[whatwg] RDFa statement consistency

[Manu Sporny]

Kristof Zelechovski wrote:
> I think RDFa has already happened: you know what it is and how to use it.

Yes, you are correct - RDFa has, more or less, already happened. It will
be an official W3C standard in the next couple of months and will be
supported in XHTML1.1 and XHTML2. Some are currently working to get it
integrated into a new HTML4 DTD as well. No putting the semantic genie
back in the bottle. :)

> You can embed it in XHTML.  Why is having it in HTML necessary for creating
> statistical models?

I was speaking generally in that case because I thought you were
speaking generally... this seems to have caused confusion. Apologies for
that.

If we are to be very specific, you do not /need/ RDFa attributes in
HTML5 to create statistical semantic models. You could build the same
models from all of the HTML4+RDFa, XHTML1.1+RDFa, and XHTML2 documents
out there. It would also be easier to check those documents for
NLP/semantic correctness with the RDFa markup embedded in the document.
Statistical models are just one approach among the many that would be
used to perform NLP correctness verification. You would not be able to
depend solely on statistical models.

So while you are technically correct, not having any sort of robust
semantic expression mechanism in HTML5 deprives the content from having
multiple paths to validating the document semantics:

- The page's NLP based semantic model verified against RDFa model.
- The page's Statistical model used to verify parts of the RDFa model.

-- manu

-- 
Manu Sporny
President/CEO - Digital Bazaar, Inc.
blog: Bitmunk 3.0 Website Launches
http://blog.digitalbazaar.com/2008/07/03/bitmunk-3-website-launches