Re: LC-1773: CC/PP 2.0 : CC/PP Structure, section 3.3

On Fri, August 24, 2007 1:39 pm, stephane boyera wrote:
> Dear Ivan,
>
> Like the WG explained for the issue LC-1772, this specification is
> concerning CC/PP not RDF. Again, yes, there are CC/PP
> specific parsers(see LC-1772 [1]) for this stuff. We're documenting the
> existing reality, not where we would go if we were starting again. We're
> trying to produce a firm foundation for the EXISTING OMA UAProf 2 spec,
> that's all.
>
> The WG still wants to reject your comment.
> Please let us know if this resolution is ok with you.
>

I am sorry, it is not...

CC/PP is an RDF vocabulary; ie, applications may use the CC/PP vocabularies in various
settings, mixing them with other RDF vocabularies, as part of the general RDF World (and
regardless of UAProf!). It is therefore important that the specification would be
correct in terms of RDF. The restriction that you put into the text is simply incorrect
(or, shall we say, meaningless) from an RDF point of view.

I would propose to separate informative and normative sections here. In an informative
section you might want to say that *if* one uses UAProf then the format of the document
is restricted here and there, in such and such a way. I have no problem at all if you do
that.

Would that work?

Ivan

> Best regards
> Stephane
> [1]
> http://lists.w3.org/Archives/Public/public-ccpp2-comments/2007Aug/0015.html
>
> Ivan Herman wrote:
>> O.k. I understand this but I think the formulation is not really correct
>> and misleading. I think what you say is that a component must have _only
>> one default_. In OWL terms, what you seem to say is that the
>> corresponding predicate is inverse functional (which, in OWL, can be
>> expressed directly).
>>
>> I maintain that the type of differentiation you make, with the text you
>> use, is incorrect from the RDF point of view because there is strictly
>> no difference between what you denote as inline and external. In other
>> words, what happens here is that you impose an extra 'semantics' (sorry
>> for the word) on the RDF/XML encoding which no off-the shelf parser
>> would understand. Ie, a CC/PP implementation will have to use its own
>> parser instead of an external one (after parsing the RDF/XML code the
>> resulting RDF Graph will bear absolutely no trace whether the default
>> was 'external' or not...)
>>
>> Ivan
>>
>> stephane boyera wrote:
>>> Dear Ivan,
>>>
>>> Thank you for your comment on CC/PP: Structure and vocabularies 2.0 ([1])
>>> Your comment on "CC/PP Structure, section 3.3" ([2]) has been referenced
>>> as LC-1773. Please use this reference for further discussion on this
>>> mailing-list.
>>>
>>> The WG decided to reject this comment.
>>> The requirement is to prevent the usage, within a component, of both a
>>> link to an external default and an inline one. Some components in one
>>> profile may have an inline ccpp:defaults, some others may have a link to
>>> an external documents. The rational for not allowing both inline and
>>> external reference in one component is to prevent conflicting
>>> information, and to avoid having to define precedence, profile matching,
>>> and rules of conflict resolutions. Such complexity is not appropriate
>>> for describing ccpp:defaults.
>>>
>>> Please, let us know if you agree with this decision
>>> Best Regards,
>>>
>>> On behalf of the UWA WG,
>>> Stephane Boyera
>>> [1] http://www.w3.org/TR/2007/WD-CCPP-struct-vocab2-20070430/
>>> [2]
>>> http://lists.w3.org/Archives/Public/public-ccpp2-comments/2007Jun/0002.html
>>> [3] http://www.w3.org/2007/uwa/editors-drafts/ccpp2/
>>>
>>
>
> --
> Stephane Boyera		stephane@w3.org
> W3C				+33 (0) 4 92 38 78 34
> BP 93				fax: +33 (0) 4 92 38 78 22
> F-06902 Sophia Antipolis Cedex,
> France
>


-- 
Ivan Herman, W3C Semantic Web Activity Lead
URL: http://www.w3.org/People/Ivan/
FOAF: http://www.ivan-herman.net/foaf.rdf

Received on Friday, 24 August 2007 14:08:27 UTC