W3C home > Mailing lists > Public > www-webont-wg@w3.org > December 2001

Serendipitous Interoperability (was: One More Use Case)

From: Ora Lassila <daml@lassila.org>
Date: Fri, 14 Dec 2001 06:52:41 -0500
Message-Id: <a05100300b83f8f2019b6@[]>
To: www-webont-wg@w3.org

after being told in yesterday's teleconference that my use case [1] 
can be likened to wanting to eliminate hunger from the world I feel I 
should explain why I chose it and why it may not, after all, 
represent na´ve idealism.

Naturally I have to apologize for choosing a research-oriented 
problem. I acknolwedge that not everyone here represents a research 
organization and may have more practical, short-term goals. Some of 
us, however, are in this working group because it has become the 
frontier in trying to realize the vision of the Semantic Web (those 
not interested in this can stop reading this message now :-).

The Semantic Web is a *hard* problem. Why? Because many of our 
simpler examples can be solved using other means - no RDF, DAML+OIL, 
reasoning etc. is required (this is anyway the response I get when 
trying to explain what the Semantic Web is about, and I have been 
trying for the past 5 years or so). Any reasonable number of 
information systems can be made to interoperate and exchange 
information by employing a group of clever XSLT and Perl programmers, 
it seems. And people will do this, and will extend this approach to 
Web Services as well.

What we are trying to do is qualitatively harder. The problem is that 
it cannot be demonstrated or illustrated using simple scenarios. I 
believe we want to make a solid case that the Semantic Web makes 
sense; otherwise we will always be faced with the folks who would 
rather believe in the "XSLT magic". As for the "uncoreographed" 
interoperability of some systems (virtual or physical), a reasonably 
powerful *standardized* representation language would be a big win, 
allowing us to expose (some of) the semantics of the systems in 
question and making it more likely a reasonable response can be 
generated in unanticipated situations.



	- Ora

[1] http://lists.w3.org/Archives/Public/www-webont-wg/2001Dec/0022.html
Ora Lassila  mailto:daml@lassila.org  http://www.lassila.org/
Research Fellow, Nokia Research Center
Received on Friday, 14 December 2001 06:52:45 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:56:40 UTC