What is a namespace ... really?

[Murray Maloney]

At 11:52 AM 1/15/99 -0500, david at megginson.com wrote:
>Borden, Jonathan writes:
>
> > I don't see a problem with this one-to-one relationship, after all,
> > a namespace *is* defined by a uri, so... and I don't immediately
> > see why this precludes inclusion of elemements from different
> > namespaces.
>
>What if I want to create a schema specifying that (for my set of
>documents) an html:p element may contain a tei:foreign element, or a
>docbook:Trademark element in addition to the regular HTML elements?

Well, if you intend to modify the content spec of an element
from a namespace over which you have no control or authority,
you might have a problem. Seems to be the same problem that
you have if you try to modify a "final" class. That's not to
say that the XML Schema work will not offer a way to do so.
>
>What if I want to create a schema specifying that (for my set of
>documents) an html:p element may *not* contain an html:font element?

Again, same problem.

But neither of these example address the question that was 
raised by Jonathan. 
>
>It doesn't make sense to have to create a new and different namespace
>for either of these -- I'm still using the individual elements in
>mostly the same way.  

I might make sense to be *allowed* to create a new *schema* for
both of these examples. The effect of doing so would be to create
a new namespace (see SOX).  

>I could, of course, use some kind of inheritance
>scheme, but I don't think the world will buy anything that requires
>retrieving 5 or 10 schemas from different servers just to figure out
>that an html:a element is from the HTML namespace.

I don't think that we know yet what the world will or will not buy 
(notwithstanding ridicyulous PE ratios).
>
> > I think the idea is that if a namespace is defined by a uri, it may
> > inherit a meaning associated with that uri, for example, suppose
> > the uri was a .DTD, would this cause a problem or work any less
> > well than a DTD which defines a default namespace and is specified
> > in a <!DOCTYPE definition?
>
>It would cause about the same set of problems as DOCTYPE (perhaps
>worse with datatyping and other niceties) -- that's why we need to get
>away from it.

Of course, David's opinion is his own. Although it may be shared
by others in the community, I hope that his opinion will not
hold sway over the design and development of better schemas and 
namespace mechanisms for XML.

Regards,

Murray


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)