From Network Failures to Offline Success: A Journey of Visible App
summary
Visible is a symptom-tracking app for long COVID and chronic fatigue that scales from a few dozen to ~20k users and hits massive network noise. Initial Hasura + GraphQL subscriptions + Hive caching produce chatty websockets, battery drain and flaky DNS errors. The team pivots to an offline-first model using SQLite via Drift, local UUID local IDs plus a queued action/synchronization system with background sync (Workmanager/iOS) and optimistic updates, cutting crashes and server bills while keeping the UI snappy.
AI-written summary · about
chapters (13)
- 0:00Introduction and speaker background
- 1:48About Visible app and its purpose
- 3:53Initial development and technical debt
- 7:04Scaling challenges and network issues
- 18:19Shift to offline-first approach
- 21:31Choosing SQLite and local storage
- 23:17Local ID generation and data model
- 25:45Decoupling actions from network requests
- 28:56Demo of offline-first functionality
- 30:21Background synchronization process
- 32:31Mindset shift on offline states
- 33:32Results and benefits of the approach
- 35:37Q&A and closing remarks
related talks
41mOffline First
35mHow Offline-First Architecture Could Save Your Apps
39mSay Hello to Offline First
21mGo Offline First to Save The World
32mLook, Mum, No Internet - Creating great experiences with offline-first apps
29mGive Apps Online Superpowers by Optimizing them for Offline
22mSurviving the Offline Apocalypse
23mDesigning for Disconnection- The Mental Model of Offline Apps
