Notes
Notes
Note: these are rough idea sketches. Nothing here is formally published. I write them down for my own archiving.
-
Re-running urban planning studies with agents
Collect every urban planning study with public data and published code, re-run them through a unified agentic codebase, and check whether the reported numbers hold up.
-
Were demolished buildings different from the start?
Street view archives let you see a building in its prime. Did buildings that were later demolished already look different from the ones still standing?
-
What makes a floor plan AI-readable?
AI keeps misreading housing floor plans. My hunch: the text annotations on a plan are more machine-readable than its lines.
-
Where urban datasets disagree
Administrative records, street view, and satellite observation describe the same city differently. I want to chart where they contradict each other and which source carries the error.