Mobile-First Architectures
summary
Argues that devices should be the primary store using Realm mobile database and an endpoint computing model: modern phones are “more powerful than a rack server,” so keeping data and business logic on-device yields instant responses, tolerance to offline, and far lower server bills. Cites Gmail offline search failure as a UX example, 60fps interactive demands, app types that work best (single-user and collaborative), and claims companies run products with 100 million users with only a tiny backend for sync and backup.
AI-written summary · about
chapters (18)
- 0:00Intro
- 1:28Request response pattern
- 2:15Benefits
- 3:40Bad user experience
- 5:51Connectivity
- 9:04Cost
- 10:40Problems
- 13:34Endpoint Processing
- 14:06Endpoint Computing
- 15:00Primary Store
- 16:00Mobile Phone
- 17:14Single User Apps
- 18:04Collaborative Apps
- 19:19Browser Apps
- 20:23Change of Mindset
- 23:51User Data
- 24:19Summary
- 27:11Collaboration
related talks
35mHow Offline-First Architecture Could Save Your Apps
34mOffline-First Mobile App Design
43mOffline-first Architecture - How to survive without the internet?
22mRapid Development on Offline First Data Sync
18mRethinking Apps with Local-First Architecture
43mLocal-First and Sync Engines, the Architecture of the Future?
32mA Tale Of Two Webs
24mGoing Local-First: A Primer