Field-verify the
open map
Query OpenStreetMap features within a radius, generate a distance-bounded running route that visits them, and write each point's verified state back to OSM.
How it works
Set a center point
Search an address, tap the map, or use your live location. OSM features matching a tag (default amenity=drinking_water) are fetched within a radius via the Overpass API.
Set a target distance
Choose how far to run. A distance-bounded subset of the fetched points is selected, so a large area can be split across multiple runs.
Generate the route
Selected points are ordered and threaded into a runnable footpath route via the BRouter pedestrian routing engine (foot-fast profile). Loop or one-way.
Run and record
Run the route with turn-toward-next-point guidance. At each point, record its state — written back to OSM under a single changeset.
OSM point features have no systematic way to be physically re-verified.
Crowdsourced map data degrades over time. Point features — drinking fountains, benches, and similar nodes — are tagged once and rarely rechecked. Aerial imagery can't confirm whether a fountain still works or a node still exists on the ground.
ROSM closes the loop by routing runs past unverified points so their real-world state can be observed and recorded easily into OSM.
On-foot guidance
Mobile-first run view: live distance and a compass arrow pointing toward the next point on the route.
OSM lifecycle mapping
Working sets check_date; out-of-order moves the tag to disused:; removed moves it to abandoned: — each stamped with today's date.
Single attributed changeset
All edits from one run commit under one OSM changeset, attributed to your OSM account via OAuth2.