Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
1 online now:
Newest Member: popoi
Post Volume: Total: 915,817 Year: 3,074/9,624 Month: 919/1,588 Week: 102/223 Day: 0/13 Hour: 0/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Evolutionary Simulators: How accurate are they?
Hyroglyphx
Inactive Member


Message 16 of 31 (431170)
10-29-2007 5:25 PM


A reply to Rrhain from another thread
What makes you think that the Monte Carlo method is only applicable to evolution?
When did I say it was? That was the pretense of Taz's rant, all of which brought up by TAZ, not me. He brought up the supposed misrepresentation of evolution, and it was he that used the MC method as a basis for providing some evidentiary support.
So why are you restricting it to evolution?
I'm not. Go back and follow the dialogue. First tell me how this conversation began, then tell me where I restricted the MC method solely to evolution.
Haven't you done any research into the subject before you dismissed it? The GOLEM project evolved walking, for example.
Rrhain, my whole point is that these simulations are specious by doling out the functions in an almost perfect state by introducing constants. All the digititalized organism has to do is haphazardly run in to another object, and, eureka, evolution before our very eyes. I mean, come on... How lame is that? That hardly resembles actual biological systems. That's like placing you in front of a tennis ball machine and then predicting that you will be hit by one of those tennis balls.
As I mentioned in your other thread, you are insinuating that evolutionary biologists are trying to pull a fast one: That their evolutionary models already have the answers pre-programmed into the system and they're only doing all this computer work to lend an air of legitimacy to their work.
I don't think they are trying to pull a fast one, or that the answers (I assume you mean the end product) are pre-programmed. What I am suggesting is that these programs are oversimplified because it does not have nearly the same amount of variables you would find in nature. Secondly, its a statistical destiny that they will evolve, not an anomaly. In that way, it isn't actually making a case for natural evolution at all, only masquerading as if it is.
As mentioned there: The Boeing 777 was not designed by humans but rather was evolved by computers. By your logic, the programmers put the design into the computer. But if they did that, why on earth bother writing a program whose sole function is to spit out an answer we already know?
I don't know anything about the development of the Boeing 777, not that it would provide any basis for the current discussion even if I did since an aircraft is not a living being capable of procreation. What I do know is that these simulators are not sufficient in providing a basis for the evolution of organisms. It would do more to advance the theory if they attempted a series of controlled experiments to mimic what it might look like in actuality.

“This life’s dim windows of the soul, distorts the heavens from pole to pole, and goads you to believe a lie, when you see with and not through the eye.” -William Blake

Replies to this message:
 Message 17 by crashfrog, posted 10-29-2007 5:31 PM Hyroglyphx has not replied
 Message 18 by PaulK, posted 10-29-2007 5:39 PM Hyroglyphx has not replied
 Message 20 by Percy, posted 10-29-2007 9:28 PM Hyroglyphx has not replied
 Message 21 by Taz, posted 10-29-2007 9:52 PM Hyroglyphx has replied

  
crashfrog
Member (Idle past 1467 days)
Posts: 19762
From: Silver Spring, MD
Joined: 03-20-2003


Message 17 of 31 (431174)
10-29-2007 5:31 PM
Reply to: Message 16 by Hyroglyphx
10-29-2007 5:25 PM


Re: A reply to Rrhain from another thread
What I am suggesting is that these programs are oversimplified because it does not have nearly the same amount of variables you would find in nature.
Random mutation and natural selection are both incredibly simple principles. How complicated do you think a simulation has to be in order to accurately model these two forces?
I don't know anything about the development of the Boeing 777, not that it would provide any basis for the current discussion even if I did since an aircraft is not a living being capable of procreation.
That was the point of the simulation, NJ - to have 777's that could reproduce and procreate under the influences of natural selection and random mutation.
How did you miss that?
What I do know is that these simulators are not sufficient in providing a basis for the evolution of organisms.
Observations of organisms evolving is what provides ample basis for conclusions on the evolution of organisms.
What these simulations do is illuminate the ability of the very simple forces of random mutation and natural selection to give rise to complicated structures and behaviors - structures and behaviors complicated enough to give the appearance of design, even - and since they're modeling fairly simple forces, why would they have to be all that complicated?
It would do more to advance the theory if they attempted a series of controlled experiments to mimic what it might look like in actuality.
Done. Directed evolution studies on actual organisms are legion.

This message is a reply to:
 Message 16 by Hyroglyphx, posted 10-29-2007 5:25 PM Hyroglyphx has not replied

  
PaulK
Member
Posts: 17822
Joined: 01-10-2003
Member Rating: 2.2


Message 18 of 31 (431178)
10-29-2007 5:39 PM
Reply to: Message 16 by Hyroglyphx
10-29-2007 5:25 PM


Re: A reply to Rrhain from another thread
quote:
What makes you think that the Monte Carlo method is only applicable to evolution?
When did I say it was? That was the pretense of Taz's rant, all of which brought up by TAZ, not me. He brought up the supposed misrepresentation of evolution, and it was he that used the MC method as a basis for providing some evidentiary support.
So why are you restricting it to evolution?
I'm not. Go back and follow the dialogue. First tell me how this conversation began, then tell me where I restricted the MC method solely to evolution.
Well I just did that and it turns out that Rrhain's reading was quite reasonable. Taz was arguing that your point was wrong in general and suggested:
quote:
Look, do yourself a favor and learn some programming. Then use the monte carlo method to solve a physical (preferably chaotic) system.
You replied:
quote:
If you're not talking about evolution, then how does the MC method even apply to the conversation?
Which implies that you think that the MC only applies to evolution. Of course the alternative is that you simply failed to understand what Taz was saying.
Edited by PaulK, : No reason given.

This message is a reply to:
 Message 16 by Hyroglyphx, posted 10-29-2007 5:25 PM Hyroglyphx has not replied

  
cavediver
Member (Idle past 3643 days)
Posts: 4129
From: UK
Joined: 06-16-2005


Message 19 of 31 (431181)
10-29-2007 5:45 PM
Reply to: Message 15 by crashfrog
10-29-2007 4:46 PM


Thanks for the heads-up, Crash. Superb video - and helped along by some great music

This message is a reply to:
 Message 15 by crashfrog, posted 10-29-2007 4:46 PM crashfrog has replied

Replies to this message:
 Message 22 by crashfrog, posted 10-29-2007 11:14 PM cavediver has not replied

  
Percy
Member
Posts: 22392
From: New Hampshire
Joined: 12-23-2000
Member Rating: 5.3


Message 20 of 31 (431202)
10-29-2007 9:28 PM
Reply to: Message 16 by Hyroglyphx
10-29-2007 5:25 PM


Re: A reply to Rrhain from another thread
I think the important point that hasn't yet gotten across is that the type of evolution simulations being talked about in this thread, ones like the example you provided, model the principles of evolution, namely mutation and natural selection. They aren't intended to model the evolution of living organisms. You provided an example of a simulation that illustrates evolutionary principles, not real-life evolution. You in essence introduced an apple into the discussion and then criticized it for not being an orange.
If you'd like to shift discussion to simulations of the evolution of real-life organisms then we could do that if the participants here are agreeable, but it would be nice to first have an example of such a simulation so that the discussion can be focused. Probably such simulations have to be of very primitive organisms in a simple environment, such as bacteria in a petri dish on a well-defined medium. My guess is that simulations of more complex organisms with populations of meaningful size in a realistic simulated environment wouldn't be practical because of computational constraints, and certainly the programming task would be immense.
--Percy

This message is a reply to:
 Message 16 by Hyroglyphx, posted 10-29-2007 5:25 PM Hyroglyphx has not replied

  
Taz
Member (Idle past 3292 days)
Posts: 5069
From: Zerus
Joined: 07-18-2006


Message 21 of 31 (431206)
10-29-2007 9:52 PM
Reply to: Message 16 by Hyroglyphx
10-29-2007 5:25 PM


Re: A reply to Rrhain from another thread
Let me clear something up.
I clearly stated in the other thrad that the reason I was pissing and moaning about the video was that the narrator kept referring to evolution as a totally and completely random process. Yes, there are random processes involved, but not all parts of evolution is random.
YOU then said that natural selection is a non-random process inside a sea of random processes, implying that it was justified for the narrator to refer to the whole thing as random and ignoring the non-random parts.
I pointed out that there are lots of processes that make pretty patterns even though the bulk of them are completely random. The example I provided (monte carlo) was just that. The bulk of a monte carlo program is nothing but calling up random numbers after random numbers and producing random results after random results. Some monte carlo programs I have produced literally tens of millions upon tens of millions of random numbers and random calculations. But part of the monte carlo method is the whole thing is controled by a single or a couple non-random parameters.
According to your logic, the monte carlo shouldn't work at all simply because the bulk of the program is just randomness. And yet I've been able to make out pretty patterns and wonderful graphs with this very simple random mutations and natural selection program.
Nem, there are a lot more to programming than just evolution simulators. And there are a lot more we can do with the concept of random mutation and natural selection than just applying it as a mechanism for biological evolution. You should really learn a thing or two about software programming and what mathematicians, physicists, and engineers do with it before you make blanket statements again. I'm getting a muscle ache on my face from all the flinches.
The main thing is it was YOU who brought up the evolution simulator, not I. Frankly, I have never bothered to look into any of the evolution simulators before. So, I'll let others who are more familiar with this territory to talk with you.

Disclaimer:
Occasionally, owing to the deficiency of the English language, I have used he/him/his meaning he or she/him or her/his or her in order to avoid awkwardness of style.
He, him, and his are not intended as exclusively masculine pronouns. They may refer to either sex or to both sexes!

This message is a reply to:
 Message 16 by Hyroglyphx, posted 10-29-2007 5:25 PM Hyroglyphx has replied

Replies to this message:
 Message 29 by Hyroglyphx, posted 10-31-2007 8:19 PM Taz has replied

  
crashfrog
Member (Idle past 1467 days)
Posts: 19762
From: Silver Spring, MD
Joined: 03-20-2003


Message 22 of 31 (431223)
10-29-2007 11:14 PM
Reply to: Message 19 by cavediver
10-29-2007 5:45 PM


and helped along by some great music
Coldplay's "Clocks", for anybody who didn't get the really obvious pun.

This message is a reply to:
 Message 19 by cavediver, posted 10-29-2007 5:45 PM cavediver has not replied

  
Hyroglyphx
Inactive Member


Message 23 of 31 (431380)
10-30-2007 8:02 PM
Reply to: Message 10 by Jazzns
10-29-2007 12:13 PM


Re: Genetic Algorithms
What genetic algorithms DO refute is the notion that biological life cannot evolve IC or specified complexity. There are examples of using gentics algorithms where the selected pressured walk of random changes have done just that in direct opposition to any enterprising IDers that claim forcefully that such processes cannot.
I'm not seeing how genetic algorithms refute IC, since all the necessary information is already extant, which is the whole point of IC. But maybe I'm not understanding your statement, in which case, can I ask you to elaborate?

“This life’s dim windows of the soul, distorts the heavens from pole to pole, and goads you to believe a lie, when you see with and not through the eye.” -William Blake

This message is a reply to:
 Message 10 by Jazzns, posted 10-29-2007 12:13 PM Jazzns has replied

Replies to this message:
 Message 24 by Percy, posted 10-30-2007 9:54 PM Hyroglyphx has not replied
 Message 25 by PaulK, posted 10-31-2007 2:32 AM Hyroglyphx has not replied
 Message 26 by Dr Adequate, posted 10-31-2007 7:02 AM Hyroglyphx has not replied
 Message 28 by Jazzns, posted 10-31-2007 11:44 AM Hyroglyphx has not replied

  
Percy
Member
Posts: 22392
From: New Hampshire
Joined: 12-23-2000
Member Rating: 5.3


Message 24 of 31 (431406)
10-30-2007 9:54 PM
Reply to: Message 23 by Hyroglyphx
10-30-2007 8:02 PM


Re: Genetic Algorithms
Nemesis Juggernaut writes:
I'm not seeing how genetic algorithms refute IC, since all the necessary information is already extant...
If by "necessary information" you mean the design that the genetic algorithm generates, then no, this information is not "extant". If it was then the programmers of the genetic algorithm would not have to actually run the program in order to create the design. In fact, if they really already had the design, why would they ever bother writing the program?
Or if by "necessary information" you are referring to, for example, the way transistors and resistors and capacitors behave when you need to design an electrical circuit, then this is the same "necessary information" that any designer of electrical circuits employs. The genetic algorithm has no more of this "necessary information" than a human designer.
What evolution simulators and genetic algorithms illustrate is that information (novel designs) can be created by applying the evolutionary principles of mutation and natural selection, neatly disproving the ID claim that only humans can create information, though this is obvious anyway to anyone who understands information theory. The ID terms "irreducible complexity" and "specified complexity" do not have any quantitative or even clear definition, and so these imprecise ideas cannot be be rationally addressed, but the solutions and designs of these programs can become arbitrarily complex, limited only by computational resources and the ingenuity of the programmers.
--Percy

This message is a reply to:
 Message 23 by Hyroglyphx, posted 10-30-2007 8:02 PM Hyroglyphx has not replied

  
PaulK
Member
Posts: 17822
Joined: 01-10-2003
Member Rating: 2.2


Message 25 of 31 (431438)
10-31-2007 2:32 AM
Reply to: Message 23 by Hyroglyphx
10-30-2007 8:02 PM


Re: Genetic Algorithms
Runs of Avida to test Behe's claim - using Behe's original definition of Irreducible Complexity succeeded. Thus they proved that Irreducible Complexity is in principle producible by Darwinian mechanisms. This refutes claims to the contrary such as Dembski's assertion that he need not consider evolution when dealing with the bacterial flagellum (in the one attempt I know of to apply his "Design Inference" to biology - a complete mess).
It should be also pointed out that Behe's whole argument is an "in principle" argument. It does not rely on the details of biology. Thus the Avida runs also discredit Behe's argument. They do not prove Behe's conclusion wrong, but they undercut it by removing the support.
It should also be emphasised that Behe's argument has nothing to do with having "information" available. It is based solely on the idea that if evolution were to construct systems by simply adding parta, one at a time, an IC system would be non-functional until it were completed - and thus could not benefit from natural selection. The information needed for the parts and their assembly could be "available" in the sense that they are in the programs we are discussing without affecting Behe's argument in the slightest.
Edited by PaulK, : Expanded

This message is a reply to:
 Message 23 by Hyroglyphx, posted 10-30-2007 8:02 PM Hyroglyphx has not replied

  
Dr Adequate
Member (Idle past 285 days)
Posts: 16113
Joined: 07-20-2006


Message 26 of 31 (431450)
10-31-2007 7:02 AM
Reply to: Message 23 by Hyroglyphx
10-30-2007 8:02 PM


Re: Genetic Algorithms
I'm not seeing how genetic algorithms refute IC, since all the necessary information is already extant, which is the whole point of IC. But maybe I'm not understanding your statement, in which case, can I ask you to elaborate?
Perhaps we could ask you to elaborate, since at first glance this looks like complete garbage.

This message is a reply to:
 Message 23 by Hyroglyphx, posted 10-30-2007 8:02 PM Hyroglyphx has not replied

  
Dr Adequate
Member (Idle past 285 days)
Posts: 16113
Joined: 07-20-2006


Message 27 of 31 (431451)
10-31-2007 7:06 AM
Reply to: Message 8 by Hyroglyphx
10-29-2007 10:48 AM


Re: Artificial selection
Insomuch that because its so simplistic with very defined parameters, its bound to produce exactly what the designers want ...
Nope.
Even if you tried to deduce your conclusion from your premise, and I notice that you haven't bothered to try, observation of the real world (remember that?) proves that you're wrong. Because in reality, people don't use genetic algorithms to prove evolution. They use them to do things that they cannot achieve by intelligent design.

This message is a reply to:
 Message 8 by Hyroglyphx, posted 10-29-2007 10:48 AM Hyroglyphx has not replied

  
Jazzns
Member (Idle past 3912 days)
Posts: 2657
From: A Better America
Joined: 07-23-2004


Message 28 of 31 (431491)
10-31-2007 11:44 AM
Reply to: Message 23 by Hyroglyphx
10-30-2007 8:02 PM


Re: Genetic Algorithms
Percy gave a good explanation but I'll give my own 2 cents.
The reason we know they refute IC is because we can create logical equivalencies between a biological entity and a computer program. That is what genetic algorithms are. In particular computer programs and biological entities can contain parts that are by definition IC.
1. So we start with a program that can mutate, reproduce, and be selected for or against based on its fitness.
2. We observe that in the beginning no IC systems exist in the program.
3. We subject the program to a genetic algorithm, allowing it to reproduce imperfectly culling out the failures and promoting the successes.
4. We observe that after time, IC systems exist in the program.
Since what these are doing is, like Percy explained, mimicing the principles of evolution, all we have to do is note that those principles exist in biology as well to refute IC.
Does it prove biological evolution? No
Does it prove that biological evolution DOES produce IC? No
Does it refute the notion that biological evolution cannot produce IC? A resounding yes!!
It is also a pretty good indicator of the general superficial quality of Behe's argument. Many people get caught up in all the bio-geek and complex language of the IC debate and fail to notice that what Behe is claiming, other than the fact that simple computers can disprove it, lives at a level of abstraction so high as to not be capable of proving anything concrete.
Behe himself admits this at the Dover trial and even subtly in his written work. He states very clearly that the strength of his argument depends on his audiences degree of belief in God. He also states that his IC argument is a plausability argument rather than a possibility argument. Many people who run around whoring Behe's work tend to try to make it a issue of possibility. In other words, they go around claiming that evolution CANNOT produce IC when in fact all Behe is saying is that he thinks it is unlikely and that a factor in that decision is the strength of belief in a higher power.
Not a very good argument when you expose it.
Edited by Jazzns, : No reason given.

Of course, biblical creationists are committed to belief in God's written Word, the Bible, which forbids bearing false witness; --AIG (lest they forget)

This message is a reply to:
 Message 23 by Hyroglyphx, posted 10-30-2007 8:02 PM Hyroglyphx has not replied

  
Hyroglyphx
Inactive Member


Message 29 of 31 (431521)
10-31-2007 8:19 PM
Reply to: Message 21 by Taz
10-29-2007 9:52 PM


Re: A reply to Rrhain from another thread
I clearly stated in the other thrad that the reason I was pissing and moaning about the video was that the narrator kept referring to evolution as a totally and completely random process.
We're all clear on that part.
YOU then said that natural selection is a non-random process inside a sea of random processes, implying that it was justified for the narrator to refer to the whole thing as random and ignoring the non-random parts.
Right. So far so good.
I pointed out that there are lots of processes that make pretty patterns even though the bulk of them are completely random. The example I provided (monte carlo) was just that. The bulk of a monte carlo program is nothing but calling up random numbers after random numbers and producing random results after random results.
Right, which is obviously in lieu of the conversation, which was about how random or non-random evolution is.
Nem, there are a lot more to programming than just evolution simulators. And there are a lot more we can do with the concept of random mutation and natural selection than just applying it as a mechanism for biological evolution. You should really learn a thing or two about software programming and what mathematicians, physicists, and engineers do with it before you make blanket statements again.
Software programs have nothing to offer biological systems Taz, which, if I haven't made it abundantly clear by now, is the angle that I've been coming from.
The main thing is it was YOU who brought up the evolution simulator, not I.
Taz, you used the Monte Carlo method as a way to typify your argument, did you not? And were we not talking about random processes within evolution? Clearly there was no other avenue or reason for you to bring up the MC method. So please tell me how I'm wrong.

“This life’s dim windows of the soul, distorts the heavens from pole to pole, and goads you to believe a lie, when you see with and not through the eye.” -William Blake

This message is a reply to:
 Message 21 by Taz, posted 10-29-2007 9:52 PM Taz has replied

Replies to this message:
 Message 30 by Taz, posted 10-31-2007 8:45 PM Hyroglyphx has not replied
 Message 31 by Percy, posted 10-31-2007 9:29 PM Hyroglyphx has not replied

  
Taz
Member (Idle past 3292 days)
Posts: 5069
From: Zerus
Joined: 07-18-2006


Message 30 of 31 (431530)
10-31-2007 8:45 PM
Reply to: Message 29 by Hyroglyphx
10-31-2007 8:19 PM


Re: A reply to Rrhain from another thread
Nem writes:
Taz, you used the Monte Carlo method as a way to typify your argument, did you not? And were we not talking about random processes within evolution?
No. We were talking about ANY process that involves lots and lots of random steps and one single non-random step. By saying that the whole thing is totally random even though parts of it is not, you've advocated a lie, which goes against one of your precious 10 commandments.
Clearly there was no other avenue or reason for you to bring up the MC method. So please tell me how I'm wrong.
We've been trying to tell you this for days now. Simple random mutation and natural selection works for other areas other than biological evolution as well. When I was pissing and moaning about the video, I was not specifically pissing and moaning about the narrator lying about biological evolution. I was pissing and moaning about the narrator lying about the process in general.
Again, the monte carlo method involves lots and lots of random steps with one or two non-random steps. I dare you to say straight to my face that it doesn't work when trying to solve physical and mathematical systems.
Software programs have nothing to offer biological systems Taz, which, if I haven't made it abundantly clear by now, is the angle that I've been coming from.
Again, I have not looked at any evolution simulation program before and am personally don't know enough about the mechanics of evolution to comment. It's called humility, something that seems to be alien to you.
Just remember that you brought up evolution simulator, not I.
Edited by Taz, : No reason given.

Disclaimer:
Occasionally, owing to the deficiency of the English language, I have used he/him/his meaning he or she/him or her/his or her in order to avoid awkwardness of style.
He, him, and his are not intended as exclusively masculine pronouns. They may refer to either sex or to both sexes!

This message is a reply to:
 Message 29 by Hyroglyphx, posted 10-31-2007 8:19 PM Hyroglyphx has not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024