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

Issue 146

From: Williams, Stuart <skw@hplb.hpl.hp.com>
Date: Mon, 5 Nov 2001 14:15:15 -0000
Message-ID: <5E13A1874524D411A876006008CD059F192727@0-mail-1.hpl.hp.com>
To: "'xml-dist-app@w3.org'" <xml-dist-app@w3.org>
Cc: fallside@us.ibm.com, "'Noah_Mendelsohn@lotus.com'" <Noah_Mendelsohn@lotus.com>
I have been asked by the the XML protocol WG to develop a clear articulation
of of Issue 146 [1].

Issue 146 arose during the closure of Issue 140 on a WG telcon [2] during to
extended discussion on an Issue 140 thread [3].

Issue 140 was about the basis upon which a SOAP Node makes a determination
that it acts in a given actor role with respect to a message that it has
received.

Extended discussion on the thread at [3], starting with [4], introduced a
new Issue, Issue 146, about whether the default/anon actor may relay a SOAP
message further along a SOAP message path. The following is an attempt to
'briefly' state the issue of Issue 146.

"The terms 'default actor', 'anonymous actor', 'ultimate recipient' and
'SOAP Endpoint' are being used loosely as synonyms. It is not clear whether
a SOAP Node acting as default actor, anon actor, ultimate recipient, or
endpoint with respect to a given SOAP message may behave as SOAP
intermediary and relay the SOAP message to further SOAP Nodes. Stated
differently: Does a SOAP Message Path always terminate at the default actor,
anonymous actor, ultimate recipient or SOAP endpoint?"

Before, substantively discussing the issue, I would first ask for
confirmation that the above sufficiently describes the issue. 
Best regards

Stuart Williams.

[1] http://www.w3.org/2000/xp/Group/xmlp-issues.html#x146
[2] http://www.w3.org/2000/xp/Group/1/10/10-minutes.html
[3] http://lists.w3.org/Archives/Public/xml-dist-app/2001Oct/0021.html
[4] http://lists.w3.org/Archives/Public/xml-dist-app/2001Oct/0057.html

--
[Exchange with Noah seeking to clarify the statement of Issue 146]

> -----Original Message-----
> From: Noah_Mendelsohn@lotus.com [mailto:Noah_Mendelsohn@lotus.com]
> Sent: 02 November 2001 03:36
> To: Williams, Stuart
> Cc: fallside@us.ibm.com
> Subject: Re: FW: HELP: Clarification of Issue 146
> 
> 
> Well, my recollection on this is a little hazy, and I'm not sure that my 
> views agree with everyone else's.  That said:
> 
> My note [1] suggested:
> 
> "We already have text, I believe (I'm on an airplane and can't easily 
> check) that makes clear that nodes acting as the anonymous actor cannot 
> further relay a message, and in that sense serve as an endpoint. I would 
> fully expect that the request/response MEP, when specified, would indicate

> that responses typically originate from the node that acted in the 
> anonymous role for the request."
> 
> In fact, the specification says in section 2.2:
> 
> "A SOAP node can establish itself as the ultimate SOAP receiver by acting 
> in the (additional) role of the anonymous SOAP actor."
> 
> and in 2.5:
> 
> "If the SOAP node is a SOAP intermediary, the SOAP message pattern and 
> results of processing (e.g. no fault generated) MAY require that the SOAP 
> message be sent further along the SOAP message path."
> 
> I think the essence of the issue to be raised is my concern that we have 
> introduced the term "endpoint", we have strongly implied that messages 
> should be relayed only in the case that the node is an intermediary, but 
> we have never come out and said what it means to be an endpoint and 
> whether indeed endpoints should not further relay a message.  My 
> preference would be that not being allowed to relay the message is the 
> defining characteristic of being an "endpoint".  Apparently there is 
> disagreement, from Henrik perhaps among others, I think. Anyway, if the 
> node acting in the role of the anonymous actor is indeed allowed to relay 
> messages, and I would think we should change the name from "endpoint" to 
> something like "main target".  In that case, we would have to answer the 
> question of whether more than one node can act as such an "main target" or

> not, and we're responses should originate in the request response pattern 
> (at the anonymous actor or potentially at whichever node finally declines 
> to relay the request message)?  As it stands, I think we are emphasizing 
> the term "endpoint" without clearly saying what it means in practice.  I 
> think that's the issue that should be 146.
> 
> You are welcome to relay this message to either of our mailing lists if 
> that would be helpful.
> 
> [1] http://lists.w3.org/Archives/Public/xml-dist-app/2001Oct/0053.html
> 
> ------------------------------------------------------------------------
> Noah Mendelsohn                                    Voice: 
> 1-617-693-4036
> Lotus Development Corp.                            Fax: 1-617-693-8676
> One Rogers Street
> Cambridge, MA 02142
> ------------------------------------------------------------------------
> 
> -----Original Message-----
> From: Williams, Stuart 
> Sent: 24 October 2001 21:50
> To: Noah Mendelsohn (E-mail)
> Cc: David Fallside (E-mail)
> Subject: HELP: Clarification of Issue 146
> Importance: High
> 
> 
> Hi Noah,
> 
> Issue 146 came up on todays telcon. Unfortunately I don't  think that when
> Issue 140 (self determination by a node that it perform the role of the
> default actor) got split into 140 and 146 I don't think Issue 146 has been
> crisply captured. Currently it is described as "How does an actor know
that
> it is the ultimate recipient?" which looks to me remarkably like Issue 140
> which we closed.
> 
> I have been actioned to produce a crisp description of the Issue 146 ASAP.
> 
> IIRC Issue 146 really arises from the discussion of whether the
default/anon
> actor in fact terminates the message path. If this indeed a correct
> recollection, then I think Issue 146 is about a behavioural constraint on
> the anon/default actor in respect to relaying SOAP Messages further along 
> a SOAP Message Path (which personnally I think is an oxymoron if the
> default/anon actor is also described as the ultimate recipient).
> 
> Best regards
> 
> Stuart
Received on Monday, 5 November 2001 09:18:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:04 GMT