At the moment I am enjoying reading Prey by Michael Crichton. I'm a slow book reader (over a decade of web-surfing has changed how I read), so I am only a little way into it (please don't reveal the plot to me).
Anyway - in-between the drama he writes about distributed processing and agent technology along with Nano-tech. I won't pretend that I full understand the science of it all but it facinated me enough that I was unable to sleep thinking about how it all works. Earlier this week I ended up getting up at 2am and reading biology sites and writing code for an hour to emulate an ant colony (one of the examples from the book) in search of food deposits.
The code kind of worked, although I think that I optimised the algorithm to the point that it wasn't interesting to watch (dots and trails on the screen). At the moment the code is spread out on the floor (so to speak) as I try to figure out how learning works.
One of the tips from the book is that you can try to break problems down into simple parts, so I am trying to model some basic instincts and devise a system where these basic instincts can form into sequences or complex behaviours in successful members of the colony (basically those who make it back with food).
Unlike Marvin, I don't have a brain the size of a planet so I have a few questions floating around my head. For example, assuming the evolution geeks are right and we all evolved from single cell organisms, how does a cell splitting effect the genes in the cell? How do single cell organisms evolve into organisms get to a point that they have something called instinct? How do they get to a point where they stop just being cells divding and start producing complete replicas of more complex structures? How do these self replicating structures start requiring binary relationships?
Do I need to get a life, or do I need to do some more reading - hang on - reading is what got me into this mess. Learning is a slippery slope . . .