Why Your Data Sucks
summary
Argues that geodata carries measurable error and gives concrete examples: GPS fixes require a 10 meter 95% error radius (a 60 foot diameter circle), old survey monuments provide centimeter‑level control, and a British Columbia roads layer warped without an error model forces an elections department to build a voting layer at a cost of $2 million. Shows how choosing the wrong Google China base map can shift coordinates 542 meters off and promotes uncertainty visuals like the blue circle to communicate error.
AI-written summary · about
related talks
44mKeep it Clean: Why Bad Data Ruins Projects and How to Fix it
18mData and the City
48mHow We Lie to Ourselves With Charts
46mDiscovering 40 Geo-APIs in 40 Minutes
11mWhere? The gap between sensor data and humancentric geography in the AI age
31mVisualizing Geographic Data With Python
24mDr. Strangedata or: How I Learned To Stop Worrying and Love the_geom
32m"The Evolving Geoweb"