10 Categories of Evidence For ID

It is understandable that IDists reject these programs, but the fact is that the modeling of natural processes via computer programs has a long history of success, from meteorology to particle physics to structural analysis to microchips and on and on. Evolution programs and genetic algorithm programs model the evolutionary process of descent with modification and natural selection, and they definitively show what is intuitively obvious anyway: information does not depend upon intelligence.Even the simplest of natural processes creates information. For example, the spectrum of light emitted by a star is encoded in its electromagnetic emissions. The star created this information. If we analyze and record the spectrum via spectroscopy all we're doing is translating information from one encoding to another. Human intelligence is not creating the information about the star's spectrum, only recording it. It does not take intelligence to create information. --Percy

This is the second time you've said this in this thread without offering any support. Finding mammals in Cambrian layers would falsify evolution. The basic process of evolution, descent with modification and natural selection, has been scientifically verified in the lab and wild both morphologically and genetically, and these experimental verifications are open to falsification. Evolution is testable and verifiable.The basic tenet of ID, as I understand it, is not that evolution is unscientific or impossible, but that it is insufficient as a theory explaining the diversity of life.The inherent problem with your version of ID is that it implicitly but inevitably invokes the supernatural. If it were really true that information can never arise via natural processes, then the first life in the universe could only have come about through supernatural means.--Percy

Of course they entered the experiment with a preset goal. The goal of the experiment was to follow a set of design constraints, and they are analogous to the constraints of the environment of a population. Without such constraints it wouldn't be a simulation of evolution.The goal of evolution is to provide a gene pool for the population that provides the best chance for its preservation. For example, a population will have constraints of temperature, food supply, predators and so on. This experiment had constraints of frequency and amplitude.In evolution, selection is performed by the environment. Those least fit for the environment produce the least offspring. In this genetic algorithm simulation, selection is performed by measuring which offspring came closest to the design goals. Those coming closest are selected to "reproduce". Offspring are "spawned" from each of the surviving designs, and random "mutations" are created in each offspring. Then the process is repeated.These genetic algorithms are simply modelled on the way evolution works by having a population of alternative designs play in an "environment", and selecting the ones that perform best to produce the next generation. There is no channel by which design information is being provided by the programmers. The experiment did not have the purpose of simulating nature. They were not trying to model evolution of simulated biological organisms in simulated environments. There are programs which do this, but that wasn't the purpose of this one. They were trying to assess the potential of a circuit design approach based upon the process of evolution. For actual life the raw materials are elements and compounds. In this experiment the analogous raw materials are design components like resisters, transisters, wires, diodes and so forth.--Percy

You're confusing matter and energy with information. Information is not governed by the laws of thermodynamics.Genetic algorithms can produce designs untouched by human minds. They do this by harnessing the evolutionary process in a design context. Random errors are created in a population of designs, oftentimes a simulation of sexual sharing of "genes" is employed, and the resulting "offspring" designs are assessed against the design goals. Those that measure up the best are selected to contribute to the next generation. If it helps you understand this, look at it as one member of the class of successive approximation approaches to problem solving. Just saying this indicates you don't yet understand how genetic algorithms work. Here's a brief and very simple example.Let us say we want to build a word guessing program using a genetic algorithm. We define the program's operational behavior from the point of view of the user like this:

1. Print "Please think of a word and I will guess it."
2. User types "OK".
3. Print "How many letters are in the word?"
4. User types in the number of letters.
5. Program prints out 10 guesses.
6. User types number of correct letters for each guess.
7. If none of the guesses were completely correct, program picks the top five guesses and allows each to produce two offspring, each different from the parent by only a single random letter, then program returns to step 5.
8. If one or more of the guesses is correct, print the word and "Thank you for playing!"

That's how simple it is (naturally it can get much more complicated). In this case the user does the selecting himself. There's no mystery to genetic algorithms. There's no secret information from the programmers.Another way to think of it is like the game of hotter/colder, where you search for an object in the room that someone is thinking of while they give you feedback about whether or not you're getting warmer. You discard your movements that brought a "colder" response, and you continue with movements that brought a "warmer" response. In the same way, genetic algorithms continue building on a design that evaluates as "Better, though still not good enough", while discarding those that evaluate as "Worse" or "Better, but not as good as some others". Well put your mind at rest. You won't have to support a negative because quite obviously something happened. What the theory of evolution actually does is propose a mechanism to explain how it happened, i.e., why fossils appear in the order they do, why life's diversity is spread across the planet in the way it is, etc. The creatures of the Cambrian explosion tended to have soft-bodied ancestors that did not fossilize well, though predecessors are slowly being found. But a mammal in the Cambrian would require millions and millions of years of hard skeletoned ancestors that would have fossilized but didn't. A mammal in the Cambrian could not have come about through an evolutionary process. It would represent a serious problem for evolution. You've drifted off your original point. If you recall, in Message 22 you claimed that evolution wasn't testable or falsifiable. What I was saying in the portion you quoted was that this isn't the traditional objection of ID to evolution. Most IDists accept evolution as a valid scientific theory, Behe most prominent among them. Rejecting evolution because you believe it is unscientific makes you a rather unusual IDist. You've somehow missed the crucial (and obvious) implication. If your claim that information can not be created by natural processes but only only by intelligence is true, then the first intelligence in the universe had to have come about by supernatural means. That conclusion is inescapable. The theistic roots of your ID "theory" are painfully obvious to everyone but IDists.As I pointed out earlier, it does not take intelligence to create information. Every process in the universe creates information. When that information reaches one of our senses we can translate that information into another form, but we don't create it.--Percy

You've missed the key reason why I provided the example I did. Jerry was claiming that the programmers *must* be providing information to the program about what to design. So I completely removed this possibility from my example - the constraints are provdied by the user.Also, you say this as if it were different in principle from what happens in nature. It isn't. Just as nature provides constraints in the form of temperature, weather, water supply, food supply, predators and so forth, the program provides constraints, too. These constraints are what provides selection, which is one of the two key components of evolution: descent and with modification and natural selection. Following the rules of grammar isn't going to produce a story, or even very many rational sentences. I can see a lot of sentences like, "The tinted cognition drove to the rusty iota." What would be your selection mechanism for choosing the "winners" in each generation?--Percy

In Shannon's seminal 1948 paper that began the science of information theory he says, "The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem."In other words, you're mistaking meaning for information. Meaning is independent from information. Your example actually consists of two possible messages from the set of messages of a channel of information 36 characters wide. The difference in meaning between the two messages is merely a human superimposed one.The number of distinct possible messages in the set, assuming each character is 8 bits wide and all characters are legal and each message in the set is equally likely, is 2^8*36 = 2²⁸⁸. With all possible messages equally likely, the amount of information produced when one message is chosen from the set is the same regardless of which message it is. Since the probability of any particular message being chosen is 1/2²⁸⁸, the amount of information represented by one message from the set is: In other words, it will take 288 bits to send one message from the set. If the probabilities of the messages weren't equal the equation is more complicated and it tells you that you can represent the message in fewer bits. But whether you send all a's or a shell script, the information content is the same. Only the meaning differs. And the meanings could be dramatically different than the obvious ones. For instance, your code could be a cipher where you tell the person who will be receiving the message that if he receives a message consisting of all the same letter that he should turn to the ordinal page in his cipher book corresponding to that letter and carry out the instructions therein. Any other message has no meaning.--Percy

Information is conceptual. You're confusing the means of representing and transmitting information with information itself. Possibly you arrived at this conclusion because you've confused informational entropy with thermodynamic entropy. This is a common mistake, so common Wikipedia even addresses it in its article on Information Theory: In other words, obvious but superficial analogies between the two caused the adoption of the term "entropy" by the science of information theory, but the two apply to distinctly different realms and have completely different rules of application. I didn't think you were dissing genetic algorithms. I was addressing the point that was the reason genetic algorithms were originally introduced into the discussion. ID claims that only intelligence can create information. Genetic algorithms are applications of evolutionary principles to design problems. Genetic algorithms create original designs, i.e., new information. Genetic algorithms are not intelligent. ID's claims about intelligence and information are not only unsupported, they're contradicted by the mere existence of genetic algorithms.And as I said earlier, this is self evident anyway. All natural processes generate information. People don't create information just by writing down observations. I'm not confusing it with real life. Someday we can discuss evolution simulation programs, but right now we're addressing genetic algoriths. Have fun with your glib comments, but don't confuse computer simulations of evolution with design tools that use genetic algorithms.The simple word guessing program I described illustrates how a genetic algorithm can solve a problem ("What word am I thinking of?") without having any way it could know that word. I invented the example to address your claim that the programmers were giving the design to the program rather than the program developing the design on its own. This should have been obvious to you anyway, since one of the examples was of a genetic algorithm that designed an oscillator using an AM radio circuit as one of the subcircuits, an approach no human designer would ever have thought of. This addresses the analogy itself rather than the principle it was intended to convey. I was trying to help you understand that genetic algorithms are a specialized subclass of problem solving techniques known as successive approximation. While it is based upon evolutionary principles, it is not a simulation of biological evolution, though there are programs that do this. The analogy was an example of a simple successive approximation.I'll skip the discussion of evolution. This thread is already having trouble staying on topic. We can pick up those issues in another thread if they come up again. I don't really see much difference between a made up god and a made up alien from another universe. "Mathematically impossible" is just an unsupported assertion, and genetic algorithms make clear this isn't the case.--Percy

Hi Jerry,About your proof of "mathematically impossible", you had claimed in Message 56 that the creation of information found in organisms is "mathematically impossible". I challenged this as an unsupported assertion in Message 96, so you've now replied by calculating "how much information nature could have produced since the inception of the universe," which you figured to be 475 bits. Then you said, "The simplest organism contains many times this amount of information."Question: How can the "simplest organism" contain "many times" the amount of information "produced since the inception of the universe"? In other words, how can a subset contain more of anything than the set of which it is a member?Genetic algorithms are one example falsifying your claim that intelligence is required to create information, and another falsification can be illustrated with a simple example from biology. Consider a gene of a population of organisms that has two alleles, call them A and B. This gene is represented by a mere three nucleotides, and A and B are encoded like this: The amount of information in this gene for the population (assuming equal probability for each allele) is -log₂1/2 = 1.Now let's say that during reproduction a mutation causes a new offspring of the population to have a mutation in this gene that produces a unique allele, call it C: The amount of information in this gene for the population (again assuming equal probability for each allele) is now -log₂1/3 = 1.585. The information within the population for this gene has increased from 1 to 1.585, and no intelligence was involved. But we're talking about your musings, not Hawkings. The point is that there is no more evidence for aliens from another universe than there is for God or gods. You said, "I'm looking at this from the aspect of hard science," but it's hard to see how. No, Jerry, information is conceptual. Conceptual does not mean subjective. You are confusing the matter or energy used to record and represent information with the information itself. Any entropic changes in the matter in which information is recorded has no effect on the information itself. For example, heat a book (but don't burn it) and the entropy of the matter making up the book will change, but the information entropy will not. Or heat the wires carrying data in your computer, such as happens over the first half hour or so after you turn it on. Again, the entropy of the matter changes, the information does not. Keeping in mind that we're talking about genetic algorithms (your link was about evolution simulations, and anyway, links should only be offered to support your argument, not to make your argument), how about you showing it now?--PercyThis message has been edited by Percy, 05-15-2005 09:28 AM

I couldn't find where this was already mentioned at EvC Forum, so I'm posting it here.The Vatican newspaper LOsservatore Romano ran an article on January 17th of this year by Fioenzo Facchini, a professor of evolutionary biology at the University of Bologna. I can't find the entire text of the article, perhaps someone else can. There's an online English language version of the vatican paper, but it doesn't have an archive that I could find. Here's what I could turn up on the net: --Percy