- From: Dan Connolly <connolly@w3.org>
- Date: Wed, 05 Dec 2001 13:30:58 -0600
- To: Lorrie Cranor <lorrie@research.att.com>
- CC: www-p3p-public-comments@w3.org
Lorrie Cranor wrote:
>
> We use the data-group syntax anywhere that we reference
> data elements. We did this mostly so that we would
> only have to define one mechanism for these data
> references. Yes, that means that there are some components
> of the syntax that don't add any value in some contexts, but
> they don't hurt anything either.
I see that this is documented in the case of
the optional attribute in data schemas...
The optional attribute is used only in policies (not in data schema
definitions)
Can you show me where the spec says that CATEGORIES
isn't supposed to appear inside ENTITY? If it's not
there, please add it. You might add some comments
to the XML schema that say, e.g. "CATEGORIES
aren't allowed inside ENTITY".
And in any case, please add a test case to your test suite
that shows this is not good syntax. (you are assembling
a test suite, I gather?)
> I won't claim this was the
> best design choice, but it seems to be an OK design choice.
As long as it's a choice that's clearly in the spec somewhere,
I don't mind too much.
> > From the XML Schema for P3P,
> > http://www.w3.org/TR/2001/WD-P3P-20010928/#Appendix_schema
> > This seems allowed:
> >
> > <POLICY>
> > <ENTITY>
> > <DATA-GROUP>
> > <DATA ref="#business.name">
> > <CATEGORIES><demographic/></CATEGORIES>
> > </DATA>
> > </DATA-GROUP>
> > </ENTITY>
> > ...
> > </POLICY>
[...]
--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Wednesday, 5 December 2001 14:31:00 UTC