xml-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brett McLaughlin <bmcla...@algx.net>
Subject Re: A mathematical vision of XML leads to interesting conclusions
Date Sat, 18 Dec 1999 18:12:43 GMT


Steve Muench wrote:
> 
> | Donald and I both came to the conclusion that such inheritance facility
> | would allow great simplification of the use of XML as data container,
> | also allowing easier data-binding with OO languages.
> |
> |  <page xml:extends="template.xml">
> |   <title>Hello world!</title>
> 
> Stefano,
> 
> In our Business Components for Java application framework
> (http://technet.oracle.com/products/jdev/info/techwp20/wp.html#app-customiza
> tion)
> we use XML metadata to represent business application-building
> components (e.g. PurchaseOrder, Customer, ChristmasPresent).
> 
> In order to facilitate layered customization of
> business components, we have implemented exactly the idea
> you are talking about for having one component "refine"
> another through this "extends" derivation.

I think Stefano is trying to get across more than just a simple extends
mechanism.  If you notice, he made a query as to whether to include this
as part of the W3C Inheritance or add it as another note.  Inheritance
tries (in the current wording) to "plus a leak" in XML, but makes no
attempt to truly orthogonalize XML.  What Stefano is referring to is
creating this orthogonality, not plugging a hole... only so many holes
can be plugged (for an example, try to use the Servlet API for things
like Cocoon!  It needs to be revised, not plugged) before something is
completely broken.

XML is _not_ orthogonal, and XML Schema, by deinition (since it is
conformed by a non-orthogonal spec, DTDs), can not be truly orthogonal. 
It can only create a simluation of that orthoganilty, through some
strectching and blurring of XML corners.  Again, that's why namespaces
only really create a unique element name, not a unqiue document space.

> 
> It has been our understanding that our current mechanism
> is a tide-me-over until XML Schema with schema refinement
> emerges.

XML Schema has nothing to address the types of problems that Stefano is
addressing, those of the merging/extension/addition of two to n XML
documents.  How are collisions defined in extensions?  How do namespaces
operate within extended documents?  How can a namespace be treated as a
document space and not an element-prefix space?  Is there a way to
extend documents without the use of XSL/T and instead simply
algorithms?  XML Schema doesn't do anything for these questions.

-Brett

> 
> Have you checked out the latest schema working draft
> section on this?
> http://www.w3.org/TR/xmlschema-1/#section-Type-Refinement-*
> 
> _________________________________________________________
> Steve Muench, Consulting Product Manager & XML Evangelist
> Business Components for Java Development Team
> http://technet.oracle.com/tech/java
> http://technet.oracle.com/tech/xml

Mime
View raw message