[OEP][SE] Semantic Interoperability & Ontology Driven Architectures

Dear All, 

 

I am catching up on the activities of this group after a long hiatus and
am thrilled with the many rich topics under discussion. Of particular
interest to me were/are the threads regarding Semantic Interoperability
and Ontology Driven Architectures. 

 

During the past months, and one of the reasons for my hiatus in this WG,
I've had the honor of interacting with many programs (of varying
maturity) that are currently engaged with OWL and RDF technology for
solving an array of business problems. When possible, I like to share
the knowledge I glean from the implementers of these many programs with
the conferences that I have the pleasure to speak at. 

 

For Dave McComb's upcoming Semantic Technologies 2005 conference in San
Francisco I am presenting on the subject of "Enterprise Semantic Web
Architectures" - which is buzzword for - "How businesses can make use of
OWL and RDF to accomplish valuable new capabilities: presented in terms
of software architecture patterns."

 

In the context of the aforementioned SWBP WG email threads, I think the
following presentation may have some ideas to contribute, or serve to
further underscore the topics already highlighted in the WG. 

 

http://www.networkinference.com/downloads/NI-Pollock-ST05.pdf

 

I would be honored if folks in this WG found the time to check this out
and offer comments in advance of the ST-05 event. Given the content of
the said discussion threads I should fully expect some measured
critiques as well, I hope to learn from everyone's feedback on the many
important questions raised by the promise - and hype - of OWL/RDF
technology within the enterprise.

 

-Jeff-

 

BTW: 

 

It should be noted and clarified that:

*	The content of this presentation has a "SOA bent" to it for
three key reasons:

	*	Industry hype for SOA at this minute
	*	Architectural influences from the NCES community
	*	Assuming XML/SOA underpinnings make the interfaces
simpler

		*	(and happens to jibe w/W3C vision/layer cake)
		*	(and bypasses many 'messy' issues with other
data formats)

*	The structure of the presentation...

	*	[Background] =>[Use Cases] =>[Services]
=>[Architectures]

...is intended to communicate in terms of UML practices

*	The "Logical Architecture" section is weak, and in a fuller form
should embody many of the OWL design patterns discussed within this WG
*	I've purposefully omitted a specific discussion of FOL/FOPC/Rule
interfaces with OWL systems

	*	Because the industry landscape is too multifaceted at
the moment
	*	And many/most of the "issues" can be resolved in the
old-fashioned way - programming

		*	(with mutual understanding of the disadvantages
that come with it)

 

 

  _____  

 

Jeff T. Pollock 
Vice President, Technology 

Main:   760-476-0650 x114
Mobile: 760-672-8444
Fax:    760-476-0648
www.networkinference.com
jeff.pollock@networkinference.com

	5963 La Place Ct., Suite 300
	Carlsbad, CA 92008 

 

Received on Monday, 28 February 2005 04:44:18 UTC