>>28
Sure, you can never truly 'know' what fundamental reality is, but that shouldn't stop you from making theories and betting on those you think are more likely (and meta-tools/heuristics like Occam's razor are good for helping you pick what theories are more probable). In that sense, it's only correct to be agnostic about most theories, however in the end, you have to bet on something if you're going to make any use of any theory, and there's always a risk that you'll be wrong.
After all, for all we know, the laws of physics could all break in 10 minutes, with objects soaring up into the sky, and dissolving into rainbows. You wont be able to confirm that it wont happen until ten minutes from now. You could say that such an event isn't very likely, as it has never happened in the past, and if you restrict yourself to a model of reality, you could say that such an event is impossible. But the model could lead to an incorrect conclusion, and just because it has never happened before in recorded history doesn't mean it can't happen in ten minutes.
I would say that is wildly improbable, but I would never claim it to be impossible. If that were to happen to me, and that I would still be conscious of these events (even less probable, consider the Anthropic Principle which states that (local) reality can't be anything that doesn't support your own existence, that is, if there are structures which don't support your own existence, you're just not going to experience them), I would try to make up hypotheses as to what is happening, my initial one would be that I might be going insane or that I'm just dreaming or hallucinating.
The only thing I will claim as impossible are concepts which contain contradictions within themselves, that is, logical impossibilities.