Responding to Justa's Comments On Evolution

[DogmaHunter]

That is absolutely false. However, I do not believe what mainstream biology asserts without evidence.

No, it's absolutely true.

The theory of evolution, as presented by mainstream biology, is a natural process that does not look forward. It's random input, followed by a natural filter (= survive and mate successfully) resulting in non-random output.

No "mind", no "purpose", no "planning", no "intention".
Instead, just what naturally happens to systems that compete for limited resources and reproduce with variation.

 

[DogmaHunter]

i have no need to.
what i posted from koonin came from the papers he himself wrote.
somehow, and don't ask me why, the cadet says i misrepresent koonin when i post his words verbatum.

Because you do. As the Cadet has proven. In great detail. For crying out loud, he even mailed the dude to ask for clarification and posted the response here for all to see...

 

[DogmaHunter]

Congratulations on completely missing the point over and over and over again.

Just for reference, here's the #1 on the "Best" page from boxcar2d for the "Speedway" course:

And here's a screenshot of the car I have after some 106 generations:

View attachment 161638
(Which, by the way, beats that other car's time by .06 seconds.)

Do you really not understand the significance of this? What this algorithm is doing is simulating the forces of evolution. Go into the program and look at the first few generations. You'll see complete messes of polygons, wheels buckled on every-which-way, absolutely nothing of meaning or structure. Sooner or later, more or less by chance, you'll happen upon something which can actually roll, and in the following reproduction, that organism will be favored. And after a while, you see organisms evolving which are not only well-adapted as cars, but which are well-adapted to the specific environment that they're in, be that environment a straight track, a parabolic curve, a sine wave that's gradually getting tighter and tighter, or just a random clutter of lines that gradually get harder and harder to navigate.

It doesn't matter that evolution does not produce cars. It doesn't matter that the program takes "already existing material". What matters is that we have a random process with no inherent design producing what appears to be clearly designed organisms. And yes, you could object that "the computer program is designed". Fair enough. However, if we could build a 1:1 simulation of the universe that perfectly hindcasts and forecasts all of the earth's history, it would surely be a valuable source of information on how natural processes work... But it would be a computer program. Designed by humans. Does that mean that it is therefore evidence that the tides were intelligently steered? That you ate that burger because an intelligent being made you?

The objections you raise either completely miss the point of the argument or are really, really wrong. Boxcar2d shows without a doubt that evolutionary mechanisms can produce a subjective appearance of design with no trouble whatsoever. Is it a perfect parallel for evolution? It doesn't have to be; we're talking strictly about the appearance of design. It provides a disproof that isn't even necessary, because your "evidence" doesn't hold up to even the slightest scrutiny anyways!

Whois, I'm going for ice cream and I will get to you next.

Well said, though I fear it will fall on deaf ears again.

Fun side fact about the boxcar2d site... You can take your car design (isn't that funny, once?) and import it unto a completely different track.

See, after a while, your car on that site will achieve, what we call in the GA field, a "local optimum". Meaning that the design of the thing being evolved has become so good that it can't really be improved upon anymore through the capabilities of the process being used.

So, for all generations after that, the designs will remain the same, since the "best" solution as already been obtained. This can happen in GA's because, typically, the organisms being evolved will undergo their fitness test against a static environment.

In this case, that static environment is the track.
In the real world though, the environment (in general, thus including other organism's migration paths etc) changes all the time.

This can be done on the site by "migrating" your cars to a new track.

The design that achieved the local optimum, now no longer finds itself in that state.
Now, the evolutionary algorithm will start to tinker once more with the design of the car and adapt it until this design once more reaches the local optimum of the NEW track.

 

[DogmaHunter]

Don't you even understand that the program is intelligently designed. So you are back to explaining design that you don't see. Amazing.

Freezers are designed to produce a freezing environment so that we can make ice in our tropic homes.

Therefor, the north pole is "unnaturally" cold and "designed" for the purpose of ice to form.

Hard to argue such solid logic, ey?

 

[The Cadet]

The boxcar2d example is an evolutionary process which starts with random polygons at generation 0 and results in highly efficient designs of cars driving the track, very successfully, after a couple of dozen iterations.

No "intelligent" interventions required. Just a "mindless, directionless, purposeless" process of "mutate, survive, reproduce".

This puts it much better than I did, honestly.

 

[whois]

In this paper some 45 years ago, he seems to argue the opposite point. That said, I don't have the fulltext, merely the first two pages, so...
In his seminal book, he speaks of exactly this issue, but I'm not sure you understand the point. He brings it up right at the start, and makes it clear that while there is no reason to expect an increase in complexity from an evolutionary standpoint (after all, bacteria today are not that much more complex than they were millions of years ago), he clearly holds that it did happen.

here is the paper i referred to concerning smith:
www.researchgate.net/publication/15314671_The_major_evolutionary_transitions
this was apparently published in nature in 1995.

"No, see, this core tenet of evolution is wrong, and here's this scientist who said so". This is a dance we've done a million times before. So we investigate. We don't just immediately cry foul, we look for the source and investigate the context!

this is where you try to interject a strawman.
in the link on my profile, koonin clearly states that darwinism and its refurbished form (the modern synthesis) belongs in a museum.
he says exactly NOTHING about evolution.
in his paper "the origin at 150" he again states ALL of the tenets of the modern synthesis has been either overturned or replaced.
it is not me saying this stuff, it is koonin.

I must have missed that. Send me a post via PM that I should send to him and I will gladly send it to him, and to you as a CC if you care to give me your email address, just to prove I did it.

sorry, you are going to have to purchase the article and send it to him and ask him if he said it.
i am not going to get banned over this.
this whole deal is ludicrous, why you are getting so stupid over this, i simply do not know.

It's an exciting time to be involved in the field of biology. Wanna bet that, 50 years from now, dinosaurs are still the direct ancestors of modern birds?

maybe.
the classification of life has somewhat subjective boundrys.
did you read the excerpt i provided on the mismatch between species and gene trees?

 

[The Cadet]

Purpose of the cell:
Re: What exactly is the purpose of a cell?
Date: Mon Nov 16 21:42:11 1998
Posted By: Jagesh Shah, Grad student, MEDICAL ENGINEERING, Massachusetts Inst. of Tech.

This, while interesting, commits what I feel is a rather dubious equivocation. When we design something, it has a purpose in the sense that we mean it to do something. We have a goal in mind when we design. We want to achieve something through our design. This paints "purpose" with a far broader brush, seeming to imply that the purpose of life is to live. Well that's interesting, but going back to how "purpose" is used in terms of design, what purpose would imply in that context is that life has some set purpose from a designer. And simply hearing about the function of various lifeforms and molecular machines therein gets you no closer to demonstrating that purpose. Because at the end of the day, what this guy is describing as "purpose" really boils down to function. The cell is not meant to do this, the cell does this. It's not purpose in the sense of design.

I believe the term is "called it".

How does this keep happening? (I'll give you a hint: it involves creationism being really, really wrong.)

 

[Oncedeceived]

I have given you countless examples of how evolutionary processes will inevitably produce things that would appear "designed". So much so in fact, that it's perfectly reasonable to say that the evolutionary process results in "design optimisation".

You brushed each and everyone of those examples aside with rather juvenile, and downright ignorant, one-liners.

Why would I put in any effort again? I know the outcome will be the same:

I gave you a very, very, very simple example of an evolutionary process producing actual designs of cars, while starting with a collection of non-designed and random polygons.


Your reaction was phenomenally ignorant.
Several people, including myself, have then explained to you why your reaction was phenomenally ignorant.

Your response?
Well.... the picture above is quite accurate.

No, sorry, I'm not going to put in that effort again.
Clearly, it doesn't matter at all what I say. You have made up your mind and nothing is going to convince you otherwise.

You say that evolution can't produce designs.
I give you a link of an evolutionary process doing exactly that.

And it doesn't make any difference to you. You simply continue to assert that evolution can't produce designs.

Well then..... yea.... Nothing much left to say at that point.

Ok lets get into the specifics that you are either unaware of or ignore or know about and want to pretend this shows exactly what you are wanting it to show.

I want to show you that your accusation of being phenomenally ignorant is absolutely unfounded and that it is my understanding of biological evolution that makes this "example" of said process a misrepresentation of that process due to factors that are not programmed into the models.

First of all, the representative of the genome is artificially small and only does one thing. The smallest real biological genome which is a parasite (depends on its host for many of its own needs) that has 0.5 million base pairs with several hundred proteins coded. The artificial program even if it could hold this million bits of information it would be only equivalent to possibly one small enzyme. IF it could be produced it would be with artificial mutation rates, artificial generation times, and other factors as well.

Another element that is ignored in these models is that in real organisms, mutations occur throughout the genome, they are not just in a gene or section that specifies a given trait. What this means then is that all the deleterious changes to other traits have to be eliminated along with selecting for the rare productive or advantageous changes in the trait being selected for. Add to that that multiple coding genes are ignored. There are systems that are co-dependent on each other which do not function if all the parts are not present at the same time, which is not reflected in the models. The information which is generated from the program does not reflect de novo generated information.

The biggest and most important element in this type of program is that the outcome is pre-set or pre-ordained. NO pre-set and intelligently designed program can simulate a purely undirected, unguided, mindless process which has no goals or plans and has no way to provide a system where the "designs" become non-functional due to harmful mutations. These programs do not represent the true biological evolution of living organisms.

 

[Oncedeceived]

I believe the term is "called it".

View attachment 161642

How does this keep happening? (I'll give you a hint: it involves creationism being really, really wrong.)

What? Do you really think that this guy thought it was designed with a purpose? I never once considered that is what he meant at all. You really misrepresented me with this. Purpose is instilled in all live forms. He can claim that the purpose is there due to evolution but that still shows that purpose is indeed in organisms and that was my point.

 

[Oncedeceived]

No, it's absolutely true.

The theory of evolution, as presented by mainstream biology, is a natural process that does not look forward. It's random input, followed by a natural filter (= survive and mate successfully) resulting in non-random output.

No "mind", no "purpose", no "planning", no "intention".
Instead, just what naturally happens to systems that compete for limited resources and reproduce with variation.

How would you conclude that there is no mind, no purpose, no planning and no intention? That is simple assertion.

 

[RickG]

How would you conclude that there is no mind, no purpose, no planning and no intention? That is simple assertion.

How would you assert that there was?

 

[Oncedeceived]

How would you assert that there was?

If you can't determine how you would assert such a concept than how can you assert that it is a mindless, unguided, unplanned process with no goals or plan?

 

[RickG]

If you can't determine how you would assert such a concept than how can you assert that it is a mindless, unguided, unplanned process with no goals or plan?

It involves two primary things, first genetic mutations, second environmental adaptation. These are physically seen. As for some guidance for the process, I see no physical evidence. Until I do......

 

[Oncedeceived]

It involves two primary things, first genetic mutations, second environmental adaptation. These are physically seen. As for some guidance for the process, I see no physical evidence. Until I do......

How would you physically see it? If organisms are predisposed for adaptation and engineered to respond to environmental pressures how would you see it?

 

[DogmaHunter]

-

Ok lets get into the specifics that you are either unaware of or ignore or know about and want to pretend this shows exactly what you are wanting it to show.

There is no pretending here.

Generation 0 is a random cluster of polygons with no "design" whatsoever.

X generations later, through the very very simple process of evolution, we have neat designs of cars optimised for successfully completing the track.

That's literally the evolutionary process producing "designs".

I don't know how you can resist this point.
It's right there, on your screen (if you leave the site open for a while).

If you are intellectually honest about what genetic algoritms are, I have no clue how you can continue to deny this with a straight face...

I want to show you that your accusation of being phenomenally ignorant is absolutely unfounded and that it is my understanding of biological evolution that makes this "example" of said process a misrepresentation of that process due to factors that are not programmed into the models.

No, it's not a misrepresentation of the evolutionary process.

There is an environment - the track.
There is a fitness test - scoring of a test drive on the track.
There is a selection process based on the fitness test - only the best performing reproduce.
There is a mutation rate - random changes to the individuals during procreation

How is this a misrepresentation of the evolutionary process??

First of all, the representative of the genome is artificially small and only does one thing.

Which doesn't matter at all.
It's a simple example, deliberatly chosen.
I could also cite the example of Boeing who used GA's to optimise their fuel distribution systems - which was ridiculously more complex with a ridiculous amount of variables.

I did that for clarity.

It doesn't matter.

The mechanics and principles are the exact same.

The smallest real biological genome which is a parasite (depends on its host for many of its own needs) that has 0.5 million base pairs with several hundred proteins coded.

....you are talking about a lifeform that has a history of 3.8 billion years of evolution.
Again, try some intellectual honesty.

And again: completely irrelevant.

The point made is that the evolutionary process is more then capable of producing neat and efficient designs without any "intelligent" intervention required.

The artificial program even if it could hold this million bits of information it would be only equivalent to possibly one small enzyme. IF it could be produced it would be with artificial mutation rates, artificial generation times, and other factors as well.

So, do you complain about every single controlled experiment that it is "controlled"?
Because if that's the case, ALL OF SCIENCE goes out the window.

Another element that is ignored in these models is that in real organisms, mutations occur throughout the genome, they are not just in a gene or section that specifies a given trait.

Mutation in the GA algoritm of the site can happen anywhere in the "chromosome". Perhaps you should first inform yourself before spouting obvious nonsense.

What this means then is that all the deleterious changes to other traits have to be eliminated along with selecting for the rare productive or advantageous changes in the trait being selected for. Add to that that multiple coding genes are ignored. There are systems that are co-dependent on each other which do not function if all the parts are not present at the same time, which is not reflected in the models. The information which is generated from the program does not reflect de novo generated information.

It's funny, because this is dead wrong.
Plenty of things in these cars are dependend on multiple things in the chromosome.

For example, the wheels...

Several variables are involved here:
- the angle of the attachment
- the force of the attachment
- the place of the attachment
- the force of the attachment of the polygon holding the wheel to other polygons
- the size of the wheels, which as impact on the force it exerts on the attachment

These 5 things need to be in balance or else the wheel comes off during driving.
I believe your hero Behe calls this "irreducible complexity". Remove/change one of them and the car no longer drives.

Yet, the first generation doesn't drive at all.
But generation 100 drives very very successfully.

The biggest and most important element in this type of program is that the outcome is pre-set or pre-ordained.

NO, it is really really not.

As said, it starts with a random cluster of polygons.

There is NOTHING in the code that "pre-sets" ANYTHING.

However, it is true that successfully driving things are inevitable.
Not because it has been "pre-set", but rather because that is what evolution does: it optimises systems to pass their fitness test until a local optimum is reached.

You are talking to a person who implemented genetic algoritms in a professional setting.

If it was known before hand what it would result in..... guess what.... nobody would pay me to write the algoritm!!!!

NO pre-set and intelligently designed program can simulate a purely undirected, unguided, mindless process which has no goals or plans and has no way to provide a system where the "designs" become non-functional due to harmful mutations

Then how come the boxcar2d is doing exactly that, along with every other implementation of GA's?????

 

[DogmaHunter]

How would you conclude that there is no mind, no purpose, no planning and no intention? That is simple assertion.

boxcar2d.

No mind, no purpose, no direction, no goal, no intention, no intelligent intervention.

Just mutate, survive, reproduce, repeat.

 

[Oncedeceived]

There is no pretending here.

Generation 0 is a random cluster of polygons with no "design" whatsoever.

X generations later, through the very very simple process of evolution, we have neat designs of cars optimised for successfully completing the track.

That's literally the evolutionary process producing "designs".

I don't know how you can resist this point.
It's right there, on your screen (if you leave the site open for a while).

If you are intellectually honest about what genetic algoritms are, I have no clue how you can continue to deny this with a straight face...



No, it's not a misrepresentation of the evolutionary process.

There is an environment - the track.
There is a fitness test - scoring of a test drive on the track.
There is a selection process based on the fitness test - only the best performing reproduce.
There is a mutation rate - random changes to the individuals during procreation

How is this a misrepresentation of the evolutionary process??



Which doesn't matter at all.
It's a simple example, deliberatly chosen.
I could also cite the example of Boeing who used GA's to optimise their fuel distribution systems - which was ridiculously more complex with a ridiculous amount of variables.

I did that for clarity.

It doesn't matter.

The mechanics and principles are the exact same.



....you are talking about a lifeform that has a history of 3.8 billion years of evolution.
Again, try some intellectual honesty.

And again: completely irrelevant.

The point made is that the evolutionary process is more then capable of producing neat and efficient designs without any "intelligent" intervention required.



So, do you complain about every single controlled experiment that it is "controlled"?
Because if that's the case, ALL OF SCIENCE goes out the window.



Mutation in the GA algoritm of the site can happen anywhere in the "chromosome". Perhaps you should first inform yourself before spouting obvious nonsense.





It's funny, because this is dead wrong.
Plenty of things in these cars are dependend on multiple things in the chromosome.

For example, the wheels...

Several variables are involved here:
- the angle of the attachment
- the force of the attachment
- the place of the attachment
- the force of the attachment of the polygon holding the wheel to other polygons

These 4 things need to be in balance or else the wheel comes off during driving.
I believe your hero Behe calls this "irreducible complexity". Remove one of them and the car no longer drives.

Yet, the first generation doesn't drive at all.
But generation 100 drives very very successfully.



NO, it is really really not.

As said, it starts with a random cluster of polygons.

There is NOTHING in the code that "pre-sets" ANYTHING.

However, it is true that successfully driving things are inevitable.
Not because it has been "pre-set", but rather because that is what evolution does: it optimises systems to pass their fitness test until a local optimum is reached.

You are talking to a person who implemented genetic algoritms in a professional setting.

If it was known before hand what it would result in..... guess what.... nobody would pay me to write the algoritm!!!!





Then how come the boxcar2d is doing exactly that, along with every other implementation of GA's?????

Can you address the points I made?

 

[Oncedeceived]

boxcar2d.

No mind, no purpose, no direction, no goal, no intention, no intelligent intervention.

Just mutate, survive, reproduce, repeat.

You have a very simplistic view of a very complicated process.

 

[DogmaHunter]

Can you address the points I made?

I quoted your entire post and responded to everything.

I have no idea what points you think I missed....

 

[florida2]

Have you researched it?

In detail? No. Is that relevant? No.

Justlooking asked a question, I provided a link for starting point of some research (which I thought was actually doing something nice). He was the one coming to the forum asking a question that he could do research about.