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

Re: Another look at namespaces

• From: "Rick Jelliffe" <ricko@a...>
• To: "XML-DEV" <xml-dev@i...>
• Date: Fri, 17 Sep 1999 15:48:51 +0800

 From: Andrew Layman <andrewl@m...>

>Rick Jeliffe wrote "Andrew Layman's comment that there
>are definitional schemas and that these set namespaces is undebated
>and can be proved wrong by the simple example of RDF, which
>defines a namespace but allows major parts of schemas undefined,..."
>
>I recommend that you return to my original post and look at what I
actually
>wrote.  You might be pleasantly surprised to find that I am both aware
of
>and support the idea of namespaces which are not defined by schemas.
:-)

I think second point in that posting is well-made, that "There is
nothing in
the namespaces specification that requires that all the elements in a
document
come from the same namespace".  I don't think that is a contentious
issue.

But it is the first point, that  I don't see:
 "If indeed there are separate languages, where the elements in one
require different processing (such as different validation) from
elements in
another, then the distinct elements of each should be distinguished by
different namespaces."

This says that a schema determines the namespace. A change in content
model
requires a change in namespace; the namespace URI is a new doctype
declaration
with attribute syntax and different combining rules.  ( I agree that
different meanings "should" be distinguished by different namespaces,
otherwise the whole
use of namespaces is subverted, but to say that different validation
requires
different namespaces goes much to far. A schema uses names in spaces;
 a document uses names in spaces; a schema does not create a namespace
and
a document does not create a namespace (at the moment, nothing formally
creates/defines a namespace, in the same way that nothing defines a
URI: the URI if syntactically correct is useable for a purpose or
unuseable.)

It follows from Andrew's second point that if I want to use <html:p> and
include in it my own element <rick:dog> then I cannot use any of the
namespaces provided by XHTML. These namespaces invoke particular
DTDs, and these DTDs have closed content models (if they were open,
we would only need one namespace).

This contrasts with the view that we can derives the items in a
namespace
by looking at all the schemas and documents which invoke that namespace.
In the absense of a namespace definition language, containing just a
simple
list, that there is




xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• Prev by Date: Re: Binary-encoding of XML for communication
• Next by Date: RE: Another try on groves
• Previous by thread: RE: Another look at namespaces
• Next by thread: Re: Another look at namespaces
• Index(es):
  • Date
  • Thread