Something can be valid, but not explainable by 'known physics'.
Gravity has always been valid, but was not understood until relatively recently.
Splitting the atom has always been a possibility, but the science wasn't known until very recently. The atom, obviously, didn't suddenly gain the ability to be split in the 1940s.
The science behind the micro-processor has always been valid etc.
But, if you use "doesn't comply with known science" as a criterion for ruling something you rule out genuinely valid things that we don't understand yet.
I actually think "isn't too outlandish" is a far easier and more useful criterion to use :-)
