Whilst simulating the universe on an IM might be fun, I suspect that chaos theory would quickly rear its ugly head in any such model. After all, we already know that it's impossible to measure the conditions of this (or any other) universe precisely...
Of course, you could generate an infinite number of universes, some (infinite) subset of which would likely look like ours (although pattern-matching of universes could be a fun problem in and of itself). You then have the problem of choosing one which is both sufficiently like ours and which will remain so for a suitable period of time.
I do wonder if you'd end up with an effect akin to that of using genetic algorithms to program FPGAs. You'd get something that worked, but which was so intricately bound to strange and minute properties of its environment that it would be totally non-portable between universes.
no subject
Of course, you could generate an infinite number of universes, some (infinite) subset of which would likely look like ours (although pattern-matching of universes could be a fun problem in and of itself). You then have the problem of choosing one which is both sufficiently like ours and which will remain so for a suitable period of time.
I do wonder if you'd end up with an effect akin to that of using genetic algorithms to program FPGAs. You'd get something that worked, but which was so intricately bound to strange and minute properties of its environment that it would be totally non-portable between universes.