"Multiple" Namespaces? (but NOT for HTML)

John Robert Gardner jrgardn at emory.edu
Fri Oct 22 18:08:53 BST 1999


One of the goals of our project--so kindly written up by Robin at
http://www.oasis-open.org/cover/atlas.html--is to allow the searching of
our records (which will be linked to XML instances of the articles in gif,
then later in keyed/tagged text) from a variety of existing search
interfaces, as transparently as possible.  In other words, someone using
DC can search for articles where Creator is Tillich, while someone using
GILS can search for Originator as Tillich, and so on.  

This grand vision hinges on a what's proving a rather thorny wicket: do we
have to have complete instances of each kind of tagged 600mb record set
for each chosen DTD?  I.E., a DC set, a GILS set, etc.?   While my
new-found joy with xslt makes _creating_ them kind of fun, the logistics
of duplication seems the long way 'round something which sounds sort of
like a namespaces kind of issue.  I ask this b/c some pretty knowledgeable
folk thought multiple namespaces would be a solution, but it seems that
can't be done--except through architectural forms which we don't want to
from Megginson):
  It's a well-proven technique for working with
architectural forms in SGML (except that architectural forms allow 0-n
while namespaces allow exactly 1).

I read this, as does those with whom I've spoken, as a sort of truncation
of the idea of multiple namesapces--at least for XML sans architectural
forms.  What would be swell is if we couldhave a tag for the example above
which is like:

	<DC:Creator GILS:Originator TEI:docAuthor>Tillich</DC:Creator
		GILS:Originator TEI:docAuthor>

I know that's a behemoth of a tag, but at least it's ONE tag in ONE
instance, and my math indicates it's still less space than 3 document
instances of the records to accomplish the same thing otherwise.

Considering that this first phase with MARC records in these various
formats of XML will be stored in and delivered via Oracle 8i, can anyone
recommend something which will address this issue (preferably a
non-proprietary solution, as our grant requires our output to be
standards-based/archivable, _not_ proprietary)?

I've scanned the xml-dev archives and have not found any other
alternatives--though they could be there only buried in a
non-namespaces-subject-line reply.



John Robert Gardner
XML Engineer

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