Re: Best practice for DTD validation of RDFa xmlns attributes

I think i am confused Dorian.  The use of xmlns:foo attributes is 
perfectly valid XML.  The W3C validation service will not complain about 
such use.  Some other validators, based upon SGML DTD technology, might 
complain but they would be in error.

Can you provide us an example of a page that 1) uses the XHTML+RDFa 
DOCTYPE declaration and 2) fails to validate against the W3C validation 
service at http://validator.w3.org ?

dorian taylor wrote:
> Greetings,
>
> I'm eager to start implementing RDFa, however I'm vacillating on one
> point - that I'm not sure what to do about DTD validation for the
> xmlns attributes. My issue is that if I'm advertising that the
> document validates against a given DTD, I feel it should actually do
> so. Ideally I'd like to do them in an internal subset but if I'm not
> mistaken, parameter entities (required to define attributes) can only
> exist in external subsets.
>
> I still have yet to find in my travels a best-practice document for
> dealing with DTD validation of XHTML+RDFa - specifically the xmlns:foo
> attributes - that's officially endorsed by the RDFa group.
>
> Anyway, the best solution I can think of is something like this:
>
> 1) Standardize organization-wide on prefix identifiers (always fun)
> 2) Produce (or generate) a DTD with the appropriate xmlns attributes
> 3) Mow over the doctype declarations with the URI of the DTD with all
> my XHTML+RDFa documents (sed ftw)
>
> Phew, enough acronyms.
>
> Is there be anything I'm missing, or perhaps a better way?
>
> Regards,
>
> --
> Dorian Taylor
> http://doriantaylor.com/
>
>   

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com

Received on Sunday, 14 September 2008 12:30:01 UTC