Jonathan Wells: Another ID Creationist Who Doesn't Understand Information Theory

I love the way these IDiots keep claiming that evolution is dead already and it's just that the body hasn't stopped moving yet. It's as if they think they're God and can simply speak it into being so.

Like many incompetent people, Wells is blissfully unaware of his incompetence.

There is a legitimate sense of "information" which is not increased by duplication. If I own, and have read, 2 copies of a book, I am not thereby better informed than if I had only owned and read one. The crucial point is that after a gene has been duplicated, and one of the duplicates has mutated and been selected for a new function, the genome then undoubtedly contains more information than before.

David Evans:

The commonsense definition of "information" which you are appealing to is not the definition used by mathematicians, computer scientists, and mathematical biologists. Although it may at first seem counterintuitive, two copies of a string really can have more information than one in the Kolmogorov sense.

But even in the informal sense, I can argue that multiple copies of a text may contain more information than simply one. For example, I can encode a message with the number of the copies - I could say: "Go to my office, and count the number of copies of the Origin of Species on my shelf. That is how much I will pay you for your doughnut."

David Evans said: There is a legitimate sense of "information" which is not increased by duplication.

If you make a copy of a message, then two people can read it at the same time. If they can then act of it faster, they might be able to make a difference in their situation.

In the same way, two copies of a gene might make twice as much RNA, which might make a difference to the organism. Quantity of information is irrelevant; it's what you do with it that counts.

John Wendt said: If you make a copy of a message, then two people can read it at the same time. If they can then act of it faster, they might be able to make a difference in their situation. In the same way, two copies of a gene might make twice as much RNA, which might make a difference to the organism. Quantity of information is irrelevant; it's what you do with it that counts.

I know diddly-zip about information theory. Where can I learn the basics? Is there an on-line Information Theory for the Stupid Dummies type site?

RBH said: A duplication of one such gene results in more of the protein being produced ... But some cell processes are dose dependent with respect to proteins, so a straight gene duplication can add 'biological information' ...

* A sidebar to the article explained how gene copy number variations, with some people having fewer or more copies of the same gene than others, are being seen as an increasingly important genetic phenomenon. Studies suggest that about 12% of the human genome consists of copy number variable regions, and that human cultures where the people have a starchy diet have additional numbers of copies of the gene for amylase, a starch-digesting enzyme found in saliva. It is increasingly apparent gene duplications may well be much more common than previously thought, and that rapid proliferation of gene copies may be driven by evolutionary selection pressures.

bigjohn756 said: I know diddly-zip about information theory. Where can I learn the basics? Is there an on-line Information Theory for the Stupid Dummies type site?

Creationists like Wells are also fond of insisting that gene mutations are frequent events and almost all such mutations are deleterious, leading to non-functional or dysfunctional proteins. Having two copies of any gene would thus greatly improve the chances of having some of the good protein. X-linked SCID (severe combined immune deficiency) is also called bubble-boy syndrome because it almost never occurs in females. It is caused by a mutation in an immune master gene located on the X chromosome. Thus, males have only on such gene and its disruption is inevitably fatal. Females have two copies of the gene and the disruption of one copy is insignificant. Now, we need Wells to come up with his definition of INFORMATION that would show these two different outcomes to be irrelevant.

As long as we have Jeffrey on the hook here, could I raise some points to see if we can find a way to clarify these concepts and confusions for the layperson? I am neither a computer scientist nor an information theorist.

A random string of characters, generated by a uniform random number generator, contains lots of “information”, in the Kolmogorov sense, because, in order to replicate that string, one must supply a tremendous amount of information (essentially the string itself). On the other hand, a string of the same character requires little information to replicate it. A string generated by a Gaussian random number generator would have some intermediate level of information – less than for a uniform random number generator but more than for the string of same character.

So, as computer scientists and information theorists understand it, the less information in a file, the more easily it can be compressed into a smaller file and re-extracted or replicated again later.

If I may belabor an example for which I attempted to get some feedback on a different thread, take that example of two gravitationally interacting bodies coming into each other’s vicinity (not a head on collision). I think this relates to biological systems as well, because ultimately there are potential wells involved in the molecular chains and complex folding membranes making up the system.

So, two bodies (planet and moon, sun and planet) swing around each other and go off onto either parabolic or hyperbolic trajectories. But how do they come to orbit each other? Energy has to be extracted from the system.

How? One possible way is by a myriad of collisions with a myriad of other bodies in the vicinity. Some of these collisions pass momentum and energy onto other bodies that carry it away from the system as they are ejected off to infinity. Other collisions turn kinetic energy into heat which gets carried away by photons or is contained in the kinetic energies of the particles making up the two bodies. Other mechanisms include tidal friction converting kinetic energy into heat, or gravitational waves carrying energy away from extremely massive neutron stars or black holes.

So we finally close the trajectories into either elliptical or circular orbits; the latter being the one with the least potential energy consistent with the angular momentum of the system.

Leave aside the question about the entropy of such a system for a moment and concentrate on the concepts of “order” and “information”; especially “information” as Jeffrey understands it.

Which of the final states, elliptical orbits or circular orbits, is more “orderly”? Which contains “more information”?

To try to clarify the question a little more, which of the orbits requires less “information” to describe? But on the other hand, which orbit tells us more (gives us more information?) about the history and formation of the system? How does order relate to information?

If I am not mistaken, much of the confusion about “information” in describing systems, from planetary systems to biological systems, hinges on just what one wants to highlight. The junk tossed out by the cdesign proponentsists simply adds to the confusion and makes the job of legitimate scientist more difficult.

In addition, the terms are used differently in different disciplines. Maybe we in the various disciplines should form a united front and see if we can clear up the problem for the general public.

Is this a good forum to attempt clarification?

Whatever Wells is, Jeffrey, he is like you---an evolutionist. He accepts species mutability, evolution. Calling a fellow evolutionist a buffoon doesn't look good.

"Hump? What hump?"

Ray Martinez said: Whatever Wells is, Jeffrey, he is like you---an evolutionist. He accepts species mutability, evolution. Calling a fellow evolutionist a buffoon doesn't look good.

I suppose that duplicating a gene (prior to mutation to one of the copies) wouldn't produce a different protein, though it might produce a larger quantity of the same protein. And that could have side effects, depending on what the protein does.

Henry

Jeffrey Shallit said: Oh, dear, we have a troll.

Mike Elzinga said: Which of the final states, elliptical orbits or circular orbits, is more “orderly”? Which contains “more information”?

@wile coyote
I knew that. Everyone knows that.

Seriously, that is an excellent and cogent explanation. I appreciate your pointing me to it. I have learned a lot tonight. Thanks.

I was apprehensive about putting it here. Shallit seems to be pretty nice guy but I know for a fact he is precise.

I frequently run into creationists who seize upon words like "information" and "code" to leap to the conclusion that there must be an intelligence who supplied the information and wrote the code. This is a most unfortunate terminology problem. In Wells' case, however, I suspect that the confusion is deliberate.

Jonathan Wells isn't necessarily ignorant, nor is he incompetent and blissfully unaware. Rev. Moon payed for Wells to get a doctorate in biology (he already had 2 doc. in religion) specifically so he could argue against evolution. He combs through evolutionary information specifically to find things he can twist into creationist propaganda. Wells is no where near as benign as you make him sound. He's a premeditated liar whose goal is to disrupt understanding of evolution.

Jeffrey Shallit said: But - all other things being equal, and choosing a point with generic coordinates - an ellipse will be more complex than a circle because an ellipse requires three numbers to encode it and a circle only two.

Wouldn't a simple demonstration be to take a word, and keep appending duplicated mutations of it?

eg. for the word 'wells', this would result in something like:

1: wells
2: wellswenls
3: wellswenlswxnls
4: wellswenlswxnlsaxnls
5: wellswenlswxnlsaxnlsaxnly

So if I understand the ID claim correctly, their claim is that string 1 has exactly the same amount of information as string 5?

Mike Elzinga said: ... As a physicist, I am not completely familiar with what the biologists who model biological systems use as “complexity” or as “information”. I don’t get a consistent picture as I read. Complexity seems to correlate with degrees of freedom, but “information” is often confusing. If Joe Felsenstein is looking in, perhaps he could offer some clarification.

Duplicating a string does add a bit of information.

Joe Felsenstein said: ... In arguing with ID types (or with standard creationists) who state that evolution cannot build information into genomes, I think the best approach would be to recognize that what is being talked about is really high fitness.

AnswersInGenitals, I would just like to say that your moniker is the funniest I have read in a while. If someone thinks about it, it is funny on so many levels; well done.

Decent post too.

This is the same Evolution News and Views post wherein Wells asked why, given the commonality of the HOX gene between them, a fly is not a horse.

I find it very hard to believe he can really be this biology-stupid and have successfully defended a PhD dissertation in molecular bio.

bigjohn756 said: I know diddly-zip about information theory. Where can I learn the basics? Is there an on-line Information Theory for the Stupid Dummies type site?

Venus Mousetrap said: Wouldn't a simple demonstration be to take a word, and keep appending duplicated mutations of it? eg. for the word 'wells', this would result in something like: 1: wells 2: wellswenls 3: wellswenlswxnls 4: wellswenlswxnlsaxnls 5: wellswenlswxnlsaxnlsaxnly So if I understand the ID claim correctly, their claim is that string 1 has exactly the same amount of information as string 5?

Creationists like Wells are also fond of insisting that gene mutations are frequent events and almost all such mutations are deleterious, leading to non-functional or dysfunctional proteins.

Mr. Shallit the IDers have only one agenda and that is to conform science to their belief system by creating social pressure on the sciences, politics and society. In a sense the IDers have created a pseudoscientific presentation that influences a large vocal minority in communities to subvert education. Wells doesn't have any interest in science and only is interested in reaching the less educated with his challenges to the scientific community, which is all agenda and no substance.

Nice coinage, Henry J. Well minted.

Scott said: 1: wells 2: wellswells 3: wellswellswellswells

Ray Martinez said: Whatever Wells is, Jeffrey, he is like you---an evolutionist. He accepts species mutability, evolution. Calling a fellow evolutionist a buffoon doesn't look good.

I'm not just a novice poster, but also a novice in terms of information theory (so be nice), but...
Are XXY and XYY syndrome relevant here? There's no doubt that having an extra copy of the sex-determinant genes has some powerful effects...

Mike Elzinga said: I suppose, as you seem to be suggesting, that the “information” line of argumentation is best avoided because it apparently shows up only in the context of arguments with ID/creationists (much like genetic entropy and entropy barriers). These are not part of the lexicon of real science.

That’s good, because I have never felt, intuitively, it was worth the effort to try to understand it in this context.

@driving analytical: good example, to go with the copy-number variations mentioned earlier.

I'm trying to visualise what it would be like going grocery shopping with Jonathan W. Apparently if you give him a list which says "six oranges", and he comes back with one orange, he's faithfully followed the list.

wile coyote said:

Mike Elzinga said: That’s good, because I have never felt, intuitively, it was worth the effort to try to understand it in this context.

It isn't. If there are applications of information theory to biology, they are very much on the margins and layfolk have no particular need to know about them. It is irritating to have to try to dope out information theory for no other reason than to figure out just how bogus the evobasher arguments really are.

Forgive me if this has been asked before, but what would Wells say if he did understand information theory? Something (including that admission of getting a PhD specifically to destroy "Darwinism") tells me it would be pretty much what he's saying now.

Joe Felsenstein said: Let me just add that while this is certainly currently true, I suspect that in the long run arguments from information theory will be of interest in biology, and that they will make use of notions very much like specified information / functional information / adaptive information.

wile coyote said: I absolutely won't contest that, but I would point out that the whole notion of "functional information" (and its like) is dodgy from a strict information theory point of view. It is tempting to make a distinction between "raw" information ("quantity") and "functional" information ("quality"), but it turns out to be really tricky to do, and nobody's come up with a "robust" definition of functional information that could be used as the basis of a fundamental law of physics. Of course there are all sorts of ad-hoc measures of information -- "base pairs", "coding base pairs" -- and they can be perfectly useful in their context. But only in their context.

I kind of studied comms theory as part of my EE degree too many moons ago to remember. It was, AFAIR, all about "Shannon's" theories. When/where does Kolmogorov come into modern information theory ? I'd love some references to read. My knowledge of Kolomogorov is related to his work on turbulence in optical paths !

Steve

The best reference for Kolmogorov's theory is the book by my colleagues Ming Li and Paul Vitanyi, An Introduction to Kolmogorov Complexity and Its Applicaitons.

“I replied that duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content.”

wile coyote said:

Like many incompetent people, Wells is blissfully unaware of his incompetence.

It's a very strange game of the crank: believing that he can hoodwink everyone else by putting a bag over his own head.

Jeffrey Shallit said: Neither Shannon nor Kolmogorov seems to quite capture what the average person "feels" is information...

Ron Okimoto said: Claiming that these guys are cranks gives them too much credit. Intelligent design is a purposeful deception.

Ron Okimoto said:

wile coyote said: It's a very strange game of the crank: believing that he can hoodwink everyone else by putting a bag over his own head.

Claiming that these guys are cranks gives them too much credit. Intelligent design is a purposeful deception.

"We don’t have to “expose” the intelligent design creationists as buffoons; they do it themselves whenever they open their mouths."

Actually, you do have to expose them: if bad ideas were self-undermining, ID and other creationisms would not be such a problem.

So instead of slinging this kind of ad hominem name-calling blah, however accurately and satisfyingly the hot feces may smack into their targets, why not buckle down and grind out the refutations and _stick_ to that? Yes, refute the errors for the thousandth time, if necessary? A rebuttal need not be dully written to be on-topic and effective. I compare the Thumb unfavorably, as regards many of its posters' willingness to sling sneers and tap-dance with glee at every new revelation of "stupidity," corruption, dishonesty, etc., to realclimate.org, where climatologists working an equally wearisome beat -- combating climate-change denialism, which displays many rhetorical homologies to creationism -- manage to sound like professional scientists defending a knowledge structure rather than like professional wrestlers hooting and pointing across the ring at the sucker punk they are about to smack down for the Nth time.

You want a better discourse, then raise the discourse.

Larry:

Go read my 54-page refutation of Dembski here, or my chapter in Why Intelligent Design Fails, or my testimony in the Dover case, and then get back to me, OK?

I wrote... “Information content” is not the only measure of useful a transmission medium is. If it were, the new York Times would only have to print one copy every day.

I know next to nothing about information theory, but would suggest this:

Suppose I have a 3-page text, digitize it and photocopy it, then digitize the 6-pages (original plus photocopy). Then I compress each digitized file using Gnuzip. Are the resulting files identical? If not, doesn't that mean one of the files contains more information?

mark said: Suppose I have a 3-page text, digitize it and photocopy it, then digitize the 6-pages (original plus photocopy). Then I compress each digitized file using Gnuzip. Are the resulting files identical? If not, doesn't that mean one of the files contains more information?

mark said: Suppose I have a 3-page text, digitize it and photocopy it, then digitize the 6-pages (original plus photocopy). Then I compress each digitized file using Gnuzip. Are the resulting files identical? If not, doesn't that mean one of the files contains more information?

Joe Felsenstein said: Let me just add that while this is certainly currently true, I suspect that in the long run arguments from information theory will be of interest in biology, and that they will make use of notions very much like specified information / functional information / adaptive information. But right now, they are not particularly important.

I agree with Mr. Coyote. Neither Shannon nor Kolmogorov seems to quite capture what the average person “feels” is information, and many people have tried to come up with a formal measure that is more like our intuitive measure. But so far, everyone has failed, in the sense that there is no definition that has achieved wide appeal.

— Jeffrey Shallit

eric said: In Bahasa (Indonesian) you often repeat a word to make it plural.

It'd be a odd business if duplicating stuff didn't increase information. How could you tell the difference between the string "Oompa" and "OompaOompa"? If you compressed those strings, would the compressed outputs ever be identical?

Even an ideal compression algorithm optimised for strings of pure Oompa, which would just be an Oompa count, would have output 1 for the first and 2 for the second.

Real world lossless compression algorithms such as LZW rather depend on duplicates involving extra information: if a new string is found, it is put in the dictionary and given a representational code so that next time it's encountered, it can be replaced with that code. Each unique string may have only one unique code, but if you don't carefully put that code in the output stream every time you find that string again, you'll lose information. How can that happen, in the Wells world of zero new information?

How would they code a lossless compression algorithm that assumed no new information in duplicates?

I believe Quechu, at least Souther Quechua, uses duplication to mean simple plural, with -kuna added as a suffix indicating plural.

And once again, all Ray can do is label something. He really seems to think that labels change the nature of the objects they're labeling. If you can't refute an argument, just stick a label on it and pretend you've magically transformed it into something else.

And he can't even get the label right! Wells, an "evolutionist?" Are you kidding me?

Ray Martinez is nothing but a sad, brittle, broken, useless failure. His rejection of the concept of species mutability says it all: he simply cannot comprehend or accept a Universe where things change over time.

Raging Bee said: And once again, all Ray can do is label something. He really seems to think that labels change the nature of the objects they're labeling. If you can't refute an argument, just stick a label on it and pretend you've magically transformed it into something else. And he can't even get the label right! Wells, an "evolutionist?" Are you kidding me?

The questioner became agitated and shouted out something to the effect that HOX gene duplication explained the increase in information needed for the diversification of animal body plans. I replied that duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content.

If there are applications of information theory to biology, they are very much on the margins and layfolk have no particular need to know about them.

Sean R. McCorkle said: FYI I've seen Shannon information analysis pop up in computational methods that attempt to refine detection of potentially weak DNA patterns, such as identifying transcription factor binding site motifs, or conserved regions.

I wish I could find it again, but I read an excellent post where the biologist pointed out our genes and the encoding process don't contain "information" per se. The genes carry out a process of replication and transcription; WE then call it encoding and label the bits used as information with tags like "AA" and such. Manipulation of Info Theory to disprove this labeling--as reinterpreted by ID--has no real meaning at all.

ID–has no real meaning at all.

Wheels said: Doubling words isn't just useful for pluralizing, it can also emphasize.

You know, in thinking about this further--when I step back and take a look at the whole picture--the logic of it just isn't. Wells et al use an objective example of a perceived gene transcription problem to disprove Darwin's Theory, (actually to demonstrate Evolution doesn't exist at all), in defiance of all other objective facts. Their solution is to apply a metaphysical concept, (Dembski's CoI--although it could be any metaphysical concept at this point), to their perceived problem as proof of their metaphysical concept in action!

This is not just bogus science, this is bad theology as well. Hell, it's not even good con game.

wile coyote said:

Ron Okimoto said: Claiming that these guys are cranks gives them too much credit. Intelligent design is a purposeful deception.

Heh! Not much of an argument, really: "Are you evobashers CROOKED or STUPID? Which are you? CROOKED? STUPID?" If I was on the receiving end, I'd be thinking I'd be given more credit to think I was CROOKED.

Re Indian and SE Asian languages - Bahasa Indonesia is v. close to Malay and a recent synthesis that has emerged over the last 80-100 years in Java. It is said to be much closer to some of the Sumatran languages and Malay than classical Javanese. It is remote from the Indian languages. In Sanskrit among the many things one memorises is the expansion of compound words (which can be a mix of as many as three different parts of speech.

Some words are compounds of twins such as

-Pratikshana (every moment) compound of Kshana Kshana (moment moment)

In Tamizh even verbs can be twinned as in

-Odi Odi uzhaikkanum

-Run Run work - or work running all the time...

It's enough to say the Jon Wells generally pulls things out the wazoo while his pet sheep obediently nod their heads.

A similar example is taking a compressed text file with a message, and making a compressed bitmap image of a screen listing of that message. Do they have the same information? No. They have the same text, but the information in them has very little to do with the contents. Almost certainly the bitmap image of the text will have a lot more information than the compressed text file.

It’s enough to say the Jon Wells generally pulls things out the wazoo while his pet sheep obediently nod their heads.

Henry J said:

It’s enough to say the Jon Wells generally pulls things out the wazoo while his pet sheep obediently nod their heads.

All the easier to pull the wool over their eyes... Henry

Philip Johnson has admitted that the ID science does not exist, and I recall seeing West quoted as saying that intelligent design wasn’t ready for prime time after the Discovery Institute ran the bait and switch scam on Ohio. After Ohio Nelson came right out and admitted that they never had a scientific theory of intelligent design to teach. Do the ID rubes think that these guys are lying now or when they claimed to have the science? What could possibly be a rube explanation for the bait and switch that has been run on them?

— Ron Okimoto

Mike Elzinga said: Then there is that story about the young philosopher pontificating in front of the famous philosopher (I can’t remember his name) about how a negative and a positive produce a negative, a positive and a positive produce a positive, and a negative and a negative produce a positive. But no one has ever shown a chase where a positive and a positive produced a negative. To which his mentor in the audience replied in a bored tone, “Yeah, yeah.”

Somebody needs some education http://bethelburnett.blogspot.com/2009/10/rottweiler-who-lost-his-teeth-his-mind.html

Dave said: Somebody needs some education http://bethelburnett.blogspot.com/2009/10/rottweiler-who-lost-his-teeth-his-mind.html

TomS said: This story is told about Sidney Morgenbesser, who was noted for his humor. I would note that in those languages in which a double negative functions as a negative it is also true that any multiple (greater than zero) functions that way.

Dave said: Somebody needs some education http://bethelburnett.blogspot.com/2009/10/rottweiler-who-lost-his-teeth-his-mind.html

Mike Elzinga said:

TomS said: This story is told about Sidney Morgenbesser, who was noted for his humor. I would note that in those languages in which a double negative functions as a negative it is also true that any multiple (greater than zero) functions that way.

:-) Thanks Tom.

Joe Felsenstein said: The same story is also told among linguists. The professor says that there is no language in which a double positive means a negative, and a student in the back row of the hall says "yeah yeah". So this one circulates among fields.

Wells: ...duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content

I went ahead and tried Wells' experiment. I wrote "idiot. Wells is an" on a piece of paper, photocopied it, put the two copies side by side, and lo and behold! New information appeared.

raulsanger said: I believe Quechu, at least Souther Quechua, uses duplication to mean simple plural, with -kuna added as a suffix indicating plural.

Dave said: Somebody needs some education http://bethelburnett.blogspot.com/2009/10/rottweiler-who-lost-his-teeth-his-mind.html

Demb Demb has responded to this post with a list of various complexity measures, claiming that Wells could have meant any of them, really.

Some of the listed complexity measures don't apply to messages (fractal dimension, tree-subgraph diversity). Clearly some of these measures have the property that XX has more information than X. One wonders if an expert might go through and see if there are any measures in Dembski's list for which X and XX have the same information/complexity.

A. Cooper said: Demb Demb has responded to this post with a list of various complexity measures, claiming that Wells could have meant any of them, really.

Note that one of the forms of information on this list is “algorithmic information content,” which is the one that Shallit attributes to Jonathan Wells. But with so many information/complexity measures floating around, why in the world does Shallit think that this is the one that Wells intended? A charitable interpretation of Wells’s remarks would suggest that he was thinking of nothing more complicated that duplicating X by X means that the probability of X given X is 1, implying that all the uncertainty from X has been removed, implying in turn that its information is zero since information can, in one incarnation, be taken as a measure of uncertainty.

TomS said:

Mike Elzinga said: Then there is that story about the young philosopher pontificating in front of the famous philosopher (I can’t remember his name) about how a negative and a positive produce a negative, a positive and a positive produce a positive, and a negative and a negative produce a positive. But no one has ever shown a chase where a positive and a positive produced a negative. To which his mentor in the audience replied in a bored tone, “Yeah, yeah.”

This story is told about Sidney Morgenbesser, who was noted for his humor. I would note that in those languages in which a double negative functions as a negative it is also true that any multiple (greater than zero) functions that way.

David Utidjian said: How about "Aye, aye."? But Mikes story is sounds better. -DU-

Mike Elzinga said: One of the problems with ID/creationism, as with any pseudo-science, is that its definitions and concepts are left to “free-float”. They can then be lubricated and slipped around to fit any situation when the heat is focused on its practitioners

David Utidjian said: How about "Aye, aye."? -DU-

Note that one of the forms of information on this list is “algorithmic information content,” which is the one that Shallit attributes to Jonathan Wells. But with so many information/complexity measures floating around, why in the world does Shallit think that this is the one that Wells intended? A charitable interpretation of Wells’s remarks would suggest that he was thinking of nothing more complicated that duplicating X by X means that the probability of X given X is 1, implying that all the uncertainty from X has been removed, implying in turn that its information is zero since information can, in one incarnation, be taken as a measure of uncertainty.

So Dembski falls back on the "there are several definitions of information" defense. Odd for an IDer, given how rabid they are about definitions, such as species.

A. Cooper said: Demb Demb has responded to this post with a list of various complexity measures, claiming that Wells could have meant any of them, really.

harold said: Dembski's comment is bizarre -

Note that one of the forms of information on this list is “algorithmic information content,” which is the one that Shallit attributes to Jonathan Wells. But with so many information/complexity measures floating around, why in the world does Shallit think that this is the one that Wells intended? A charitable interpretation of Wells’s remarks would suggest that he was thinking of nothing more complicated that duplicating X by X means that the probability of X given X is 1, implying that all the uncertainty from X has been removed, implying in turn that its information is zero since information can, in one incarnation, be taken as a measure of uncertainty.

In other words, if the biochemical event of gene duplication resulted in a second gene that was exactly the same as the first gene in some way - say nucleotide sequence - then we could know the sequence of the "copy" by knowing the sequence of the "original". That's trivial but irrelevant, since it would be unlikely for gene duplication to produce such a result (even a small gene is a lot of nucleotides to be replicated without a single "error"), and it would be impossible to know that it had produced such a result without comparing the sequences of the "original" and "copy" at any rate. This can't possibly be what Wells meant. The context was that Wells was arguing that duplication of a Hox gene does not represent an increase in "information", using the insanely wrong analogy of photocopying a piece of paper. Wells' comment clearly makes no sense. It's point blank wrong in any formal treatment of information theory. It's also, despite desperate attempts to state otherwise, point blank obviously wrong at an intuitive level. Wells response to a discussion of Hox genes was to repeat an idiotic Jack Chick level YEC sound bite.

I would suggest we use Dembski's message as an opportunity to ask Wells which type of information he meant.

And as a more broad follow-up, since we now have IDers going on record as listing several types of possible information, it is perfectly reasonable to ask them what type CSI is.

...not that I have high expectations of a useful response...

eric said: I would suggest we use Dembski's message as an opportunity to ask Wells which type of information he meant. And as a more broad follow-up, since we now have IDers going on record as listing several types of possible information, it is perfectly reasonable to ask them what type CSI is. ...not that I have high expectations of a useful response...

Prof. Shallit,

Yes, a duplicated gene adds a bit of information to the genome.

Literally one bit. Take the symbol string potato. If duplicated it's potatopotato. All the information that's added is an increment to the number of potato strings.

I'm not sure if it still rhymes but let's illustrate it with the children's poem using binary numbers:

1 potato, 10 potato, 11 potato, 100,
101 potato, 110 potato, 111 potato more.
Icha bacha, soda cracker,
Icha bacha boo.
Icha bacha, soda cracker, out goes Y-O-U!

Not even sure it's a single bit added but I'm not an information theorest like you are. How many bits of information by each new potato? A half a bit?

Suppose it's a recipe, and the string is "add a 1/2 tsp baking soda", or... hmm... how about... "make some amylase".

Repeating the string changes the recipe.

Not even sure it’s a single bit added but I’m not an information theorest like you are. How many bits of information by each new potato? A half a bit?

Well, we can add Dave Springer to the list of people who don't understand information theory.

Duplicating a string doesn't have to introduce new information in the Kolmogorov theory. But it can. It depends on the string, and on the model of universal machine chosen.

We can precisely quantify how much each new copy adds, on average, though. If we make n copies, we add approximately log n bits. So each new copy adds approximately logn - log (n-1) bits, on average.

By the way, Dave, the important part of gene duplication and divergence is not the duplication, which, as you note, only increases the information content of the genome by a constant number of bits. It's the divergence. A random walk will then approximately double the information in the genome.

Jeffrey Shallit said: By the way, Dave, the important part of gene duplication and divergence is not the duplication, which, as you note, only increases the information content of the genome by a constant number of bits. It's the divergence. A random walk will then approximately double the information in the genome.

Dave Springer said: Prof. Shallit, Yes, a duplicated gene adds a bit of information to the genome. Literally one bit.

Prof. Shallit,

You wrote that a gene duplication doesn't have to add information to a genome. It would seem then that Jonathan Wells didn't have to be wrong.

So now you're off on a tangent saying it isn't the duplication, per se, but rather the potential for the duplicated gene to diverge from the original. Well, I can't argue against that. If I'm writing a book and I run out of paper it would be really handy if I could duplicate a previous page, erase the old material, and thus have room to write something new.

It was the admission that the duplication in and of itself doesn't add much if any new information that I was looking for from you.

Thanks for playing.

Good luck with students of yours who can't figure out how duplicating a string infinitely many times adds Kolgomorov complexity. In fact it adds infinite complexity, but it adds it at a diminishing rate as the number of repetitions increases. I always loved the concept of larger and smaller infinities.

Yours in potatoes,
Dave

Dave Springer said: It was the admission that the duplication in and of itself doesn't add much if any new information that I was looking for from you. Thanks for playing.

Oh, dear, I see poor Dave Springer is terrifically confused. That's the problem with being a braying jackass; he brays so loud he can't understand when someone else is worth listening to. It's a shame, because Springer is obviously a bright guy who sometimes has interesting things to say.

Wells' statement implied a universal claim: that for all strings x, the string xx cannot have more information than x. This is the claim that is false. To prove a universal claim false, all one needs to do is provide a single counterexample. In fact, there are infinitely many counterexamples, which is the exercise I give to my students.

The moral is that, contrary to Wells' claim, you can indeed accumulate arbitrarily large amounts of information by gene duplication. Of course, the divergence that follows lets you accumulate information even faster. But both mechanisms work to increase information.

Furthermore, my example has nothing at all to do with "duplicating a string infinitely many times". Nor does it have anything to do with "the concept of larger and smaller infinities", because the cardinality of all strings is countable.

If a start codon is spliced in front of a sequence of preexisting "junk" DNA, so that it can now be read and transcribed, would that be an "increase of information"?

Or, if one gene is spliced onto the end of another gene, would that be an "increase of information"?

It was the admission that the duplication in and of itself doesn’t add much if any new information that I was looking for from you. Thanks for playing.

Prof. Shallit,

I see think name calling is going to buttress your arguments. Do you teach playground science in addition to information theory?

Here is what Wells said:

"I replied that duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content."

I don't see where he made any absolute claim there was no information increase at all. I'm calling straw man.

In fact Wells described gene duplication by analogy pretty much the same way I did. I didn't go look up what he said until just now. Great minds think alike. Instead I had taken took you at word that he made an absolute claim. Therein was my only mistake.

I see think name calling is going to buttress your arguments. Do you teach playground science in addition to information theory?

Pretty rich - from the guy that once called Denyse O'Leary a "morphodyke" and "canadian crossdresser"!

Dave, I'm really sorry I don't have the time to teach you English comprehension in addition to information theory - but really, it's simply not possible to argue with someone who thinks "duplicating a gene doesn’t increase information content" isn't making a univeral claim. One can only roll one's eyes and sigh.

Elzinga,

While you were teaching a group of high school students in your physics class I was teaching a group of hardware and software engineers at one of the world's largest computer manufacturers. Spare me.

At least Shallit is teaching at the university level. I might have even hired one of his students and completed their education but that's unlikely considering the hires where I worked usually came from places like MIT and Stanford.

Kolmogorov complexity increase is simply illustrated in a trivial psuedo-code fragment.

cardinal c = 1;
string s = "gene";

to_infinity:
write s;
increment c;
loop to_infinity;

The ONLY increase in Kolgomorov complexity is the increasing number of bits needed to store the cardinal number n.

Shallit, Wells, and I all know this. It seems no one else here does.

Prof. Shallit,

Be a sport and explain to me (add name calling if you must) how you translated this:

"I replied that duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content."

into an absolute claim that no information is added by gene duplication. "Any more than" is the operative phrase here. Maybe the problem is you don't speak quite the same language that Wells and I do. I'll ask O'Leary for a translation into hoser dialect.

Sigh [rolls eyes].

And Springer thinks I'm a Canadian, too. How droll.

Everybody here knows the bit from the physicists about "not even wrong". Well, that's normal for intruders here, but they usually have some amusement value.

Unfortunately, this one rates "not even funny".

Prof. Shallit,

No, I didn't think you were Canadian. I thought that after almost 20 years teaching at the University of Waterloo you might have gone native.

With your pedigree coming out of Princeton and Berkeley you could have been high up on the food chain at Intel or Microsoft. We raided IBM Boca Raton mercilessly for guys like you back in the early 1990's. If you don't mind me asking a personal question, what inspired you to take up teaching computer science at Waterloo? You aren't old enough to be a draft dodger. Was a woman involved? Berkeley anti-establishment politics? You love rotten cold wet weather?

I suppose I'm being unfair about the rotten weather. I grew up 2 hours' drive southeast of Waterloo and love the area. There's a lot to be said for having four real seasons. I wouldn't have loved this past summer though. Usually I spend the summers there but took a pass this year.

Dave Springer said: Elzinga, While you were teaching a group of high school students in your physics class I was teaching a group of hardware and software engineers at one of the world's largest computer manufacturers. Spare me. At least Shallit is teaching at the university level. I might have even hired one of his students and completed their education but that's unlikely considering the hires where I worked usually came from places like MIT and Stanford.

wile coyote said: Everybody here knows the bit from the physicists about "not even wrong". Well, that's normal for intruders here, but they usually have some amusement value. Unfortunately, this one rates "not even funny".

He's probably half-drunk and just about ready to cry about Dembski ditching him. Eventually he'll fall asleep.

Mike Elzinga said:

wile coyote said: Everybody here knows the bit from the physicists about "not even wrong". Well, that's normal for intruders here, but they usually have some amusement value. Unfortunately, this one rates "not even funny".

He seems to be another Keith Eaton; over-inflating his credentials, and attempting to use that to intimidate people here. Yet he speaks gibberish with no apparent awareness that he is doing so. It would have been really funny to watch those bright high school students run him out the door screaming with a persecution complex. They were good at stripping idiots naked and flailing them.

Stanton said: "evolander"

Stanton said: No, he's nowhere near as loopy as Keith Eaton: no hate-filled non sequitors, blatant projectionism or wistful longing for "evolander" pogroms.

Wheels said:

Stanton said: "evolander"

Anyone know where/when that term popped up? Google seems to think our beloved William Wallace is the guy who uses it the most.

Anyone know where/when that term popped up?

Hey Davey, you gutless disgrace to the Corps, how about finally answering the questions that got me banned from UD:

{ begin old comment }

DaveScot wrote:

"ID predicts that no evolution of complex structures will occur by chance & necessity within the temporal and geographical constraints imposed by the earth due to the statistical improbabilities involved."

What is the ID theory that predicts this? The reason I ask is the following statement from a senior member of the Discovery Institute, Paul Nelson:

"Easily the biggest challenge facing the ID community is to develop a full-fledged theory of biological design. We don't have such a theory right now, and that's a problem."

DaveScot continues:

"ID can be falsified by the observation of a single complex structure built by mechanisms of chance & necessity."

What, exactly, constitutes a "complex structure"? Michael Behe has admitted that a new viral protein-viral protein binding site has evolved in HIV. This resulted in new function (an ion channel), a form of complexity that Behe claimed was beyond the "Edge of Evolution". That falsifies your prediction.

DaveScot continues again:

"This prediction appears to have been confirmed by the observation of P.falciparum over the last 50 years during which time it replicated billions of trillions of times, which represents more opportunities for mutation than the entire sequence of reptile-to-mammal evolution, and nothing beyond trivial changes were observed."

Could you please provide a cite with more details? For example, how was this experiment performed? By whom? what selection pressures were the populations subjected to? What calculations support the claim of "more opportunities for mutation than the entire sequence of reptile-to-mammal evolution"?

{ end old comment }

You're an intellectual lightweight and a moral midget, Davey. Let's see what you've got when you can't hide behind the moderators' skirts.

Maya

Dave Scot should be waking up right about now.

There'll be lots of empty bottles and beer cans strewn about the Floating Command Center. In the bathroom, the mirror will be shattered, touches of blood on the glass. A crumpled photo of Dembski in the sink that Dave will tenderly try to smooth out, but fail.

He'll slump against the toilet and have a good cry.

Dave Springer said: Elzinga, While you were teaching a group of high school students in your physics class I was teaching a group of hardware and software engineers at one of the world's largest computer manufacturers. Spare me. At least Shallit is teaching at the university level. I might have even hired one of his students and completed their education but that's unlikely considering the hires where I worked usually came from places like MIT and Stanford. Kolmogorov complexity increase is simply illustrated in a trivial psuedo-code fragment. cardinal c = 1; string s = "gene"; to_infinity: write s; increment c; loop to_infinity; The ONLY increase in Kolgomorov complexity is the increasing number of bits needed to store the cardinal number n. Shallit, Wells, and I all know this. It seems no one else here does.

All I know is that UD is seriously dumber without Dave. The fundies are in charge over there now old buddy I sure wish you would dust off the cheesy poofs and put down the mushroom farm and get back in the command center swivel seat. The morphodyke is walking around with a shaved face and Barry and niwrad have mutual diddle sessions where they blame Darwin for Hitler and for their unrequited man-love. When jerry is the sanest SOB in the room you know that there is a loudspeaker in the ceiling that is going unmanned.

Do it for the kids old pal.

Your frind

Maya said: Hey Davey, you gutless disgrace to the Corps, how about finally answering the questions that got me banned from UD: { begin old comment } DaveScot wrote: "ID predicts that no evolution of complex structures will occur by chance & necessity within the temporal and geographical constraints imposed by the earth due to the statistical improbabilities involved." What is the ID theory that predicts this? The reason I ask is the following statement from a senior member of the Discovery Institute, Paul Nelson: "Easily the biggest challenge facing the ID community is to develop a full-fledged theory of biological design. We don't have such a theory right now, and that's a problem." DaveScot continues: "ID can be falsified by the observation of a single complex structure built by mechanisms of chance & necessity." What, exactly, constitutes a "complex structure"? Michael Behe has admitted that a new viral protein-viral protein binding site has evolved in HIV. This resulted in new function (an ion channel), a form of complexity that Behe claimed was beyond the "Edge of Evolution". That falsifies your prediction. DaveScot continues again: "This prediction appears to have been confirmed by the observation of P.falciparum over the last 50 years during which time it replicated billions of trillions of times, which represents more opportunities for mutation than the entire sequence of reptile-to-mammal evolution, and nothing beyond trivial changes were observed." Could you please provide a cite with more details? For example, how was this experiment performed? By whom? what selection pressures were the populations subjected to? What calculations support the claim of "more opportunities for mutation than the entire sequence of reptile-to-mammal evolution"? { end old comment } You're an intellectual lightweight and a moral midget, Davey. Let's see what you've got when you can't hide behind the moderators' skirts. Maya

... duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content.

stevaroni said:

... duplicating a gene doesn’t increase information content any more than photocopying a paper increases its information content.

I'll be in New York. I'll be in New York, New York. Yup, no difference at all.

Or cheese puffs.

didymos said:

Wheels said:

Stanton said: "evolander"

Anyone know where/when that term popped up? Google seems to think our beloved William Wallace is the guy who uses it the most.

I've always liked "evolander". It makes me laugh because it sounds like some SUV model or something: "The new Nissan Evolander. Performance and style...at a reasonable price (trade-ins welcome)."

Even though I know creationists mean the term [evolander] to be perjorative, in my mind I can't help but think the opposite. That 'lander' at the end always makes me think the term refers to "inhabitant of the real world." Which brings a smile to my face.

Dave Springer said: Prof. Shallit, Yes, a duplicated gene adds a bit of information to the genome. Literally one bit. Take the symbol string potato. If duplicated it's potatopotato. All the information that's added is an increment to the number of potato strings. I'm not sure if it still rhymes but let's illustrate it with the children's poem using binary numbers: 1 potato, 10 potato, 11 potato, 100, 101 potato, 110 potato, 111 potato more. Icha bacha, soda cracker, Icha bacha boo. Icha bacha, soda cracker, out goes Y-O-U! Not even sure it's a single bit added but I'm not an information theorest like you are. How many bits of information by each new potato? A half a bit?

Dave Springer said: Prof. Shallit, You wrote that a gene duplication doesn't have to add information to a genome. It would seem then that Jonathan Wells didn't have to be wrong. ... It was the admission that the duplication in and of itself doesn't add much if any new information that I was looking for from you. ... Yours in potatoes, Dave

If you want my opinion (or even if you don't), it's not the amount of information, it's the adaptive value of the genome that matters.

Henry

Henry,

Shhhh! Be vewwy quiet.... I'm huntin' wackos..

No need to be quiet for that. ;)

slpage said: Henry, Shhhh! Be vewwy quiet.... I'm huntin' wackos..

Wheels said: Doubling words isn't just useful for pluralizing, it can also emphasize. If being emphatic carried no useful information, I doubt we'd ever ever ever ever utilize it so ubiquitously.

Stanton said: If a start codon is spliced in front of a sequence of preexisting "junk" DNA, so that it can now be read and transcribed, would that be an "increase of information"? Or, if one gene is spliced onto the end of another gene, would that be an "increase of information"?

Toidel Mahoney said: If you splice a gene you add information coming from your mind. All information must come from some mind. That is what ID states.

Toidel wrote:

"All information must come from some mind."

No, actually information is found everywhere in nature, whether there is a mind to create it or not. It doesn't require a mind to produce information, it requires a mind to interpret information. Organisms can evolve by natural means without intelligent intervention, however it takes a mind to understand how that evolution actually occured. Whether anyone is intelligent enough to figure it out, it still occured.

As for creolanders, they apparently lack the intelligence to even figure out that evolution did occur, let alone determine how. Perhaps if they were less obsessed with the sexual behavior of others they would at least have more time to study the information in nature.

Creationist moron Dave Springer said: Prof. Shallit, You wrote that a gene duplication doesn't have to add information to a genome. It would seem then that Jonathan Wells didn't have to be wrong.

Perhaps if they were less obsessed with the sexual behavior of others they would at least have more time to study the information in nature.

Toidel Mahoney said: If you splice a gene you add information coming from your mind. All information must come from some mind. That is what ID states.

phantomreader42 said:

Creationist moron Dave Springer said:

Or are chimps really the ‘higher’ life form?

http://www.evolutionnews.org/2009/10/jonathan_wells_hits_an_evoluti.html

Henry J said:

Or are chimps really the ‘higher’ life form?

Certainly. They live in trees; we live on the ground. Trees are higher. ;)

But in his case, what does the skyscraper have to do with it? :p

Henry

"We can precisely quantify how much each new copy adds, on average, though. If we make n copies, we add approximately log n bits. So each new copy adds approximately logn - log (n-1) bits, on average."

Anyone know the argument behind this? Seems pretty trick to extrapolate to something useful if we can't know duplication will always increase the k-complexity of a string. For instance, it is quite plausible that duplication will dramatically *decrease* the KC of a string, and it doesn't seem possible to predict when this will happen.

Simple counting arguments can't be used since duplication picks out a very small, particular subset of strings, which may not have properties similar to the average string of that length.

It was a very nice idea! Just wanna say thank you for the information you have shared. Just continue writing this kind of post. I will be your loyal reader. Thanks again.

Jeffrey Shallit said:

Toidel Mahoney said: If you splice a gene you add information coming from your mind. All information must come from some mind. That is what ID states.

So, Toidel: when weather forecasters predict the weather, they use information that they collect from the environment: temperature, pressure, wind speed, etc. Which mind does that information come from?

oops!

"id it is non-random"

should be

"if it is non-random"

Toidel Mahoney said: Well, if weather happens by random chance, none, but id it is non-random information must be present so it must come from the mind of an intelligent designer.

DS said: Toidel wrote: "All information must come from some mind." No, actually information is found everywhere in nature, whether there is a mind to create it or not. It doesn't require a mind to produce information, it requires a mind to interpret information. Organisms can evolve by natural means without intelligent intervention, however it takes a mind to understand how that evolution actually occured. Whether anyone is intelligent enough to figure it out, it still occured.

As for creolanders, they apparently lack the intelligence to even figure out that evolution did occur, let alone determine how. Perhaps if they were less obsessed with the sexual behavior of others they would at least have more time to study the information in nature.

Toidel Mahoney said: ... but id it is non-random information must be present so it must come from the mind of an intelligent designer.

Toidel Mahoney said: How do you know this, from the Barnumesque third rate art projects such as Piltdown Man, Archaeaptor and Tikkalik?

Well, ignoring that embarrassing lapse ... yes, we all know there are only two fossils ever found (or maybe only one since I can't find what an "archaeapter" is), which means that the numerous natural history museums I've visited that are full of detailed fossils are frauds ...

Daniel J. Andrews said: Or maybe none because I couldn't find Toidal's "Tikkalik" either ...

It seems that Davey posted and ran away, in a manner indistinguishable from that of an intellectual coward. What a surprise.

Maya said: It seems that Davey posted and ran away, in a manner indistinguishable from that of an intellectual coward.

wile coyote said:

Toidel Mahoney said: ... but id it is non-random information must be present so it must come from the mind of an intelligent designer.

Ah, the infamous Law Of Conservation Of Information! One of the fundamental laws of nature taught in EVERY physics book! Oh, we must be clueless to have forgotten it!

wile coyote said:

Toidel Mahoney said: How do you know this, from the Barnumesque third rate art projects such as Piltdown Man, Archaeaptor and Tikkalik?

Believe it or not Piltdown Man was ... A HOAX! And everybody knows that, it seems EXCEPT FOR YOU, despite the fact that it was known as a hoax for probably LONGER THAN YOU HAVE BEEN ALIVE! Well, ignoring that embarrassing lapse ... yes, we all know there are only two fossils ever found (or maybe only one since I can't find what an "archaeapter" is), which means that the numerous natural history museums I've visited that are full of detailed fossils are frauds ... as are the fascinating reports we get of new fossil finds EVERY MONTH these days ...

Daniel J. Andrews said:

Well, ignoring that embarrassing lapse ... yes, we all know there are only two fossils ever found (or maybe only one since I can't find what an "archaeapter" is), which means that the numerous natural history museums I've visited that are full of detailed fossils are frauds ...

Or maybe none because I couldn't find Toidal's "Tikkalik" either...well, I did find Reinhardtius hippoglossoides or Greenland Halibut so maybe he's saying this bizarre but existing living fish with its eye on the dorsal ridge (cyclops-like) is also a third-rate art project? If so, I can't imagine you would endear yourself to any Creator by saying his/her/its fish looks like a third-rate art project even if you add that s/he/it made it taste good. If Toidal can't get the basics like names right then what are the odds that anything else s/he says will be right? --Oh, Toidal's post was an example of Poe's Law, right?

yters said: "We can precisely quantify how much each new copy adds, on average, though. If we make n copies, we add approximately log n bits. So each new copy adds approximately logn - log (n-1) bits, on average." Anyone know the argument behind this? Seems pretty trick to extrapolate to something useful if we can't know duplication will always increase the k-complexity of a string. For instance, it is quite plausible that duplication will dramatically *decrease* the KC of a string, and it doesn't seem possible to predict when this will happen. Simple counting arguments can't be used since duplication picks out a very small, particular subset of strings, which may not have properties similar to the average string of that length.

Yes, Piltdown Man was actually exposed as a hoax.

The other hoaxes have yet to be exposed.

In addition, this is Archaeoraptor.

Anti-Christian bigotry on the part of the global epicenter for sodomy is the only reason Dembski has not won the Nobel Prize in physics.

Toidel Mahoney said: Anti-Christian bigotry on the part of the global epicenter for sodomy is the only reason Dembski has not won the Nobel Prize in physics. You are clueless to have forgotten his sublime discovery!

wile coyote said:

Anti-Christian bigotry on the part of the global epicenter for sodomy is the only reason Dembski has not won the Nobel Prize in physics.

It's the CONSPIRACY! Well, just put another layer of tinfoil in your hat to block out the mind-control beams, and you'll be all right.

Another candidate for the bathroom wall.

DS said: Another candidate for the bathroom wall.

Aw, this was a really quality post. In theory I'd like to write like this too - taking time and real effort to make a good article... but what can I say... I procrastinate alot and never seem to get something done