Open data and attribution
Where the data comes from
The parking facilities shown on this map are derived from OpenStreetMap. © OpenStreetMap contributors.
Share-alike
The derived database we produce (the tileset used by this map and by the app) is offered under the same ODbL 1.0 licence as the source data.
User-contributed facts
Factual corrections sent from the app (for example “this car park is free”) are OpenStreetMap-type data: our intent is to push them upstream to OpenStreetMap.
These facts live in a separate database and are distributed through a separate channel. When they are baked into a tile set, each one carries its provenance attribute: the distributed tile set itself states what comes from OpenStreetMap and what does not. They remain separable at any time — a pure OpenStreetMap database can be regenerated on demand, and the provenance mode is written into the produced tile set's metadata and then verified by our quality gate: a tile set announced as pure that would not be cannot be published.
Regenerating a pure OpenStreetMap database
The production pipeline accepts a mode that forbids any injection of contributed facts:
NO_OVERLAY=1 pipeline/bin/run_daily.sh
The provenance mode is stamped into the produced tileset's metadata (an “overlay” token) and checked by the quality gate: a database claimed to be pure but that is not cannot be published.