Quote Originally Posted by veti View Post
Is there any such thing as a "correct" theory? The only thing we can test is whether the theory is "good enough" for whatever practical test we can put it to.
I use this kind of caveat when talking about artificial universes induced by a theory to head off responses of the form 'but what if that's wrong, what if the universe isn't like that?'. In this case, that would distract from the point that you can have a physics in which you can e.g. build a finite approximation to a Turing machine, but for which something like the Halting Problem isn't actually an issue when it comes to that physics predicting definite things for all initial conditions.

Basically, you can pose questions within a universe whose subject exceeds the bounds of the universe in which they're posed. That doesn't mean that those questions have answers which can be obtained by making further observations of that universe - quite the opposite, really.

If you could come up with a theory that was "correct", i.e. perfect, I guess that particular branch of science - would have ended. I don't know if that's theoretically possible, but it seems unlikely - if a model meets all the tests we can put it to, that just means we need to get more inventive in coming up with new tests. And that might take a while - a couple generations, maybe, to develop the technologies to implement those tests. So even if the theory was perfect, it would still be subject to inquiry. And I don't see any plausible end point for that process.
At a certain point you tend to hit barriers of the form 'no experiment that can be performed within the universe can distinguish between these two theories'. At which point you've got a philosophical choice to make - do you care, even if the only way for the answer to be meaningful is for you to choose to give it meaning?

I think you can extend that more or less infinitely - well, at least to the bound of the total information storage possible using the entirety of the mass energy of the universe - but I would say its up to personal taste whether it still qualifies as science. Basically the same question as 'is mathematics a science?'.