I really hate to say this:  This just doesn't work as it stands.  I think I can see where it's going, I appreciate the effort, but right now, it just doesn't work.  But at least it's concrete text and we can work with it.

First, it loses one of the main points of CR 15, namely that using anonymous for a response endpoint in a request means to send back the response in the response message of the same message exchange.  This seemed a bit pedantic the first time round, more a belt-and-braces thing, but given that AcksTo is meant to mean the response message of other message exchanges (whichever they may be), the distinction seems crucial.  If I use anonymous in [reply endpoint], I don't want to get a 202 back and get the actual reply in response to some later message.  This may seem clear now, but once WS-RX is deployed the line will blur and later implementors may well either have questions or (worse) scratch their heads and say "Hmm, that seems odd, but the spec says it's OK and we should be consistent with acks."

However, this is easily fixed by sharpening "sending it as the response message of the MEP" to "sending it as the response message of the same message exchange as the request," as we did in resolving CR 15.

Second, this text no longer says that using anonymous response endpoints means use the SOAP 1.1/HTTP binding unchanged, but this probably follows from the text as it stands.  If not, the text from the resolution of CR 15 can be re-inserted.

My larger concern is how one would build on this.  I may use anonymous as the address of any EPR.  What this means depends on the MEP and definition of the EPR in question (i.e., whether it's [destination], [reply endpoint], AcksTo or whatever, IIUC).  The SOAP request-response MEP defines a channel for response messages.

Is this meant to say that, in the context of a SOAP request-response MEP, use of anonymous MUST refer to use of the response message, or that it MAY refer to it?  If it's MUST, this disallows the proposed use for cases where a TCP connection is kept open.  If it's MAY, then referring to anonymous doesn't guarantee use of the back channel.  If it's neither, just exactly what are we saying?

If we're trying to say that, in the context of request-response, anonymous always and only refers to the response message of request-response, then fine, but then let's say it in as many words.

Looking back at CR 4 and CR 15, it appears that CR4 was proposed as a slight tweak to the text on anonymous, while CR 15 was the result of a more lengthy review and rewrite.  As such, and given that CR 15 was adopted more recently, I believe it would be better to hew more closely to that text.  Assuming that we want to state that anonymous MUST [or MAY] refer to the response channel (or MAY do), I believe we can amend the resolution to CR 15 as follows.

For those reading this in HTML format, additions are in green and deletions are in strikethrough.  The text inserted at the beginning is based on, but not identical, the text proposed for the action item.

5.1 Use of Anonymous Address in SOAP

The "http://www.w3.org/@@@@/@@/addressing/anonymous" URI MAY be specified as the [address] of an EPR.  The exact meaning of this depends on the underlying SOAP protocol binding, the Message Exchange Pattern (MEP) in use and on whether the URI occurs in a response endpoint, as the [destination] property, or in some other context.

5.1.1 SOAP 1.1/HTTP

When "http://www.w3.org/@@@@/@@/addressing/anonymous" is used in the context of the SOAP 1.1/HTTP binding, it MUST [or MAY] refer to the use of the HTTP response message.  When in particular it is specified for the response endpoint then there is no change to the SOAP 1.1/ HTTP binding.

5.1.2 SOAP 1.2

When "http://www.w3.org/@@@@/@@/addressing/anonymous" is used in the context of  specified for the response endpoint and the request is the request part of a SOAP request-response MEP [SOAP 1.2 Part 2: Adjuncts], then it MUST [or MAY] refer to the use of the response message of the exchange.  When in particular it is used in a response endpoint in a request message, any response MUST be the response part of the same SOAP request-response message exchange [SOAP 1.2 Part 2: Adjuncts].

Since we're here, here's how the resolution of CR 18 would work, as a delta to the above.

5.1 Use of Anonymous Address in SOAP

The "http://www.w3.org/@@@@/@@/addressing/anonymous" URI MAY be specified as the [address] of an EPR.  The exact meaning of this depends on the underlying SOAP protocol binding, the Message Exchange Pattern (MEP) in use and on whether the URI occurs in a response endpoint, as the [destination] property, or in some other context.  Except as defined in this section, this binding assigns no particular semantics to the use of "http://www.w3.org/@@@@/@@/addressing/anonymous".

When an anonymous response endpoint is used for a response, the rules in section 3.4 of the WS-Addressing Core dictate that the [destination] property of the response MUST also be "http://www.w3.org/@@@@/@@/addressing/anonymous". In this case, the anonymous address refers to the use of the underlying response message.

5.1.1 SOAP 1.1/HTTP

When "http://www.w3.org/@@@@/@@/addressing/anonymous" is used in the context of the SOAP 1.1/HTTP binding, it MUST [or MAY] refer to the use of the HTTP response message.  When in particular it is specified for the response endpoint then there is no change to the SOAP 1.1/ HTTP binding.

5.1.2 SOAP 1.2

When "http://www.w3.org/@@@@/@@/addressing/anonymous" is used in the context of  a SOAP request-response MEP [SOAP 1.2 Part 2: Adjuncts], then it MUST [or MAY] refer to the use of the response message of the exchange.  When in particular it is used in a response endpoint in a request message, any response MUST be the response part of the same SOAP request-response message exchange [SOAP 1.2 Part 2: Adjuncts].

Katy Warr wrote:

Further to my action on Monday to propose resolution text for CR23, here is a suggestion.  I've tried to maintain the previously agreed text from CR4 and CR15 where possible.  

The intention is that the first paragraph enables the use of anonymous EPR in semantics including (but not exclusive to) replyTo and faultTo (such as WS-RX acksTo).  This satisfies CR4.

The second paragraph gives the semantics specific to replyTo/faultTo - this is very similar to the resolution of CR15 but specific to the replyTo/faultTo scenario so that it does not undo the resolution to CR4 in the first paragraph.

This text is applicable to section 5.1 of the SOAP binding document.

-------------------------
The "http://www.w3.org/@@@@/@@/addressing/anonymous" URI MAY be specified as
the [address] of an EPR to designate that the target endpoint is
reached by a channel of the underlying SOAP protocol binding.
The specification of the channel to which the
"http://www.w3.org/@@@@/@@/addressing/anonymous" URI refers depends on the
Message Exchange Pattern (MEP) and on the
defined semantics of the EPR in question.  Any underlying protocol binding supporting the SOAP
request-response MEP provides such a channel for
response messages.
                                                 
In the context of a SOAP request-response MEP, sending a response message to a
ReplyTo or FaultTo
EPR whose [address] is "http://www.w3.org/@@@@/@@/addressing/anonymous"
means sending it as the response message of the MEP.
For instance, the SOAP 1.2 HTTP binding[SOAP 1.2 Part 2: Adjuncts]
puts the reply message in the HTTP response.
--------------------------

We may be able to omit the sentence:  " Any underlying protocol binding supporting the SOAP
request-response MEP provides such a channel for
response messages. "

This second paragraph requires restructure to fit with the SOAP 1.1/HTTP and SOAP 1.2 split in the editors'
draft.  I suggest that we defer this until we have reached agreement on the general structure of the text above.

thanks
Katy