Building OpenStreetMap Validation Processes in the Open
summary
Recap of seven years: the August 30, 2018 renamed‑city incident forces companies to run closed pipelines like Meta Daylight and Apple’s Atlas to produce vetted snapshots, while community tools such as OsmCha and Mouette improve but lack a shared workflow. Real problems include nontrivial versioning (a node moved from Lisbon to Brazil didn’t bump a way version) and many error types. Proposal: publish validation outputs in a common GeoJSON schema, feed them into realtime ML (an ORES‑style score) and tools like osm‑compare to prioritize fixes.
AI-written summary · about
chapters (6)
- 0:00Introduction and context
- 3:13Historical shift in validation focus
- 6:50Corporate validation responses
- 8:55Challenges of map validation
- 11:20Building real-time ML solutions
- 13:07Future of open validation
related talks
48mOpenStreetMap - Balancing structure with grassroots mapping
45mThe OpenStreetMap Project = Data + Community
32mThe OpenStreetMap Community
40mHow do you harness crowd sourcing for a community that also delivers for business?
21mFIXME: Documenting the Uneven Distribution of OSM Data Quality Issues
15mBeyondTracks: an OpenStreetMap case study
45mMapping the Future: The Intersection of Geospatial, Open Source, and Open Data
41mEmpowering the Map Ecosystem With Open Data