W3C home > Mailing lists > Public > xml-dist-app@w3.org > November 2001

RE: Issue 146 proposed resolution

From: Nilo Mitra (EMX) <Nilo.Mitra@am1.ericsson.se>
Date: Thu, 15 Nov 2001 09:20:31 -0600
Message-ID: <C358DED30DFED41192E100508BB392278C3A3C@eamrcnt716.exu.ericsson.se>
To: "'Henrik Frystyk Nielsen'" <henrikn@microsoft.com>, Noah_Mendelsohn@lotus.com
Cc: skw@hplb.hpl.hp.com, xml-dist-app@w3.org
In your scenario below, is it possible for the front end to add something to the body before forwarding it to the back end? I ask because I unwittingly did an example for the Primer with just your architecture (shown below) but needed to add a body block in the "front end". (I can elaborate separately if you wish)

Nilo Mitra			
Ericsson Internet Applications Inc.
phone: +1 516-677-1073
mobile: +1 516-476-7427
email: nilo.mitra@ericsson.com

> -----Original Message-----
> From:	Henrik Frystyk Nielsen [SMTP:henrikn@microsoft.com]
> Sent:	Wednesday, November 14, 2001 5:39 PM
> To:	Noah_Mendelsohn@lotus.com
> Cc:	skw@hplb.hpl.hp.com; xml-dist-app@w3.org
> Subject:	RE: Issue 146 proposed resolution
> >The text you suggest doesn't seem to prevent a node playing 
> >the role of 
> >the anonymous actor from acting as an intermediary and relaying the 
> >message.
> Well, it is an attempt of indicating *what* it means to act in the role
> of the default/anonymous actor rather than *how* to do it. The reason
> for this is that *who* is the ultimate recipient depends on from what
> side one is looking.
> In the scenario that I brought up some time ago about a front-end server
> and a back-end server like this:
> 	sender  -->  front-end  --> back-end
> the *sender* believes that front-end it the ultimate destination. While
> this is true, the front-end has off-loaded the actual processing to the
> back-end. All three parties are SOAP nodes, it is just that the way the
> front-end has decided to process the message is to forward the message
> to the back-end server.
> In short, this model implicitly supports both intermediaries as well as
> gateways. I am concerned that if we don't allow this then we will have
> to define a SOAP node as an abstract entity that can contain multiple
> nested SOAP nodes which in my mind is much more complicated.
> Henrik
Received on Thursday, 15 November 2001 10:21:17 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 22:01:16 UTC