Re: HTML+RDFa issue 120

I am not sure how the HTML5 process works, ie, whether that page can still be improved. 

I think it is worth adding more and significant deployment proofs ... Like:

- All dbpedia, eg, http://dbpedia.org/page/Paris; today they are in xhtml but probably served as text/html; there are as many pages as there are wikipedia pages
- All the library of congress authorities pages, eg, http://id.loc.gov/authorities/sh95002791 for HTML

I am sure that Manu has some good examples in ecommerce, too. Of course there is facebook. I think the argument of facebook could be used a little bit more forcefully. There are millions of facebook pages that today use RDFa and prefixes. This is more important than Google or Yahoo: while these guys allow RDFa to be used alongside microformats or microdata, the Facebook advise is RDFa only!

The point that has to be made I think even stronger: all these pages would become illegal HTML5 pages if the prefix facilities were removed. Put it another way, an upgrade to HTML5 would be significantly more difficult if backward compatibility was broken.

Another issue that might be worth adding. The group is currently *considering* adding default prefixes and terms to an HTML5 profile, using the @profile mechanism. That approach may significantly reduce the 'cognitive load' that the original submission referred to, because the prefix URI-s would be automatically valid. In other words, only experts, who need more complex vocabularies, would need to worry too much about extra terms and/or prefixes...

Thanks!!!!

Ivan

On Jan 21, 2011, at 16:44 , Toby Inkster wrote:

> Heads up!
> 
> There is an issue filed against HTML+RDFa with the HTML WG on the use
> of CURIEs within RDFa. Ian Hickson has written a change proposal for
> HTML+RDFa:
> 
> http://lists.w3.org/Archives/Public/www-archive/2011Jan/att-0022/cp-120.html
> 
> This proposal, if adopted, would make the @profile and @prefix
> attributes non-conforming in HTML+RDFa, and require user agents to
> ignore them. It would also require user agents to ignore @xmlns:*
> attributes. In summary, there would be no mechanism for authors to
> define prefix mappings.
> 
> In my capacity as invited expert to the HTML WG, I've submitted a
> counter-proposal:
> 
> http://www.w3.org/html/wg/wiki/ChangeProposals/RDFaPrefixesNoChange
> 
> This proposes that no technical changes be made to HTML+RDFa, but
> editorial changes be made to promote the use of full URIs in RDFa,
> and make it clearer that CURIEs are simply an abbreviation for the
> full URIs.
> 
> The proposal also suggests that the HTML+RDFa editor (Manu) work with
> the RDFa Working Group to make sure examples with full URIs are
> similarly featured more prominently in XHTML+RDFa and RDFa Core.
> 
> -- 
> Toby A Inkster
> <mailto:mail@tobyinkster.co.uk>
> <http://tobyinkster.co.uk>
> 


----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf