If you want the failure-mode and testing path through the catalog, start here. Recommended route: Service Worker Failure Modes in Offline-First PWAs Rollback Patterns in Offline-First PWAs Testing IndexedDB Schema Migrations in Offline-First PWAs Offline Queue Replay and Idempotency in Offline-First PWAs If you want privacy-first, offline health tech to exist without surveillance funding it: sponsor the build → https://paintracker.ca/sponsor If you want the privacy boundary underneath those failures, add: Trust Boundaries in Client-Side Health Apps Service workers look elegant on paper. They make offline apps feel solid. They cache the shell, keep the interface alive, and let the app keep moving when the network starts falling apart. In diagrams, they look like resilience. In real life, they can behave more like a trapdoor. Because once an offline-first app starts updating in the wild, you are no longer dealing with a clean release path. You are dealing with timing. Old tabs. Partial installs. Stale assets.…