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

Re: issue 168 proposal: xsi:type of external references in Encoding

From: Jacek Kopecky <jacek@systinet.com>
Date: Wed, 12 Dec 2001 17:24:34 +0100 (CET)
To: Jean-Jacques Moreau <moreau@crf.canon.fr>
cc: "John J. Barton" <John_Barton@hpl.hp.com>, xml-dist-app <xml-dist-app@w3.org>
Message-ID: <Pine.LNX.4.33.0112121722260.6731-100000@mail.idoox.com>
 Jean-Jacques,
 can you show me a scenario where you need to know in
SOAP-Encoding-encoded data the type before trying to dereference?
Anyway, the "other means" can include guessing from the URL. 8-)
 I'm strongly against saying "the application decides whether or
not to dereference a particular SOAP Encoding href".
 Best regards,

                   Jacek Kopecky

                   Senior Architect, Systinet (formerly Idoox)
                   http://www.systinet.com/



On Wed, 12 Dec 2001, Jean-Jacques Moreau wrote:

 > Jacek,
 >
 > I am not so much concerned by the local href, but by the external one, and would
 > prefer not to have to derefence the (external) href to find out the type of the
 > resource at the other end.
 >
 > Jean-Jacques.
 >
 > Jacek Kopecky wrote:
 >
 > >  Jean-Jacques,
 > >  again, what would you do with the following?
 > >    <a id="1"    xsi:type="xsd:string"  >42</a>
 > >    <b href="#1" enc:hrefType="xsd:ing"/>
 > >  I'm opposed to enc:hrefType attribute because it puts typing
 > > information on the accessor, whereas SOAP Encoding types values,
 > > not accessors.
 > >
 > >  I believe the current Encoding tries to comply to a
 > > contradictory set of requirements using one mechanism:
 > >
 > >  1) being able to share data and therefore enabling circular
 > > graph structures,
 > >  2) being able to point to anything out there,
 > >  3) typing all values.
 > >
 > >  We can remove the requirement 2) or 3), the former IMO results
 > > in a cleaner and nicer beast. Or we can try other paths, like
 > > adding other linking mechanism (like gref which was I think
 > > proposed at the f2f), or like the following one that I just
 > > thought about:
 > >  "The href attribute can in fact point to anything at all and
 > > it's the pointee's responsibility to provide it's type
 > > information. (In case of XML, type can be gotten from a schema or
 > > from the xsi:type attribute inf. item, in other cases by other
 > > means.)"
 > >  IMO this simple text, if present somewhere in the Encoding spec,
 > > would close issue 168 (relying on third-party specifications or
 > > implementors' practices for the "other cases" as SOAP does
 > > already in places like ordering of header processing, for
 > > example).
 > >
 > >                    Jacek Kopecky
 > >
 > >                    Senior Architect, Systinet (formerly Idoox)
 > >                    http://www.systinet.com/
 > >
 > > On Wed, 12 Dec 2001, Jean-Jacques Moreau wrote:
 > >
 > >  > Congratulations for finding out what the "S" now stands for in "SOAP"! :)
 > >  >
 > >  > On a more serious note, I think Jacek was ruling this out in his proposal [1]
 > >  > (ante-penultimate paragraph), the reason being it is "too complicated".
 > >  >
 > >  > Personnally, I would prefer us to say instead that "the type of the referenced
 > >  > entity MAY be indicated by the enc:hrefType attribute".
 > >  >
 > >  > Jean-Jacques.
 > >  >
 > >  > [1] http://lists.w3.org/Archives/Public/xml-dist-app/2001Dec/0129.html
 > >  >
 > >  >
 > >  > "John J. Barton" wrote:
 > >  >
 > >  > > At 07:48 PM 12/11/2001 -0500, Mark Baker wrote:
 > >  > > >An enthusiastic +1
 > >  > > >
 > >  > > > > I agree and on a slightly more serious note, I would from a general Web
 > >  > > > > architecture point of view suggest that we not go down this particular
 > >  > > > > slippery slope.
 > >  > >
 > >  > > Somewhere in this thread I lost with whether or not
 > >  > > my SOAP application would be able to predict the
 > >  > > datatype of an external reference without accessing
 > >  > > the resource.  I gather that the answer is "not through
 > >  > > SOAP".  I can guess from the URL. I can make an extra
 > >  > > element. But the system doesn't say.  Its Smalltalk,
 > >  > > not Pascal.  (I'm ok with that, just unclear).
 > >  > >
 > >  > > ______________________________________________________
 > >  > > John J. Barton          email:  John_Barton@hpl.hp.com
 > >  > > http://www.hpl.hp.com/personal/John_Barton/index.htm
 > >  > > MS 1U-17  Hewlett-Packard Labs
 > >  > > 1501 Page Mill Road              phone: (650)-236-2888
 > >  > > Palo Alto CA  94304-1126         FAX:   (650)-857-5100
 > >  >
 >
Received on Wednesday, 12 December 2001 11:24:42 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 22:28:13 UTC