Re: microdata use cases and Getting data out of poorly written Web pages

On May 14, 2009, at 17:21, Ben Adida wrote:

> Except, as I already responded, this was a bug in one of Yahoo's  
> user-facing pages they had, not in the actual RDFa parser they use  
> for their actual SearchMonkey pages (which is, in fact, based on one  
> of the RDFa task force's parsers.)

My point is that there's a problem that CURIEs enable that class of  
bugs.

> My point remains: there have been long, verbose claims of namespace  
> aware vs. non-namespace aware API calls, how it will require browser  
> changes, etc... and in practice, *it doesn't matter*. If browsers  
> choose not to fix this, RDFa parsers can address this with *one*  
> "if" statement.

I think the connotations of the word "fix" are inappropriate here,  
since DOM Level 2 predates RDFa.

As for one 'if' statement, experience with the lang vs. xml:lang issue  
suggests that violation of DOM Consistency cause more code than one  
'if' statement in practice.

> I have trouble seeing these so-called problems you raise as anything  
> more than theoretical. In all of the actual RDFa implementations to  
> date, no one's complained.

When I last checked RDFa implementations, they either used a namespace- 
unaware API or supported only XML. I don't know if this is a  
coincidence or the effect of the RDFa syntax poking a syntactic area  
whose infoset mapping differs for XML and HTML.
http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2009Feb/0106.html

-- 
Henri Sivonen
hsivonen@iki.fi
http://hsivonen.iki.fi/

Received on Monday, 18 May 2009 08:44:25 UTC