Re: [hybi] Fwd: The MessageBroker WebSocket Subprotocol

Hi Mark, Clebert,

What I *really* like about the proposal is that unlike other messaging 
protocols it takes care to cleanly separate the reliability from the 
broker behaviour. The latter is explicitly defined as out-of-scope.

Some comments follow:

1. Given the explicit focus on reliability, it's strange that the 
proposal deals with message metadata, which have to do with broker or 
even application behaviour and have absolutely nothing to do with 
reliability. What ensues is a spec where metadata are defined as opaque 
syntactic placeholders with no associated semantics. The semantics are 
meant to be defined on the broker or application level. The question is 
whether it doesn't follow that the syntax should be defined on those 
layers as well.

2. AFAICS there's nothing in the spec that presumes existence of the 
broker. It can be as well used for direct communication between 
applications. Thus, I would suggest replacing messaging broker/messaging 
client terminology with WebSocket server and WebSocket client wording.

3. As for reliability itself, it should be clear what kind of error 
conditions is the protocol meant to handle. Possible options:

a. Network failure. If so, how does it differ from simply turning off 
TCP keepalives?

b. Client failure and restart?

c. Server failure and restart?

Martin

On 10/02/2011 10:40 AM, Salvatore Loreto wrote:
> for some reason the mail below didn't go through the HyBi mailing list,
> so I am forwarding it (and of course working on try to figure out the
> reason of why it didn't go)
>
> cheers
> /Sal
>
> -------- Original Message --------
> Subject: 	The MessageBroker WebSocket Subprotocol
> Date: 	Sun, 2 Oct 2011 00:33:52 +0200
> From: 	Hapner mark <hapner.mark@huawei.com>
> To: 	hybi@ietf.org <hybi@ietf.org>
> CC: 	csuconic@redhat.com <csuconic@redhat.com>, Salvatore Loreto
> <salvatore.loreto@ericsson.com>
>
>
>
> Hi,
>
> Let me introduce myself since I have not posted to the list before.I've
> been involved with message brokers for some time. I was the engineer
> behind the creation of the Java Message Service API. I current work for
> Huawei's US Software Lab in Santa Clara, CA.
>
> Clebert is the JBoss HornetQ lead.
>
> Clebert and I are excited about the potential to bring the functionality
> of message brokers to the internet and we see WebSocket as the best
> message transport to enable this. We believe the WebSocket subprotocol
> described in the attached document provides the minimal extension to
> WebSocket needed for this purpose. We hope this subprotocol reflects
> well on the expectations the WG has for the use of WebSocket extensibility.
>
> We are submitting this document informally to the WG for its comment. It
> is our hope that it will mature into a subprotocol that can be submitted
> for either inclusion into the WebSocket specification; or, as a separate
> RFC. We would like to get the WG's thoughts on how best to proceed in
> this direction.
>
> http://www.ietf.org/id/draft-hapner-hybi-messagebroker-subprotocol-00.txt
>
> Mark Hapner
> Clebert Suconic
>
>
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi