Clojure on the cyberpunk frontier of democracy
summary
Polis is an open-source platform that uses a machine-learning engine in Clojure to turn votes on freeform statements into a vote matrix, apply PCA + clustering to find consensus, and surface unexpected common ground. It starts as an R prototype, moves to JVM Clojure for concurrency, uses core.matrix (Neanderthal when needed), and Vega-Lite/Oz for visualizations. Deployed in Taiwan, it helps resolve policy fights — including regulation of Uber/Airbnb and breaking a four-year gridlock on online alcohol sales — despite early library issues (Incanter, Storm).
AI-written summary · about
chapters (21)
- 0:00Intro
- 0:50Obligatory etymology slide
- 1:14Democratic social behavior is ancient
- 2:02Athenian democracy
- 3:23Limitations
- 4:30Cybernetics
- 5:41Generalized abstract nonsense
- 6:07Digging deeper
- 8:37How can we use data science to make the opinion landscape transparent to everyone?
- 13:52Consensus is the secret sauce
- 14:19Sunflower Revolution
- 20:55Initial math engine
- 21:19Functional Programming
- 22:00The Downfall of Imperative Programming
- 24:46Data driven programming
- 26:20Deep thought in design
- 27:06Challenges
- 32:06System component alternatives
- 32:52Machine learning & stats
- 37:11Dataviz
- 40:26Closing case: Clojure for Data Science
related talks
28mScaling deliberation with data science and Clojure
27mHow to Increase Clojure Outreach
44mData Science in Clojure
29mPanel with Christophe Grand and Adam Helins
41mHelping voters with Pedestal, Datomic, Om and core.async
27mReal World AI Integration
40mClojure Emerging Data Ecosystem
40mBuilding Machine Learning Models with Clojure and Cortex