> Evolution is, like Ford, an unsentimental optimizer. I certainly won't argue with this position, but note that evolution is a peephole optimizer, rather than

> Evolution is, like Ford, an unsentimental optimizer. I certainly won't argue with this position, but note that evolution is a peephole optimizer, rather than trying newton's method, much less trying to optimize with a goal in mind. It does seem most people draw the wrong conclusion from the ape-to-man drawing (e.g. https://www.kindpng.com/picc/m/408-4084640_evolution-ape-to-man-theory-of-evolution-illustrated.png https://www.kindpng.com/picc/m/408-4084640_evolution-ape-to-... ) It's a random walk: each set of die rolls is continually tested on local (i.e. contemporary) conditions. You may have superior lifespan but live where people commonly die in floods. Only broad signals (e.g. ambient temperatures ranging between -10 and 40 C) really determine widespread attributes. So what? Well 300 My of human evolution has included feedback loops uncommon and later unknown to other animals: speech, writing, and the social structures they afforded; hygene and other medical intervention; etc, and they do feed back into evolution even at the slow rate at which humans reproduce. Now we can (poorly) do genetic intervention, it's quite possible to subject humans to more rapid evolutionary pressure (most of it likely driven by fashion), with an attempt to move to goal-directed rather than peephole optimization.