Computing through time

1950s

• Single-user
• Single-process
• Non-networked
• Completely synchronous
• Very technical users
• Default expectation is success

2015

• Multi-user
• Multi-process
• Highly-networked
• Very asynchronous
• Non-technical users
• Unable to assume success

Asynchrony

Why is it important?

Common problems with Asynchrony

• Memory Leaks
• Race Conditions
• Spaghetti Code
• Complex State Machines
• Callback Hell
• Error Handling

Asynchrony is hard

Let's think about it differently

Some definitions

• Immutable: Unable to be changed after creation
• Mutable: Alterable after creation
• Array: A collection of items in memory
• Event: A collection of items over time
• Functional Programming: a style of building the structure and elements of computer programs—that treats computation as the evaluation of mathematical functions and avoids changing-state and mutable data.
• Reactive Programming: a paradigm oriented around data flows and the propagation of change

Cause and effect

Proactive

Cause →         Effect

Reactive

Cause         → Effect

Cause and effect

Proactive

Cause →      (  Effect)

Reactive

Cause        (→ Effect)

Benefits of Reactive Programming

• Similarity between synchronous and asynchronous collections
• Rather than treating an HTTP request as different than a mouse drag, they can be seen as similar concepts.
• 70+ easy utilities for managing observables
• Mental manageability of complex events
• Built-in concept of disposability, leading to easy memory management with minimal effort
• Nice error handling with simple retry/throttle/abort semantics

An example

const searchResult$ = keyPress$
    .throttle(250)
    .map(event => event.target.value)
    .map(query => getJSON('/searchResults?q=' + query)
        .retry(3)
        .takeUntil(keyPress$))
    .switchLatest()