xml-dev - RE: XML Schemas: Best Practices

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Wouldn't the case for the instance exposing the namespaces 
for each vocabulary be when the component implementations 
for the property sets used in those vocabularies must be 
explicit?  Consider the case for a hypothetical real time 
3D rendering system that includes an interface to a voice 
board.  I might want to combine specific grammars for 

1.  Geometry, lighting, the scene description
2.  Character abstractions for different types 
    of roles (eg, Puck is of type fairy)
3.  Tagged speeches where each speech has been 
    tagged for emotional delivery but not the speaker

I would then want to include each of these into a 
final rendering language, but also indicate which 
component is required that can interpret say the 
speech data.  So do I want to hide that from the 
author or do I want the author to ensure, and 
the final rendering system to ensure these components 
are available?  Certainly I may not need a namespace 
to cite the component but that is one means to 
indicate the language in use is from a vendor who 
also provides the component.

Len Bullard
Intergraph Public Safety
clbullar@ingr.com
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Roger L. Costello [mailto:costello@mitre.org]

This discussion has argued for keeping hidden in the schema the location
(namespace) of the components.  However, there are scenarios where it is
desirable to make such namespaces explicit, i.e., we want the instance
documents to explicitly show where the components come from.  Would
anyone care to make a case for that?