(React)ive State Management: Enter RxJS
summary
Claims that react is not reactive and shows two concrete approaches: use rxjs with react-rxjs bindings to create signal-like observables, bind them to components and use operators like shareReplay for unicast and takeWhile to cap updates (example stops at 9), but React still re-renders child subtrees; or use Legend State observable which uses proxies for fine-grain reactivity, yields single renders for subscribed components and benchmarks as fastest. Notes practical limits: debugging and unit testing rxjs in React remain tricky.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 3:16Why React isn’t truly reactive
- 5:44RxJS fundamentals
- 7:32Reactive UIs with React RxJS
- 10:48Fine-grained reactivity with Legend State
- 13:42Legend State vs Solid JS signals
- 17:19React vs fine-grained reactivity
- 18:24Q&A
- 20:13Q&A
- 21:21Q&A
- 22:29Q&A
- 23:55Q&A
- 24:58Q&A
- 27:11Q&A
- 27:51Q&A
- 29:15Q&A
- 30:21Closing and Q&A wrap-up
related talks
35mManaging State in React Apps with RxJS
36mRxJS: A Better Way to Write Frontend Applications
27mWhy React is not Reactive
36mModern React and the case for Reactive State Management
40mReactive Programming: A Better Way To Write Frontend Applications
27mEmbrace Reactive Programming in React and Vue with RxJS
49mRedux, beyond React
21mThinking Reactively: Most Difficult