I was watching a documentary based on Brian Greene's "The Elegant Universe" at http://www.pbs.org/wgbh/nova/archive/int_phys.html a while back.
One thing that I've been pondering about is the notion of mathematically sound, but supposedly unfalsifiable hypotheses. Would you say that being mathematically sound is reasonable evidence for belief? I think a while back somebody posted here that there was a mathematical model being published soon that showed that as we go back towards the time of the big bang, the universe never actually gets infinitelly small, and I think it was supposed to make some statements about what form the universe was in "before" that. Again, how is it testable? How is it falsifiable? And how much does it matter? Presumably it is taken more seriously than mere wild speculation, but what do the resident scientists have to say about this kind of thing?