RE: [xml-dev] Identify PUBLICID outside of dtd?

["Chris Wilper" <cwilper@cs.cornell.edu>]

> AFAIK, any element can be a root in an appropriate 
> document: there is no way to privilege some elements 
> or to prohibit others.

FYI, in W3-XML-Schema, I can define a single global element 
'root' that has a complexType 'myType' in which I declare that 
'myElement' may be used.   If myElement and associated
constraints are only defined within the context of of 
'myType', then by definition, 'myElement' may not 
occur as a root element.

In other schema languages I could express it differently,
-- but the point is, the notion exists... and it doesn't
need to be expressed with a schema language in order to exist.

> ...support for referencing a type (i.e. "declaring" a type 
> by canonical name only, as in "This is a Docbook document".)...

Thanks for the link -- an informative discussion.  According
to that, what I am looking for was not really (even attempted
to be -- but was often assumed to be) provided by DTDs, but 
is more along the lines of what was conceived of in 
SGML Architectures (ISO/IEC 10744:1997 Annex A3).

I don't know enough about architectural forms (and how
people have concieved of an XML equivalent) to determine
whether what I'm looking for can be found there...

> ...I think you mean "XML namespace", which makes it true 
> by definition...

Right, I'm talking about XML namespaces.

>| a corresponding doctype could similarly be defined -- and would naturally 
>| fit within the same namespace,
> What does this mean?  
> Wouldn't it be better to have a name for the taxonomy as a whole, within
> which components such as "elements" could be identified?

A "document type" as I conceive of it crosses vocabularies (talking
xml namespaces here).  Documents of that type start with an element 
belonging to one namespace, and beyond that may delve into 
elements/attributes belonging to others.  Existing schema languages
might be used to express validity constraints on a document type,
but the notion exists outside of the context of validation.

- Chris

-----Original Message-----
From: Arjun Ray [mailto:aray@nyct.net]
Sent: Tuesday, October 15, 2002 2:36 PM
To: xml-dev@lists.xml.org
Subject: Re: [xml-dev] Identify PUBLICID outside of dtd?


"Chris Wilper" <cwilper@cs.cornell.edu> wrote:

| It seems that a doctype exists for every element that can be used as a 
| root element in a document.

AFAIK, any element can be a root in an appropriate document: there is no
way to privilege some elements or to prohibit others.  The notion of a
"document type" is underspecified in SGML too; in particular, there is no
support for referencing a type (i.e. "declaring" a type by canonical name
only, as in "This is a Docbook document".)  See, e.g.

 http://groups.google.com/groups?as_umsgid=34E9CBC9.401B6BB0@isogen.com

| Since an element can be defined as belonging to a namespace,

I think you mean "XML namespace", which makes it true by definition.  In
the ordinary sense of "namespace", a SGML/XML document has many of them in
a single document, all of them taken together comprising a taxonomy (or,
if you like, "vocabulary".)

| a corresponding doctype could similarly be defined -- and would naturally 
| fit within the same namespace,

What does this mean?  

Wouldn't it be better to have a name for the taxonomy as a whole, within
which components such as "elements" could be identified?

I sense some old debates cropping up again.


-----------------------------------------------------------------
The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
initiative of OASIS <http://www.oasis-open.org>

The list archives are at http://lists.xml.org/archives/xml-dev/

To subscribe or unsubscribe from this list use the subscription
manager: <http://lists.xml.org/ob/adm.pl>