Spatial Programming for Environmental Monitoring
summary
Argues for treating space as a first-class programming concept using bigraphs to encode containment and links, implements a runtime called byrost that derives graphs from OpenStreetMap, adds spatial indexing and location-based naming with TLS via ACME, and supports spatial multicasting and delegated edge operation for network partitions. Demonstrates rule-triggered side effects (chainsaw audio → wake collocated camera), claims real deployment performance: 10,000+ nodes with under 1 ms reaction on constrained hardware.
AI-written summary · about
chapters (10)
- 0:00Spatial programming intro
- 1:46What are biographs
- 3:32Programming space with rules
- 5:18Extending biographs for state
- 6:44Networking spatial devices
- 8:09Spatial multicasting
- 8:51Querying by proximity
- 9:54Delegating local control
- 10:58Future work
- 11:39Q&A
related talks
05mSimplifying access to OpenStreetMap data for urban planning
21mAlgorithmic Planning, Simulation and Validation of Smart, Shared Parking Services using Edge HW
22m"How We Built a Rust-y Real-Time Public Transport Map"
01h 05mGraph Algorithms in the MPC Model
15mTowards automated mapping of pedestrian path networks at scale
22mToward a Live, Rich, Composable, and Collaborative Planetary Compute Engine
18mA Planet-Scale Computer – Abstract Away Regions via Global Service Placer (GSP)
41mDeclarative, Convergent Edge Computation