Indeed. Until we can quantify nature's computational power and the computational complexity of evolving solutions to problems in nature, the only sensible respo

Indeed. Until we can quantify nature's computational power and the computational complexity of evolving solutions to problems in nature, the only sensible response to the fact that evolution works extremely well is no response because we lack comparisons. Evolution only surprises us when we falsely (or at least without good reason) regard nature and our own creative abilities to belong to the same reference class. There are also already plenty of clues for why evolution works, e.g. distributed representations, parallelization, composition counteracting combinatorial explosions, selection for structures that evolve faster/better (kind of like learning to learn) etc.