Author
|
Topic: Evolution in action? (Read 38711 times)
|
|
|
RTyp06
*Smell* controller
Offline
Posts: 491
|
Come back when you've run the experiment over a few millennia and I'll be a bit more accepting of the "never" part. How long did that ability take to evolve in the first place? How probable is getting that specific ability? There are many wild bacteria that cannot produce the enzyme to digest lactose even though it certainly would benefit them as a new food source. And if it takes a millennia to evolve a *single* gene that produces an enzyme for the simple task of digesting lactose (not to mention bacteria are considered some of the oldest species on earth), how am I supposed to believe that evolution can produce anything that uses multiple genes for more complex tasks? Especially when you get to the level of hearing, seeing tasting etc where many genes are responsible for the protiens and programming that make the parts fit like a finely tuned machine. Let alone the big picture that we humans evolved from single celled life? Is my thinking here *really* that flawed?
One other thing to keep in mind is that Hall spent 25 years trying to evolve functions in bacteria and this one expiriment, to my knowledge, was the only success. His reason for E-Coli not evolving the function back after the double gene knockout is the lack of "evolvability" of the species. Another band-aid excuse so we can all go on happily knowing that evolution is indeed "fact". Could it just possibly be that (gasp!) perhaps the premise is flawed?
I will refrain from repeating my objections to specified complexity, as you didn't seem to understand them the last time. As far as I can tell (and it seems there's even more wrong than I stated), Dembski's reasoning is based completely on misuse of probability and information theory. Regardless of your scientific methodology, this means Dembski's work is invalid. All right, I'll re-read your objections and the criticisims...
|
|
« Last Edit: February 28, 2007, 12:16:05 am by RTyp06 »
|
Logged
|
|
|
|
|
|
|
|
RTyp06
*Smell* controller
Offline
Posts: 491
|
Way to miss the point. We are more complex than bacteria. If you're willing to grant a new gene every thousand years, then the bacterial genome could be generated in far far less time than they actually did the job in. Like, a hundred times faster.
Actually I didn't miss your point. I'm not arguing if there is enough time to evolve or not, just that if there are so many bacteria with so much evolutionary potential, why are we not seeing it happen? And if it takes thousands of years to evolve one small component with so much evolutionary potential in the trillions upon trillions upon trillions of bacterial generations a millinea represents, wouldn't the problem be compounded once evolution arrived at the multicellular level with a much slower reproduction rate?
And I understand the human gene count is not known for sure and estimated anywhere between 40k and 100k. Oh and Novus said a few mellinnia, not just one .. Okay, so what? My math was so obviously order-of-magnitude that this is insanely quibblesome. Of course it is, I was just being an ass and wasn't really serious. Hence the winky emoticon.
Ok, Novus, it seems your problem with Dembski is that plutonium supposidly fits the description of specified complexity. As for your quantum physics argument, I'll admit I do not understand what you are driving at. And I'll admit I know very little about quantum physics.
Your understanding of plutonium seems different than mine. Plutonium as I understand it, is randomly complex in metallic composition yet specified in it's radioactivity. Radioactivity comes in waves with clocklike precision yet you claim it's random? The radioactivity is subject to the 2nd law of thermodynamics and thus has a half life and will eventually run it's course.
Now, many things can be contaminated with radiation. So the plutonium itself and the radiation it emits is not mutually exclusive. Also, plutonium with it's radioactivity *togther* doesn't preform a specifc function. Thus is not specified. Instead, plutonium acts as a container and radioactivity simply leaks out.
|
|
« Last Edit: March 03, 2007, 04:11:37 pm by RTyp06 »
|
Logged
|
|
|
|
|
|
RTyp06
*Smell* controller
Offline
Posts: 491
|
Genetic Algorithms:
I find this particularly disturbing...
An ideal fitness function correlates closely with the algorithm's goal, and yet may be computed quickly. Speed of execution is very important, as a typical genetic algorithm must be iterated many, many times in order to produce a usable result for a non-trivial problem.
And this:
In some cases, it is very hard or impossible to come up even with a guess of what fitness function definition might be. Interactive genetic algorithms address this difficulty by outsourcing evaluation to external agents (normally humans).
So here, since "fitness" cannot be defined, human intelligence intervenes... Tsk, Tsk, Tsk... Do these seem like a realistic representation of natural selection to anybody? C'mon now...
From,
http://en.wikipedia.org/wiki/Fitness_function
...which linked off Novus' wiki link to genetic algorithms.
http://en.wikipedia.org/wiki/Genetic_algorithm.
Ok back to Dembski.. I get the critisim. They feel Dembski (as do you) that he is shooting for a specific goal where evolution has no direction and any number of possibilities. Described this way:
Another criticism refers to the problem of "arbitrary but specific outcomes". For example, if a coin is tossed randomly 1000 times, the probability of any particular outcome occurring is roughly one in 10^300. For any particular specific outcome of the coin-tossing process, the a priori probability that this pattern occurred is thus one in 10^300, which is astronomically smaller than Dembski's universal probability bound of one in 10^150. Yet we know that the post hoc probability of its happening is exactly one, since we observed it happening. This is similar to the observation that it is unlikely that any given person will win a lottery, but, eventually, a lottery will have a winner; to argue that it is very unlikely that any one player would win is not the same as proving that there is the same chance that no one will win.
The problem with this critisim as I see it is that as you climb the ladder of complexity, there become less and less viable options. Expotentially growing vast seas of neutral, meaningless possibilities.
Here's a interesting analogy from Sean Pitman's great site.
" Regis Philbin is the host of a game show called “Millionaire or Not” and you are the next contestant. In front of you is a safety deposit box with a million dollars in it. On the front of the box is an apparatus that looks like a slot machine. It has 15 rotating wheels, each with the 26 letters of the alphabet on it. Regis tells you that there are one million different winning combinations of fifteen letters that will open the safety deposit box. You can rotate each wheel at will and then press a button to see if the combination that you chose is one of the one million winning combinations. You can keep doing this until you give up. You think that this game is a synch. With one million winning combinations possible, you are practically guaranteed to win. However, if you never choose the same combination twice and if you test a new combination every second, how long will it take you on average to find any one of the one million correct combinations? It would take you a bit over 53 million years on average.
http://www.detectingdesign.com/thecatandthehat.html
5th paragraph
|
|
« Last Edit: March 09, 2007, 03:20:45 am by RTyp06 »
|
Logged
|
|
|
|
|
Death 999
Global Moderator
Enlightened
Offline
Gender:
Posts: 3874
We did. You did. Yes we can. No.
|
Before looking at the link: Well, yeah, no theory answers all questions, duh.
After reading the link: What's the problem here? This sort of calculation is done very easily by something called a 'neural net', and furthermore is fairly stable against perturbation (i.e. not being exactly right). Which means it's not even trivially irreducibly complex
No 'riddle' here.
|
|
|
Logged
|
|
|
|
jucce
Frungy champion
Offline
Posts: 95
|
Also answered, many times... a species which is well adapted to its present situation is under very little evolutionary pressure and tends to stay largely the same for very long periods, in no small part because it turns down its mutation rate. This has been directly observed in bacteria; and sexual selection is a great stabilizer when it does not run amok (peacocks). Stuff like this sounds very reasonable and looks good on paper in a college text, but can you point to real world living examples? I can think of species that are under enviornmental pressues. Global warming is shrinking the polar ice caps and causing starvation of polar bears. Do you think polar bears will evolve to adapt? Why or why not? How about south american parrots that are losing their ancient breeding grounds due to removal of the tropical rain forest. Will they evolve to adapt? How come the passenger pigeons didn't evolve a fear of man and were hunted to extinction? Sometimes there is not enough time for a suitable mutation to appear. When it comes to the pidgeon for example, it would probably have taking evolution over a very long time and a transformation of the species for them to survive. There can only be so much the species can adapt to within a reasonable amount of time. And species that don't adapt will die out.
Then what about crocs and alligators? Crocs supposidly have been a stable species for millions of years, yet lived just fine through global ice ages that killed off many other species?
Yes, what is the problem with that?
Here's a interesting analogy from Sean Pitman's great site. " Regis Philbin is the host of a game show called “Millionaire or Not” and you are the next contestant. In front of you is a safety deposit box with a million dollars in it. On the front of the box is an apparatus that looks like a slot machine. It has 15 rotating wheels, each with the 26 letters of the alphabet on it. Regis tells you that there are one million different winning combinations of fifteen letters that will open the safety deposit box. You can rotate each wheel at will and then press a button to see if the combination that you chose is one of the one million winning combinations. You can keep doing this until you give up. You think that this game is a synch. With one million winning combinations possible, you are practically guaranteed to win. However, if you never choose the same combination twice and if you test a new combination every second, how long will it take you on average to find any one of the one million correct combinations? It would take you a bit over 53 million years on average.http://www.detectingdesign.com/thecatandthehat.html5th paragraph Well 53 million years is nothing compared to the age of earth so if anything that proves that with sufficient time and random events unlikely results will be achieved.
http://english.pravda.ru/science/tech/15-03-2005/7885-hiv-0
Researchers at the University of Liverpool in the U.K. believe that some 10% of Europeans became immune to the human immunodeficiency virus (HIV). They argue that new immunity characteristics were caused by the mutation of some genes as a result of the plague epidemics during the Middle Ages. In particular, a Delta-32 mutation affecting a cellular receptor of the gene CCR5 makes the human body protected against the HIV. The above genetic mutation is most frequently found in the residents of Scandinavian countries and Russia. Beneficial mutation?
|
|
« Last Edit: March 16, 2007, 09:15:13 am by jucce »
|
Logged
|
|
|
|
Novus
Enlightened
Offline
Gender:
Posts: 1938
Fot or not?
|
Genetic Algorithms:
I find this particularly disturbing...
An ideal fitness function correlates closely with the algorithm's goal, and yet may be computed quickly. Speed of execution is very important, as a typical genetic algorithm must be iterated many, many times in order to produce a usable result for a non-trivial problem.
And this:
In some cases, it is very hard or impossible to come up even with a guess of what fitness function definition might be. Interactive genetic algorithms address this difficulty by outsourcing evaluation to external agents (normally humans).
So here, since "fitness" cannot be defined, human intelligence intervenes... Tsk, Tsk, Tsk... Do these seem like a realistic representation of natural selection to anybody? C'mon now...
You're comparing apples to oranges. If you're trying to use genetic algorithms to achieve a specific goal, you need to use a fitness function that corresponds to what you want. For example, if you want to evolve a sorting program, the fitness of your program is (typically) how close to the right order you get; if you choose something unrelated, there's nothing to favour programs with the desired trait. The biological equivalent would be selective breeding of animals (e.g. dogs) for a specific task (e.g. herding sheep).
The long-term natural evolution scenario could be considered to be (very roughly) equivalent to a genetic algorithm where the fitness function is, essentially, capability to survive and reproduce in whatever environment the organism is in.
The point here, anyway, was to illustrate that a wide range of goals can be met by evolutionary processes, not that genetic algorithms are always equivalent to the evolution of life on Earth.
The problem with this critisim as I see it is that as you climb the ladder of complexity, there become less and less viable options. Expotentially growing vast seas of neutral, meaningless possibilities.
The problem with that counter-argument is that it's not really true. Considering the sheer amount of genetically different individuals of our species alone (with different strengths and weaknesses) compared to the large amount of species of bacteria with billions of identical individuals, I'd say that complexity does not limit diversity. However, I agree that it can limit the speed with which change occurs due to the amount of neutral and harmful changes available (why is why evolutionary experiments stick to simple organisms).
Dembski's reasoning (or at least this particular part) would be a sensible argument against evolution if, for example, bacteria must have flagella to survive (many don't). The argument against Dembski is that life needn't be like we see it; there is a huge amount of possibilities, and the probability of a specific mechanism appearing is irrelevant.
|
|
|
Logged
|
|
|
|
|