XML Namespace

[Rick Jelliffe]

James Robertson wrote:
> 
> At 07:26 31/08/1998, John Cowan wrote:
 
>   | Learn to relax and live without DTDs.  The new namespace proposal
>   | makes them useless anyway, so people are now trying to develop
>   | various replacements for them.
> 
> Doesn't this statement make people scared?!?
 
No, because it simply isnt so.  Having namespaces merely expands the
richness of the vocabulary available to document system designers. 
Applications for which XML DTDs are best suited will continue to use
DTDs.  Applications which can stand 100s of Ks of schema data will use
instance syntax.  Applications for which existing AFDR syntax
(attributes for architectural forms) are clearly appropriate will use
AFDR syntax.  People who find that PIs fit will use them; people who
like embedding markup in comments or creating their own embedded markup
landuages will find some excuse to do that.  

The important thing is that document system designers understand when it
is appropriate to use each one.  Namespaces as now proposed do not make
DTDs useless: as far as I can see the proposal does not really impact on
whether a full DTD is useful or not much at all (except for the limited
case of fixed or defualting attributes for the prefix and namespace name
themselves, which has hardly been the prime use of DTDs up to now).  The
fact that DCD takes the namespace URI and uses it to key which DTD or
schema to use is no different from what people do with attributes all
the time: they use attribute values to figure out what processing is
required for an element.

XML wasnt designed on a minimalist principle (what is the smallest we
can get away with?) but more on the optimalist principle (can we satisfy
80% of needs with something only 20% of the complexity?). As far as I
can see, the XML WG has placed quite a high emphasis on keeping (SGML's)
richness; the provision of Namespaces only confirms this impression
(that richness is more important to the WG than terseness or
minimalism). DTDs have proved themselves in the field for many years and
dont require more defense; if other syntaxes arrive which are more
appropriate in their own circumstances than DTDs, that is no slur on
DTDs. 

Namespaces certainly provide a good set of hooks for external schemas,
but such hooks can be provided in fixed attributes in DTDs just as
readily as in attribute values in the instance.  In fact, for size
reasons, having namespace attributes declared in fixed attributes in
DTDs may be more preferable than bloating the instance with repeated
attributes. Counter to John Cowan, namespaces may, for some document
types, make an explicit DTD *more* attractive, not less!


Rick Jelliffe

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)