Posted by Todd Rowland

After just having listened to Michael Schreiber's talk, I was thinking that I'd like to say something to the blogience. It is kind of a neat idea, though it is hard to understand and explain what he is talking about.

As he said in the talk, some people find it a logically impossible dream, while others see it as completely trivial. Given those reactions, it means he is on to something.

Allow me to paraphrase, and then I will say that he wants to find simple descriptions of rules that lie within large rule spaces. From the global context of complex systems, where one might have a complicated rule, but at the same time one wants to use the power and methods of NKS. Then one has to have a simple description, or a compression of the rule. Going even further, if one allows that all natural processes are computations, then one wants to compress the complicated rules in terms of simpler ones.

So that might explain why one might think it is impossible. But he has some ideas of realizing this idea using what he calls formal coordinates. Interesting, but there would need to be a bit more concrete work done. He showed a few examples, but to be more convincing, we'd probably need hundreds or thousands of such examples.