Comments on Gpuccio: Functional Information Methodology

[Art (Arthur Hunt)]

@Dan_Eastwood

Likewise, any thermodynamic system will have the same sort of complexity.
I anticipate the claim that a thunderstorm has no function.

But it has a clear, obvious specification. Which provides the function for the individual water molecules.

[colewd (Bill Cole)]

@Dan_Eastwood

It was a long setup, but we finally got there.

There are a couple issues with Josh’s analogy. He is not looking at a translated sequence. He is also not looking at a changing or potentially changing configuration over time.

This change allows you to look at comparative data and make a judgement about cause of the configuration.

In much the same way the observation of 500 heads is significant in that we have something to compare it to. That is the normal statistical expectation of the quantity of heads and tails after tossing a coin 500 times.

[Faizal_Ali]

It appears you do not understand the argument @gpuccio is trying to make.

[colewd (Bill Cole)]

@Faizal_Ali

t appears you do not understand the argument @gpuccio is trying to make.

I just posted my argument on UD let’s see the response. How do you understand his argument as being different?

[glipsnort (Steve Schaffner)]

@Rumraket

This wall of substance-less but technical-sounding fog left me still wondering what it has to do with evolution or the evolution of protein coding genes.

Or with @gpuccio’s argument, which makes no reference to these claims about information. (I note that a google search for “physics of symbol systems”, quotation marks included, returns only three hits, two of them to Uncommon Descent. )

[Joe_Felsenstein]

I believe that if a protein family originates by translation of a piece of random DNA that previously had gone untranslated, the resulting protein sequences would be a family isolated in protein sequence space.

[Faizal_Ali]

So, from the ID perspective, wouldn’t that be a functional protein that arose de novo, the kind of thing they say can only happen by “design”?

[Giltil (Gilbert Thill)]

@Dan_Eastwood

Well for starters, I can use the same argument to prove that it’s impossible (…beyond probabilistic resources…) to flip a coin 500 times

This seems quite a strange statement ! Can you explain?

[Giltil (Gilbert Thill)]

@glipsnort

Every functioning human immune system finds a target with FI>500 bits. The target is a set of antibodies that can effectively respond to hundreds of different pathogens

I don’t think you can say that every human immune system FINDS a target with FI>500 bits in the case the target is a the set of antibodies for the genes coding for these antibodies are already there in every human being.

[Giltil (Gilbert Thill)]

@Timothy_Horton

That also fails because biological systems show such high FI values without being designed

You are guilty of « question begging » here.

@Timothy_Horton

Objects produced with evolutionary algorithms also show the process can increase FI with the amount only being limited by how long the algorithm is allowed to run.

ID theorists have shown that evolutionary algorithms that are able to produce high FI can do the trick only when active information is smuggled in the system by the algorithms’s designers.

[Ashwin_s]

@Timothy_Horton

Objects produced with evolutionary algorithms also show the process can increase FI with the amount only being limited by how long the algorithm is allowed to run.

This is not true. EAs give diminishing returns over time and stagnate over long periods of time. This is established in computer science.
Have you read the article linked you to?

If evolution is like EAs, then as far as we know, there is a limit to the innovation possible.

[Giltil (Gilbert Thill)]

@nwrickert

Have you ever considered the chance that vibrating air molecules would assemble a tornado. The probability is really small. Yet it happens often.

If it happens often, then, by definition, the probability isn’t small.

[Giltil (Gilbert Thill)]

@Rumraket

Yes, I guess that "the chance that a tornado sweeping through a junkyard might assemble a Boeing 747 » is not null. But we would rather not consider this type hypothesis if we are to remain in the domain of science!

We have to consider all hypotheses and evaluate their relative likelihoods, you can’t just dismiss things out of hand.

If we want to remain sane, we would rather dismiss out of hand that a tornado sweeping through a junkyard might assemble a Boeing 747. If you dispute this point, there is nothing I can do.

[Timothy_Horton]

@Giltil

ID theorists have shown that evolutionary algorithms that are able to produce high FI can do the trick only when active information is smuggled in the system by the algorithms’s designers.

LOL! I wondered how long it would be before someone brought up Dembski’s incredibly lame and completely discredited “smuggled in information” hand wave.

[Timothy_Horton]

@Ashwin_s

If evolution is like EAs, then as far as we know, there is a limit to the innovation possible.

Any specific EA may provide diminishing returns because the environment is usually kept constant. In real life the environment is always changing so evolutionary processes can just keep churning out the new FI. That is established in both computer science and actual biology.

[Timothy_Horton]

@Giltil

If we want to remain sane, we would rather dismiss out of hand that a tornado sweeping through a junkyard might assemble a Boeing 747. If you dispute this point, there is nothing I can do.

Actual evolutionary processes utilize the feedback from selection to produce increasing complexity over time. Where in your 747 junkyard analogy is the feedback from selection?

[Timothy_Horton]

@Giltil

If it happens often, then, by definition, the probability isn’t small.

Proteins have evolved often enough we can safely say the probability isn’t small either.

[Ashwin_s]

@Timothy_Horton

Any specific EA may provide diminishing returns because the environment is usually kept constant. In real life the environment is always changing so evolutionary processes can just keep churning out the new FI. That is established in both computer science and actual biology.

Not established by any empirical studies in computer science.
It is assumed to have happened in biology.

By the way, a constantly changing environment could as well be more of a deterrent to optimisation. In most cases it will be just that. Unless of course the environment change is designed in such a way as to produce the best result. Who told you a constantly changing environment helps in finding more innovative solutions?

The reason why EAs are not as effective is given in the paper I had shared with you. Two of the possible reasons is that evolutionary theory is not developed well enough to generate an algorithm or that Darwinism is just plain wrong.

There are other possibilities. End of the day, since evolutionary theory is probabilistic in nature, there will always be a probability of the right no: accidents coming together again and again.
The fact that no one knows enough to calculate these low probabilities and dismiss the theory is one reason it still stands.

[Giltil (Gilbert Thill)]

@Timothy_Horton

If we want to remain sane, we would rather dismiss out of hand that a tornado sweeping through a junkyard might assemble a Boeing 747. If you dispute this point, there is nothing I can do.

Actual evolutionary processes utilize the feedback from selection to produce increasing complexity over time. Where in your 747 junkyard analogy is the feedback from selection?

It does no good to quote me out of context.