[Fwd: Opps... wrong URL]

"Williams, Stuart" wrote:

> Opps... sorry wrong URL in subject and body should have been:
>
> http://www.w3.org/TR/2000/WD-CCPP-vocab-20000721/
>
> -----Original Message-----
> From: Williams, Stuart
> Sent: 01 September 2000 15:11
> To: 'GK@ACM.ORG'
> Cc: 'Johan Hjelm'
> Subject: http://www.w3.org/TR/2000/WD-CCPP-ra-20000721/
>
> Hi Graham,
>
> For what it's worth I've taken a brisk read through
> <http://www.w3.org/TR/2000/WD-CCPP-ra-20000721/> and have a couple of
> comments. I'll probabbly be giving it a few more closer looks and will offer
> more feedback as it arises.
>
> General:
>
> 1) Upstream and Downstream directions for proxy chains. Is the direction
> defined and is it used consistently. Reading differents bits of text I was
> lead to different conclusions about which direction is which. An explicit
> statement would be helpful. Most of the time I think that downstream is from
> the origin server toward the device. If it's already there... I missed it.
>
> Section 3.3.1 (and 2.3.1)
>
> I find the notion that there are multiple (well two) ways in which represent
> set valued CC/PP attributes a little unfortunate. It would be ok if it were
> merely RDF syntactic differences that yielded the same RDF model. However,
> as far as I can tell this is not the case. The RDF model constructed by both
> forms will be different and therefore it will not be possible to test for
> equivalence simply by comparing the RDF models. Using the (simple) example
> in 3.3.1:
>
> <Component>--type-->[rdf:Bag] --rdf:_1--> "text/html"
>                               --rdf:_2--> "text/xml"
>
> and
>
> <Component>--type--> "text/html"
>            --type--> "text/xml"
>
> are not the same when subject to comparision as RDF and yet ARE the same
> when regarded as a set-valued CC/PP attribute.
>
> Testing the equivalence of two Bags in RDF I guess is already problematic,
> because ordering is (of the rdf:_?? attributes) is unimportant.
>
> >From the examples given later in the document it looks like there is a
> preference for the Bag form of encoding set valued CC/PP attributes (which
> would be my preference too). Also, reading through the RDFS scripts at the
> end of the document (torture) which I am not totally proficient with it
> looks like the schema's really only admit the single (Bag based) set value
> representation.
> regards
>
> Stuart Williams
> HP Labs

--

ERICSSON*RESEARCH*ERICSSON*RESEARCH*ERICSSON*RESEARCH*ERICSSON*RESEARCH

     Johan HJELM, Ericsson Research, T/K User Applications Group
                   johan.hjelm@era-t.ericsson.se
       GSM Mobile +46-708-820315 (works everywhere but in Japan)

                W3C Advisory Committee Representative
                      Chair CC/PP Working Group

                   Read more about my recent book
               Designing Wireless Information Services
                 http://www.wireless-information.net

      OPINIONS EXPRESSED ARE PERSONAL AND NOT THOSE OF ERICSSON

ERICSSON*RESEARCH*ERICSSON*RESEARCH*ERICSSON*RESEARCH*ERICSSSON*RESARCH