Piltri — Explore

Compare cities and countries on a weighted Piltri Score (Safety, Economy, Environment, Quality of Life) for ~140,000 places (every one with 1,000+ people) in 245 countries.

How it works

• Data is computed offline by pipeline/ from free bulk sources (GeoNames, World Bank, WHO, WorldClim, NASA POWER, satellite PM2.5, GHS-POP, Ookla, Overture Maps, Natural Earth, USGS) and published as a versioned bundle to Supabase Storage. See pipeline/README.md.
• The site (Next.js, Vercel) copies the current dataset into public/data/<version>/ at build time and serves it from its own CDN. Everything - city pages, search, pin mode, Advanced Search - reads those static files in the browser; there's no server-side data code and no third-party API call when a page loads.
• Maps: MapLibre GL + OpenFreeMap tiles (free, no key).
• Database: only the curated Resources links (lib/supabase/schema.sql).

Local setup

npm install
cp .env.example .env.local   # fill in the values - see DEPLOYMENT.md
npm run dev                  # downloads the current dataset first (predev)

Open http://localhost:3000.

Project structure

app/                  Pages (the public ones in app/(explore)/, addresses in
                      lib/urls.ts); app/api/ only has Resources links + /admin
components/ui/        Design system building blocks
components/explore/   Explore pages' components (map, panels, charts)
lib/dataset/          Dataset schema (shared with the pipeline), browser loaders,
                      score assembly, search, pin mode
lib/advancedSearch/   Criteria registry + in-browser search engine
lib/kpiRows.ts        What each section panel shows
pipeline/             Offline data pipeline (npm run pipeline)
scripts/              sync-dataset.mjs - installs the dataset before dev/build

See DEPLOYMENT.md for production setup and HANDOFF.md for history and design decisions.