A)Search space for 100 1-bit safes:
Search space for 1 safe: 2
Search space for 2 safes: 4
Search space for 3 safes: 6
…
…
Search space for 100 safes: 200
B) Search space for a 100 bits safe: 2^100
Great. Spell it out, though. What are the 6 possible targets for 3 safes?
2 Likes
Water runs downhill in a blind, mindless and purposeless search to maximize entropy.
GA’s generate randomness in population of test solutions, and then expel excess randomness as determined by some fitness function, resulting in a population of overall higher fitness. GA’s search in random directions, but not all of those searches are allowed to continue. A human programmer may tune this to some useful purpose, but purpose is not a requirement for the GA.
2 Likes
So does that mean FI is not uniquely defined for a given sequence? If a protein can serve two purposes, two functions, then there are two FIs, right? The search space would be the same but the target space would be different. I find it hard to understand how we could use measures of FI as an argument, one way or another, if the first step is to define a singular function (so that we can get an unique FI value). Functions can and do change, right?
3 Likes
As usual this an interesting idea to ponder 
Oy vey! You’re killing me here Bill.
3 Likes
As far as two FI’s I don’t think so. There is one FI that can create a protein with two functions.
Cytochrome C is one of the proteins in the electron transport chain yet is also involved in skull development in mice. Two mission critical processes of animal development and sustainability. The function in this case would be both ATP production and embryo development. This requires a higher level of specificity than for a single function.
OK, I think that makes some sense, but doesn’t it make some of the criticisms about the lack of knowledge of the target space worse? To truly calculate the FI, wouldn’t we need to know every possible function? Or is the actual FI value not interesting to you, but rather you are using it to place something like a lower bound (as in, “the FI is at least X”)?
2 Likes
Yet I can shuffle two standard decks of cards together, deal them out, and the resulting sequence will have a probability of 1.03 x 10^166 which blows right past Dembski’s UPB.
3 Likes
Your point is well taken. Gpuccio and Kurt are estimating FI and its accuracy is assumption dependent. I think Gpuccio’s FI calculation is pretty accurate but that is because I think the design argument is right. If you live in the evolutionary paradigm where lots of function exists and we are observing lucky accidents when we see preservation you will see the estimate differently.
I see such a small window for evolution to work that I have discounted it as a viable explanation and so given the design hypothesis a protein that is preserved over 400 million years shows very high FI as it has to be in that configuration for a healthy functioning animal that can compete in the reproduction battle.
What makes you think that these are 2 separate functions, rather than the role in skull development being a result of its role in the ETC?
Good question. The experiment I am pulling from memory was a single mutation to cytochrome c causing no forehead to development in mice.
Of course not! This is polemics, not science.
I’ll bet that no one who claims that FI is a meaningful thing will choose any such protein
They aren’t trying to measure target space, just claim that it’s in a range that works polemically
It’s a lame argument. Even if we accept its flawed premise, it fails.
Indeed they do. A primary evolutionary mechanism is duplication and diversification, leaving families of proteins that have partially-overlapping functions, something we don’t see in intelligently-designed mechanisms.
Something as simplistic as FI will never address that reality.
1 Like
Good point.
I forgot to add the most important word, ie., the word « specified ». So let me correct:
« according to ID theory, a single SPECIFIED event of that probability (1/10^150), or put differently, a single SPECIFIED event producing 500 bits of FI cannot be attributed to chance. In that case, design is the right explanation ».
Thank you.
So show us the before-the-fact specification for anything in biology. All I see are ID-Creationists making after-the-fact measurements and calling them specifications.
4 Likes
What about this one?
https://images.app.goo.gl/DdazKbnTAAGkuiRAA
So show us the before-the-fact specification for it.
If you look at these two images side by side:
Can you really tell that one was designed and the other was not? I really couldn’t. The one made entirely out of H2O is periodic and complex. When you look at many of them you see that their shapes are not uniform but have common (maybe irreducible even) elements. It looks like a work of art compared to the more utilitarian gears.
As a chemist who has worked on molecular machinery (including design of artificial molecular motors), I can tell you that chemistry is really quite good at producing periodic, often intricate patterns at the microscopic scale. I would not use a picture to determine much more beyond how beautiful nature can be, however it came about.
You have the additional complication that this is a picture, a snapshot. I know from experience that Brownian/thermal motion can easily scramble nice “arrangements” that look so intentionally placed.
And thirdly, you left out the other picture from that article:
All of a sudden the toothed gear looks much less like:
8 Likes
and how is that matter? gears can be made in many different shapes and sizes. but they are still gears. so they are still evidence for design.
It matters because the look of the object was used as the argument.
I really can’t see how. Two snowflakes could act as a gear, that doesn’t prove anything does it?
2 Likes