Status of 2.0

Greetings. A quick progress report on LISA development.

Work on 2.0 is proceeding. The reference rete implementation is being
rewritten, from the ground up. It's still basically CLOS-based but the new
version makes extensive use of closures and function objects. An immediate
benefit from this approach is an extremely simplified code base; the entire
messy binding mechanism of 1.x is gone (!), along with the multitude of
little slot classes that were used by the 1.x compiler. The network itself
is also simpler. Pattern-node sharing is in place, but I've no plans to
introduce join-node sharing (at least in the reference implementation). As
of today, the 2.0 algorithm is mostly complete, processing generic, negated
and test patterns. Simple test rules seem to function correctly. And, I
finally grok rete.

The project has picked up its second developer. Dean Allemang joined the
effort a few weeks ago. His credentials are on the sourceforge site so I
won't go into details here; lots of Lisp and AI experience make him a
valuable asset. I should learn much from him. Dean will probably be working
on truth maintenance initially, then perhaps some overall performance
improvements.

There will be a 2.0 alpha release some day; don't know when yet. That's it
for now.

Cheers,
--
------------------------------------------
David E. Young
[email protected]
http://lisa.sourceforge.net

"Those who expect to reap the blessings of liberty
 must undergo the fatigues of supporting it."
  -- Thomas Paine

"But all the world understands my language."
  -- Franz Joseph Haydn (1732-1809)



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf