Re: [hybi] Call for interest: multiplexing dedicated for WebSocket

[Tobias Oberstein <[e-mail]>]

> Sure. But I fail to see why you need MUX for your use case. One duplex
> connection seems enough to me.

Example: Say I want to write a single-page browser app that has both low-latency chatty traffic and mass-data traffic (e.g. chat + file transfer).

I want the chatty traffic maintain it's low-latency, while mass-data transfers are concurrently happening.

This could be done by muxing 2 logical WS over a physical WS with the first WS having high-prio vs the 2nd WS.

So a small auto-fragment size plus flow-control and scheduling would be helpful.

Doing this in JS seems suboptimal the least ..

> 
> > So I'd say a WS-MUX with flow-control (draft there) plus
> priorization/scheduling (not there) is "the last missing piece". Lets finish it.
> 
> The problem I see is that the 2 features you mention here are quite
> complicated to specify and implement correctly, i.e. we duplicate the larger

I agree .. those are tricky. FWIW I am willing to invest time and effort to extend Autobahn testsuite for WS extensions. We are working now on permessage-deflate.

> effort undertaken by HTTP/2.

It all depends on perspective, goals and use cases .. but I can follow your view .. guess Jetty wants to have both WS and SPDY.

/Tobias