Should Versus Is

Conundrum: how do you solve a problem, when that problem is defined as outside the bounds of the possible and is therefore innately unsolvable? Do you need to venture beyond the boundaries of Euclidean space-time? Or just re-think your assumptions about the (im)possible?