|
Register | Sign In |
|
QuickSearch
Thread ▼ Details |
|
Thread Info
|
|
|
Author | Topic: Micro v. Macro Creationist Challenge | ||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2
|
There is another interesting aspect to all this complexity talk. Something that any professional designer (eg, an engineer) could tell you: Complexity is a designer's worst enemy and spells doom for the design.
Most of the life cycle of a design lies in the maintenance phase. That is where bugs get fixed and features get added. When a design becomes overly complex (eg, "irreducibly complex"), then maintenance becomes extremely difficult if not impossible. Therefore, the amount of complexity in a design can be used to measure how badly designed it is. However, it turns out that complexity is an expected feature of a design that has evolved. Evolutionary processes and methods naturally generate complexity. Many engineers, especially software engineers, have accidentally employed evolutionary processes -- primarily in the "copy something that performs one function and modify it to perform a slightly different function" manner -- and they have learned from bitter experience that the result is a near-exponential growth of complexity in the overall design. The complexity of that "evolved" design increases to the point that hardly anybody can figure out anymore just exactly how it works. And the code has become so intertwined that a very simple change in one place can cause catastrophic changes in totally unrelated parts of the code. Hence this photo of a t-shirt in an engineer's office:
Complexity is anathema to design. More formally, there have been experiments using evolutionary processes to "evolve" useful designs. In some, there are extra parts that don't do anything, basically "vestigial remains" in the more classical sense (to short-circuit standard creationist quips, I mean parts that serve no purpose at all, not parts that still serve some kind of purpose, just not the primary purpose it used to serve). But in some experiments, they ended up with a highly complex, "irreducibly complex" even, design which would have been impossible for any human designer to have created. The one I remember is evolving the design of a particular kind of amplifier using a field-programmable gate array (FPGA). Now, in my professional work for over 25 years, most of our designs included an FPGA -- before that, the US Air Force had trained me in 1977 as an electronic computer systems repairman (AFSC 305x4 -- USAF uses a different designation now), so I do have some understanding of digital electronics (eg, in one civilian job, the electrical engineer only knew analog electronics, so he consulted with me, the software engineer, regularly about digital electronics). Basically, an FPGA is an array of logic circuits which you can program by loading into it a file that tells each element in that array what kind of logic circuit it is (eg, AND gate, OR gate, NOT gate (AKA "inverter"), flip-flop) and exactly how it is connected to all the other elements in that array. I have no direct experience with that, since it's the electrical engineers who work directly with FPGAs (my only involvement is that they define read and write ports into that FPGA that my software then communicates with to control it and to read its status. I also learned a few things about electronics, both analog and digital, in Air Force tech school. The supposed dichotomy between analog and digital electronics is purely artificial. All electronics is analog. What digital electronics chooses to do is to define only two narrow voltage ranges as valid. Depending on the actual logic definitions (a very salient point in the Data Systems Technician Chief Petty Officer advancement exams, one which it took me a second time to finally figure out), you have two and only two binary values, 0 and 1, both represented by a rather narrow range of voltages. What about the voltages between them? My Air Force training called that "The Forbidden Zone", meaning that those are voltage levels that have no real digital meaning and should never happen -- AKA "ambiguious", which is death to digital. So now back to this experiment. They were using this one FPGA. And they were "evolving" a programming file to download into this FPGA so that they could then evaluate its performance as an amplifier of whatever type they were shooting for -- obviously, their measure of fitness was how well it performed. The result of this experiment was an FPGA that functioned well as the kind of amplifier they were seeking Funny things about that design:
OK, so here's a bugaboo that electrical engineers have to deal with all the time. It is never just pure electronics. Every wire contains some internal resistance. In addition, every wire contains some inductive reactance (ie, you pass a current through a wire, it's going to generate a magnetic field). And everywhere that any two wires come close enough to each other, such as in an inductive coil, you also have some capacitance. You thought that biology was messy? Try some time to get down to the lowest levels of electrical nitty-gritty (actually, electronics is still orders of magnitude less messy than biology is). Despite all the quality control we can throw at it, the production of digital circuits still includes some variances. So long as you take these circuits and use them in the prescribed manner (eg, digitally), none of those variances will ever mean anything, will make no difference whatsoever. But the moment you open those minute variances up for exploitation (as in those experiments), then all bets are off! The thing about that FPGA design is that it naturally, with absolutely no intelligent intervention whatsoever (since what human could have possibly worked out all those analog operations of the digital circuits of the FPGA), created a working design that was "irreducibly complex." Therefore, complexity, even "irreducible complexity", is the natural product of evolution. Not of "design".
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2
|
JonF writes:
"Elementary statistical theory shows that the probability of 200 successive mutations being successful is then () , or one chance out of 10 . The number 10 , if written out, would be "one" followed by sixty "zeros." In other words, the chance that a 200- component organism could be formed by mutation and natural selection is less than one chance out of a trillion, trillion, trillion, Show your calculations or references.trillion, trillion!" - Henry M. Morris, Ph.D. However, such a probability model does not describe what happens in evolution, which renders such creationist probability arguments moot and completely irrelevant. Creationists' probability arguments serve no purpose other than to deceive their audience. Evolution doesn't work on a single individual, but rather on a population of individuals. Thus, instead of a single path, evolution uses multiple parallel paths. Thus the probability of a step (AKA "a generation") succeeding would be expressed by a calculation like P=1-(1-p)s, where p is the probability of success and s is the size of the population. That is to say, what is the probability that at least one individual in the population succeeds, which is the inverse (ie, q=1-p) of the probability of every single member of the population failing. So in a population of 1000 where the probability of success is 0.5, that means the probability of failure is 1-0.5 = 0.5. The probability of every individual failing is then 0.51000 = 9.3310-302. The probability of at least one individual succeeding is then 1-9.3310-302 = 1 approx. A probability of 1 is dead certainty. Let's change the values a bit: let p=0.01 and s = 100. q = 1-p = 0.99. qs = 0.366 . 1 - 0.366 = 0.634. Since others have named Morris' faulty argument as a lottery argument, let's look at the probabilities in a lottery. In California's Super Lotto Plus, five numbers are drawn from 47 and then one super-number is drawn from 27 balls. The probability of winning is 1 in 1,533,939 or 6.51910-7. Let's assume that every person in Calfornia buys one lottery ticket, which amounts to 41,416,353 tickets. We already know how unlikely it is for a given person to win, but what is the probability that somebody will win? Well the probability that a person will lose, q = 1-p = 0.99999935 (virtual certainty). The probability that 41,416,353 people will all lose, q41,416,353, is 1.879510-12, fairly small. So subtract from 1 to get the probability that somebody will win and you get 0.999999999998120, virtual dead certainty. If it's a slow half-week and only a million tickets are sold, then the probability that someone will win drops to 0.478. About 50/50, but still fairly likely, unlike the odds of a specific person winning. The object lesson here is that when you try to use math to prove or disprove something, you must develop a math model that accurately describes that something. We know from experience that the probability of a creationist constructing an accurate mathematical model of evolution is virtually zero. Edited by dwise1, : Added 1,000,000 ticket case. Edited by dwise1, : Added "the probability of a creationist constructing an accurate mathematical model of evolution"
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2 |
My page about MONKEY is at cre-ev.dwise1.net/monkey.html. I have also written about it in this forum, so you could do a search.
Such programs therefore don’t simulate realistic biological populations. Neither WEASEL nor MONKEY were ever intended to. I highly recommend that you read my MONKEY page before you make more false assumptions. When Dawkins described WEASEL, I couldn't believe it, so I had to test it. Since Dawkins didn't provide any source code (I think it was a form of BASIC running on a MacIntosh), the only way I could test it was to write my own. In order for it to run as exactly like his as possible, I used his description of the program to come up with a specification for mine. I wrote it in Turbo Pascal, since that was what I was working in at the time (1990). It took Dawkins' program the lunch hour to run, but I think that was because he used an interpreted language. Mine succeeded in less than a minute. I ran it repeatedly and it succeeded repeatedly, every time without fail. I still couldn't believe it, so I calculated the probabilities involved. As improbable as each individual step is, the probability that every single step would fail becomes smaller and smaller until it is virtually impossible for it to fail. I wrote those calculations up in a text file, MPROBS.DOC, which I included in a PKARC package and uploaded it to CompuServe. The rest is history, including the part where I ported it over to C. Again, it is not a simulation of evolution! Rather, it is a comparison of two different selection methods. Re-read the first half of Chapter 3, "Accumulating Small Changes", from Dawkins' book, The Blind Watchmaker, where he describes two different kinds of selection to use to get a desired string:
So what WEASEL and MONKEY do is to compare how well those two selection methods work by giving them both the same problem to solve. Everything is kept the same except for the selection method. Single-step selection fails abysmally while it is virtually impossible for cumulative selection to fail. And the probability calculations in MPROBS explain why. I recently added to the end of the page a brief discussion of programs which do try to model evolution. I also found that some creationists misrepresent how WEASEL works; I discuss that too. If you want to claim that Royal Truman is correct, then I invite you to show me in my source code and in my probability calculations where I am supposed to have done what he claims we do. Like I say, "Everybody's got something to hide, except for me and my monkey!" Keep in mind that when I wrote MONKEY, we were still using XT clones. Most PCs now are at least hundreds of times faster, if not thousands of times. So when you run MONKEY, it will appear to succeed instantaneously. In order to observe the progressing towards and regressing away from the goal, pick a smaller number of offspring generated per iteration. Here is Ian Musgrave's email to me about it back when all this was going down. As you can see, that was nearly 20 years ago. I remember having received a xerox copy of what Remine wrote about me and my MONKEY and I remember being amazed at how much he had misunderstood it. BTW, I had written MONKEY about a decade before this and posted it in a library on CompuServe where it was constantly downloaded at least once a month for years thereafter. I explicitly asked for feedback and got none except that my numbering of the Markovian chain steps in MPROBS was off by one, and spurious complaints about teleological assumptions in my fitness test which both I and even Dawkins in his original presentation discussed and showed to be irrelevant -- ie, since the single-step selection model made the exact same fitness test, when why was it such an abysmal failure when cumulative selection was such a resounding success?
quote: The problem he reported in the PS was apparently an overflow error in the Turbo Pascal startup code, probably in a timer calibration loop, because PCs had gotten too fast. I was able to find a patch for it and fixed it. That's covered in the first section of my MONKEY page, Development History and Issues. Again, that is at cre-ev.dwise1.net/monkey.html. Edited by dwise1, : Changed subtitle
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2 |
But will you now read the source and see what it's really about?
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2
|
But single-step selection versus cumulative selection is very relevant. As such, it needs to be examined.
I forget. Were you one of those deluded creationists who tried to force single-step selection upon evolution with a terminally false creationist probability argument? Sorry, but it's hard to keep you all apart.
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2 |
CRR at ~6200 years. Is that just because you believe that your religion requires it? Or do you think that the physical evidence supports it? Id est, do you hold that religious belief despite the evidence? Or do you claim to hold it because of the evidence? I have seen many young-earth creationists insist that if their young-earth claims do turn out to be wrong, then they would have to through their Bibles into the dust bin and become atheists. Kind of extreme, but that is literally what a number of them insisted was the case, so I'm not making this up. What would your reaction be?
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2
|
Or were you asking why I think dogs would be genetically closer? Because we have in common many body tissues, organs, etc. that are built of similar proteins. From a common designer we would expect the genomes to contain many similar sequences. Except that is not how it works. Those proteins being compared are functionally identical even though many of the individual amino acids can vary widely such that their amino acid sequences can differ greatly. It is the patterns of those differences that we are interested in. Yes, a common designer (or even a decent one, for that matter) would be expected to reuse certain proteins based on their functionality, but why would we ever expect that designer to also use such a wide variation of actual sequences. And why would he do that in such a deliberate manner so as to support the idea that those genomes are related to each other in exactly the way that we would expect if evolution actually happened? Is your trickster god's name Loki by any chance? Here's a little something along those lines that I wrote in an email in 1996 to a creationist who had rehashed that tired old false probability argument about the probability of a protein 80 amino acids long just having them all fall into place at random in just that exact order that is needed for the protein to work. Two major problems with that claim:
quote: That just describes an active site. I believe we will find that most of a proteins sequence consists of a some active sites connected with a lot of structural sequences. Thwaites and Awbrey didn't come out and say it here, but I would expect that most of the locations in those structural sequences would accept any amino acid.
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2 |
CRR writes: I don't see any evidence that those people are creationists. Professor Dr Bernard Brandstater, Prof. Stuart Burgess, Professor Dr Ben Carson, Dr Raymond Damadian, Dr John Hartnett, Dr Raymond Jones, Dr Felix Konotey-Ahulu, Dr John Sanford, Dr Wally (Siang Hwa) Tow You would need to research into the biographies and writings of each and every person in that list in order to make that determination. Since CRR presented that list and claimed them all to be creationists, it is incumbent on CRR to provide that information. I do know that plant geneticist Dr. John Sanford is and IDist and a YEC. The only reason I know that is that a local creationist cited him in a debate as a scientific source, making sure to repeatedly point to his PhD, without making even a single mention that he was also a YEC. He did the same thing with another PhD with absolutely no mention that she's a professional creationist and spokesperson for AiG. IOW, that creationist was deliberately lying to the audience. You should have recognized Dr. Ben Carson, the one who believes that the pyramids were built to store grain and now heads HUD, a post for which he is ill suited.
CRR writes: Yeah, not to mention. Scientists who were unaware of the ToE or did not attempt to use Biblical-based creationism in their scientific work were not creationists in the modern sense. Not to mention all past scientists such as Faraday and Maxwell. Excellent point. One that CRR completely ignored in his reply, Message 131. So what's his point? ABE:Though the story of Dr. John Sanford raises a pertinent question. A person can be both religious, deeply religious even, and a scientist with no problem, just so long as he actually does science when serving as a scientist. That may require some compartmentalization. Certainly there is no problem with pursuing science motivated strongly by one's religious beliefs, but there is a problem with having religious motivation to subvert science. So the question is whether being a creationist in the modern sense would interfere with one's ability to do science, to make significant scientific contributions. In the Wikipedia article on Sanford he is described:
quote:Reviewing his significant work and accomplishments, they are all in the 1980's and 1990's, before he became a YEC. That must mean something. Edited by dwise1, : ABE
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2 |
But was he a creationist in the sense of our discussion? Was he also a YEC who denied real-world evidence in order to promote a narrow and false theology in order to proselytize and to try to have laws passed for that narrow and false theology to be taught in school as science?
There is no inherent conflict between science and religion unless you misuse either or both. There is no inherent conflict between creation and evolution unless you misrepresent and misuse either or both. You seem to think otherwise. You seem to equate evolution with atheism, evidence by your calling talkorigins an atheist site. You are a YEC and equating your "evolution model" with atheism is an article of faith. So if you truly think that Faraday was a modern-day YEC-type creationist, then please present the evidence. Or admit that Faraday has nothing to do with the discussion.
|
||||||||||||||||||||||||||||||||||
dwise1 Member Posts: 5949 Joined: Member Rating: 5.2 |
Yeah, those creationist "scientists who were creationists" lists are so stupid that they're just tiring.
Those lists are kind of along the lines of the arguments that strain all logic to arrive at a conclusion that there must be a god, at which point suddenly that indistinct "a god" becomes their specific god accompanied by their entire intricately detailed theology. How blazingly stupid! Just because a scientist was Christian doesn't mean he followed creationists' particular narrow theology. Did they ever bother to ask him to be associated with them and their false theology? Edited by dwise1, : Replaced impersonal "you" with "they" for clarity sake
|
|
|
Do Nothing Button
Copyright 2001-2023 by EvC Forum, All Rights Reserved
Version 4.2
Innovative software from Qwixotic © 2024