W3C home > Mailing lists > Public > public-sws-ig@w3.org > November 2003

Introduction - Geoff Arnold

From: Geoff Arnold <Geoff.Arnold@Sun.COM>
Date: Sun, 09 Nov 2003 20:46:58 -0500
To: public-sws-ig@w3.org
Message-id: <C59CE700-131F-11D8-ABF5-000A95718676@sun.com>


I'm Geoff Arnold, of Sun Microsystems. Until recently I worked in
Sun Microsystems Laboratories; I've just shifted over into the
office of the Chief Engineer. For several years I was involved
in multiagent systems research in SunLabs, and was Sun's
representative to (and board member of) FIPA, the Foundation
for Intelligent Physical Agents. During 2000-2002, a number of FIPA
members were involved in work on standards for the software
infrastructure for multiagent systems. As this work progressed,
a number of us realized that the most promising way of exploiting
the work in multiagent systems might be to marry it to the emerging
web services movement. Accordingly, a number of us proposed
the establishment of a "semantic web services" interest group
within W3C. Our intent was to create a forum which would
accomodate a variety of groups, including the DARPA agents community, 
multiagent research community, and those working on the application
of web services technology to applications using autonomous agents

More recently I have been Sun's alternate representative
to the W3C Web Services Architecture working group.

My primary interest within SWS-IG is the software engineering
problems associated with the creation of agents that are capable
of semantically rich interoperation. When we worked in FIPA
on agent communications language (ACL) and the FIPA abstract
architecture (AA), it became apparent that the biggest obstacle
to progress in interoperation was the lack of shared, open source
software systems for implementing the BDI semantics of ACL. We had
several sophisticated implementations of the "plumbing" (agent
communications, advertisement, brokerage, and so forth). What we
lacked were the tools to develop and gain practical expertise
with multiagent applications, and this (from my perspective)
made it impossible to really evaluate the utility of much of
the theoretical work on agent communications. I want to encourage
the adaptation of software tools for reasoning and planning to
support semantically rich web services interactions.

Received on Sunday, 9 November 2003 20:46:58 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:32:43 UTC