Re: [xml-dev] Victory has been declared in the schema wars ...

[Henri Sivonen <hsivonen@iki.fi>]

On Nov 29, 2006, at 14:31, Elliotte Harold wrote:

> Henri Sivonen wrote:
>
>> I think not having a normative schema for HTML5 is a very good  
>> idea. None of the current schema languages (not even Schematron)  
>> are adequate for expressing all the conformance requirements of  
>> HTML5.
>
> More to the point, HTML5 is not XML. It is not well-formed, and in  
> general an HTML5 document cannot be made well-formed without  
> crippling it by removing quite a few elements. Therefore XML schema  
> languages really don't fit it at all.

They fit quite nicely if the HTML parser reports an equivalent XHTML  
document to the schema layer, which is what my parser does. Jing  
doesn't care what calls the ContentHandler methods.

> Frankly I think this is insane, but it's what the HTML5 folks have  
> told me quite clearly.

Parsing text/html from the real Web using an XML parser is an  
exercise in futility. Therefore, a real-world-compatible parsing  
algorithm is needed. With that in place, there's no good reason for  
text/html suppliers to make their stuff XML. And having people  
pretend that they are sending XML when they aren't wouldn't do any  
good to XML.

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