating agency or, where applicable, a private operator, we show the evidence behind the attribution and how confident we are. An inferred attribution is shown as inferred, not as an established fact. We do not state that a named business or agency operates cameras at a place as a settled conclusion unless the record supports it; where the link is inferred, we say so, and the correction channel is one tap away.
Provenance you can check
Provenance travels with every record. A journalist, a researcher, or an attorney can open any marker and follow it back to the public document it came from. Citation is free and always available; it is the trust moat of this project. If you find a record you believe is wrong, every record is disputable (see Corrections).
Licensing and attribution
Map data and community sightings are from OpenStreetMap. Map data (c) OpenStreetMap contributors, licensed under the Open Database License (ODbL). See https://www.openstreetmap.org/copyright. Corrections to OpenStreetMap-sourced records are routed upstream to OpenStreetMap so the public map improves for everyone.
Our procurement and public-records dataset is compiled by Crossroads Technologies, LLC from public sources, with each record citing its origin.
Questions
Methodology questions: [email protected]. To dispute or correct a specific record, see our Corrections page.