Data sources, licenses & attribution

Facts vs. expression — how media aggregation works

Copyright protects the expression of a story — its words, photos and structure — not the underlying facts. CityAlert stores only incident facts: a headline, a publication time, an approximate geocoded location, an inferred category, and a link back to the original article. We never copy, store or republish article text or images. Every media-sourced incident links to the publisher's site, driving readers to the original reporting. If you are a publisher and would like your feed removed or adjusted, see our takedown process.

Feed refresh & status

CityAlert.live polls 690 feeds, each on its own schedule — dispatch and alert feeds every few minutes, batch open-data portals every few hours. “Last fetched” is when this server last pulled the feed, not when the source published; a source's own lag can add minutes to days on top. 8 feeds are currently failing — the map keeps serving each one's last good data.

Government & open-data feeds

Media & RSS aggregation

In regions without incident-level police open data (Vernon / North Okanagan, Los Cabos, Metro Manila), CityAlert geocodes safety-relevant headlines from local RSS feeds. Aggregation is limited to headline + derived location + link — the copyright-defensible core described above.

Maps & geocoding

Place search and article geocoding use the public Nominatim service under the OSM Foundation Nominatim usage policy (max 1 request/second, cached permanently). At scale we plan to self-host or move to a paid geocoder.

Platform & planned sources

Use of this data

CityAlert is an informational tool only — do not rely on it for safety decisions, and always follow official emergency channels. Media-sourced incidents are approximate and selective. See the Terms of Use for the full disclaimer, and Contact for correction or takedown requests.