Mike Linksvayer writes:
> Ever since reading Vernor Vinge's "True Names" I've thought a distributed
> artificial intelligence would be one of the neatest things one could do
> with distributed computing (the "Mailman" from the aforementioned story
> was a distributed AI that got its name because it answered questions
> slowly, as if by post).
I don't know about distributed artificial intelligence, but there is a project for distributed artificial life. This is Thomas Ray's Network Tierra, a version of his Tierra alife program which will run on a distributed network across the Internet. Follow the links from:
http://www.hip.atr.co.jp/~ray/tierra/tierra.html
In addition to providing more computing cycles for the simulation, this is intended to create a heterogenous environment, with different workstations having different properties. The organisms are supposed to be able to sense aspects of the various systems they can run on, and perhaps evolve differently on different machines. There could be "ecological niches" analogous to tide pools or caves where specific organisms would dominate.
At least, that's the theory. The project's web pages haven't been updated for a year. They had a lot of technical glitches with the software, and then they ran into balancing problems, where trivial strategies would dominate (often a problem with alife simulations). At last report they were introducing various ad hoc rules and limitations to try to get robust evolutionary behavior.
Hal