I note once again that this is an abandonment of your original claim, which was that we know a mind can produce de novo proteins.
We know a mind can produce information. So what? We know that non-minds can also produce information. Thus far you have provided absolutely no evidence that a mind is a sufficient cause for the data under discussion, still less a necessary cause.
I see two troubles here. The first is the “linear functional information”, whatever that means, almost certainly exists in any environment capable of sustaining life. The second is the requirement to finish - perhaps necessary for a program but irrelevant to life.
Looks like an interesting discussion at TSZ tho - I will review. 
Linear functional information examples:
2 and 3 have been empirically observed to contain new biological information produced by known evolutionary processes. Again what’s the problem?
How can you possibly identify this type of information? You know it’s a phone number and information after the fact. You couldn’t tell if it was a number or random string of numbers. So it all seems pretty subjective and useless.
Can you tell us how RV + NS can account for ATP synthase?
We are not going to have a mutation-by-mutation pathway for every single protein over the last 4 billion years, and it isn’t required. What we do have is plenty of evidence for many genes that carry the signature of RV + NS, including the evidence I already supplied:
We could also talk about phylogenetic signal, divergence of intron and exon sequence, and Ka/Ks ratios if you like. We find the signature of evolution throughout genomes.
My alter-ego has posted, but it looks like I’ve already missed the fun bits.
I really hope I never lose that thread.
I’d be interested in learning about some of this.
No, it doesn’t do that at all. It incrementally mutates a sequence at random, and selects mutants that are more similar to the target, as the basis for further mutation and selection. It doesn’t “simulate” any “linear functional sequences” at all.
Note, however, that there are other types of programs that more accurately simulate evolution, but which do NOT need any targets to evolve functional sequences. Such as Avida. And Boxcar2D.
These programs are more realistic, because in these programs it is the phenotypic effects of the mutations that determine their fitness, not because they’re being compared to some pre-specified target.
It doesn’t matter that Dawkins chose to design his WEASEL program that way, when there are other more realistic programs out there. There is no need to set any targets to have evolution result in rare and unlikely functional sequences of information.
You’ve asked this question before, and you were provided answers to it. You didn’t respond in any substantive way back then. So what happened, you hit some sort of cognitive reset meanwhile?
So now I want to ask you back: Can you tell hos how ATP synthase was designed, and what is your evidence for that, and how is the explanation I (really don’t) expect to see from you not just some just-so story you made up?
They are not realistic at all. Cellular changes are observed and required for phenotypic changes. All living diversity is driven by changes to cellular components.
They are much, much more realistic than Dawkins’ WEASEL program. For the reasons I stated. They actually simulate functions, and those functions affect fitness.
Cellular changes are observed and required for phenotypic changes.
No, cells are not the only things that have phenotypes. Even single molecules have phenotypes. The lifeforms in Avida have phenotypes. They have genes that encode instructions, and those instructions have fitness effects in the simulated environment. Fitness effects that are not pre-specified, but are emergent from the interactions between simulated entities.
The cars in the Boxcar2D simulation also have phenotypes. In particular in Boxcar2D, there is an actual physics simulation going on. With gravity, friction, and momentum. The specific attributes such as size, weight, shape, and so on of the car components each contribute to the car’s final performance.
All living diversity is driven by changes to cellular components.
The life known on Earth is cellular, sure. So what? That changes nothing about what I said. Both Avida and Boxcar2d are way more biologically and physically realistic simulations than Dawkins’ WEASEL program is, and both of them function without any pre-specified targets, and can evolve a priori unlikely complex genetic sequences, on the basis of the differential reproductive success of mutants.
That’s a rather meaningless claim given you have shown no idea how the reality of actual biological evolution works.
Even worse is, if the degree of realism is important to Bill, then why is he even considering Dawkins’ WEASEL program when there are so many other, clearly much more realistic programs out there?
None of them might be COMPLETELY realistic, but on that question of realism it seems Bill wants to focus on one of the worst and pretend it’s the pinnacle of evolution simulations out there. The only reason he likes it is that it has a target. That way he can pretend this is necessary for evolution to “work”.
But there are no targets in real life. The bacteria in Lenski’s flask are not being selected towards any target by anyone. They are competing, and the phenotypic effects of mutations determine what sequences are more likely to dominate the population in coming generations. Obviously, computer programs that more accurately simulate this demonstrable reality should be considered instead. Programs such as Avida does that.
Because of his religious beliefs Bill knows, just knows somehow, someway, evolutionary theory must be wrong. He picks strawman targets because they are the only ones he can argue against. Keep in mind his goal is not to convince others his beliefs are true, it’s to convince himself.
Why don’t programs like Boxcar and Avida simulate novel gene formulation without a target?
Because that’s not what they were designed to produce. Why do you ignore so much evidence that evolutionary processes can produce new information and new features?
That’s an important point to remember. It reminds me of the Birthday Problem:
That is a counterintuitive result. Most people would not say that you only need 23 people to have a 50/50 chance of two of those people having the same birthday, but that is what the math shows. The very same probability applies to evolution since you are trying to match a change in phenotype with an increase in fitness. There are many mutations and many targets for adaptive phenotypes, and finding a match between them can occur at much higher rate than trying to calculate the probability of just the one adaptation that does emerge.