Anders Sandberg wrote:
> Well, an entity composed of multi-million light-year positronium pairs
> and faint gravity waves, thinking thoughts over eons in a silent and
> cold universe doesn't strike me as that bizarre. Much weirder things
> are already happening in mathematics :-) Whether this is implementable
> is another question, we need something like a billiard ball computer
> CA example to see if it is feasible according to known laws of
> physics, and then of course arguments for the practical
To make that work you would have to use isolated subatomic particles to substitute for the molecules that no longer exist. Now, I'm not going to say that an SI couldn't make it work, but from a human perspective it seems improbable - it would be like trying to make a gaseous, chemical-based life form that could live in deep space without any kind of protective membrane.
> You seem to be assuming that if you remove the big bang model, you
> will also need to get rid of everything associated with it - including
> the expansion of the universe, apparently the dynamics of spacetime
> and everything else done in cosmology.
No, just one particular feature. The current model holds that there was one big bang, which is now over, and no similar event can ever occur again. This is a very unusual claim suggested by nothing more fundamental than the limits of our current observations. There is no deep, underlying theory explaining why things should be that way - we just accept it because it comes fairly close to fitting the evidence.
The problem is, we don't know anywhere near enough to make that call with any authority. We don't know why the big bang happened. We can't see far enough into space to confirm that it was a singular event. We can't really confirm the model's predictions by other means - when we try, we find that it agrees with our observation in some cases, but not in others.
Any successor theory is going to have to explain a host of problems, from the age of globular clusters to the existence of mega-scale structure (not to mention those supernovae with *increasing* recession velocities). I don't have any idea what such a theory would look like, but I doubt that it will have much resemblance to our current one.
> It is a bit like saying
> "because we don't have a correct formulation of quamtum gravity
> everything we know about quantum mechanics is wrong, so it is
> impossible to speculate on the future of solid state circuits".
We can productively speculate about the future of solid state circuits because we have an underlying theory that can correctly predict the behavior of the matter that the circuits are made of. There is nothing remotely comparable in the field of cosmology. Instead, we have a morass of untestable theories and after-the-fact explanations.
> Besides, I seriously doubt that the big bang theory is in that much
> trouble. So far none of the alternatives seems to get along without
> even weirder epicycles or have strong observational evidence..
One contrary observation disproves a theory.
The big bang/inflation model survives, not because there is no contrary evidence, but because no one has a better idea. When contrary evidence turns up, the typical response is "OK, how can we tinker with the model to make it produce that outcome?". The problem, of course, is that you can make any theory fit any set of observations if you get to do that.
Now, in the absence of a viable alternative, I suppose it makes sense to continue using it as a working hypothesis. We should not, however, pretend that the theory should have the same status as the Standard Model of physics. "Your theory has been disproven more times than mine" is not exactly a glowing testament of reliability.
> Hmm, why are we debating uploading and the singularity on this list?
Because such discussions don't stray nearly as far from our actual knowledge base. Modern physics, chemistry, engineering and computer science provide us with large areas of well-tested knowledge, which we can use to make reasonable projections about technological advances in the near future. The soft sciences aren't as amenable to testing, but they are close enough to ordinary experience that we can reason by analogy with some chance of success. Besides, we mostly only talk about the next century or so - past a certain point, we start saying "I don't know" instead of "IMO, X must be true".
Billy Brown, MCSE+I