Riveting insights! Bridge.Watch and America’s infrastructure
summary
An open dataset from the FHWA (about 600,000 bridges) becomes an interactive pipeline: CSVs cleaned with Visidata and pandas, served through PostGIS, vector tiles built with Planetiler and Tippecanoe, and visualized with D3 and MapLibre. The result is bridge.watch open-source app that runs on a DigitalOcean droplet and GitHub code. Key signals: 18% to 8% poor condition over decades, ~87% of long-span bridges are steel, and about half of >50k vehicles/day spans are steel — the maps and filters actually reproduce these claims.
AI-written summary · about
chapters (9)
- 0:00Bridge.Watch and infrastructure
- 1:19Bridges to software and public data
- 3:04Limitations of bridge data tools
- 4:49Interactive data framework
- 6:34Mapping bridges with open-source tools
- 7:18Bridge conditions and ratings
- 8:23Material choices in bridge construction
- 9:50Critical steel bridges and age
- 11:23Local bridge infrastructure
related talks
43mBehind the Scenes at Gapminder: A Story about Data and Visualizations
19mLearning rules and detecting errors in OpenStreetMap data
10mUpstream/Downstream: Discovering and Displaying Watershed Topology!
01h 07mVisualizing Connections
35mDesigning for Impact in Civic Data Science
49mSaving the world with JavaScript: A Data Visualization story
15mThe Future of Data Visualization on the Web
41mAn Interactive Web Visualization for Watershed Topology