Since the discussion is about biological evolution not of computing per se, you are agreeing with my statement (and Alamo-Girls - and Yockey's!) about the virtual impossibility of creating functional genes.
Biochemistry is computationally probable for the most part, and we can compute specific results with relative ease.
No. You have just gone through a long exegesis on how the computing faculties are strained trying to find a simple change. Unlike with computers which work fast and do not die if they do not find the answer, organisms do not reproduce at megabytes per second. They also die if they get the wrong answer.
Further the computers have been given intelligent directions which cut down the number of tries required to get success. This is not the case in nature.
To change at random a single DNA bit correctly will take numerous tries. This claiming that there are 'pathways' which cut down the chances is not correct because there is no chemical reason for the sequence of DNA. What the 'pathways' do is exclude out of hand a tremendous amount of possible changes, it does not cut down in any way the random tries it takes to achieve those changes. You are indulging in the usual evolutionist fallacy of the future predicting the past. When put this way it is obvious nonsense. When put as 'pathways' determine the outcome, it does not sound as silly but it is the same logic - that what will be successful in the future is the cause for the events in the past.
In short: difficult to compute is utterly unrelated to probability. We aren't just analyzing what happens from a specific known starting point, we are reverse engineering the entire phase space of possible starting points and possible end points.
You are again giving support to my statement above. What you are speaking of is the reverse of how things actually happen. The future does not determine the past (except in the Terminator movies!).
I know of no evolutionist, nor any evolutionary theory, which relies in any way on "the future predicting the past". If someone's argument appears that way to you, I submit that you've misunderstood what they're actually saying and you should give it another look.
When put this way it is obvious nonsense.
Yes, that *would* be nonsense. But that's not what he's saying.
When put as 'pathways' determine the outcome, it does not sound as silly but it is the same logic - that what will be successful in the future is the cause for the events in the past.
I find no statement in any of his posts which can fairly summarized in the way that you have. Could you quote the portion of his post(s) which you feel rely on the "future" being the "cause" of events in the past?
[In short: difficult to compute is utterly unrelated to probability. We aren't just analyzing what happens from a specific known starting point, we are reverse engineering the entire phase space of possible starting points and possible end points.]
You are again giving support to my statement above. What you are speaking of is the reverse of how things actually happen. The future does not determine the past (except in the Terminator movies!).
He most certainly is not. Computing an entire phase space is in no way an exercise in "the future determining the past".