-> Actually, it would be very interesting, and probably quite simple, to
-> write a Quixote-based tuple space engine: a bare-bones write/take
-> interface with a pluggable backend (in-memory, ZODB, Berkeley, ...).
-> Would anyone be interested in such a thing?

It's trivial with PostgreSQL; I've written one on top of cucumber, my
O/R mapping for PostgreSQL <--> Python.  I just didn't know it was called
a Linda tuple space ;).

It really is trivial to write, and doing it with e.g. Pyro is probably
the simplest bet.  I may write one in the next few days for some genome
stuff I'm doing; it's a veeerry handy and simple way to write a queueing
system.

cheers,
--titus