What makes this a source-backed food map?
This site uses real OpenStreetMap-derived records instead of mock narrative labels. Each location points to a source URL for verification.
A Data-Driven Hong Kong Food Map
Explore real restaurant points from OpenStreetMap. Each marker links to verifiable source records and is filterable by district, amenity type, and cuisine tags.
Data source strategy
The dataset is sourced from OpenStreetMap-compatible fields (name, location, amenity, cuisine, district-related address tags) and can be refreshed through Overpass API.
Restaurant cards intentionally keep source labels minimal for readability; detailed source provenance is documented here. Current map records are derived from OpenRice and OpenStreetMap Nominatim/Overpass workflows.
Data quality limits
Open data quality varies by neighborhood and contributor activity. Missing cuisine or opening hours are kept as null values rather than inferred.
How to interpret filters
Filters combine verifiable source fields (district, amenity, cuisine) with a lightweight discourse coding layer (source lens and frame label) to support comparative interpretation.
Project purpose
The aim is a reproducible map workflow: collect open restaurant records, clean them into a stable schema, and explore patterns without opaque labeling.