[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Arbitrary Infoset boundaries (was Re: Common XML -FinalReviewDraft)

• From: Wayne Steele <xmlmaster@h...>
• To: xml-dev@l...
• Date: Wed, 02 Aug 2000 20:20:02 -0700 (PDT)

>"Simon St.Laurent" wrote:
> >
> > At 09:32 AM 8/3/00 +0800, Rick JELLIFFE wrote:
> > >This agrees with XML's premise that it is a resolved document format,
> > >rather than a format for documents in progress.
> >
> > Is that premise formally stated anywhere?
>

It seems to me much of this comes down to the question of "What are DTDs 
for"?

Everyone agrees about Validation.

Everyone agrees that DTDs declare Entities and Notations, and allowance 
needs to be made for preserving this information. Likewise for attributes of 
type ID.

Is that it?

The infoset takes the view that once you have validated the document, and 
you get your entities & notations, you have no further need for the DTD. 
It's about as useful as storing the HTTP headers you got when you slurped 
the document off the web. You can have it, if you want, but it's outside 
their scope.

Simon (and myself, and others) dissents from this view.

I think I've summed up most of this Infoset debate right here. DTDs: In, or 
Out?

Weirdly, PIs and Comments in the DTD are in the Infoset.
But the 90% job of the DTD, declaration of elements and their content 
models, is left out.

I think it should be put back.


-Wayne Steele




________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com

• Prev by Date: 'resolved' SGML
• Next by Date: DOCTYPE replacement kit
• Previous by thread: ANN: Dr. Dobb's audio interview with David Orchard on XLink
• Next by thread: DOCTYPE replacement kit
• Index(es):
  • Date
  • Thread