MediaObserver vs BreakpointObserver - what are the differences?

Question

After doing some searching, I came to realize that I'm supposed to use the Flex-Layout library to make Material-themed UI responsive (as answered here). And according to the documentation, this library provides the MediaObserver class to programmatically detect media query activation.

I used the material schematics command -

ng g @angular/material:navigation shell/layout

to add a Sidenav component, and noticed the following generated code -

export class LayoutComponent {
    isHandset$: Observable<boolean> = this.breakpointObserver.observe(Breakpoints.Handset).pipe(
        map(result => result.matches),
        shareReplay()
    );

    constructor(private breakpointObserver: BreakpointObserver) { } 
}

It is using the BreakpointObserver class and predefined Breakpoints from @angular/cdk/layout package to detect viewport changes. And the documentation says -

The layout package provides utilities to build responsive UIs that react to screen-size changes.

So, apparently, there are two different libraries available for making Material-themed UI responsive and they use different approaches. (Please correct me if I'm wrong).

I'd like to know -

• From usage perspective what are the differences between these two libraries/packages, that is, when to use which?
• Does MediaObserver uses BreakpointObserver under-the-hood?
• If I use Flex-Layout, should I avoid @angular/cdk/layout utilities altogether? Are there some best practices to follow in this regard?

Edit: 2020.05.11
According to the source code of MediaObserver, it doesn't use the BreakpointObserver under-the-hood. It uses the native Window.matchMedia() API instead.

Answer 1

From : https://github.com/angular/flex-layout/issues/685

@angular/cdk BreakpointObserver will not replace @angular/flex-layout MediaObserver. MediaObserver is an enhanced version that notifies subscribers of activations for standard AND overlapping (lt-xxx, gt-xxx) breakpoints.