I'd phrase it slightly differently to differentiate the types of
protocols (akin to differentiating the types of bindings...)
WS-RX is most definitely defining a protocol. It has a state machine,
etc. But it is not defining an underlying SOAP protocol binding as you
say. The WS-A use of protocol is in that binding of SOAP to what SOAP
calls an underlying protocol. BTW, that's one reason why many of us
SOAP-heads keep saying "don't call it a transport, call it an underlying
protocol". The use of the term "underlying" is crucial to differentiate
from "overlaying" protocols like RX, Application protocols, etc. The
WS-RX protocol is a protocol encoded in SOAP headers and body. No more
or less than any other application protocol like submitPO, getPOStatus,
cancelPO.
Cheers,
Dave
> -----Original Message-----
> From: Richard Salz [mailto:]
> Sent: Tuesday, March 14, 2006 7:35 PM
> To: Marc Goodner
> Cc:
> Subject: RE: [ws-rx] AnonURI and Offer and WS-Addressing
>
> > WS-A provides the latitude to define
> > what anonymous means when a given protocol is in the context of an
> > EPR. I suggest that we do so if we need to.
>
> When WS-Addressing talks about a protocol, they mean the "underlying
SOAP
> protocol binding," such as SOAP over HTTP.
>
> WS-RX isn't defining a protocol, therefore it cannot change the
meaning of
> anonymous.
>
> /r$
>
> --
> SOA Appliance Group
> IBM Application Integration Middleware
