[
Lists Home |
Date Index |
Thread Index
]
On Tue, 2005-04-05 at 09:04 +1200, Fraser Crichton wrote:
> It was only scary because I didn't think anybody else would be doing it
> that way and I'll probably have to refactor some other schemas to make
> it easier to do this, otherwise it sounds like a good approach.
>
> How do you handle versioning in that case? Or, rather, how do you
> decide when you have created a new version of a schema? Does it make
> sense to ask that??
Versioning is very hard in this scenario, but unfortunately, schema
versioning is hard in any scenario. We use a version attribute
(similarly to XSLT) defined in schema and expressed in instance. Master
schemata have the same version number as derived schemata. One still
needs to be vigilant about the consequences of version bumps. The
subset schema framework increases the things of which one has to be
vigilant.
> > The key is that we have a formal means of
>
> > annotating the relationship between the master and the derived.
>
> I'd be interested to know a bit more about this as I could see it being
> useful for my company's client.
I briefly describe the overall methodology I've developed for formalized
schema annotations here:
http://www-128.ibm.com/developerworks/xml/library/x-tipdict.html
In this particular case rather than OWL, we tried to suppress the
semantic load a bit. In the derived schema we have something like:
<ns:master-link href='http://schema-library/master-foo.rng'/>
Nothing so complex, but the tools know how to look for these notations
and match them against the relevant rules.
Interestingly enough, we're not in talks with that client to use OWL far
more broadly in the project, so these annotations would probably be
updated to OWL in the process. And we should be able to express more of
the processing pipeline's business rules declaratively.
--
Uche Ogbuji Fourthought, Inc.
http://uche.ogbuji.net http://4Suite.org http://fourthought.com
Use CSS to display XML, part 2 - http://www-128.ibm.com/developerworks/edu/x-dw-x-xmlcss2-i.html
Writing and Reading XML with XIST - http://www.xml.com/pub/a/2005/03/16/py-xml.html
Use XSLT to prepare XML for import into OpenOffice Calc - http://www.ibm.com/developerworks/xml/library/x-oocalc/
Be humble, not imperial (in design) - http://www.adtmag.com/article.asp?id=10286
State of the art in XML modeling - http://www.ibm.com/developerworks/xml/library/x-think30.html
|
