james mccartney wrote:
> 
> The hard part will be SMP AND real time GC. I know of no good SMP
> and real time garbage collection algorithm. The only one that I've seen
> is in a paper by Herlihy and Moss, but would be very inefficient  since
> it requires making a copy of an object every time it is modified.
> Thus would only seem to be useful for functional languages.

Given your e-mail address, you may already know this- but
Paul Wilson at the University of Texas has papers on the
web about garbage collection algorithms and such... I believe
that his group has been looking at distributed garbage collection.
They are also known for the Texas Persistent Store, for those
interested in object-oriented database additions to Squeak.
( that brings up a whole other issue... orthogonal persistence.
Anyone making progress on that? ). Anyway, his web page is at:

	http://www.cs.utexas.edu/users/wilson/

les