[whatwg] Table integrity and conformance

[Henri Sivonen]

On Aug 29, 2006, at 10:44, Henri Sivonen wrote:

> Should one expect HTML table row/column integrity to become an  
> HTML5 conformance requirement?

My plan is to spend November prototyping conformance checker parts  
that don't belong in the parser, don't belong in schemas (RELAX NG or  
Schematron 1.5) and don't belong in a RELAX NG datatype library. That  
is, I intend to implement a couple of such features and figure out  
which such features need to be implemented later.

(Stuff specified in the parsing section belongs in the parser.  
Element nesting and attribute occurrence that is nicely expressible  
as a grammar belongs in RELAX NG. Assertions that can be tested with  
IDness-unaware XPath belong in Schematron. The format of the values  
of individual attributes belongs in a datatype library. I'm talking  
about checking conformance requirements that aren't covered by any of  
those.)

I think table integrity checking would be sufficiently complex to  
work as a technical proof of concept and useful enough compared to  
SGML-based validators to work as a nice demo. However, it doesn't  
make sense to work on it as part of a conformance checker if table  
integrity won't be a conformance requirement. I'd appreciate some  
indication on what I should expect in this area, i.e. whether it  
makes sense to prototype a table integrity checker based on HTML 4  
forms.

I am also open to suggestions on what non-schema checks to prioritize  
in prototyping if it turns out that it is too early to work on tables.

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