Is natural selection an algorithm that is "looking for" humans?
Nope. We are just a happy accident in the history of life.
That inheritance improves the odds slightly was not in question. The point was that there are other unaccounted forces (survival, environmental changes, luck) working against the unaccounted filtering effect, such that you cannot assume progress just because a filtering effect exists.
Given that the evidence supports the fact that this is in fact what happened, that assumption is more than justified.
Do you believe that 3.1 Gbs of human genome is filler? If true, you should be able to irradiate and randomly mutate somewhere around 99% of a human's DNA sequence with no ill effect.
See, this is where your simplified version gets you into trouble. Genes are added to genomes all the time, from gene copying mistakes to interspecies breeding to bacterial gene interchange to endogenous retroviruses. Entire chromosomes are sometimes duplicated. Random events are adding genetic material all the time, and natural selection is winnowing out the non-working combinations.
You assumed that humans evolved from chimpanzees, extrapolated that rate to the past, and found it reasonable that humans evolved from chimpanzees and fish. That's circular reasoning.
Nope. I demonstrated that, if evolutionary theory is true, the rate of change as demonstrated by our closest living relatives is consistent with the amount of time that change had to have happened, which is precisely what you are denying.
Can you support the assumed human mutation rate with a measured human mutation rate?
Sure. Start here:
The average mutation rate was estimated to be ~2.5 × 108 mutations per nucleotide site or 175 mutations per diploid genome per generation
6 million years divided by an average generation time(20 years for humans, 15 for chimpanzees, so 17.5 average for both species) gives us 342,857 generations. This yields 59,999,975 mutations.
Can we reverse engineer all those intermediate organisms? (Can't do that now, but should be possible if evolution is true!)
Can we extrapolate the likely form and location of the intermediate organisms? Sure. That's how we discovered Tiktaalik, for example. The researchers worked from an understanding of the development of the tetrapods and the population distribution in the fossil records and predicted that they would find a proto-tetrapod in Devonian strata in Canada. And they did. As for whether they can be recreated exactly, then no, and there is no reason to expect it to be possible at all. We can compare related organisms to determine which genes were likely conserved or developed independently. That can get us in the ballpark(which is why we can infer a lot about intermediate species), but recreating an entire extinct genome is likely never going to be possible.
Filters always remove information. At best, with an analog signal carrying digital encoded information, you can remove the noise without harming the information.
Hey, you are almost there! Congrats!
Filters can be a part of an overall creative process
Bingo! That's what I was saying!
A random string generator has the potential to "write" a novel.
Indeed. And I have generated the first paragraph of "Romeo & Julliette" using one that starts with a small string of random gibberish and uses random mutations and selection to generate it. Of course, the couplet is just a model used to represent a theoretical genome which is fit to survive in a particular environment.
Having solved that simpler problem, does natural selection really behave like the filter you built?
Natural selection works on a higher level, so to speak. Genes live or die in the organisms which they build. So, while the filters represent a high level view of how natural selection works, it would take a few more steps to make it more realistic: interpreting the "genome," building an "organism" from that genome and allowing it to compete in a simulated environment. This is, coincidentally, one of my current personal projects.
Every organism has a sequence: be born, mature, survive, reproduce. That is a meaning that natural selection has to find, even though it doesn't know how to look for it.
OK, let's go with that, noting that we are ignoring a large part of the organisms on earth. Most catastrophic genetic combinations are selecting against before birth. If an organism has a problem with low level protein synthesis, it probably won't make it through gestation. This process continues throughout the organism's lifetime. If they can't compete for resources, they are less likely to live long enough to pass on their genes. If they can't escape predation, they are less likely to live long enough to pass on their genes. If they can't compete in mating, they are less likely to live long enough to pass on their genes. And so on. Natural selection isn't "looking" for something. There is no "correct" genome which is the goal. There is no goal. There is no intelligence. There is just life, death and adaptation.
Yes, I get that you're saying it. The math you have offered is insufficient to support it though. Using simplified models that assume things in your favor is not how you "proof of concept" the numbers.
Funny, coming from someone who started with a ludicrous "tornado in a junkyard" style argument to pick apart a straw man representation of evolutionary theory.
"It works in theory" does not mean "it works in practice".
You have it backwards. We saw that it worked in practice, and constructed a theory to fit the evidence.
On the flip side, I can "disproof of concept" with a model that assumes in your favor; if you can't do it in favorable conditions, you obviously can't do it in disfavorable conditions.
No, your model disproves a model of evolution which exists only in your head.
All barbs aside, I would say that you should seriously consider taking some courses in population biology and evolutionary biology.