W3C home > Mailing lists > Public > public-ws-addressing@w3.org > November 2004

Re: NEW Issue - Reference to WSDL definition in an EPR [i033]

From: Mark Nottingham <mark.nottingham@bea.com>
Date: Wed, 10 Nov 2004 21:06:17 -0800
Message-Id: <6B378BCB-339F-11D9-9807-000A95BD86C0@bea.com>
Cc: "<public-ws-addressing@w3.org>" <public-ws-addressing@w3.org>, Stephen Vinoski <Steve.Vinoski@iona.com>, Eric Newcomer <Eric.Newcomer@iona.com>
To: Rebecca Bergersen <Rebecca.Bergersen@iona.com>, Anish Karmarkar <Anish.Karmarkar@oracle.com>

OK. This is now i033;

> The focus of Issue 27 is whether the existing PortType and ServiceName 
> elements of an EPR are meaningful.  It also suggests that if the 
> consumer has the WSDL, the PortType and ServiceName elements are not 
> needed.  The focus of the issue revolves around the PortType and 
> ServiceName elements.  These are good comments and they make a worthy 
> issue, but it is not the issue I have defined.  A possible resolution 
> to issue 27 would be to remove the portType and serviceName elements 
> as being meaningless (and I do NOT advocate such a resolution!)

I didn't read that into 27 (the references to portType and 
serviceName's utility seemed to me to be more rhetorical motivation for 
defining such a property), but I can see that it could be read 
otherwise, and that the resolution could take a different path.

Anish, please note that a possible resolution to i027 is also the 
proposed resolution to i033; this might save you some work.

> Additionally, I would like to own the issue I have been diligently 
> pursuing.  I believe both issues are valuable and should be addressed 
> independently.

Since you've expressed interest, I've already assigned it to you.

Regards and thanks for your patience,

>> -----Original Message-----
>> From: Mark Nottingham [mailto:mark.nottingham@bea.com]
>> Sent: Wednesday, November 10, 2004 9:01 PM
>> To: Bergersen, Rebecca
>> Cc: Vinoski, Stephen; public-ws-addressing@w3.org; Newcomer, Eric
>> Subject: Re: NEW Issue - Reference to WSDL definition in an EPR
>> Rebecca,
>> Issue 23 became a discussion of optionality in EPRs because that was
>> the most well-defined aspect of it discussed at the F2F; your action
>> item was specifically intended to assure that the other
>> aspects of the
>> original issue, as you saw them, were captured.
>> Your proposed issue did not "disappear"; I asked for clarification, in
>> http://lists.w3.org/Archives/Public/public-ws-addressing/2004Nov/
>> 0109.html
>> Thank you for providing that clarification, and for capturing the
>> remainder of your original issue.
>> Can you differentiate what you describe below from the
>> existing issue
>> 27?
>>    http://www.w3.org/2002/ws/addr/wd-issues/#i027
>> What's there *appears* to be the same issue that you're raising.
>> If it is essentially the same issue, I'll augment the existing issue
>> with the information and proposal you've provided, rather
>> than open a
>> duplicate issue knowingly.
>> Thanks,
>> On Nov 10, 2004, at 4:31 PM, Bergersen, Rebecca wrote:
>>> In case this issue looks familiar to the twenty or so people who
>>> attended the NYC face-to-face, it should. This "new" issue is a
>>> restatement of an issue that was defined at the second day of that
>>> face-to-face meeting in New York; it was discussed for two
>> hours on
>>> the third day of that meeting and discussion was continued to the
>>> teleconference.  However, when the teleconference occured,
>> the issue
>>> had been framed as the optionality of metadata - certainly
>> a point of
>>> view on a link to a WSDL service definition, but not the
>> actual topic
>>> of the issue defined at the face-to-face.  However, at the
>>> teleconference I was given the action item to redefine both
>> this issue
>>> and the multiple ports issue.  I did that, publishing both
>> issues the
>>> following morning.  The ports issue made it to the issue
>> list, but the
>>> WSDL reference issue disappeared.  Instead an issue dealing with
>>> WSDL:location that was submitted later by another
>> individual appeared.
>>> This reference to WSDL definition in an EPR is a
>> restatement of the
>>> issue in the formal manner that was defined after the sequence of
>>> events described above.  Please discuss this issue based on
>> the formal
>>> definition I have presented.
>>> With respect,
>>> Rebecca Bergersen
>>> Principal Architect, Middleware Standards
>>> rebecca.bergersen@iona.com
>>> -------------------------------------------------------
>>> IONA Technologies
>>> 200 West Street Waltham, MA 02451 USA
>>> Tel: (781) 902-8265
>>> Fax: (781) 902-8001
>>> -------------------------------------------------------
>>> Making Software Work Together TM
>>> -----Original Message-----
>>> From: Bergersen, Rebecca
>>>  Sent: Wednesday, November 10, 2004 7:11 PM
>>> To: 'public-ws-addressing@w3.org'
>>> Cc: Vinoski, Stephen; Newcomer, Eric; Bergersen, Rebecca
>>> Subject: NEW Issue - Reference to WSDL definition in an EPR
>>> Title: Reference to WSDL definition in an EPR
>>> Description:  According to the ws-addressing submission, "Endpoint
>>> references
>>>   convey the information needed to identify/reference a Web service
>>>   endpoint, and may be used in several different ways: endpoint
>>>   references are suitable for conveying the information needed to
>>>   access a Web service endpoint...."  However, in order to
>> assure that
>>>   the information needed to access a Web service endpoint,
>> a reference
>>>   to the WSDL definition of a service is sometimes required and in
>>>   those cases must be included as part of the EPR construct.
>>> Justification: This requirement derives from several common use
>>>   cases. For example, in a communication chain there may be
>>>   intermediaries that can accept incoming messages and, in a fully
>>>   dynamic manner, further dispatch or route those onward.
>> This is what
>>>   we do with our products.  The trick is that the next
>> recipient might
>>>   use a completely different protocol/transport/format than what the
>>>   message came in on. For this case it is necessary to
>> perform a fully
>>>   dynamic dispatch by using the target's WSDL definition
>> and to build
>>>   dynamic proxies and to bind to the service over one of the
>>>   protocol/transport/format combinations it supports. The whole
>>>   definition is required so there is access to all the possible
>>> bindings
>>>   for the service. The WSDL definition is also used in cases where
>>>   consumer applications want to avoid compiling in static port type
>>>   information, and instead want, for flexibility purposes, late
>>>   (runtime) binding to the service.  
>>> Target: Core
>>> Proposal:
>>>   1. Extend section 2.1, Information Model for Endpoint
>> References, to
>>>   include the following:
>>>     [definition] : URI (0..1)
>>>       The optional element that provides an link to the WSDL service
>>>       definition.
>>>   2. Extend section 2.2, Endpoint Reference XML Infoset
>>>      Representation, to include the following:
>>>      Example 2-1. @@@
>>>        <wsa:EndpointReference>
>>>           ...
>>>           <wsdl:serviceDefinition>xs:anyURI</wsdl:serviceDefinition>
>>>           ...
>>>        </wsa:EndpointReference>
>>>        and to include the following as a description of the
>> additional
>>>       information:
>>>       /wsdl:serviceDefinition
>>>       This optional element provides a link to the WSDL service
>>>       definition.
>> --
>> Mark Nottingham   Principal Technologist
>> Office of the CTO   BEA Systems

Mark Nottingham   Principal Technologist
Office of the CTO   BEA Systems
Received on Thursday, 11 November 2004 05:06:27 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:28:21 UTC