On Sat, 30 Sep 2006, padawan12 wrote:
anything that is Turing computable can be computed on any Turing machine (like a modern Von Neumann computer) but that isn't the same as saying it can compute "anything".
In the real world, there is no machine that can compute as much as any theoretical Turing machine can, because there no such thing as infinite tape.
Beware of the discrepancy between the real world and mainstream mathematics, which is a kind of hall of mirrors, only worse (how many so-called "Real" numbers don't exist?).
Constructivist mathematics were invented in order to prevent problems due to fairy tales such as infinity being taken too seriously. I don't know whether that movement didn't catch because it was too difficult to think about or because it was too realistic; prolly both.
All this to say that I don't care what computability theorists believe it means to "be able to compute anything".
Penfield with Churchland if you want to grapple this issue vis determinism and computability. The contoversy is still powerful in AI research on the equivilence of neural and finite state computing.
What, do humans have an infinite number of neurons? does any neuron in there have an infinite degree of precision? let's forget about neurons: what's the computing power of a soul?
No you could not implement the "internet" in Pd because the internet is not a program, it is a distributed data store.
What? The internet is not a data store.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada