W3C home > Mailing lists > Public > public-ws-addressing@w3.org > March 2005

Re: wsa:Action in responses

From: Conor P. Cahill <concahill@aol.com>
Date: Fri, 18 Mar 2005 07:30:36 -0500
To: "Hugo Haas" <hugo@w3.org>
cc: "Nilo Mitra (TX/EUS)" <nilo.mitra@ericsson.com>, WS-Addressing <public-ws-addressing@w3.org>
Message-ID: <423AC9EC.104@aol.com>



Hugo Haas wrote on 3/18/2005, 6:50 AM:

 > I interpret this as being opaque to everybody but the issuer of the
 > identifier. IOW, either your application knows about it and therefore
 > understands the carried meaning, or it doesn't, and it cannot guess.
 >
 > I am not sure what value it brings and whether it's really useful to
 > have this specified here, as it may be the source of confusion.

We were planning on using this for both routing and dispatching on
the invocation side (e.g. intermediaries can use this value to route
this request to the "real" endpoint within a cluster that is ready to
handle that type of request and within a "real" endpoint, the framework
layer can use this value to decide on which callback to invoke within
the service).

The main reason that we saw for having this in a header is that it
allows this work to be done without requiring such parties to interact
with the body of the message.

So, it can be opaque and useless for those who don't use it, but it
can be useful for those who want to be able to do this kind of stuff
without having to step into the body of the message.

Conor
Received on Friday, 18 March 2005 12:31:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:35:04 GMT