KEYNOTE: Big Opportunities in Small Data
summary
Demo of an open-source tool that uses SQLite to publish and explore data shows a 1,700-row San Francisco assets CSV rendered on a map, interactive read-only SQL API queries that return CSV/JSON and bookmarkable URLs, and an automated newsletter pipeline built by SQL + a Observable notebook feeding Substack via copy‑paste that actually works. A browser build, Datasette Lite, runs entirely in WebAssembly; analysis of RedPajama’s urls.txt (2,048 URLs, 2.7 TB total) surfaces a 111 GB Wikipedia file.
AI-written summary · about
chapters (10)
- 0:00Introducing Simon
- 1:31Why talk about Small data?
- 2:23Demo of Datasette
- 8:03Read-only SQL queries via API
- 11:57Datasette built on SQLite
- 13:37Baked Data architectural pattern
- 16:30SQL + HTTP as integration tool
- 18:44Baked Data with PostgreSQL?
- 19:48Demo of Datasette in WebAssembly in browser
- 26:26Q&A
related talks
57mVisualization Tools for Data Discovery
41mVisualizing big data
40mArchitectures for Exploring Your Personal Data
11mData analytics in browser with AlaSQL.JS
30mBig Data and Small Data Everywhere, All at Once
27mUsing embedded SQL engines for plotting massive datasets on a laptop
43mKeynote
47mHow to get Open Data into the Hands of Activists