A Plea for Schemas

Matthew Gertner matthew at praxis.cz
Tue Nov 2 10:14:02 GMT 1999

Tim Bray wrote:
> I share the doubt that a grand unified schema repository will be tractable
> to build & maintain.  I also am enthusiasm-challenged because in my (lengthy)
> experience with XML and its ancestors, I have only ever seen machine-readable
> schemas put to use in one application: namely the use of DTDs in
> hand-authoring XML documents.  Clearly the addition of datatypes and so on
> should enable all sorts of other goodies but we are all placing bets on
> an as-yet-unrealized future.  So some degree of skepticism is in order.

This is an easy issue to be sceptical about. It is worth keeping in
mind, however, that these historical arguments ("never needed it so I
probably never will") neglect the fact that many of the pieces of the
puzzle have been in place only recently (or are still being worked on):

* The right mindset: until very recently generic markup has been seen as
primarily useful for documents, where the utility of schemas is
admittedly limited for anything but authoring applications. The idea of
using XML as a software engineering tool to model data structures has
now become very prevalent.

* Critical mass: I'll put more effort into something if I can leverage
it across a wide user base. It's worth the effort to craft XML Schemas
and distribute them  because the community is so much bigger than it was
in the good old SGML days.

* Communications infrastructure: Something like an automatic discovery
mechanism for schemas would not be possible without widespread access to
a common network infrastructure.

* The schema language itself: IMHO DTDs are the weakest part of the
current XML standard. Hats off to the SGML inventors (I already got told
off for giving Dr. Goldfarb all the credit): most of what they started
doing in the 60s works very well in the late 90s. DTDs don't, as the
many "how do I combine these DTDs in a single instance" messages to this
list attest. A metalanguage that makes it easier to work with, combine
and extend schemas might plausibly increase willingness to use schemas.

So yes, some degree of skepticism is certainly in order, but this could
end up making the difference of an order of magnitude in the influence
that XML ends up having on information technology. We'll see.


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/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe 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)

More information about the Xml-dev mailing list