Current state of UIs
• Extremely complex
• Full of async events, client/server communication
• UIs are always composed based on a state
• State is hard to reason about
Slide 3
Slide 3 text
Imperative
programming
Slide 4
Slide 4 text
No content
Slide 5
Slide 5 text
User action UI Responds User action …..
Uses imperative programming
Uses imperative programming
UI Responds
time (s)
Slide 6
Slide 6 text
Caveats of IP
• Isolates state from time
• You don’t see the bigger picture
• Mutable values
Slide 7
Slide 7 text
Is there a way … ?
Slide 8
Slide 8 text
Reactive
programming
Slide 9
Slide 9 text
– Andre Staltz
“Reactive programming is programming with
asynchronous data streams.”
Slide 10
Slide 10 text
Streams
• Sequence of ongoing events ordered in time
• Cheap, ubiquitous
• Allow you to react to new data
• Think event busses, typical click events, etc.
Slide 11
Slide 11 text
But in RP …
Slide 12
Slide 12 text
No content
Slide 13
Slide 13 text
Observable/Observer
• Observables are push-based, observable
sequences that can send data to anyone who is
interested
• Observers are entities who register an interest
through a subscription to an Observable
Slide 14
Slide 14 text
Observers
• Register via subscribe method
• You get back a Disposable object (a handler for the
subscription)
• onNext, onError, onCompleted
Slide 15
Slide 15 text
Demo time with RxJS
Slide 16
Slide 16 text
When to use RxJS
• Orchestrate event-based computations
• Deal with async data
• Complex UIs with lots of user interaction