A Tale of Two Threading Models
summary
Compares two concrete threading patterns — copy-on-write immutable objects (persistent data structures, streams) versus transactional objects that expose immutable snapshots per event-loop cycle — and shows Realm’s implementation inside the new Kotlin Native SDK (Multiplatform) that supports background writes, frozen handoffs, fine‑grained observability and instant exceptions for cross-thread misuse. Demonstrates fewer race conditions, no full copies for large graphs, and production‑ready behavior via Realm’s open source GitHub.
AI-written summary · about
chapters (10)
- 0:00Introduction and background
- 3:24Why threading matters on mobile
- 7:18Race conditions and their dangers
- 13:17Traditional locks and their problems
- 18:32Immutability and functional reactive programming
- 26:01Problems with immutable streams
- 35:52Limitations of current models
- 36:59Transactional objects solution
- 43:24Summary and call to action
- 44:41Q&A and closing
related talks
49mWhat Mom Never Told You About Multi-Threading
30mExtra! Extra! Thread all about it!
42mAdvanced techniques for concurrency & memory management (en)
33mUntangle your App: Efficient Threading in Android
25mKotlin Multiplatform Mobile in production: key takeaways from developing two cross-platform apps
25mThe ever increasing convergence of native Android and iOS development
34mFunctional Android
28mThe ever increasing convergence of native iOS and Android mobile development