xml-dev - RE: interoperability (was Re: Obfuscating XML with namespaces)

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

Defining interoperability as distinct from portability is 
great mind teaser.  I've watched some excellent 
efforts bog down in not making these distinctions:

1  Data is portable (XML) Note, I did not say, information is 
portable.  Teasing namespace apart from semantic is tough.

2  Systems interoperate (APIs)  Can you encapsulate a system 
definition in terms other than interfaces without an implementation?
  
3  Vocabularies enscribe domains (Languages).  Can you say a 
language is portable or interoperable without 1 and 2?  In fact, 
is a language ever portable or interoperable?

Is the InfoSet the commmon definitional framework to enable all three of
these?

XML can't fix the nattering need of developers and designers to keep 
pushing the edge of any definition into its neighbor's yard, so I never 
expect the semantic web to be more than a noisy neighborhood.  On 
the other hand, arc systems have been around for years and in 
neighborhoods with high fences or lots of empty lots, they work.
  
Consider density and overlaps as quality properties of the 
semantic neighborhood.

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: Simon St.Laurent [mailto:simonstl@simonstl.com]

I'm not sure interoperability as commonly understood is really a priority
at the W3C.  It's no wonder that developers are confused about which tools
to use and how to integrate them, even without the additional possibilities
of JDOM, RELAX, and others.