Shared Planning and Activity Representation (SPAR)

I think the terminology now being suggested for SWSL:

   perform some process or activity

   invoke some service

gives the right flavour of a higher level description of a process, 
activity or its sub-processes and sun-activities... before getting down to 
a specific service (its really a resource that provided the service which 
executes the [atomic] activity)

-----------------------------

With respect to discussions of naming things in SWSL like 
performs/uses/invokes, objectives/goals, plans, processes, activities, 
etc.. We could also maybe use some of the nomenclature from the experience 
of drawing on the PIF, NIST PSL and other plan/process representation work 
for an initial stab at a shared planning and activity representation (SPAR) 
language for DARPA research programs... It involved a large team of people 
commenting and critiquing this.. including some active folks on the current 
SWS-IG list. See http://www.aiai.ed.ac.uk/project/spar/

The SOAR work, which I chaired, started off with very large aspirations and 
a long term engineering approach to shared representations that would alter 
over time.  version 0.1 tried this and went down like a lead brick.  It was 
just way beyond anyone's interests to be looking at CHANGING ontologies 
when they did not even have one yet!  This reflects my own background in 
data bases where the first question is often how do I get the data out in 
30 years when the system no longer runs!

But SPAR ended up with a more limited but very useful core ontology and 
extensions... aligned with NIST PSL and other work on <I-N-C-A> for 
example... See
http://www.aiai.ed.ac.uk/project/spar/spar-doc-02.html

The UML diagram at 
http://www.aiai.ed.ac.uk/project/spar/IMG/spar-model-19mar99.gif near the 
end of this document gives a quick overview.


The roots of SPAR in terms of earlier plan and process representations, and 
its relationship to PIF and NIST PSL which are suggested as the basis for 
the SWS process language is documented in a paper from the Knowledge 
Engineering Review at
http://www.aiai.ed.ac.uk/project/spar/DOCS/spar-roots.html

Austin

P.S. I am acting as corporate memory for the people who tried hard with 
work on these standards 4 or 5 years ago. it was a lot of work with many 
man years of effort and thousands of messages and heated debates. It would 
be nice to see some of it used in the new SWS world.

Received on Thursday, 12 February 2004 12:49:54 UTC