Complexity Improbability and Design

[Furrowed Brow]

In the debate Winning Life’s Lotteries 4gold made the following point:

Complexity is special because it is a method by which we use to determine whether a phenomenon is random or designed.

Question 1: Is it too improbable to believe that some complex natural phenomena do not require a designer?

Question 2: Are improbability and complexity two measures by which we can validly conclude a designer.

[micatala]

I think you are confusing purpose with function.

I might be out camping and have forgotten or lost my tent pegs. I find some short sticks lying around that are strong enough to 'serve the purpose.' However, these sticks had no 'intended purpose' at all. I have merely been able to employ them so that they can function as tent pegs.

The same is true of animals or plant products we use for food or other 'purposes.' We are simply looking at things from our own anthropromorphic point of view if we tell ourselves the purpose of these things is to feed and cloth us, etc. However, as with my stick tent pegs, just because we have used these things to fulfill a particular function, does not mean that these functions were their intended purposes.

So when you are walking through the woods and you see some 'sticks' holding up a tent, you would conclude these are not manufactured tent pegs but only natural sticks therefore there is no evidence that an intelligent designer has camped out here.
I am not asking anyone to see things according to my point of view, but if I tell you that I can see God everywhere that I look and you say you can't see him anywhere, then either you are blind or I am a liar or I am delusional.

I am a believer in God as well. I certainly consider what we see around us to be ultimately attributable to God. However, I view evolution as part of Gods creation, as one of the mechanisms of His creation, just like the law of gravity, the laws of chemistry, or quantum mechanics.

I am also not saying there is no such thing as design. However, you are infering design where there are perfectly reasonable 'non-design' explanations and no evidence for design other than "I don't see how this could not be intentionally designed". Subjective determinations of what is designed and what are not are not particularly reliable.

Just look at what DNA looks like. We can denoted it by a string of letters

AACTGAGGTCACCTGAGATACCTGT . . .

If you look at this, it looks just like a scrambled deck. It might as well be a shuffled deck (or thousands or millions of decks all shuffled together) where we only keep track of the 4 suits. A 'designed DNA' would look more like

ACTGACTGACTGACTG . .

or

AAAAAAATTTTTTTGGGGGGGCCCCCCC . . . .

In addition to confusing function with purpose, I think ID proponents confuse structure with design.

A snowflake has an incredibly complex structure. Anyone looking at a magnification of a snowflake could be forgiven if they thought it was designed. It is more complex than a lace doily or embroidery done by one of my aunts. However, the structure is the result of completely random processes which are constricted by the simple laws of chemistry.

There is no 'intelligence' or 'design' of a particular snowflake, except perhaps in that the underlying laws of chemistry and physics might reflect the intelligence or planning of a creator.

\
Just to let you know, I haddn't thought about the difference between structure and design. A light bulb went off in my head after this. Thought I would let ya know.

Well, thank you.

Now I can go to bed thinking I did something useful today.

[olavisjo]

I am not asking anyone to see things according to my point of view, but if I tell you that I can see God everywhere that I look and you say you can't see him anywhere, then either you are blind or I am a liar or I am delusional.

*looking at bolded part*, well, no insult intended, but yeah.


You seem to look at the world in a skewed view. The world is not this way because of US, but WE are the way we are BECAUSE of the world. Do you understand what i mean?

No insult taken.
I do understand what what you mean and I would add to it.
The world is not this way because of US, but WE are the way we are BECAUSE of the way God created the world and all the things that he created for US.
A good analogy is music. There is an infinite way to create sound, but if the sound lacks harmony (and all that stuff that musicians put into music which I am clueless about) it becomes nothing but noise. All the music we like can be described by big black dots in sheet music.
It is the same way with DNA, it describes the music of life.
If we could look at the landscape of life, I would imagine a lot of peaks separated by deep valleys.
The peaks represent areas where the DNA combinations are in harmony and convergence, and where a viable life form can exist.
For example, take the cockroach, it is so good at survival that its peak on the evolutionary map is very high and large. So there are a lot of ways that roaches can be manifested, but in order for it to evolve into another type of life form, its DNA needs to descend into the valley separating it from the nearby species. The problem is that it becomes significantly less able to survive in its environment and the intact cockroaches will out compete it for food and other necessities and it will die out and become extinct.
If you have a spreadsheet program, I can tell you how to simulate evolution, it is not as complicated as it sounds and one day when I get some time I will make a computer program that does all the leg work for you but for now you have to do it yourself. If any Java or C++ student reads this, feel free to make a program out of this evolutionary algorithm.
First click in cell A1 (top left cell) and type =SUM(B1:E1)
B1 =INT(G1/1000)
C1 =INT((G1-1000*B1)/100)
D1 =INT((G1-1000*B1-C1*100)/10)
E1=INT(G1-1000*B1-100*C1-10*D1)
F1 leave it blank
G1 =1
H1 =INT(G1+RAND()*99-49)

Then highlight the top 30 rows by dragging your mouse from the top to row 30. (make sure all the cells you entered information into are highlighted)
Then from your edit menu chose 'Fill Down' and all the cells should have either a 1 or 0 and column H should have random numbers.
Column H represents the children of column G with a generous amount of mutations.
Now copy all 30 of the children in column H1-H30 by dragging your mouse and select Edit Copy.
Then click G1 then Edit, Paste Special (you should get a dialog box with 'Values only' selected) then click 'okay'.
Now all of your children have become parents.
But before we allow these children to have young, we will filter them for survival of the fittest.
Each of the digits represents an attribute of the creature, and the stronger that attribute the better its chance of survival. For example a 9 is more fit than a 2.
For animals with more than one digit we add the digits to determine its ability to survive. For example a 99 has a fitness of 18 so it has a better chance of survival than 333 with a fitness sum of 9. It is like saying that the cockroach has a better chance of survival than a dodo bird, even though it appears to be a simpler life form.
(loop) Column A has the sum of all the digits in our new parents, so we select all the cells and click Tools, Sort (you should get a dialog box with 'Sort by' set it to column A, if it is not already there, then click the radio button for 'Descending') then click sort.
Now all the fittest critters are at the top of the spreadsheet.
So now you are going to reproduce only the top 15 of the new children, the other 15 died before reproducing.
With your mouse select H1 to H15 and choose copy.
Then click G1 and paste special, values only.
Then click G16 and paste special, values only.
repeat (loop)
You paste the same values twice since the stronger critters reproduced twice to make up for the ones that died before reproducing.
Then keep repeating the same thing from where it says (loop).
What this algorithm will show you is that a small amount of mutations will not get you to where you want to go, but a large amount of mutations can get you there but it also destroys the good stuff that you need when you get there.
If you have any algorithm that describes your version of evolution I would love to see it.
Like I can imagine a program that picks a note at random and you can click either 'keep' or 'try again'. When you hit try again it will repeat the notes you liked from before and add a different random note at the end, but when you hit keep it will add a new random note to the end of that string of notes, and eventually you will have a nice song created by the 'natural selection' of notes.
When I have time I will make that program too, this is how delusional people spend their spare time.

[Beto]

I am not asking anyone to see things according to my point of view, but if I tell you that I can see God everywhere that I look and you say you can't see him anywhere, then either you are blind or I am a liar or I am delusional.

*looking at bolded part*, well, no insult intended, but yeah.


You seem to look at the world in a skewed view. The world is not this way because of US, but WE are the way we are BECAUSE of the world. Do you understand what i mean?

No insult taken.
I do understand what what you mean and I would add to it.
The world is not this way because of US, but WE are the way we are BECAUSE of the way God created the world and all the things that he created for US.
A good analogy is music. There is an infinite way to create sound, but if the sound lacks harmony (and all that stuff that musicians put into music which I am clueless about) it becomes nothing but noise. All the music we like can be described by big black dots in sheet music.
It is the same way with DNA, it describes the music of life.
If we could look at the landscape of life, I would imagine a lot of peaks separated by deep valleys.
The peaks represent areas where the DNA combinations are in harmony and convergence, and where a viable life form can exist.
For example, take the cockroach, it is so good at survival that its peak on the evolutionary map is very high and large. So there are a lot of ways that roaches can be manifested, but in order for it to evolve into another type of life form, its DNA needs to descend into the valley separating it from the nearby species. The problem is that it becomes significantly less able to survive in its environment and the intact cockroaches will out compete it for food and other necessities and it will die out and become extinct.
If you have a spreadsheet program, I can tell you how to simulate evolution, it is not as complicated as it sounds and one day when I get some time I will make a computer program that does all the leg work for you but for now you have to do it yourself. If any Java or C++ student reads this, feel free to make a program out of this evolutionary algorithm.
First click in cell A1 (top left cell) and type =SUM(B1:E1)
B1 =INT(G1/1000)
C1 =INT((G1-1000*B1)/100)
D1 =INT((G1-1000*B1-C1*100)/10)
E1=INT(G1-1000*B1-100*C1-10*D1)
F1 leave it blank
G1 =1
H1 =INT(G1+RAND()*99-49)

Then highlight the top 30 rows by dragging your mouse from the top to row 30. (make sure all the cells you entered information into are highlighted)
Then from your edit menu chose 'Fill Down' and all the cells should have either a 1 or 0 and column H should have random numbers.
Column H represents the children of column G with a generous amount of mutations.
Now copy all 30 of the children in column H1-H30 by dragging your mouse and select Edit Copy.
Then click G1 then Edit, Paste Special (you should get a dialog box with 'Values only' selected) then click 'okay'.
Now all of your children have become parents.
But before we allow these children to have young, we will filter them for survival of the fittest.
Each of the digits represents an attribute of the creature, and the stronger that attribute the better its chance of survival. For example a 9 is more fit than a 2.
For animals with more than one digit we add the digits to determine its ability to survive. For example a 99 has a fitness of 18 so it has a better chance of survival than 333 with a fitness sum of 9. It is like saying that the cockroach has a better chance of survival than a dodo bird, even though it appears to be a simpler life form.
(loop) Column A has the sum of all the digits in our new parents, so we select all the cells and click Tools, Sort (you should get a dialog box with 'Sort by' set it to column A, if it is not already there, then click the radio button for 'Descending') then click sort.
Now all the fittest critters are at the top of the spreadsheet.
So now you are going to reproduce only the top 15 of the new children, the other 15 died before reproducing.
With your mouse select H1 to H15 and choose copy.
Then click G1 and paste special, values only.
Then click G16 and paste special, values only.
repeat (loop)
You paste the same values twice since the stronger critters reproduced twice to make up for the ones that died before reproducing.
Then keep repeating the same thing from where it says (loop).
What this algorithm will show you is that a small amount of mutations will not get you to where you want to go, but a large amount of mutations can get you there but it also destroys the good stuff that you need when you get there.
If you have any algorithm that describes your version of evolution I would love to see it.
Like I can imagine a program that picks a note at random and you can click either 'keep' or 'try again'. When you hit try again it will repeat the notes you liked from before and add a different random note at the end, but when you hit keep it will add a new random note to the end of that string of notes, and eventually you will have a nice song created by the 'natural selection' of notes.
When I have time I will make that program too, this is how delusional people spend their spare time.

This algorithm has the information that feature A is better than feature B, before they are actually tried out? During biological evolution, features have to be experimented with... or don't they?

[ENIGMA]

If you have a spreadsheet program, I can tell you how to simulate evolution, it is not as complicated as it sounds and one day when I get some time I will make a computer program that does all the leg work for you but for now you have to do it yourself. If any Java or C++ student reads this, feel free to make a program out of this evolutionary algorithm.

I believe that there is a java applet based off of a similar principle currently online:

Weasel Applet

What it does is it generates random strings of characters, scores those strings based upon how closely they match the target string ("Methinks it is like a weasel" by default), and then takes the top scoring results, randomly change letters in those result and use those strings to create the next generation of strings, then repeat. It is a pretty good illustration of how mutation + selection can generate strings that are wildly improbable to occur by chance.

The two main shortfalls of the weasel applet for illustration purposes that I see:

1) It relies on an absolute rather than a comparitve metric for determining success. Such a process need not rely on knowing the absolute value of a given string, but merely needs the ability to determine which ones are better than others (with "By how much?" being a useful but ultimately optional answer).

2) It implies a static fitness landscape. In other words, the big goal is always pointing towards that one phrase and never diverges from that final goal. This is not the case with biological evolution as seen in numerous examples. The major one that comes to mind is the meteorite that landed 65 million years ago that created a cloud of dust that limited the ability of sunlight to reach the earth. This limited the amount of plant growth and thus limited the food supply for the large herbivores (whose location on the fitness landscape sunk like a stone due to the large body mass that they needed to maintain), and consequently the large carnivores (who were running out of large herbivores to eat and thus also had a big decline). Such changes in the fitness landscape can happen catastrophically as well as gradually over time. (The Sahara wasn't always a huge desert you know...)

[Beto]

If you have a spreadsheet program, I can tell you how to simulate evolution, it is not as complicated as it sounds and one day when I get some time I will make a computer program that does all the leg work for you but for now you have to do it yourself. If any Java or C++ student reads this, feel free to make a program out of this evolutionary algorithm.

I believe that there is a java applet based off of a similar principle currently online:

Weasel Applet

What it does is it generates random strings of characters, scores those strings based upon how closely they match the target string ("Methinks it is like a weasel" by default), and then takes the top scoring results, randomly change letters in those result and use those strings to create the next generation of strings, then repeat. It is a pretty good illustration of how mutation + selection can generate strings that are wildly improbable to occur by chance.

The two main shortfalls of the weasel applet for illustration purposes that I see:

1) It relies on an absolute rather than a comparitve metric for determining success. Such a process need not rely on knowing the absolute value of a given string, but merely needs the ability to determine which ones are better than others (with "By how much?" being a useful but ultimately optional answer).

2) It implies a static fitness landscape. In other words, the big goal is always pointing towards that one phrase and never diverges from that final goal. This is not the case with biological evolution as seen in numerous examples. The major one that comes to mind is the meteorite that landed 65 million years ago that created a cloud of dust that limited the ability of sunlight to reach the earth. This limited the amount of plant growth and thus limited the food supply for the large herbivores (whose location on the fitness landscape sunk like a stone due to the large body mass that they needed to maintain), and consequently the large carnivores (who were running out of large herbivores to eat and thus also had a big decline). Such changes in the fitness landscape can happen catastrophically as well as gradually over time. (The Sahara wasn't always a huge desert you know...)

Unfortunately, the weasel program is constantly misrepresented by creationists, who attempt to use it as an argument against biological evolution, despite Dawkins saying way back in 1986, that it was just intended as a tutorial demonstration. The above described algorithm is the same old straw man, and subsequent fallacious argument: "If this particular evolutionary algorithm doesn't accurately simulate biological evolution, than biological evolution must be false".

[olavisjo]

If you have a spreadsheet program, I can tell you how to simulate evolution, it is not as complicated as it sounds and one day when I get some time I will make a computer program that does all the leg work for you but for now you have to do it yourself. If any Java or C++ student reads this, feel free to make a program out of this evolutionary algorithm.

I believe that there is a java applet based off of a similar principle currently online:

Weasel Applet

What it does is it generates random strings of characters, scores those strings based upon how closely they match the target string ("Methinks it is like a weasel" by default), and then takes the top scoring results, randomly change letters in those result and use those strings to create the next generation of strings, then repeat. It is a pretty good illustration of how mutation + selection can generate strings that are wildly improbable to occur by chance.

The two main shortfalls of the weasel applet for illustration purposes that I see:

1) It relies on an absolute rather than a comparative metric for determining success. Such a process need not rely on knowing the absolute value of a given string, but merely needs the ability to determine which ones are better than others (with "By how much?" being a useful but ultimately optional answer).

2) It implies a static fitness landscape. In other words, the big goal is always pointing towards that one phrase and never diverges from that final goal. This is not the case with biological evolution as seen in numerous examples. The major one that comes to mind is the meteorite that landed 65 million years ago that created a cloud of dust that limited the ability of sunlight to reach the earth. This limited the amount of plant growth and thus limited the food supply for the large herbivores (whose location on the fitness landscape sunk like a stone due to the large body mass that they needed to maintain), and consequently the large carnivores (who were running out of large herbivores to eat and thus also had a big decline). Such changes in the fitness landscape can happen catastrophically as well as gradually over time. (The Sahara wasn't always a huge desert you know...)

Thank you very much for the weasel.
And I agree with you about the two shortfalls of this program.
If I understood you correctly, you said...
1. Man was a goal to be achieved.
2. All combinations of letters were viable, but the most viable were the ones that were closest to the goal.
So my simulation is better in that it has neither of those shortfalls.

[The Duke of Vandals]

1. Man was a goal to be achieved.

Evolution doesn't have goals. It just has changes.

Are you familiar with how mountain ranges are formed? By two tektonic plates forcing land upwards where they meet? Stating man is a "goal" in evolution is like saying "The Alps are a goal of plate tektonics." They're not. They're just a type of change.

[olavisjo]

1. Man was a goal to be achieved.

Evolution doesn't have goals. It just has changes.

Are you familiar with how mountain ranges are formed? By two tectonic plates forcing land upwards where they meet? Stating man is a "goal" in evolution is like saying "The Alps are a goal of plate tectonics." They're not. They're just a type of change.

Thank you for agreeing with me.
The weasel program does not simulate evolution, as you know it, because it knows what its goal is, evolution by natural selection does not have a goal, therefore the weasel program does not simulate evolution.
However, it may support intelligent design.

[Goat]

If you have a spreadsheet program, I can tell you how to simulate evolution, it is not as complicated as it sounds and one day when I get some time I will make a computer program that does all the leg work for you but for now you have to do it yourself. If any Java or C++ student reads this, feel free to make a program out of this evolutionary algorithm.

I believe that there is a java applet based off of a similar principle currently online:

Weasel Applet

What it does is it generates random strings of characters, scores those strings based upon how closely they match the target string ("Methinks it is like a weasel" by default), and then takes the top scoring results, randomly change letters in those result and use those strings to create the next generation of strings, then repeat. It is a pretty good illustration of how mutation + selection can generate strings that are wildly improbable to occur by chance.

The two main shortfalls of the weasel applet for illustration purposes that I see:

1) It relies on an absolute rather than a comparative metric for determining success. Such a process need not rely on knowing the absolute value of a given string, but merely needs the ability to determine which ones are better than others (with "By how much?" being a useful but ultimately optional answer).

2) It implies a static fitness landscape. In other words, the big goal is always pointing towards that one phrase and never diverges from that final goal. This is not the case with biological evolution as seen in numerous examples. The major one that comes to mind is the meteorite that landed 65 million years ago that created a cloud of dust that limited the ability of sunlight to reach the earth. This limited the amount of plant growth and thus limited the food supply for the large herbivores (whose location on the fitness landscape sunk like a stone due to the large body mass that they needed to maintain), and consequently the large carnivores (who were running out of large herbivores to eat and thus also had a big decline). Such changes in the fitness landscape can happen catastrophically as well as gradually over time. (The Sahara wasn't always a huge desert you know...)

Thank you very much for the weasel.
And I agree with you about the two shortfalls of this program.
If I understood you correctly, you said...
1. Man was a goal to be achieved.
2. All combinations of letters were viable, but the most viable were the ones that were closest to the goal.
So my simulation is better in that it has neither of those shortfalls.

No, I believe what is said is that creationists assume that Man is a goal to achieve. From an evolutionary standpoint, that is not true.

And, since there is no 'goal' , other than fittingi into the filter, the assumption the 'most viable ones were the ones that were closest to the goal' is invalid, since you are assume a goal, where there isn't any

[olavisjo]

Weasel Applet

No, I believe what is said is that creationists assume that Man is a goal to achieve. From an evolutionary standpoint, that is not true.

And, since there is no 'goal' , other than fitting into the filter, the assumption the 'most viable ones were the ones that were closest to the goal' is invalid, since you are assume a goal, where there isn't any

Are you talking about the Weasel Applet that ENIGMA referenced?
The filter compares the current critter to the goal, and bases its fitness on how closely it maches that goal, which is what all those pesky creationists keep saying.
If you go to the link provided by clicking the 'Weasel Applet' it will give you a link to look at the Java source code and you will see for yourself, that this simulation goes against what you belive.

Here is a snippet...

// Score the supplied (Weasel)Critter.
public int score(Critter c_)
{
WeaselCritter c = (WeaselCritter) c_;

int wrong = 0;
int len = goal.length();
for (int i = 0; i < len; ++i)
if (c.charAt(i) != goal.charAt(i))
++wrong;

return wrong;
}