What we learned when a user tried to load a 1 GB GML file in a browser
A user attempted to load a 1 GB GML file directly in a browser, highlighting the impracticality of handling large GIS datasets with traditional vector formats. The experience demonstrated that vector tiles are far more efficient for web-based GIS, as they enable faster rendering, reduced bandwidth usage, and scalable performance with large geospatial data.
Background
GML (Geography Markup Language) is a verbose XML-based format for geographic data, often used in government and enterprise GIS systems. A 1 GB GML file is considered large—it will choke most web browsers and typical GIS tools.
Vector tiles are a modern alternative: they chop geographic data into small, pre-rendered chunks (tiles) that a browser can load incrementally, like map tiles from Google Maps or OpenStreetMap. This makes large datasets viewable without bogging down the user's machine.
Traditional GIS workflows often rely on desktop software (QGIS, ArcGIS) that can handle large files natively. But the web is increasingly the default interface for sharing and visualizing geodata, so formats matter for performance.
The piece reflects a broader shift in geospatial tech: moving from "give the user the whole file" (GML, Shapefile, GeoJSON) to "give the user only what's visible" (vector tiles, cloud-optimized formats like COGs).