Re: CDR: style and precision problems in definitions appendix

I am satisfied with all resolutions to these issues except the  
following:

On Mar 13, 2006, at 8:53 AM, Kevin E Kelly wrote:

> - Stating that "The DOM that is formed must be tree-based" is vacuous
> - the DOM is always tree-based.
>
> [KEK] No change was with respect to this comment.

So what does it mean to say "The DOM that is formed must be tree- 
based"? What is an example of a non-tree-based DOM? Since this is  
stated with "must", it appears to be a conformance criterion, but how  
would one test for conformance? I think the WG should either explain  
what this sentence means or remove it. To me it reads like "Water  
must consist of hydrogen an oxygen atoms," i.e. a statement of fact  
misstated as a conformance requirement.


>  "CDI
> A document which includes other documents by-value.
>
> For example: A single XML 1.1 document making use of XML grammars
> defined independently, normally making use of multiple namespaces."
>
> - Instead of "by-value" it may help to use a more clear term, like
> "textually include" or "directly include".
>
> - CDR expands to Compound Document by Inclusion, it may be worth
> including this expansion in the definition.
>
> [KEK] CDI expanded to define acronym
> [KEK] definition changed to "A document which directly includes  
> other documents and or namespace markup within the same physical  
> document."

I'm not sure what this means. What is directly including another  
document? If there's a separate document in the DOM sense, you have  
CDR, not CDI. What exactly is "namespace markup"? Surely not any XML  
namespace declaration, since just declaring the xhtml namespace in an  
XHTML document doesn't make it a CDI. I'd even suggest that declaring  
the xlink namespace and using xlink:href in an SVG document doesn't  
make it a CDI.

I'll see if I can suggest better wording.

Regards,
Maciej