A peer-reviewed article that supports ID . . . or something else

"So, on this view, natural selection does not create information, it just transfers it into the genotype. The information is out there already, lying around."

Exactly. So as long as there is an environment, no intelligence is needed in order to create information, in genomes, complex, specified or any other. This guy has just disproven all of his own previous nonsense. And of course, now that it is in a journal for all to see, I'm sure that many people are going to point this out to him.

Why would someone with absolutely no knowledge whatsoever about genetics, population genetics, selection or fitness try to make such obviously flawed arguments? Who exactly does he think he is fooling? Who does he think is going to read this journal? Does he think that he can get away with calling this evidence for ID? Next thing you know he will be trying to patent phylogenetics!

In Dawkins’s Weasel case, a monkey with a typewriter finds the solution after about 1040 tries, while one version of Dawkins’s program would take only about 728 tries. The active information is the log of the ratio of these numbers, about 124 bits.

What weasels. Seems that engineers are now poised to take over biology. So how does an accurate description of ID creationism's failure have to be worded now? "There are no actual peer reviewed papers on ID creationism, or any other form of creationism, or anything that actually supports creationism, in biological research journals." Would require explicit definitions of what "peer reviewed" and "biological research journal" means. Peer viewed doesn't mean fellow unqualified wingnuts, and "journal" doesn't include newsletters, or an unrespected obscure journal edited and published by one wingnut.

it said nothing about evolution and was a purely theoretical paper (negative use of theoretical)

So what's the fuss about?

It is in the computer engineering journal IEEE Transactions on Systems, Man and Cybernetics A, Systems & Humans in the September 2009 issue.

So, on this view, natural selection does not create information, it just transfers it into the genotype.

Is it only my browser that is rendering the number of tries for the random version as just a trifle over one thousand? :)

I see that D&M are on about "ratchetting" again. Isn't the apparent retention of correct characters in the Weasel program is merely a consequence of the simplistic nature of the problem acting in conjunction with the draconian level of selection in the implementation. (Am I correct in thinking of this behaviour as "emergent"?)

Is it not ironic that Dembski sees "Design" in the supposed "latching".

The latching is an illusion. Therefore "Intelligent Design" is also an illusion. QED.

Ravilyn Sanders said:

In Dawkins’s Weasel case, a monkey with a typewriter finds the solution after about 1040 tries, while one version of Dawkins’s program would take only about 728 tries. The active information is the log of the ratio of these numbers, about 124 bits.

A typo here? A monkey would take way more than 1040 attempts and the log of the ratio is way more than 2^124. Missed a gillion,zillion or a gazillion there?

Kevin B said: Is it only my browser that is rendering the number of tries for the random version as just a trifle over one thousand? :) I see that D&M are on about "ratchetting" again. Isn't the apparent retention of correct characters in the Weasel program is merely a consequence of the simplistic nature of the problem acting in conjunction with the draconian level of selection in the implementation. (Am I correct in thinking of this behaviour as "emergent"?) Is it not ironic that Dembski sees "Design" in the supposed "latching". The latching is an illusion. Therefore "Intelligent Design" is also an illusion. QED.

fixed.

Kevin B said: I see that D&M are on about "ratchetting" again. Isn't the apparent retention of correct characters in the Weasel program is merely a consequence of the simplistic nature of the problem acting in conjunction with the draconian level of selection in the implementation. (Am I correct in thinking of this behaviour as "emergent"?) Is it not ironic that Dembski sees "Design" in the supposed "latching". The latching is an illusion. Therefore "Intelligent Design" is also an illusion. QED.

Dembski and Marks’s argument ends up leaving us to argue about where the laws of physics ultimately come from, and most evolutionary biologists will not feel too worried.

— Joe Felsenstein

So what’s the fuss about?

— Michael Roberts

I forwarded this page to the IEEE Transaction's editor for an FYI. I suspect they don't consider biological aspects of such articles. I also used to be an IEEE member, and I'm surprised (ugh) to see that Dembski & Mark are also members.

The WAD's blog said: Question: When Dawkins introduced this example, was he arguing pro-Darwinism? Yes he was. In critiquing his example and arguing that information is not created by unguided evolutionary processes, we are indeed making an argument that supports ID.

I loved the way that William Dembski after only about 7 comments, several of which were his, stated that he was "weary of the quibbling" and closed the comments (I can't imagine that Richard Dawkins or PZ Myers would do that).

e-dogg said: For the millionth time, the weasel was cited as a simplistic example of a process. He used it to illustrate a point, not to say, "See, it works in the computer so it must work in biology!"

Does the smoothness of fitness surfaces come from this weakness of long-range interactions in physics? If so, then Dembski and Marks’s argument ends up leaving us to argue about where the laws of physics ultimately come from, and most evolutionary biologists will not feel too worried.

— Joe Felsenstein

Mike Elzinga said:

Does the smoothness of fitness surfaces come from this weakness of long-range interactions in physics?

— Joe Felsenstein

If these guys actually learned some physics and chemistry, let along some biology, they might have some hope of publishing a paper that made sense. Won’t happen.

D&M have mischaracterized the entire process of evolution right from the beginning of their paper. Therefore, any argument they want to make about “endogenous” or “exogenous” information is irrelevant and simply bogus. So is their invocation of “active information”

They should start with a simpler problem to see the bogusness of their critique. Water is sitting within a glass tube. Calculate the probability of finding a water molecule on the glass surface a given distance above the mean level of the surface of the water. Do this problem with and without a gravitational field.

How close will their answers come to reality if they simply employed their search algorithms with none of what they apparently disparage as “active information” injected into the search?

Why would not putting such information into the search be justified? Why is putting it in “cheating”?

It seems like D&M don’t think it is fair to put any knowledge of how the universe works into the search algorithm of a computer program. This is anti-science as near as I can tell.

The universe is full of potential wells; it’s what we know. If it weren’t, nothing would exist except elastically scattering particles at most.

A fitness landscape with potential wells about which solutions cluster is a perfectly valid representation of how nature searches for “solutions” to “problems” from adjacent sets of solutions. It’s what evolution does.

Sheesh! Why is that “unfair”?

Mike Elzinga said: How close will their answers come to reality if they simply employed their search algorithms with none of what they apparently disparage as “active information” injected into the search? Why would not putting such information into the search be justified? Why is putting it in “cheating”? A fitness landscape with potential wells about which solutions cluster is a perfectly valid representation of how nature searches for “solutions” to “problems” from adjacent sets of solutions. It’s what evolution does. Sheesh! Why is that “unfair”?

If these guys actually learned some physics and chemistry, let along some biology, they might have some hope of publishing a paper that made sense.

— Mike Elzinga

Mike said: What weasels. Seems that engineers are now poised to take over biology. So how does an accurate description of ID creationism's failure have to be worded now? "There are no actual peer reviewed papers on ID creationism, or any other form of creationism, or anything that actually supports creationism, in biological research journals." Would require explicit definitions of what "peer reviewed" and "biological research journal" means. Peer viewed doesn't mean fellow unqualified wingnuts, and "journal" doesn't include newsletters, or an unrespected obscure journal edited and published by one wingnut.

Xpost Pharyngula: More on Ev. This paper is open source, available to anyone with an internet connection. Rather than babbling on about "information", they actually measure Shannon information content before and after. Information increases by evolution. Nucleic Acids Res. 2000 July 15; 28(14): 2794–2799. PMCID: PMC102656 Copyright © 2000 Oxford University Press Evolution of biological information Thomas D. Schneidera National Cancer Institute, Frederick Cancer Research and Development Center, Laboratory of Experimental and Computational Biology, PO Box B, Frederick, MD 21702-1201, USA Received March 7, 2000; Revised May 25, 2000; Accepted May 25, 2000. This article has been cited by other articles in PMC. AbstractHow do genetic systems gain information by evolutionary processes? Answering this question precisely requires a robust, quantitative measure of information. Fortunately, 50 years ago Claude Shannon defined information as a decrease in the uncertainty of a receiver. For molecular systems, uncertainty is closely related to entropy and hence has clear connections to the Second Law of Thermodynamics. These aspects of information theory have allowed the development of a straightforward and practical method of measuring information in genetic control systems. Here this method is used to observe information gain in the binding sites for an artificial ‘protein’ in a computer simulation of evolution. The simulation begins with zero information and, as in naturally occurring genetic systems, the information measured in the fully evolved binding sites is close to that needed to locate the sites in the genome. The transition is rapid, demonstrating that information gain can occur by punctuated equilibrium.

Joe Felsenstein said: Well, in the paper they don't say it is "cheating" or "unfair". All that happens when their concepts get used in ID debates.

Raven wrote:

"Rather than babbling on about “information”, they actually measure Shannon information content before and after. Information increases by evolution."

So why don't they? Why did the editor let them get away with this when the standard had been set nearly ten years ago for research in this area?

Maybe they just couldn't bear to actually quantify information, cause you know, then they would have had to quantify complex specified information and then they would have had to admit that that could produced by exactly the same mechanism. Still no intelligence in sight.

James F said: It's the same definition as ever. No data in support of ID in peer-reviewed scientific research papers. In particular, no mechanism of how an "intelligent agency" influences biological complexity.

So why don’t they? Why did the editor let them get away with this when the standard had been set nearly ten years ago for research in this area?

AbstractHow do genetic systems gain information by evolutionary processes? Answering this question precisely requires a robust, quantitative measure of information. Fortunately, 50 years ago Claude Shannon defined information as a decrease in the uncertainty of a receiver. For molecular systems, uncertainty is closely related to entropy and hence has clear connections to the Second Law of Thermodynamics. These aspects of information theory have allowed the development of a straightforward and practical method of measuring information in genetic control systems. Here this method is used to observe information gain in the binding sites for an artificial ‘protein’ in a computer simulation of evolution. The simulation begins with zero information and, as in naturally occurring genetic systems, the information measured in the fully evolved binding sites is close to that needed to locate the sites in the genome. The transition is rapid, demonstrating that information gain can occur by punctuated equilibrium.

So it’s the same old false dilemma, proving ID by only disproving evolution? Evidence against you is evidence for me?

Why would not putting such information into the search be justified? Why is putting it in “cheating”?

Dembski’s complex specified information without a definition and a way to quantitate it is just bafflegab, meaningless. So what is Active Information or Injected Information. Without a definition and quantitation, just more meaningless bafflegab.

Slightly off-topic: Dembski gets fooled by an "Urban Legend" - http://www.uncommondescent.com/humor/off-topic-school-answering-machine/

poor dr dr d d

Mike Elzinga said: But information is not the same as entropy; it never has been.

stevaroni said:

Why would not putting such information into the search be justified? Why is putting it in “cheating”?

How does one even do a search without at least some information? Isn't the first question in any search "What am I searching for?", a question whose answer automatically provides at least some tidbit of information. (Admittedly, for evolution at least, the answer is usually something terribly generic, along the lines of "gets laid more than the other ones", but still, that's information.)

Eric Finn said: Now, I would like to hear your opinion on the following reasoning: Entropy is an extensive quantity, which means that it is proportional to the size of the system under study. http://en.wikipedia.org/wiki/Intensive_and_extensive_properties#Examples_2 Thus, a person loosing weight is decreasing his or her entropy. What would be the prediction of the thermodynamics or the information theory regarding information gained by the person in this example?

As is the phrase “intelligent design” - to me that phrase looks like it was designed to trick people into not thinking about the engineering implications of what they’re trying to imply without actually saying.

— Henry J

The ID crowd will very likely cite D&M’s paper, just as Joe Felsenstein surmises. They will not pick up on the mischaracterization of evolution and the misuse of information that D&M make about the process of evolution in this paper.

Dembski about his paper: “in critiquing his example and arguing that information is not created by unguided evolutionary processes, we are indeed making an argument that supports ID.”

Tom Schneider: The simulation begins with zero information and, as in naturally occurring genetic systems, the information measured in the fully evolved binding sites is close to that needed to locate the sites in the genome. The transition is rapid, demonstrating that information gain can occur by punctuated equilibrium.

Joe Felsenstein said:

Mike Elzinga said:

Does the smoothness of fitness surfaces come from this weakness of long-range interactions in physics?

— Joe Felsenstein

If these guys actually learned some physics and chemistry, let along some biology, they might have some hope of publishing a paper that made sense. Won’t happen.

I am going to make myself unpopular by being less hard on them. I don't think getting the "latching" issue wrong matters much, and their suggestion of associating an amount of information with the shape of the fitness surface is not necessarily silly. It depends on what you can then do with it. Unfortunately, about all it seems you can do with it right now is to simply use it to say "see, information was already there, nyah, nyah, nyah". And imply (without being explicit, and wrongly) that this somehow validates ID instead of either theistic or nontheistic evolution. I don't think the paper is really bad, just lacking a section that explains what all this is good for.

stevaroni said: Don't forget, Dembsi's paper argues for things that we've been saying for years, 1) Creationists whine that genetic information can only be destroyed, never created Dembski says false. Information in the genome can increase. 2) Creationists claim that increasing genetic information violates the 2LOT. Dembski says false. Information can, in effect, be concentrated concentrated in the genome by "appropriating" it from the environment. True, the environmental entropy may increase, but environment is big and can afford it (and don't forget, 2LOT establishes that, universally, entropy is increasing?) the genome gets more complex to compensate.

3) Creationists claim that only intelligence can provide useful information - that any such information must be "pre-loaded" by design. Dembski calls bull-pookey. If information can be obtained by the environment, then not only is the need for an intelligent organizer eliminated, but it's a dynamic proscess. The environment changes, the local fitness surface changes. The fitness surface changes, the organism tracks accordingly, like a rubber duckie in the sea tends to naturally flow toward the trough of the waves, even though the waves are always changing.

John Kwok said: Am disappointed you're willing to cut some slack on their behalf. Neither one deserves it IMHO and am dumbfounded that, of all people, Michael Ruse regards Bill as a "friend":

Joe Felsenstein said:

Mike Elzinga said:

Does the smoothness of fitness surfaces come from this weakness of long-range interactions in physics?

— Joe Felsenstein

If these guys actually learned some physics and chemistry, let along some biology, they might have some hope of publishing a paper that made sense. Won’t happen.

I am going to make myself unpopular by being less hard on them. I don't think getting the "latching" issue wrong matters much, and their suggestion of associating an amount of information with the shape of the fitness surface is not necessarily silly. It depends on what you can then do with it. Unfortunately, about all it seems you can do with it right now is to simply use it to say "see, information was already there, nyah, nyah, nyah". And imply (without being explicit, and wrongly) that this somehow validates ID instead of either theistic or nontheistic evolution. I don't think the paper is really bad, just lacking a section that explains what all this is good for.

John Kwok said: Am disappointed you're willing to cut some slack on their behalf. Neither one deserves it IMHO and am dumbfounded that, of all people, Michael Ruse regards Bill as a "friend":

Joe Felsenstein said:

stevaroni said: Don't forget, Dembsi's paper argues for things that we've been saying for years, 1) Creationists whine that genetic information can only be destroyed, never created Dembski says false. Information in the genome can increase. 2) Creationists claim that increasing genetic information violates the 2LOT. Dembski says false. Information can, in effect, be concentrated concentrated in the genome by "appropriating" it from the environment. True, the environmental entropy may increase, but environment is big and can afford it (and don't forget, 2LOT establishes that, universally, entropy is increasing?) the genome gets more complex to compensate.

Excellent points. I agree that the increase of information in the genome corresponds to an increase of entropy in the whole system, more than compensating. I suspect that something along these lines, rather than calculating an "active information" for the fitness surface, is the right way to think of these processes.

I have explained in my original post that his argument does not insist on a role for a Designer in all of this, but doubt he would draw the same conclusion. However, he is wrong when he says that the D&M argument is pro-ID. It is just as compatible with theistic evolution or even nontheistic evolution.

Ravilyn Sanders said:

In Dawkins’s Weasel case, a monkey with a typewriter finds the solution after about 1040 tries, while one version of Dawkins’s program would take only about 728 tries. The active information is the log of the ratio of these numbers, about 124 bits.

A typo here? A monkey would take way more than 1040 attempts and the log of the ratio is way more than 2^124. Missed a gillion,zillion or a gazillion there?

The question of "latching" does seem rather like a sideline to the whole subject of the WEASEL program, particularly when the WEASEL itself is not and never was a model of biological evolution. ("Latching" would make the WEASEL search even less biologically realistic.) However, on a second look, I think it's worth spending a moment on. First, whether the program "latches" letters or not determines whether it can be described as a partitioned search; second, Dembski's history of ignoring corrections on this point illustrates his [ahem] character. It is not a wholly minor matter when a purported academic misrepresents the content of the sources he cites.

Roberto said:

Ravilyn Sanders said:

In Dawkins’s Weasel case, a monkey with a typewriter finds the solution after about 1040 tries, while one version of Dawkins’s program would take only about 728 tries. The active information is the log of the ratio of these numbers, about 124 bits.

A typo here? A monkey would take way more than 1040 attempts and the log of the ratio is way more than 2^124. Missed a gillion,zillion or a gazillion there?

I think YOU misread it - 10 to the POWER of 40

While mathematically, it may well be that the subset of all possible adaptive landscapes is minuscule compared to the rest, the real world, on the space and time scales that biological organisms interact with it, is comparatively smooth.

WMD apparently finds the idea that the universe is not some over glorified devil's staircase incomprehensible. Long range interactions like electromagnetism and gravity are essentially smooth continuum theories. And the universe is pretty smooth except for the occasional hole and dimples where we find solar systems and life.

Joe Felsenstein said: I have tried to grapple with the implications of D&M's basic argument for debates about ID. It doesn't help to declare everything in their paper wrong -- that simply isn't true.

James F said: Good point, Prof. Felsenstein. I'm curious, though, about your thoughts on an issue raised by PZ Myers at Pharyngula, that D&M get the "Methinks it is like a weasel" algorithm wrong.

the points you make from it are the same.

I think you should rephrase that to:

"One could make the same points from a modified version of weasel as from the original."

However, that's not what Dembski et al are in fact doing, so it's kind of misleading to say it the way you did.

You have to understand that many of us have followed Dembski's "arguments" for years, and they've always been in the "not even wrong" category simply because the assumptions made to begin with were so far off.

I rather think you yourself are dissecting the presentation, looking for sense, when there really is none.

to coin an old cliche:

You're missing the forest for the trees.

Ichthyic said: I rather think you yourself are dissecting the presentation, looking for sense, when there really is none. to coin an old cliche: You're missing the forest for the trees.

Ravilyn Sanders said:

In Dawkins’s Weasel case, a monkey with a typewriter finds the solution after about 1040 tries, while one version of Dawkins’s program would take only about 728 tries. The active information is the log of the ratio of these numbers, about 124 bits.

A typo here? A monkey would take way more than 1040 attempts and the log of the ratio is way more than 2^124. Missed a gillion,zillion or a gazillion there?

Mike said: Sorry, I read that as 10^40 tries, not 1040.

Ravilyn Sanders said:

In Dawkins’s Weasel case, a monkey with a typewriter finds the solution after about 1040 tries, while one version of Dawkins’s program would take only about 728 tries. The active information is the log of the ratio of these numbers, about 124 bits.

A typo here? A monkey would take way more than 1040 attempts and the log of the ratio is way more than 2^124. Missed a gillion,zillion or a gazillion there?

I've always used the rough approximation that the information contained in a genome is information about the environment. I thought that natural selection demanded that idea, but now Dembski says that since God designed the environment, he gets to take credit for the genome, too. Makes me wonder if God gets to take credit no matter what the evidence is.

John Vreeland said: (snip)Makes me wonder if God gets to take credit no matter what the evidence is.

I had a quick flip through the D&M paper and am very grateful for Joe Felsenstein's easy to read analysis of it.

Hasn't Dembski shot himself in the foot here? He seems to show that biological information can be entirely explained by random mutation and a selection algorithm. Whether this is described as information creation or information gathering seems like a semantic quibble.

If Dembski uses this in the future to argue that a designer was required to put the information in the environment in the first place then that's very close to the theistic evolution position and a considerable improvement on ID.

Pete said: Hasn't Dembski shot himself in the foot here? He seems to show that biological information can be entirely explained by random mutation and a selection algorithm. Whether this is described as information creation or information gathering seems like a semantic quibble. If Dembski uses this in the future to argue that a designer was required to put the information in the environment in the first place then that's very close to the theistic evolution position and a considerable improvement on ID.

Pete wrote:

"Hasn’t Dembski shot himself in the foot here?"

Absolutely. And he did it in the scientific literature. Whenever he brings up information, this reference can be used to show that no intelligence is needed in order for information to increase in the genome. Man, you don't even need to quote mine.

Of course then Dembski will probably complain that the paper does not show that no intelligence is needed, since it did not define or quantify information! Let's face it, shooting himself in the head would not have done as much damage. Goos thing for him none of his devout followers read the scientific literature.

If Dembski uses this in the future to argue that a designer was required to put the information in the environment in the first place then that’s very close to the theistic evolution position and a considerable improvement on ID.

— Pete

So it does not validate ID over theistic evolution, and might not even validate theistic evolution over naturalism.

— Joe Felsenstein

Joe F.: So, on this view, natural selection does not create information, it just transfers it into the genotype. The information is out there already, lying around.

I've always been a bit skeptical of computer models. What are these models, models of. The real world.

If the data input is not good, and/or the model is not an accurate one of the real world, then the output is no good.

One computer model I've dealt with was not accurate and cost several thousand people several tens of millions of dollars.

A good model will represent the real world and result in testable predictions. Other than these internet commentaries, not aware that Dembski's programs have made any predictions or that they have even been tested experimentally.

The other problem with computer models is that we also have the real thing, reality, to study and run experiments in and on. Any computer model that proves evolution is impossible has to deal the the fact that we see it going on every day all around us.

The newest (and predicted!) pandemic, swine flu might be impossible but it is here and is about to scare most, sicken many and kill a few creationists (and everyone else) whether they accept evolution or not.

Joe F.,

I wrote to Bill the day he announced the paper was published and said essentially the same thing you did including me not understanding what point was being made. It's essentially correct, the ratcheting or lack thereof in Weasal is inconsequential, but it doesn't really tell us anything new and simply leads to the question of where the information came from in the first place.

Perhaps where the authors are heading with this is in establishing some kind of metric for how much work the genetic mechanism had to do in order to extract the information from the environment. Just because it's possible for an energetic process to decrease entropy in one place at the expense of increasing entropy in another doesn't mean it's plausible or likely that any given change in entropy will occur. It's possible that the library of congress would have come into existence without intelligent agency but it's rather unlikely. Is the information concentrated in living things like the library of congress? Is a simple trial & error with feedback mechanism sufficient (statistically reasonable) to extract the information contained by living things from the environment with the time and opportunity it had to do it? That is and remains the essential question.

John Vreeland said: I thought that natural selection demanded that idea, but now Dembski says that since God designed the environment, he gets to take credit for the genome, too. Makes me wonder if God gets to take credit no matter what the evidence is.

“When you are attributing the wonders of nature to these mindless material mechanisms, God’s glory is getting robbed.” He further added, “And so there is a cultural war here. Ultimately I want to see God get the credit for what he’s done — and he’s not getting it.”

Makes me wonder if God gets to take credit no matter what the evidence is.

Dave Springer said: Perhaps where the authors are heading with this is in establishing some kind of metric for how much work the genetic mechanism had to do in order to extract the information from the environment.

Just because it’s possible for an energetic process to decrease entropy in one place at the expense of increasing entropy in another doesn’t mean it’s plausible or likely that any given change in entropy will occur.

Perhaps where the authors are heading with this is in establishing some kind of metric for how much work the genetic mechanism had to do in order to extract the information from the environment.

Cross-posting. Mike said it better.

Of course, even if this could be accurately characterized as a "pro-ID" paper, the proper response is, "OK, there's one. Whoop-de-do."

MPW said: Of course, even if this could be accurately characterized as a "pro-ID" paper, the proper response is, "OK, there's one. Whoop-de-do."

Recognition of the inability of search algorithms in themselves to create information is “very useful, particularly in light of some of the sometimes-outrageous claims that had been made of specific optimization algorithms” [7]. Indeed, when the results of an algorithm for even a moderately sized search problem are “too good to be true” or otherwise “overly impressive,” we are faced with one of two inescapable alternatives. 1) The search problem under consideration, as gauged by random search, is not as difficult as it first appears. Just because a convoluted algorithm resolves a problem does not mean that the problem itself is difficult. Algorithms can be like Rube Goldberg devices that resolve simple problems in complex ways. Thus, the search problem itself can be relatively simple and, from a random-query perspective, have a high probability of success. 2) The other inescapable alternative, for difficult problems, is that problem-specific information has been successfully incorporated into the search.

Christensen and Oppacher [7] note the “sometimes outrageous claims that had been made of specific optimization algorithms.” Their concern is well founded. In computer simulations of evolutionary search, researchers often construct a complicated computational software environment and then evolve a group of agents in that environment. When subjected to rounds of selection and variation, the agents can demonstrate remarkable success at resolving the problem in question. Often, the claim is made, or implied, that the search algorithm deserves full credit for this remarkable success. Such claims, however, are often made as follows: 1) without numerically or analytically assessing the endogenous information that gauges the difficulty of the problem to be solved and 2) without acknowledging, much less estimating, the active information that is folded into the simulation for the search to reach a solution.

Mike quotes from the Dembski paper... Indeed, when the results of an algorithm for even a moderately sized search problem are “too good to be true” or otherwise “overly impressive,” we are faced with one of two inescapable alternatives. 1) The search problem under consideration, as gauged by random search, is not as difficult as it first appears. Just because a convoluted algorithm resolves a problem does not mean that the problem itself is difficult. Algorithms can be like Rube Goldberg devices that resolve simple problems in complex ways. Thus, the search problem itself can be relatively simple and, from a random-query perspective, have a high probability of success. 2) The other inescapable alternative, for difficult problems, is that problem-specific information has been successfully incorporated into the search.

I said that inexpertly because Dembski's already spinning with "simple problem - convoluted search".

What I meant to say is "3) The problem is demonstrated to be not nearly as insurmountable by simple iteration as it first appears."

stevaroni said: I said that inexpertly because Dembski's already spinning with "simple problem - convoluted search". What I meant to say is "3) The problem is demonstrated to be not nearly as insurmountable by simple iteration as it first appears."

Joe Felsenstein said:

James F said: Good point, Prof. Felsenstein. I'm curious, though, about your thoughts on an issue raised by PZ Myers at Pharyngula, that D&M get the "Methinks it is like a weasel" algorithm wrong.

I suspect D&M did get that wrong, but it doesn't matter much which variant on the Weasel algorithm you use, the points you make from it are the same. Is the point to find something they are Wrong about and crow? Or is it to figure what their basic argument is? To me, the latetr.

Ian Musgrave said: The point here is scholarly honesty. Dembksi knows that Dawkins' "Weasel" program is not a latching program, he's not only been told several times, he's had it graphically illustrated to him. Yet he still attributes latching to Dawkins. If he can't cite this minor thing correctly, what are we supposed to think about the rest of his work? His treatment of Yockey's work is misleading as well. And characterising the "Monkey at a typewriter" as an evolutionary search algorithm they are well off the rails. The presence of such basic errors must speaks to the overall quality of Dembski's work in this paper, which bears directly on the trust we can place in the main work.

Gary Hurd said:

John Vreeland said: I thought that natural selection demanded that idea, but now Dembski says that since God designed the environment, he gets to take credit for the genome, too. Makes me wonder if God gets to take credit no matter what the evidence is.

Dembski gave a talk at the Baptist Fellowship Church in Waco, TX (March 7, 2004), that was taped recorded. Relevant to the current topic, Dembski, in response to an audience member's question said,

“When you are attributing the wonders of nature to these mindless material mechanisms, God’s glory is getting robbed.” He further added, “And so there is a cultural war here. Ultimately I want to see God get the credit for what he’s done — and he’s not getting it.”

Frank J said: Is there no end to Dembski's chutzpah? It's IDers like Dembski who refuse to name the designer, while ID critics like Ken Miller, Francis Collins and John Haught freely admit that it's God. It's IDers who refuse to say what the designer did, when and how, while ID critics who do the hard work to find out what God did, when and how, and even speculate on why He might have done it that way. Dembski knows darn well who is robbing God of His glory.

MPW said: Of course, even if this could be accurately characterized as a "pro-ID" paper, the proper response is, "OK, there's one. Whoop-de-do."

Mike Elzinga said: The difficulties come when semantics and conflated concepts begin to obscure the science when such a search algorithm is used on a representation of a physical system. Your speculation that there is some measure of “work” involved here is a step in the direction of conflation. Work has a specific meaning in physics, and such an interpretation begins to suggest that, in the context of evolution, there is work being done to “climb uphill” and that this makes evolution more improbable.

Dembski and Marks’s argument would then at most favor theistic evolution and could not be used to favor ID over that.

Dave Springer said: Joe F. ID is perfectly compatible with deism and deism is perfectly compatible with a deity who is long gone or otherwise disengaged from the universe today.

Just because a convoluted algorithm resolves a problem does not mean that the problem itself is difficult. Algorithms can be like Rube Goldberg devices that resolve simple problems in complex ways.

Robin said:

Dave Springer said: Joe F. ID is perfectly compatible with deism and deism is perfectly compatible with a deity who is long gone or otherwise disengaged from the universe today.

As I see it, ID is perfectly compatible with anything because it explains nothing. How is ID even remotely useful as a conceptual tool? To say, "it seems that there is some intelligence" behind the workings of the universe when that intelligence seems to impact nothing about our study of this universe, it amounts to nothing other than perhaps a statement some people's ego needs in order to feel there's some purpose to their getting out of bed in the morning.

"Why would someone with absolutely no knowledge whatsoever about genetics, population genetics, selection or fitness try to make such obviously flawed arguments? Who exactly does he think he is fooling? Who does he think is going to read this journal? Does he think that he can get away with calling this evidence for ID?"

This doesn't seem to me to be about credibility for ID in the scientific community. I'm guessing it's more about being able to bamboozle the existing unsophisticated grass-roots scientifically illiterate creationists who will be trumpeting Dembski's "Pro-ID article in a peer reviewed journal" in their future message board and blog arguments. It'll become one more piece of misinformation for the rest of us to have to deal with along with the existing arsenal of misinformation that we're accustomed to being spammed with.

It'll also become one more piece of reinforcement for the "mountain" of anti-evolutionary "evidence" that these grass-roots creationists have been accumulating to bolster their misguided view that "evolution has been thoroughly debunked".

The enormous amount of active information provided by partitioned search is transparently evident when the alphabet is binary. Then, independent of L, convergence can always be performed in two steps. From the first query, the correct and incorrect bits are identified. The incorrect bits are then complemented to arrive at the correct solution. Generalizing to an alphabet of N characters, a phrase of arbitrary length L can always be identified in, at most, N ? 1 queries. The first character is offered, and the matching characters in the phrase are identified and frozen in place. The second character is offered, and the process is repeated. After repeating the procedure N ? 1 times, any phrase characters not yet identified must be the last untested element in the alphabet.

Mike E writes.... Basically what they do is calculate the probability of hitting by an entirely random search on, say, the sequence of digits necessary to open a combination lock (one of their examples). The “distance” to, or what they call the “baseline difficulty” .... is essentially the negative of the log to base 2 of the probability of hitting on the solution by a strictly random search. They claim that this “active information” “assesses numerically the problem specific information incorporated into the search.”

fnxtr said: Indeed. As has been asked before, what would a science that incorporates supernatural or teleological concepts even look like? How would it work? It's fine to say "God did it", but at some point if you have any curiosity, you're going to ask "How?". If you just shrug and say "It was a miracle", well, the science is pretty much out the window by that point, wouldn't you say?

Conflating all ID proponents with YEC is not intellectually honest.

William Dembski: The implications of intelligent design are radical in the true sense of this much overused word. The question posed by intelligent design is not how we should do science and theology in light of the triumph of Enlightenment rationalism and scientific naturalism. The question is rather how we should do science and theology in light of the impending collapse of Enlightenment rationalism and scientific naturalism. These ideologies are on the way out…because they are bankrupt.

I yammered log ( (all valid solutions / specific solution as found) / random states )

It does need to be understood that it does indeed require work (and work is the correct term to use) to move information from the environment to the genome.

Dave Springer said: I meant "work" with the specific meaning in physics. Moving matter or energy across a boundary from lower concentration to higher concentration requires work be performed. This also applies to information going across the boundary.

It does need to be understood that it does indeed require work (and work is the correct term to use) to move information from the environment to the genome. Dembski’s “active information” is a measure of the difference in the work required between a search of randomly distributed information versus information that is not randomly distributed.

Felsenstein understands this and explains it terms of the smoothness of a fitness landscape. He explains that the fitness landscape in Weasel is perfectly smooth and thus requires a minimal amount of work to find the target. The smoothness in Weasel’s fitness landscape of course was configured by an intelligent agent. If the fitness landscape in the real world is smoothed to some degree then it raises the question of how and when it was smoothed.

Personally I prefer to cut right to the chase and focus on the fine tuning problem in cosmology where the smoothness is much more amenable to precise measurement compared to the biological fitness landscape.

The hope of ID opponents, including Susskind, is pinned on faith that a theory explaining why the universe is so finely tuned to support the evolution of organic life will be found. If not string theory then something else.

stevaroni said: One other thing I realized only after a few days, Mike, is that this paper once again implicitly perpetuates the "lottery winner" fallacy. That is, as applied to evolution, it perpetuates the idea there's something unique about the specific solution at hand, and ignores the fact that there are many, many other possible solutions that are equally valid, and any one of them would solve the problem (albeit, not in the 'specified' way, but for evolution, the 'specification' is an artificial construct).

What mass and/or energy are contained in “information”. By which of the four known forces does “information” interact with other matter and energy in the universe.

raven said:

What mass and/or energy are contained in “information”. By which of the four known forces does “information” interact with other matter and energy in the universe.

My understanding is that information is an attribute of mass/energy like shape. It makes no sense to talk about information without realizing that it exists as a property of a tangible substrate. It is not a disembodied entity existing free of mass/energy. If anyone knows more, post it.

Dave Springer said: Personally I prefer to cut right to the chase and focus on the fine tuning problem in cosmology where the smoothness is much more amenable to precise measurement compared to the biological fitness landscape. The cosmological fitness landscape is widely acknowledged by physicists as being extraordinarily smooth and there's no theory to explain it.

The hope of ID opponents, including Susskind...

The problem with string theory is no one knows how it can be experimentally falsified or verified. I read recently that string theory predicts a slight negative curvature of space/time and that the next generation of instrumentation may have enough sensitivity to measure the predicted negative curvature.

In the meantime cosmological ID is a competing hypothesis...

Dave Springer said: Many of us on the ID side consider TE to be a specific case of ID.

Conflating all ID proponents with YEC is not intellectually honest.

I meant to write: "...so there’s nothing to rule out flicking a magic wand to make it happen, there’s nothing to rule out sneezing complete systems out of a gigantic schnoz in a perfectly intentional manner to produce the Universe or the bacterial flagellum (leaving some things oddly random?), there's nothing in short to prevent any ridiculous Creation scenario from being accepted, or facilitate accepting one over another."
Got lost under all the editing tape before it went to press.

Mike Elzinga said: Obviously matter and energy are involved, but so is the “eye of the beholder”. What is seen as “significant” or as “information” has little relationship to the physics of matter and energy involved.

This is a paper submitted to a journal for electrical engineering. It covers the dilemma of how the best way to solve a problem is to know the answer in advance. The question is, how much of the answer is needed?

The paper addresses information retrieval and processing in an artificial and controlled environment.

There is no natural analog for this. It does not occur in Nature. If there was a natural analog we would already be using it. If it occurred in nature Dr. Dembski would be submitting the paper to Nature or a similar journal, not a journal for electrical engineering.

Attempting to apply artificial constraints and their solutions to the natural world is delusional at best. Suggesting the results of a study on information retrieval indicates the intervention of a supernatural force is of little merit and even worthy of scorn.

The texture of the "landscape" (ontology) doesn't determine the functioning of the real world (phylogeny)--no matter how wholistic Dr. Dembski's metaphysics may be.

Mike Elzinga said: It is not a good idea to just make up crap. If you don’t know anything about physics, at least have the decency to admit it instead of bluffing and starting a mud wrestling contest. What mass and/or energy are contained in “information”. By which of the four known forces does “information” interact with other matter and energy in the universe. What potential wells result when “information” is brought into close proximity to other “information” or matter?

Mike Elzinga said: It is not a good idea to just make up crap. If you don’t know anything about physics, at least have the decency to admit it instead of bluffing and starting a mud wrestling contest. What mass and/or energy are contained in “information”. By which of the four known forces does “information” interact with other matter and energy in the universe. What potential wells result when “information” is brought into close proximity to other “information” or matter?

Dave Springer replied: Hi Mike, Leo Szilard said that Maxwell's Demon would need to have information about the speed of the molecules moving across the boundary in order to admit the fast ones and block the slow ones. He further went on to say that acquiring information takes an expenditure of energy. So I guess you think Szilard doesn't know anything about physics either and is just making crap up. It appears you’re the one without any appreciable understanding of physics.

Maxwell's demon was a fiction invented to illustrate a point, so I'm not sure what the point is of arguing about what it would need to actually work. That its existence would break the rules was kind of the point of that story in the first place.

Dave Springer said: It appears you're the one without any appreciable understanding of physics.

Eric, you do realize that Dave can't answer that question? I find it amusing that someone is claiming a knowledge of physics by quoting a hypothetical discussion of a hypothetical demon.

That's frickin' hilarious.

Where are the actual answers to the questions posed, Dave? I guess being a retired hack coder doesn't automatically give you much of an education.

Dave Springer said: Interesting. Szilard is one of the 20th century's more renowned physicists. He is credited with One of his academic advisers was Albert Einstein. His credits include being the first person to conceive of the nuclear chain reaction and he shares a patent on nuclear reactors with Nobel Prize winner Enrico Fermi. It appears you're the one without any appreciable understanding of physics. Take care.

Dave, I'd appreciate some sort of acknowledgment on my points.
Also, do you know why I never get confirmation e-mails when I try to register at Uncommon Descent?

Hi Eric

Try this little experiment. Turn off your computer so it's using zero energy then see if you can read those morse codes.

I'd suggest trying that with your brain but that would be mean spirited.

Dave Springer said: Hi Eric Try this little experiment. Turn off your computer so it's using zero energy then see if you can read those morse codes. I'd suggest trying that with your brain but that would be mean spirited.

Dan said: The funny part is that you ended by insulating that Eric is stupid!

Dave Springer said: Hi Eric Try this little experiment. Turn off your computer so it's using zero energy then see if you can read those morse codes. I'd suggest trying that with your brain but that would be mean spirited.

Stanton said:

Dan said: The funny part is that you ended by insulating that Eric is stupid!

You mean "insinuated"?

Henry J said: Maxwell's demon was a fiction invented to illustrate a point, so I'm not sure what the point is of arguing about what it would need to actually work. That its existence would break the rules was kind of the point of that story in the first place.

Stuart Weinstein said: There is a fair bit of literature regarding the demon, and from what I remember it explains why such a demon even it did exist, and was bounded by other laws like conservation of mass-energy would fail to avert the second law. Imaginary yes. So were Einstein's Gedaneken experiments. I don't think anybody refers to those as a waste of time. Imaginary or not, it is instructive to consider why the demon fails. "It breaks the rules" is a poor answer. It is these types of thought experiments that allow us to probe physics with our minds.

Dan said:

Stanton said:

Dan said: The funny part is that you ended by insulating that Eric is stupid!

You mean "insinuated"?

Yep, thanks. Spelling was never my forte -- that's why I went into science, where the ideas are more important than the expression -- but that mistake was especially bad.

p. 293: "Here's my version of the law of conservation of information: "In a nonbiological context, the amount of specified information initially present in a system S_i, will generally equal or exceed the specified information content of the final system, S_f." This rule admits only two exceptions. First, the information content of the final state may exceed that of the initial state, S_i, if intelligent agents have elected to actualize certain potential states while excluding others, thus increasing the specified information content of the system. Second, the information content of the final system may exceed that of the initial system if random processes, have, by chance, increased the specified information content of the system. In this latter case, the potential increase in the information content of the system is limited by the "probabilistic resources" available to the system."

— Jeffrey Shallitt quoting Meyer