Websockets & messaging: IETF draft

[Martin Sustrik]

http://www.ietf.org/id/draft-hapner-hybi-messagebroker-subprotocol-00.txt

Martin

[Paul Colomiets]

On Sun, Oct 2, 2011 at 3:17 PM, Martin Sustrik <sus...@250bpm.com> wrote:
> http://www.ietf.org/id/draft-hapner-hybi-messagebroker-subprotocol-00.txt
>
> Martin
>

It's remarkable that this protocol is doing things you explicitly
said you don't want to do in SP, namely connection recovery
and using extensions for labels.

--
Paul

[Martin Sustrik]

Yes. And actually, doing recovery -- if it'll be able to define it in
some consistent manner -- can possibly make it a good underlying
protocol for SP.

What we'll get would be nice separation of layers:

1. web socket extension does recovery
2. SP on top does routing

As for the labels, I still believe defining the syntax without
associated semantics is not a good idea and I have already said so at
the hybi list.

Martin