Quote:
The wikipeida article lays out the various forms of an eye pretty well. The most basic version would be a clump of photosensitive cells that can tell the difference between day and night. Then evolving into a pit with photosensitive cells along the wall that can discern the direction of the sun. Brains aren't always required. The article lists a jellyfish with elaborate eyes, but no brain.
Do you agree with Darwin's quote Thaddeus73 cut in half to make his point, or the full quote that is more responsible?
Imagining binary addition evolution to an OS isn't a terrible analogy for biological evolution. Basic evolutionary steps multiplied by millions of years and billions of generations isn't far off from basic binary addition across billions of bits and operations in a modern processor.
Yes, I can read wikipedia also. I don't think you understood my objection. It isn't about evolving an eye because a very simple organism with the eye in the bottom right would be significantly disadvantaged vs one with the "eyes" it "needs". It would be expending a great deal of "biological energy" (both in the literal sense as in bloodflow and the figurative sense) to sustain a physiological feature that generates little benefit - which is a net loss, it would be less fit or competitive.
That doesn't mean I object to the model. It means I find the model to be not compelling because it doesn't explain nearly enough of the whole system. It's like telling me how our expertise in aircraft developed over time by focusing on wing shape. While no doubt there is significant progress there, utilization of that requires leaps of technology of the system as a whole, in an integrated way. In order to explain the wing shape of a modern aircraft you also have to understand how the engine came about to make that wing shape useful.
The eye is the same way. There's a lot of other things that have to happen simultaneously in order to move from a to f, and in terms of complexity those aren't remotely equal steps. In a-e you have more or less one change happening. In f you've added an intricate muscular focusing mechanism in the iris, a muscular directional mechanism (with its requisite skeletal and vascular support), a new vitreuos feature, a secondary aqueous feature between the lens and the cornea... never mind the complexity of corneal epithelial cells, or the lubrication system - tears, nasolacrimal ducts, lacrimal glands, and all of the host of reflexes and neural patterns to make all that stuff actually work together. And it seems like that all has to progress simultaneously, which makes the individual steps difficult. You get one mutation at a time, but it
seems like that won't work very well... any more than taking a WWI biplane and putting a Merlin engine on it.
The problem with basic binary addition on computers is that random selection sucks. Even with AI which maybe is the closest analogue I can think of (because it's not strictly random given that it has feedback) outcomes are extremely sensitive to how you structure the initial array, feedback conditions, etc. With a "fuzzer" (random input) you can train things, but its not very efficient. And maybe evolution isn't very efficient either.
The biggest objection I have, again as a layman as I'm sure this isn't novel or anything, is that even if you get a beneficial mutation some random chance can preclude that mutation from being passed on. Or it could be precisely the mutation that's required...but the one that you need in two steps, not in this one.
