Re: [xml-dev] Re: determining ID-ness in XML

[David Brownell <david-b@pacbell.net>]

> Who said I was quoting the XML specification?  

You wrote something about validation behavior, which
is defined by the XML specification.  So you implicitly
referenced it, whether you intended to do so or not.


>    The point I'm making is 
> that if you use a DOCTYPE and cite an element in an ATTLIST that isn't 
> otherwise declared, a validation pass should minimally warn you of this, 
> if not totally puke.  Though it's legal as far as the XML spec is 
> concerned, it's ugly and generally generates bad karma.

Before, you said something different:

> > > a validating parser won't validate a document that has no DOCTYPE 
> > > references (treats it simply as well-formed), but *will* try to validate 
> > > a document in an all-or-nothing fashion if there is.

That was the bit that I disagreed with.  I can't read the XML spec
(section 5 in particular) as supporting that statement at all.

> > >      Simply identifying 
> > > an ATTLIST won't be enough because the 'rootType' element won't be 
> > > defined.

Won't be enough to create valid documents, true -- not that you can ever
just have an ATTLIST decl without also declaring what you call "rootType".
I could have picked a bone with that statement too.

But creating valid documents wasn't the issue at hand.  And whether it's "ugly
and generally generates bad karma" is a policy question, on which people
will disagree.


> > You may be confusing the XML standard with what Microsoft does
> > in some of their implementations.
> 
> Ehem...  Don't insult me.  Seriously.

Microsoft _has_ defined validation differently, and much the same
way you have.  They've taught people, who've taught people, ...

It puzzles me how (why) you would construe that as an insult.
Seriously.

Have a nice day,

- Dave