W3C home > Mailing lists > Public > public-ws-addressing@w3.org > February 2006

Re: WSA From

From: Francisco Curbera <curbera@us.ibm.com>
Date: Mon, 13 Feb 2006 12:24:53 -0500
To: John Kemp <john.kemp@nokia.com>
Cc: Christopher B Ferris <chrisfer@us.ibm.com>, "ext Mark Baker" <distobj@acm.org>, public-ws-addressing@w3.org, public-ws-addressing-request@w3.org
Message-ID: <OF01278804.BC733CC6-ON85257114.005F273B-85257114.005FA9B0@us.ibm.com>

WSA uses EPRs to support Web services message addressing. If a URI is all
you need for that, good for you, you just need to fill the address field in
the EPR. If you need additional information (session state for example,
interaction policies etc.) that is also supported by the EPR.

Paco



                                                                                                                                  
                      John Kemp                                                                                                   
                      <john.kemp@nokia.        To:       Francisco Curbera/Watson/IBM@IBMUS                                       
                      com>                     cc:       Christopher B Ferris/Waltham/IBM@IBMUS, "ext Mark Baker"                 
                                                <distobj@acm.org>, public-ws-addressing@w3.org,                                   
                      02/13/2006 10:05          public-ws-addressing-request@w3.org                                               
                      AM                       Subject:  Re: WSA From                                                             
                                                                                                                                  




On Feb 12, 2006, at 1:41 PM, ext Francisco Curbera wrote:

> WSA identifies parties sending and receiving messages using endopint
> addresses.

A *physical* endpoint address? Can I contact that endpoint? What can
I send it, and how do I know? Can I, the client, send multiple values
of wsa:To? If not, why not? I don't see answers to any of that
written anywhere in WS-A. If it's simply an "extension point" for
implementations or other specifications, then why not just allow EPRs
other than those defined in WS-A to be sent as extra items of
wsa:EndpointReferenceType (which, as I understand it, is already
possible)? Any usage of wsa:To as a physical endpoint is clearly not
apparent from the WS-Addressing specification itself.

If it is not a physical endpoint, then surely a URI would suffice.

- JohnK

>
> Paco
>
>
>
>
>                       John Kemp
>                       <john.kemp@nokia.com>           To:
> "ext Mark Baker" <distobj@acm.org>
>                       Sent by:                        cc:
> Christopher B Ferris/Waltham/IBM@IBMUS, public-ws-addressing@w3.org
>                       public-ws-addressing-req        Subject:  Re:
> WSA From
>                       uest@w3.org
>
>
>                       02/10/2006 04:56 AM
>
>
>
>
>
>
> On Feb 9, 2006, at 9:29 PM, ext Mark Baker wrote:
>
>
>>
>> I've seen this too.  HTTP "From" works similarly;
>>
>> http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.22
>>
>
> Quoted from the referenced link:
>
> "The From request-header field, if given, SHOULD contain an Internet
> e-mail address for the human user who controls the requesting user
> agent." [...]
>
> Clearly an identifier, not a physical endpoint.
>
> And:
>
> On 2/9/06, Christopher B Ferris <chrisfer@us.ibm.com> wrote:
>
>
>>
>>
>>>
>>> In many B2B scenarios with which I am familiar, the "From" is
>>> used to
>>> identify the party that
>>> sent the message. It is not intended to be some sort of physical
>>> endpoint
>>> (typically) but a logical
>>> identifier that serves to identify the party (e.g. http://
>>> www.ibm.com/)
>>>
>
> Indeed.
>
> So, shouldn't wsa:From be simply a URI, rather than an EPR? And
> having used such a syntax, shouldn't we imbue it also with the
> semantics of an identifier, in a manner similar to that of the above-
> referenced section of RFC2616?
>
> - JohnK
>
>
>
>
Received on Monday, 13 February 2006 17:25:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:35:11 GMT