Slashdot Mirror
Biomorphic Software
CowboyRobot writes "From the molecular structure of spiders' silk to the efficient use of energy by insects and fish, we can learn many things from Nature and apply them to our engineering tasks.
One thing that nature is particularly good at is the development of dynamic, self-organizing systems.
Ken Lodding is a software engineer at NASA and is currently developing 'swarm algorithms for groups of wind-driven, remote exploratory vehicles'.
He has a six-page article at Queue on 'biologically inspired computing', how to develop 'algorithmic design concepts distilled from biological systems, or processes.'"
After all its just an attempt to reproduce human though and decision making processes in machines.
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
Seems logical to me, especially for multiuser/processor networking. Nature has been "networking" bugs, fish, packs of mammals, etc. for many more years than we've been around. All that extra research time has to count for something. Now that I think about it, a hive of insects are somewhat similar to a group of computers. The individuals posess little (or no) independant thought, only giving responses to electrical or chemical signals. Interesting...
That's right, I read at +2 and post at +1. Not even I care what I have to say.
I have to disagree with the plausability.
The secret weapon they use to kill the rogue swarms of psycho nano cameras is a gunk impurity that got into the STERILE nano-construction area. Like that woudl never occur naturally in non-sterile (i.e Everywhere) areas of the world.
The other thing which got to me was the amount of processing power these nano clouds were assumed to have. A sophisticated predator-prey model that would be CAPABLE of evolving into what those evolved into would need tremendous processing power.
So, lets see, what they would have to have? They'd need high bandwodth that couldn't be jammed (they'd be pretty worthless if you could just turn on a jammer and have them fall apart). They'd need non-volitile memory, because they're solar powered, and if they didn't have it, they'd be stupid again every morning. They'd need a sophisticated distributed processing alogrythm with massive failure tolerance and freakishly complex load balancing (this is more possible than most of it). And beyond all this, they'd need to be able to be microscopic flying cameras that could kill people.
In biological terms, most species have a "specialization". Which means that most species have ONE thing that they do really well. Birds aren't too smart because flying is hard to do. Same with cheetas, because running that fast requires really specific evolution.
Those little nano-bots would have to do the thing they're specialized by the design to do...And everything else as well. Christ, he's got them mimicing human behavior by the end! That is such an incredible stretch! I love sci-fi, but that book had me sneering almost from the very beginning.
ad logicam Claiming a proposition is false because it was presented as the conclusion of a fallacious argument.
Your entire nervous system works like this, as do things like active transport protiens in your cells, the majority of organelles in your cells. A whole lot of nature follows, "The individuals posess little (or no) independant thought, only giving responses to electrical or chemical signals" plan. The interesting stuff comes from emergent properties which still seem to baffle scientists. For example, your brain is a collection of basically binary gates - few than are in current CPUs - and yet we (and several other animals) exibit fascinating emergent properties like emotion, abstract thought, and pooping, that computers don't have yet. I think figuring out the mechanism behind emergent properties (besides saying, "oh, well...uhh...there's a bunch of things interacting...and this happens because they're...uh...interacting") will really propel biocomputing. Hopefully whatever engineers implement the science have an eye for ethics, as well.
Except for the fact that "genetic" algorithms only sometimes produce better results. Often times, they fall short of well designed algorithms. Genetic algorithms are often a fascinating curiosity rather than something useful.
I have not read the book, but I think the grand parent have a good point. If you view the collective as one object, the complexity of that object greatly depends on the capability of its components to communicate, differentiate in task, retain memory...etc.. all of which requires very tight binding between the components. Human isn't a collection of very simple cells, we have very differentiated cells. While all cells have the same DNA master plan; they communicate with each other via mulplitude of complex physical, chemical and electrical channels. The most amazing organ we have, the brain, is highly organized with very tight and complex binding. Loosing the communication (the white matter) inside the brain, as in Alzheimer, and the brain is dead. Cellular automata also have very tight binding requirement. Very simple rule, but very tight binding nonetheless, enforced by the environment code.
Birds aren't too smart because flying is hard to do.
This doesn't make any sense no matter how many times I read it.
First of, birds are the most intelligent animals after mamals. Flying for a bird is no more difficult then running for a human. Despite their small brains, birds learn to fly way faster then humans learn to walk. Insects also fly and they are definetly dumber then birds. I can make a paper airplane fly and it has no brain power at all. Basic auto pilot on a light aircraft have about as much processing power as a pocket calculator. I've learned to fly airplanes and don't think I've become dumber in the process.
There is almost no correlation between flying and intelligence or processing power. Any correlation that does exist would be positive, not negative.