new arrangements and the flash of genius

Came across this nugget from Wittgenstein's Philosophical Investigations:

The problems are solved, not by giving new information, but by arranging what we have long since known.

Another way of stating this in terms of information theory, I think, is to say that new information consists now of applying past knowledge into new arrangements.  Consider this interesting exchange from the movie Flash of Genius:

You have a Ph.D in electronic engineering, is that correct?
Uh, that's correct. I've taught for the past several years...
No, that's fine, sir. Your credentials are already part of the record. Now, when you said earlier that Mr. Kearns didn't create anything new, could you explain what you meant by that?
 Yes. As you can see, Dr. Kearns's basic unit consists of a capacitor a variable resistor and a transistor. Now, these are basic building blocks in electronics. You can find them in any catalog. All Mr. Kearns did was to arrange them in a new pattern, you might say. And that, that's not the same thing as inventing something new, however.
 Did Mr. Kearns invent the transistor?
 No, sir, he did not.
 Did Mr. Kearns invent the capacitor?
 Again, no, he did not.
 Did Mr. Kearns invent the variable resistor?
 No, he did not.
Thank you, Professor.

Or more amusingly, in an episode of the series "3rd Rock From the Sun," High Commander Dick Solomon tries to discredit a novelist by pointing out that he has used information that is accessible to anyone:

[to Jeff] You think you're pretty clever, don't you? I happen to know that every word in your book was published years ago! [to everyone in the room] Perhaps you've read...The Dictionary!

Jeff wrote his novel by merely recombining meaningful words, and Robert Kearns merely rearranged the previously discovered transistors, capacitors, and variable resistors into new parts.  In the first instance, plagiarism would consist of stealing Jeff's particular improbable combinations of words, and in the second instance, intellectual theft consisted of stealing Kearn's ingenious arrangement of standard components.

Consider now that Charles R. Marshall and other biologists have argued that the Cambrian Explosion might have resulted from "simply rewiring the existing genetic networks." To say that this solves the informational problem is like saying that all I need to do to create useful computer programs is to "tweak" how the procedures invoke one another and how data flows between them.

Pierre-Paul Grassé wrote, "If to determine the origin of information in a computer is not a false problem, why should the search for the information contained in cellular nuclei be one?"  Indeed.

Heuristics and Mechanical models

Following the tradition of the 19th century mathematical physics in Great Britain, James Clerk Maxwell extensively employed mechanical models in the study and representation of electromagnetic phenomena.
http://www.ifsc.usp.br/~cibelle/arquivos/SCED%202006.pdf

Pierre-Paul Grassé, French biology, and mechanism

Pierre-Paul Grassé is the past President of the French Academie des Sciences and editor of the 35 volume "Traite de Zoologie" published by Masson, Paris. [+]

I think it was Berlinski that hypothesized that the French identification of Darwinism as an inherently Britannic idea allowed their biology to be free of the Darwinian stranglehold.  I almost love the avant-garde stubbornness of it.  Read here about Pierre-Paul Grassé.  He could make you become a Francophile.  

"The opportune appearance of mutations permitting animals and plants to meet their needs seems hard to believe. Yet the Darwinian theory is even more demanding: a single plant, a single animal would require thousands and thousands of lucky, appropriate events. Thus, miracles would become the rule: events with an infinitesimal probability could not fail to occur .... There is no law against day dreaming, but science must not indulge in it."

I would like to sit at a sidewalk cafe in Paris, sip some good coffee (okay, the French do that really well), and talk to Grassé.  Or listen, mostly. "Dr. Grassé, what do you think of the fact that most American biologists think that you can't actually do biology without a commitment to the truth of the Modern Synthesis?"  

Alas, it is too late to have such a conversation with Dr. Grassé. much as it is with Dr. Marcel Schutzenberger.  Though perhaps Dr. Schutzenberger may have said similar sorts of things.

This all reminds me of a Chinese researcher quoted by Dr. Jonathan Wells:

"In China we can criticize Darwin but not the government. In America you can criticize the government but not Darwin." [+]

Hey, something has to be sacred.

I just saw that there is a TalkOrigins article that tries to address the anti-Darwinism of the French, which they write off as mere cultural bias (and how much pro-Darwinism in England and America circa 1880-1920 was motivated by trends of atheism and scientism in university culture?) They clarify that it was Prof. Louis Bounoure who said:

That, by this, evolutionism would appear as a theory without value, is confirmed also pragmatically. A theory must not be required to be true, said Mr. H. Poincare, more or less, it must be required to be useable. Indeed, none of the progress made in biology depends even slightly on a theory, the principles of which are nevertheless filling every year volumes of books, periodicals, and congresses with their discussions and their disagreements.

while it was biologist (and atheist, for what it's worth) Jean Rostand of the Académie Française who said, "Evolution is a fairy tale for adults." French atheist and scientist Paul Lemoine who was director of the National Museum of Natural History, wrote in the Encyclopedie Francaise:

Evolution is a kind of dogma which its own priests no longer believe, but which they uphold for the people. It is necessary to have the courage to state this if only so that men of a future generation may orient their research into a different direction.

TalkOrigins explains that Paul Lemoine believed in descent with modification, and so, like Michael Behe presumably, was only referring to a theory about actual evolutionary mechanism, not to the theory that some sort of change over time occurred.  Of course, so rendered, such a theory devoid of mechanism doesn't fit with the umbrage taken over describing neo-Darwinism as "just a theory."

For the sole purpose of false economy of explanation, if there is some evidence of change over time, that is, a presumed relationship of organism A at time X and organism B at time Y, it is presumed that NaturalSelectionDidIt, and any diachronic difference between otherwise similar organisms becomes prima facie evidence of the power of natural selection.

Consider Dr. Paul Nelson's 1995 recollection of an exchange with ardent neo-Darwinist Kenneth Miller (the guy with the mouse-trap tie-clip).

The critical question is one of mechanism. Is it possible for mammals to vary sufficiently for - to take the case of whales - their skulls to be completely remodelled, so that the nostrils (nares) move all the way up to the top of the head? (Indeed, all the other cranial bones must change their size and shape as well.) If not, we're looking at an apparent, not actual, continuum. The designed forms are transformationally discrete, with independent histories. 

That's what I said to Ken, as he was putting away his slides. He replied, "Yeah, Paul, I know, with you it's always 'mechanism, mechanism, mechanism.' But there's more to it than that." 

What the "more" is ain't clear to me.

It isn't clear to a lot of us.  Especially after Eugenie Scott's highly publicized "theometer" comments and the desperate, desperate plea for a mechanism for intelligent design.  Tu quoque?

Back to Prof. James M. Tour:

Although most scientists leave few stones unturned in their quest to discern mechanisms before wholeheartedly accepting them, when it comes to the often gross extrapolations between observations and conclusions on macroevolution, scientists, it seems to me, permit unhealthy leeway. When hearing such extrapolations in the academy, when will we cry out, “The emperor has no clothes!”? 

…I simply do not understand, chemically, how macroevolution could have happened. Hence, am I not free to join the ranks of the skeptical and to sign such a statement without reprisals from those that disagree with me? … Does anyone understand the chemical details behind macroevolution? If so, I would like to sit with that person and be taught, so I invite them to meet with me.

How about Dr. Robert Hazen?  Dr. Hazen co-ao-authored an article with Dr. Jack Szostak on "Functional Information" which had a formula that reminds me wuite a bit of some of Dr. William Dembski's formulas.  Dr. Hazen states:

A fundamental law of nature, the law describing the emergence of complex ordered systems (including every living cell), is missing from our textbooks. ¹

Why do Robert Hazen and Stuart Kaufman and so many others think we need to specify another fundamental law of nature when we have a perfectly adequate, fully specified mechanism in NaturalSelectionDidIt?

McCabe complexity of the polymerase machine

Most software engineers have probably heard of McCabe Complexity.  I can't help but wonder how Thomas McCabe would evaluate the complexity of DNA/RNA decoding and transcription process:

He [Thomas McCabe] is taking a much more achievable approach [than attempting to compute the complexity of the cell].  At the risk of oversimplifying his idea, we will say that instead of looking at all the metabolic processes in the cell, he is looking at just one. Specifically, there is a process in living cells that decodes the genetic information in the DNA molecule and builds biological structures accordingly. Conceptually, this process is not much different from the software program in a CD player that reads a compact disk and converts the information into music. Since we can compute the complexity of a program that reads a CD, one should also be able to compute the complexity of the biological process that reads and processes genetic information.

Sounds exciting.
 

islands of functionality and the flash of genius

Hazen et al illustrate some alternative ideas of functional protein accessibility in protein space, where the plane represents the dimensionality of protein space (2 is much, much fewer than what would be needed) and the E axis represents the catalytic usefulness of the various points in protein space (in essence, a fitness landscape):

D has more of a needle-in-a-haystack problem than A, B, or C, due to its relatively small hypervolume (area in the figure) of protein space.  But it's not the only aspect that makes in inaccessible.  The relative distance from other islands of functionality.  Using the above figure a little out of its intended representation, we see that B and C are relatively close together, so that not as vast a distance of neutral mutation would have to be crossed to get from B to C as from A to D.

But vast oceans of neutral mutation to be crossed are not the only impediment to finding the points of especially high functionality ... There is also the fact that less optimal peaks might serve as attractors that divert computational resources away from the brass ring.  In this case, the good is the enemy of the best.

D of Hazen et al's figure above corresponds to this diagram from one of Douglas Axe's papers, where sequence/protein space is represented by only one dimension:

Here the white noise of suboptimal adaptations might be considered negligible, all of the being more or less neutral in that they don't change the survivability rate of the organisms enough to inhibit the traversal of sequence space.  It is possible that the neutral is filled with many little hills and valleys, a so-called rugged landscape.  In the Picasso-esque landscape below, it may be that the difficulty in finding the high peak of innovation is compunded, both by the volume of sequence/protein space to search but also by the attractive "force" of suboptimal solutions.

The size of the relevant space to be searched along with the distractive force of more accessible (more "obvious") solutions might contribute to the Non-Obviousness of the more optimal solution.

It would seem that both of these have relevance to Bennett's concept of "logical depth", as they both may drive up the necessary computational resources (or, the amount of brute force "tinkering") to realize the non-obvious solution -- where a flash of genius might render all that brute force tinkering unnecessary.  

In Shadows of the Mind, in which Roger Penrose argues for mathematical insight requiring something beyond computation, Penrose has a section on "Things that computers do well -- or badly":

Conscious understanding is a comparatively slow process, but it can cut down considerably the number of alternatives that need to be seriously consideredand thereby greatly increase the effective depth calculation.  

In other words, a flash of insight can cross large distances of "logical depth" a la Charles H. Bennett.  Insight is like a wormhole, a directed wormhole, through solution space.

The Land Bridges of Solution Space

Once upon a time, before the scientific community as a whole really started taking continental drift seriously, doctrine was that flora and fauna distribution was brought about by some sort of land bridge network that spanned the oceans.  There was no evidence for these land bridges other than the terrible need for them to have existed, the explanandum itself.

Now, in sequence space, there is also a terrible need for bridges.  To a certain extant bridges can be made through sequence space, "wormholes" that connect remote regions of the space, but it can be very tricky to make the ends of the wormhole open at useful locations.

One kind of such wormhole, maybe Kenneth Miller's favorite kind, is the frame shift mutation.  Where these wormholes open up is determined by the genetic code itself, but the chances of a useful mutation on one part of sequence space being bridged to a much more useful  (for a completely different purpose) mutation in a remote part of solution space does not seem like something that can be relied upon for a pervasive mechanism.

Crossover is another, more plausible source of wormholes through sequence space.  Again, there is a precipitous element of contingency about this, not about what is predetermined by the genetic code, but in the chances of a useful part of gene P being spliced onto a useful part of gene Q so that gene P'Q' does something that is of significant value to the organism.

As with the land bridges that were presumed ubiquitous (and explanatorily sufficient) before continental drift, the chance in a quintillion bridges between islands of measurable utility in sequence space are evidenced to occur in the right locations and frequency by whatever explanatory gap calls for them to exist.  It is not clear that hyperadaptationists necessarily depend on them (it's all smooth fitness functions like Methinks It Is Like A Weasel), but if they do, you can be sure that all sorts of enzymes can be produced as easily as "nylonase."

NaturalSelectionDidIt

NaturalSelectionDidIt is one of the masterstrokes and trump-cards that Darwinists or other neo-positivists have at their disposal when debating points with skeptics (who they call "deniers") and ID proponents.  It proposes that anything is and was possible because of the omnipotence of natural selection - specifically the ability to bend the laws of time, logic, and physics.  This means that arguments that focus on the feasibility of the de novo ability of microevolution to bring about macroevolution -- the development of complicated body plans or to change an artiodactyl ("no more amphibious than a tapir") into a highly adapted marine mammal -- can be swept away and ignored.

The concept of NaturalSelectionDidIt can be used to create unfalsifiable* theories. A neo-Darwinist need never doubt the creative power of chance and selection because Natural Selection could have made anything that was made.  (What is the proof that Natural Selection can do it? -- why, the fact that it was done!)  It may also be used as a euphemism to indicate something that cannot be explained in terms of natural laws, most likely due to having no idea how evolution works anyway.)

This was inspired by the Goddidit entry at RationalWiki.  :-)

antibiotic resistance: is evolution a fact?

Sometimes an ID proponent gets caught up in hyper-adaptationists' conflation of evolution with Darwinism, that he/she starts to borrow the use of "evolution" as a shorthand for neo-Darwinism.  Descent with modification, on one hand, has a compelling case that can be made for it, much more compelling than neo-Darwinism, and I'm not sure how much more compelling the case for DwM has been made since Darwin made it, but it is still compelling.  What it means is another story entirely, and there are many ways to get descent with modification.  For example, Michael Behe believes so many mainstream things about evolution that it seems utterly absurd to call him a creationist -- I'd say, it is absurd for any other reason than a political one, and I'm sure there are some obvious political reasons for Eugenie Scott and Barbara Forrest to call him a creationist.   However, Behe essentially believes in descent with modification and in some version of common ancestry.

So on the one hand, I would prefer ID proponents not get caught up in the game of "Duck Season, Rabbit Season" and use "evolution" as a shorthand for undirected evolution with complexity built via natural selection.  Evolution is so vague to be almost a useless word in biology except for the most general concept of apparent vast changes of ecosystems over the billenia.

Still, the vast awe-inspiring wisdom of Judge Jones the Third notwithstanding, those that are less than impressed with DwM often make good points themselves, and unlike Dawkins and others like him infected with The Smug, I think they often make good points.

On that note, from "Is Evolution A Fact":

Nobel laureate Sir Ernest Chain (credited with purifying penicillin in a way that made it possible to employ it as an antibiotic) wrote in agreement.  

“To postulate that the development and survival of the fittest is entirely a consequence of chance mutations seems to me a hypothesis based on no evidence and irreconcilable with the facts. These classical evolutionary theories are a gross oversimplification of an immensely complex and intricate mass of facts, and it amazes me that they are swallowed so uncritically and readily, and for such a long time, by so many scientists without a murmur of protest” (1970, p. 1, emp. added).

Now consider Sir Chain's statement as you read this bit of Doonesbury satire:

I don't understand.  How did Sir Chain do any science with penicillin without true belief in natural selection?  He isn't a true believer like cartoonist Gary Trudeau.  Trudeau has the sense to believe in antibiotics.   Actually, exactly how do we treat resistant strains other than just trying different antibiotics until we find one to which the strain isn't resistant?  Exactly how did neo-Darwinism.  How does Fisher's mathematics guide my primary care physician's choice of antibiotics?  Trudeau seems to know.

Although Trudeau seems to actually believe the false dichotomy between neo-Darwinism and Biblical literalism, and he seems completely unaware of the difference between microevolution and macroevolution. (Once you have an idea of artificial selection--as any ancient animal breeder would--does antibiotic resistance really seem counter-intuitive?)   But he is a cartoonist, and he seems bursting with pride over his anti-religious sensibilities.  So believe like Trudeau.

Dr. Sean D. Pitman in "Stepping Stones" offers a meaningful explanation:

In the case of de novo antibiotic resistance, such rapid evolution is made possible because there are so many beneficial "steppingstones" so close together, right beside what the bacterial colony already has. Success is only one or two mutational steps away in many different directions since a multitude of different single mutations will result in a beneficial increase in resistance. How is this possible? 

In short, this is made possible because of the way in which antibiotics work. All antibiotics attack rather specific target sequences inside certain bacteria. Many times all the colony under attack has to do is alter the target sequence in just one bacterium by one or two genetic "characters" and resistance will be gained since the offspring of this resistant bacterium, being more fit than their peers, will take over the colony in short order. A simple "spelling change" made the target less recognizable to the antibiotic, and so the antibiotic became less effective. In other words, the pre-established antibiotic- target interaction was damaged or destroyed by one or two monkey-wrench mutations. As with Humpty Dumpty and all the king's men, it is far easier to destroy or interfere with a pre-established function or interaction than it is to create a new one, since there are so many more ways to destroy than there are to create.

Here is an interesting statement as well by Prof. James M. Tour:

Let me tell you what goes on in the back rooms of science – with National Academy members, with Nobel Prize winners. I have sat with them, and when I get them alone, not in public – because it’s a scary thing, if you say what I just said – I say, “Do you understand all of this, where all of this came from, and how this happens?” Every time that I have sat with people who are synthetic chemists, who understand this, they go “Uh-uh. Nope.” These people are just so far off, on how to believe this stuff came together.  I’ve sat with National Academy members, with Nobel Prize winners. Sometimes I will say, “Do you understand this?”And if they’re afraid to say “Yes,” they say nothing. They just stare at me, because they can’t sincerely do it. 

I was once brought in by the Dean of the Department, many years ago, and he was a chemist. He was kind of concerned about some things. I said, “Let me ask you something. You’re a chemist. Do you understand this? How do you get DNA without a cell membrane? And how do you get a cell membrane without a DNA? And how does all this come together from this piece of jelly?” We have no idea, we have no idea. I said, “Isn’t it interesting that you, the Dean of science, and I, the chemistry professor, can talk about this quietly in your office, but we can’t go out there and talk about this?”

I don't understand.  I thought evolution was so simple that even a million monkeys sitting at typewriters could understand it...  !!!   Maybe Prof. Tour has a science-poor background.

Professor James M. Tour is one of the ten most cited chemists in the world. He is famous for his work on nanocars (pictured above, courtesy of Wikipedia), nanoelectronics, graphene nanostructures, carbon nanovectors in medicine, and green carbon research for enhanced oil recovery and environmentally friendly oil and gas extraction. He is currently a Professor of Chemistry, Professor of Computer Science, and Professor of Mechanical Engineering and Materials Science at Rice University. He has authored or co-authored 489 scientific publications and his name is on 36 patents. Although he does not regard himself as an Intelligent Design theorist, Professor Tour, along with over 700 other scientists, took the courageous step back in 2001 of signing the Discovery Institute’s “A Scientific Dissent from Darwinism”

What a dunce!  I don't know.  I think Prof. Tour just can't grasp a heady concept like macroevolution through natural selection.  No doubt explainable by the Salem Hypothesis.

insights from directed evolution

Is their any reason for using the term "directed evolution" instead of the old "artificial selection"?  Well, "natural selection" was an analogy to "artificial selection" -- and the analogies in evolutionary computing continue.  The term "directed evolution" attempts to reverse this and treat "undirected evolution" as the original inspiration and merely be a speeding up of the natural process.

However, "directed evolution" of necessity must, to be effective in a large number of cases, involve a means of skipping between the islands of functionality in protein space:

Though mutation and screening is the essence of directed evolution, there are many factors that combine to turn the process into as much of an art as a science. These considerations include initial protein choice, mutant library construction, and method of screening.(here)

Initial protein choice: starting off the process at a point that is likely to get you what you want, a key location in what you hope is an archipelago of related functionality.  Mutant library construction: this is actually the method for not being bogged down by the typical Darwinian problems with searching nucleotide sequence space -- the sort Michael Behe discusses at length in The Edge of Evolution.  Method of screening:  In natural selection, nature is a terrible screener for the once-in-100-millenia innovation.  Just because an organism has a prize-winning genetic innovation, this doesn't mean that the noise in the innovation-detection system won't overwhelm it.  The survival/reproductive advantage may be washed out in the collective effect on survival, not the least of which is natural danger.  Walk by a tiger, and the innovation may be lost to bad luck, and there is no shortage of this sort of bad luck in the natural world.  So a protein engineer must have a better detection system, in order to ensure that he recognizes what he's looking for once it appears.

Colin Patterson questions biologists on evolution

Colin Patterson's question to biologists:

...I mentioned a question ('Can you tell me anything you know about evolution?') that I have put to various biologists, and an answer that had been given: 'I know that evolution generates hierarchy.' In the framework of phylogenetic reconstruction and our current problems with it, another answer comes to mind: 'I know that evolution generates homoplasy' [or "convergence," in the older jargon of systematics]. In both cases, the answer is not quite accurate. It would be truer to say, 'I know that evolution explains hierarchy' or 'I know that evolution explains homoplasy.' We must remember the distinction between the cart--the explanation--and the horse--the data. And where models are introduced in phylogenetic reconstruction, we should prefer models dictated by features of the data to models derived from explanatory theories.

Careful there, Patterson.  You are starting to sound like a secular creationist.

not only unwelcome but unneeded

Berlinski:

My own view, repeated in virtually all of my essays, is that the sense of skepticism engendered by the sciences would be far more appropriately directed toward the sciences than toward anything else. It is not a view that has engendered wide-spread approval. The sciences require no criticism, many scientists say, because the sciences comprise a uniquely self-critical institution, with questionable theories and theoreticians passing constantly before stern appellate review. Judgment is unrelenting. And impartial. Individual scientists may make mistakes, but like the Communist Party under Lenin, science is infallible because its judgments are collective. Critics are not only unwelcome, they are unneeded. The biologist Paul Gross has made himself the master of this attitude and invokes it on every conceivable occasion. 

Now no one doubts that scientists are sometimes critical of themselves. Among astrophysicists, backbiting often leads to backstabbing. The bloodletting that ensues is on occasion salutary. But the process of peer review by which grants are funded and papers assigned to scientific journals, is, by its very nature, an undertaking in which a court reviews its own decisions and generally finds them good. It serves the useful purpose of settling various scores, but it does not -- and it cannot -- achieve the ends that criticism is intended to serve. 

If the scientific critic finds himself needed wherever he goes, like a hanging judge he finds himself unwelcome wherever he appears, all the more reason, it seems to me, that he really should get around as much as possible.

[Emphasis mine]  From The Deniable Darwin

diffusion through conformational space?

http://link.springer.com/static-content/lookinside/371/art%253A10.1007%252FBF00163809/000.png

Protein evolution as diffusion through conformational space?

Shouldn't it be diffusion through typographical space?

Another thing:  If it is an easy thing to make a protein for any particular effect, why isn't it exceptionally easy to make a protein that really gums up the works?  either ties itself into a useless little knot or turns into something that binds with all sorts of things it shouldn't and kills its organism (slowly or quickly)?

But then again maybe it isn't that easy:

The need to maintain the structural and functional integrity of an evolving protein severely restricts the repertoire of acceptable amino-acid substitutions^{1, 2, 3, 4}. However, it is not known whether these restrictions impose a global limit on how far homologous protein sequences can diverge from each other. Here we explore the limits of protein evolution using sequence divergence data. We formulate a computational approach to study the rate of divergence of distant protein sequences and measure this rate for ancient proteins, those that were present in the last universal common ancestor. We show that ancient proteins are still diverging from each other, indicating an ongoing expansion of the protein sequence universe. The slow rate of this divergence is imposed by the sparseness of functional protein sequences in sequence space and the ruggedness of the protein fitness landscape: ~98 per cent of sites cannot accept an amino-acid substitution at any given moment but a vast majority of all sites may eventually be permitted to evolve when other, compensatory, changes occur. Thus, ~3.5 × 10⁹ yr has not been enough to reach the limit of divergent evolution of proteins, and for most proteins the limit of sequence similarity imposed by common function may not exceed that of random sequences.

the worm or the spaghetti

http://www.teemings.net/series_2/issue_01/worm_or_the_spaghetti.html

Ways to construct protein space

This is similar to the example that Stu Kauffman uses in his book on complexity.  It's a geometry based on only using two amino acids and gets as complicated as 4 sites of change or 4 degrees of freedom --or, in the verbiage below, a protein exactly 4 aminos long.

This scheme can't really be used correctly for nucleotide sequences because the connectivity is all wrong, but the idea is mostly there.  It works for the toy example in protein space but one quickly gets lost if he imagines 20 choices  at each step of construction instead of 2.  And because protein sequence depends on nucleotide sequence, it becomes important, from an evolutionary point of view, exactly what is the Hamming distances between the corresponding nucleotide sequences.

It's still a useful picture (from here).

tightly patterned

For example, consider Dembski's claims about DNA. He implies that DNA has CSI [19, p. 151], but this is in contradiction to his implication that CSI can be equated with highly-compressible strings [19, p. 144]. In fact, compression of DNA is a lively research area. Despite the best efforts of researchers, only minimal compression has been achieved [36, 84, 12, 56, 2, 59].

Most highly patterned sequences have considerable redundancy, and the redundancy is often (though not always) a clue to a pattern.  However, a JPEG file is hard to compress but it hardly is random.  Or a more trivial example maybe, a Zip file is hard to compress but is certainly not random.  These are tightly patterned. We do not know how tightly patterned portions of DNA are because we don't know how much of it works.  The most well understood part of DNA -- what specifies protein structure -- has a lot of redundancy built into it.  Many of the codons that make up protein DNA can be compressed to 4 or 5 bits if we knew for a fact that they were only being used to specify one protein, i.e. they only had one specificational role in one gene.  But that isn't necessarily true, meaning that many genes are more tightly patterned than can be ascertained at first glance.  For those that aren't, codons that can be replaced with a codon for a more or less equivalent amino acid often are.  This replaces one kind of redundancy (one that with less variation might allow more compression of a genome) with another (one that increases the Shannon entropy of the genome).

We know so little, despite the amazing things that have been learned, about the language of development.  Ironically, even if the neo-Darwinian paradigm were true, it gives so little information about what to expect that it actually works against understanding.  Biologists going in expect things to be put together in an incoherent way.  They have a nearly behaviorist model of the DNA code and do not approach it as a coherent program, even as they more or less describe it as a coherent program that arose through an incoherent process.  It must be hopelessly cobbled together because the Modern Synthesis tells us that it was hopelessly cobbled together.

One study suggest a lot of internal structure to the genome, making one wonder just how tightly patterned it is.

2^219 states

2²¹⁹ states

But that's just one star, and a measly one at that. A typical supernova releases something like 10⁵¹ ergs. (About a hundred times as much energy would be released in the form of neutrinos, but let them go for now.) If all of this energy could be channeled into a single orgy of computation, a 219-bit counter could be cycled through all of its states.
    https://www.schneier.com/blog/archives/2009/09/the_doghouse_cr.html

macroscopically describable

If we repeat an experiment 2^k times, and define an event to be “simply describable” (macroscopically describable) if it can be described in m or fewer bits (so that there are 2m or fewer such events), and “extremely improbable” when it has probability 1/2n or less, then the probability that any extremely improbable, simply describable event will ever occur is less than (2^(k+m))/(2^n). Thus we just have to make sure to choose n to be much larger than k + m. If we flip a billion fair coins, any outcome we get can be said to be extremely improbable, but we only have cause for astonishment if something extremely improbable and simply describable happens, such as “all heads,” or “every third coin is tails,” or “only every third coin is tails.” Since there are 10^23 molecules in a mole of anything, for practical purposes anything that can be described without resorting to an atom-by-atom accounting (or coin-by-coin accounting, if there are enough coins) can be considered “macroscopically” describable.
   - Granville Sewell, Entropy, Ev and Open Systems, note 5

Coydogs and the Meaning of 'Species'

While speciation is an important concept, especially with regard to evolutionary theories, I don't think there is a precise idea of species, and the value of the idea in measuring the similarity between types of animals is dubious.

If one of two previously interbreeding subspecies populations undergoes a mutational shift, where a single mutation catches on in it, such that the populations will find it difficult to interbreed, are the two populations necessarily as different as any two subspecies of the same species are to each other?  Something slight change to A has finally tipped the scales and now B can freely mix with it any more. And could some closely related pairs of species (that have diverged from each other tens of thousands of years ago) be much more genetically similar than A and B are to each other?

The ability for two species to hybridize in vitro and produce functional offspring is one measure of their similarity.  The ability for two species to form chimeras is yet another, more distant, measure, but an important one.  Chimeras require embryonic cells that follow essentially the same body plan along with sharing important biochemical compatibilities.

Hybrids themselves test the boundaries of species.  We are told now that only one wild canid was the ancestor of the domesticated dog, but some species of coyote and wolf reflect considerable gene flow between what are considered separate species.  Some wild coyotes were determined to have more than half wolf ancestry.  Some scientists seem to speculate that the some of this is due to coydogs mating with wolves, but I don't believe that accounts for all of it.  At any rate, if some dogs have coyote blood, and some are part jackal, have they merely found evidence that most dog breeds have exclusive wolf ancestry?  And do wolves have exclusive wolf ancestry?

There are also species of frogs and lizards that are not independent species but require repeated interbreeding with one or more of its parent species to keep the line going.

And there is ongoing speculation that in order to resolve certain discrepancies in the molecular clocks of human and chimp DNA we must allow that the proto-humans mated with proto-chimps long after the species had become separate species.  This was apparently a desparate act at one of the several apparent severe genetic bottlenecks that the human species has endured.   And of course, a small percentage of Europeans supposedly share some heritage with the Neanderthal subspecies of human, which was apparently not interbreeding with the rest of humanity so much.

Then there are various subspecies of squirrels around a national park such that subspecies A can breed with B, B with C, C with D, ... and E with F.  But A can't breed with F.  Take out all the subspecies in the middle and A and F suddenly are different species, even though there is gene flow thoughout the species, even between A and F, if only indirectly.

Coincidences

I've been thinking about how coincidences are evaluated in forensics.  How does one separate the pure coincidences from the can't-be-coincidences?

Supposedly, a detective does not believe in coincidence.

Let's say George has some mysterious charges on his bill.   Let's say that something comes to light that the exact digits of his credit card number appeared in a text from a Hans whom George doesn't know, but say George's daughter knows Hans.  There might be enough digits in that number to warrant thinking that it is not a coincidence -- that is, that those doesn't represent something else: a book-of-the-month club user registration, a hotel confirmation number, etc.  A long enough number is supposed to use up probabilitic resources.  After all, it should be the case that a random sequence of digits probably NOT be valid credit card number.  There could be a preponderance of evidence that Hans

To make this more interesting, let's say that these mysterious charges occurred shortly after George died under mysterious circumstances and his credit card went missing.  There is the proximity to George: two degrees of separation.  There is the time proximity: near the time of George's death.

To a detective, this forms a pattern, a specification.  Now the small probability is much larger than the Universal Plausibility Bound, but it is still compelling in many cases, once motive and alibi are dealt with.

The quality Dembski would call tractability, gets into the degrees of separation along with the specificational resources.  What are the chances that someone would have the credit card number and also that this person be only two degrees of separation from the credit card owner, close to the time the card went missing?   Too coincidental?   How to put some firm numbers behind that assessment?  I suspect that departments of justice deal with tractability all the time, and somehow it is not pseudoscience  despite the problem resisting really precise calculations in a lot of cases.

The Grammar of DNA

Similar to David Abel's(?) distinction of change contingent and choice contingent variation is Lila Gatlin's D1 and D2 distinctions.  D1 is "context-free" variation in the statistic Shannon sense and D2 is "context sensitive" in a grammatical sense.  Gatlin is/was a biochemist/biophycisist and is not using these terms in their specific computer science meanings, where a grammar-generated language is specifically referred to as context-free.

Gatlin referred to a Law of Conservation of Information, not too coincidently similar to Dembski's interpretation of Wolpert's.  Since at that time, the modern ID Movement had not taken shape, and the guardians of the light of science were only concerned with creation science based on religious axioms, Gatlin wasn't derided, marginalized, and silenced the way that ID proponents now are.  She was published in Nature and the Journal of Molecular Biology (as has Douglas Axe in the latter, if I recall).  Jeremy Campbell leaned quite a bit on her ideas for his Grammatical Man, which garnered praise from mathematician and Skeptic (in the anti-supernatural and anti-creationist sense) Martin Gardner.  Surely, Gardner would be expected to have a negative reaction to a speculative book that depended on pseudoscience.  Way back in the 70s, you must realize, these ideas weren't pseudoscience.

This gets back to my Conjecture:  The opposition leveled at an idea in biology will be in proportion to the perceived utility of the idea to supporting theistic belief.  Professional skeptics would not close ranks against this idea until it was associated with the idea of Intelligent Design.  Note here that all 2nd Law of  θ∆ics arguments are summarily dismissed with a "guilt by association" argument (you know, one of them there fallacies) of the Barbara Forrest kind.

Anyway, Gatlin recognizes that the statistical measure of information of the Shannon sort is insufficient to reason about the interrelatedness of grammatically organized information that has semantic value.  Where here analysis possibly falls short, is that "grammatically correct" strings, or strings that follow grammatical rules, do not necessarily convey meaning, and in fact, are usually nonsense.   But it is certainly a step  in the right direction.  There is a limited context-sensitivity in grammar that is relevant to protein structure.  Subject must have its predicate and vice versa.  In English, the choice of a plural noun will affect the form of the predicate verb.  In protein space, particular "choices" in one part of the protein will have consequences for residue choices in remote parts of the same protein (or possibly, in a part of a co-enzyme).  The context-freedom in grammar is that rules only get you so far:

Greenness is a square light bulb, that grazes in the death of fragrance.  A bus, on the other shoe, does not swim with the lug nuts, but flies in a river of boots.

Grammatically correct, more or less, but not likely to inform the hearer about much of anything.   And for every meaningful sentences, there are many combinatorially more that would be unlikely to be useful sentences in any context.  The birth of tables preceded the art of melting dishes by a trillion dollars.  My sofa isn't feeling well.

See far ogle.

space walk

What I took away from Yockey's book most of all was the idea that mutational walks (the neutral diffusion through phase space) are not walking from residue to residue but from codon to codon.  Yockey said that it was important that the substitutable amino acids form a Hamming chain.  If one of the links in the Hamming chain was not substitutable in the sense of the others in the set of functionally similar amino acids (i.e. the residues that fell within the same sphere in Borstnik-Hofaker space) then the walk was truncated.  It would imply that even a double-mutation can't be expected to happen regularly over evolutionary epochs.

And it calls into question whether evolutionary walks through protein space can go against the grain.  Just how sensitive is the process to loss of function?  For any population, there is some threshold over which the effect of the residue substitution is deleterious enough to not be ignored.  The tendency of the mutation to spread is not only stunted but negatively selected against.  Some "neutral" mutations are only neutral because they are masked by the noise of thousands of other mildly deleterious alleles.  And then there is the thing that is still largely ignored:  Some mutations simply make the protein non-functional; not worse than comparable its neighbors in protein space, but no good at all.

It would be interesting to see what comes of studies in programmable matter with regard to random programming.

Also need to consider in what sense substitutability affects the information in each codon.

Schützenberger quotes

Schützenberger:

Biology is, of course, not my specialty. The participation of mathematicians in the overall assessment of evolutionary thought has been encouraged by the biologists themselves, if only because they presented such an irresistible target. Richard Dawkins, for example, has been fatally attracted to arguments that would appear to hinge on concepts drawn from mathematics and from the computer sciences, the technical stuff imposed on innocent readers with all of his comic authority. Mathematicians are, in any case, epistemological zealots. It is normal for them to bring their critical scruples to the foundations of other disciplines. And finally, it is worth observing that the great turbid wave of cybernetics has carried mathematicians from their normal mid-ocean haunts to the far shores of evolutionary biology. There up ahead, Rene Thom and Ilya Prigogine may be observed paddling sedately toward dry land, members of the Santa Fe Institute thrashing in their wake. Stuart Kauffman is among them. An interesting case, a physician half in love with mathematical logic, burdened now and forever by having received a Papal Kiss from Murray Gell-Mann. This ecumenical movement has endeavored to apply the concepts of mathematics to the fundamental problems of evolution -- the interpretation of functional complexity, for example. [from here]

Okay, one more:

Speaking ironically, I might say that all we can hear at the present time is the great anthropic hymnal, with even a number of mathematically sophisticated scholars keeping time as the great [Darwinian] hymn is intoned by tapping their feet. The rest of us should, of course, practice a certain suspension of judgment.

not protein space exactly

Stuart Kauffman writes about the protein space idea of Smith's.  It's an interesting concept but more primary is that a string of codons inhabits configuration (genotype) space as well as protein (phenotype) space.  Both spaces together constitute the hyperspace or phase space of the string.

The "walk" through phase space is not a traversal where each site is a dimension of 20 interconnected nodes.  Each base pair choice is directly connected to three others, making each codon directly connect to three single point mutations.  In typographical space, these nine other values are one "jump" away; three being slightly closer than the others (transitional mutations).  In protein space, the alternate codons may be near or far, depending on the effect that the change to the functional utility of protein itself has on the probability that mutation will be propagated.  This could be approximated, for a simple model, in terms of average substitutibility, in three dimensions.
http://www.evolutionnews.org/2006/07/mathematicians_and_evolution002387.html

What is the combined measure of distance between nodes in phase space if typographical space has Hamming distance and protein space has some king of functional dissimilarity measure?  Probability is the common denominator.  Hamming distance may be reconceived as log probability and so may functional dissimilarity.

In the big picture, it is not simply the ability for the search through phase space to discover a "winning" amino acid sequence but the utility of it to be sensed by the sensing apparatus of natural selection.

Note: Long jumps through phase space is part of a bigger picture, but one that does a lot less to describe the actual behavior as a Markov process.  Intragenic recombination and frame shift mutations are little wormholes through phase space, but they do little to describe evolutionary behavior.  It would be interesting to be able to model these dei ex machini but my guess is that their probabilities are very poorly understood.
http://www.sciencedirect.com/science/article/pii/S0888754306001807

Treat:  Schutzenberger's "Algorithms and the neo-Darwinian Theory of Evolution":
http://www-igm.univ-mlv.fr/~berstel/Mps/Travaux/A/1967-3NeoDarwinianFullPaper.pdf

Bonus:  Negoro's explanation of why he doesn't think that nylonase resulted from a frame-shift mutation.
http://www.sciencedirect.com/science/article/pii/S0022283607005347

Perakh and free lunch

http://dennisdjones.wordpress.com/2013/01/05/response-to-the-mark-perakh-essay-there-is-a-free-lunch-after-all-william-dembskis-wrong-answers-to-irrelevant-questions/

http://dennisdjones.wordpress.com/2013/04/18/does-evolution-alone-increase-information-in-a-genome/


http://www.infidels.org/library/modern/mark_vuletic/dembski.html

overview of the treatment of "information"

Excellent overview:
http://www.discovery.org/a/14251


I'm trying to remember which Dembski critique was claiming that genetic algorithms are a dark art.  And which was saying that genetic algorithms have a solid mathematical foundation in the work of Fisher.  Is the work of Fisher a red herring for the fact that genetic algorithms have to manipulated into provide specifc sorts of answers?

Update:  Ok... the first quote is actually Wein quoting Geoffrey Miller's "Techonological Evolution As Self-Fulfilling Prophecy" (intriguing title, n'est pas?):

The trick in genetic algorithms is to find schemes that do this mapping from a binary bit-string to an engineering design efficiently and elegantly, rather than by brute-force.... The genetic operators copy and modify the genotypes from one generation to the next.... Getting the right balance between mutation and selection is especially important.... Finally, the evolutionary parameters [such as population size and mutation rate] determine the general context for evolution and the quantitative details of how the genetic operators work.... Deciding the best values for these parameters in a given application remains a black art, driven more by blind intuition and communal tradition than by sound engineering principles.²⁴ 

which I quoted here on this blog.

The first quote, I must've been thinking of the "eandsdembski" paper.  Elsberry and Shallitt actually try to avoid the problematic claims about genetic algorithms and imply that we know that the amazing functional complexity we see in nature simply follows from the math:

Dembski asserts that \evolutionary algorithms" represent the mathematical underpinnings of Darwinian mechanisms of evolution [19, p. 180]. This claim is egregiously backward. A large body of scholarly work is completely ignored by Dembski in order to make this claim, including Ronald Fisher's 1930 book, The Genetical Theory of Natural Selection.[16]  It is evolutionary computation which takes its underpinnings from the robust mathematical formulations which were worked out in the literature of evolutionary biology.

They draw a distinction between genetic algorithms and artificial life.  They seem to be implying that none of the fine tuning done for genetic algorithms applies to evolutionary computing in artificial life, as it's general target of survival doesn't predispose it to solving particular, goal-directed problems (such as Schneider's ev program?).

Aside:
It occurs to me that the information going from the environment to the population in question should be represented as the logarithm of the decrease in probability of death before reproduction.  Given all the bits of information being absorbed by a population about property X, what would the signal to noise ratio be?

Mark Perakh R.I.P.

Just found out recently that Mark Perakh died this year.  I might have to update some of this years posts in case they refer to Perakh in the present tense.

macroscopically describable

If we repeat an experiment 2^k times, and define an event to be “simply describable” (macroscopically describable) if it can be described in m or fewer bits (so that there are 2^m or fewer such events), and “extremely improbable” when it has probability 1/2^n or less, then the probability that any extremely improbable, simply describable event will ever occur is less than (2^(k+m))/(2^n). Thus we just have to make sure to choose n to be much larger than k + m. If we flip a billion fair coins, any outcome we get can be said to be extremely improbable, but we only have cause for astonishment if something extremely improbable and simply describable happens, such as “all heads,” or “every third coin is tails,” or “only every third coin is tails.” Since there are 10^23 molecules in a mole of anything, for practical purposes anything that can be described without resorting to an atom-by-atom accounting (or coin-by-coin accounting, if there are enough coins) can be considered “macroscopically” describable.
    Granville Sewell, "Entropy, Evolution and Open Systems", note 5

Note:  Relate Methodological equivalence to experimental math and P=NP...

Does information equal entropy?

I think both Schneider and Yockey have a problem with the term "negentropy", and it is somewhat understandable.

If someone says that information = uncertainty = entropy, then they are confused, or something was not stated that should have been. Those equalities lead to a contradiction, since entropy of a system increases as the system becomes more disordered. So information corresponds to disorder according to this confusion. --Schneider

This is in one of the several pages where Schneider makes it quite clear that information can and does only mean one thing:  some reduction in uncertainty defined on some receiver somewhere.

What you will see in many anti-ID writings are assertions that "meaningful" text is always less jumbled than random character sequences.

Sending compressed messages maximizes the uncertainty that is reduced with the reception of each character (Shannon information per character), and also minimizes the ratio of Kolmogorov information to length of message.  But the compression will undo the sense of the message.

Ultimately, the "sense" of the message is the macroscopic effect it has on the receiver of the message, and if that isn't what Schneider means by saying that random messages have zero information, that such messages tend to put the receiver in the highest entropy state (no net reduction of uncertainty)?

Like I said compression undoes the sense of the message unless it can be decompressed into the form where the inherent redundancies in the text allow the text to "mean something" to the receiver, i.e. exhibit operational semantics.

Shannon information is maximized through a comm channel as each transmitted symbol tends to offer about the same amount of "surprisal" as any other symbol.  If a symbol has a lower probability of being received, it conveys more information than the other symbols.  It perhaps represents a much more interesting macrostate in the transmitter than they other symbols do, but in that case each of the other symbols will pay a cost of conveying less information.  If symbols {a,b,c} convey the same information but {d} represents a rare event/macrostate, then the information conveyed by a or b will be closer to log 3 bits than to 2 bits.  That's a hit to the communication rate.

The leveling of symbol frequencies for optimized comm rates is analogous to the leveling of macrostate probabilities in a phase space.  Schneider seems to acknowledge this.  Reading an unlikely character conveys a lot of information, and a system/machine being in a tiny/rare macrostate in its phase space indicates that an event more more interesting than equilibrium has occurred.  There is a correlation even though Schneider is terse about this and much more concerned with the sin of conflating entropy and information.

Going back to his idea of a random text conveying no information, . . .

TO BE CONTINUED   

Another blogger's thoughts on the _ev_ program

I forget whether I had already logged godandscience's thought-provoking article:

    http://www.godandscience.org/evolution/specifiedcomplexity.html

I also thought Dembski's point about ev's handling of the equality condition seemed relatively weak, but I've had some other thoughts on that since.  I have since had the feeling that the treatment of ev was a late decision in the development of Dembski's book, and he had the choice of either quickly shoehorning something about it into his book or deal with the postings that the book was sadly behind the times since there exist papers in which simulations are purported to create information.

Like the mound of papers theatrically stacked in front of Behe in Judge Jones' courtroom, there are constant allegations of refutations and counterexamples.

I'm less sure that ev demonstrates something real about the biological world, though the more I've thought about the less sure I am what exactly it does demonstrate.  (I think I know what Schneider was attempting to simulate, but that is a different question ultimately from what was actually demonstrated.)

secretor system evolved from the more elaborate flagellum?

http://www.ncbi.nlm.nih.gov/pubmed/23028376

The secretor system evolved from the more elaborate flagellum?

Sure, it makes more sense that the less comes from the more, rather than the more from the less, but ...  Wasn't the opposite story part of the reason Judge Jones gave for dismissing Michael Behe?  Because Behe was allegedly too obtuse to realize that the flagellum might possibly could've maybe developed from the secretor?

Uh-oh, these researchers have thrown a bit of candy to the creationists.  Which makes them secular creationists.  (Unless they are Christian, and then they are simply creationists.)

Of course, there are severe problems with exaptation, which even Allan Orr recognizes, and these problems don't go away if the secretor continues to be trotted out as the missing link, or if another less complicated nanomachine is.

Irremedial Complexity

Behe writes here about the fuzzy hypothesizing (don't worry--it's tolerated because it doesn't threaten neo-Darwinism) on the complex interactions within cells--particularly catalytic networks.

Irremedial complexity tries to fuse the non-selective aspects of Kimura's neutral mutations with the "information for free" notinos of Kauffman's hypothesis of random catalytic networks.  Kauffman is hoping to fill as large a gap as (or much more than) Kimura's idea did.  As Behe notes, "irremedial complexity" is a possible solution to a possible problem (i.e. it will only be acknowledged that there was a problem if it turns out to be a solution) that provides a narrative for the seemingly gratuitious complexity we observe in cells--especially eukaryotic cells.

Seemingly gratuitous?  Like junk DNA was seemingly useless?  Sounds like a vestiges-of-the-gaps argument to me.

Irremedial complexity:  Autologous name? Is this an irremedial concept? Who knows?  Maybe it is an idea worth considering.  But considering the vitriol against ID for not coming up with exact figures for everything with exact justifications (is anything really that rigorous in evolutionary biology? the figures are all over the place...) that every biologist accepts, isn't it curious that "the literature" is tolerant of possible solutions to possible problems?

It is funny that they describe bureaucracy as an ratcheting, irreversible complication of control.  But that analogy isn't encouraging that anything good would come out of it.

Logical Depth and Physical Complexity

Very interesting stuff:

"Logical Depth and Physical Complexity" by Charles Bennett

    http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.70.4331&rep=rep1&type=pdf

Selection Equals Creativity

I can't remember whether I specifically posted about how, in spite of other evolutionary factors being recognized by some leading biologists as driving adaptation and speciation and diversity and even innovation, and selection is still commonly held to be the key driver of creativity and information

Take Darwinian evolution evangelist Kenneth Miller:

Just three things: selection, replication, and mutation... Where the information 'comes from' is, in fact, from the selective process itself. [quoted here]

Stephen Jay Gould would've called this hyper-selectionism.  Gould seemed to think though that selection drove the changes and the really neat stuff (or a lot of it) resulted from side effects of the adaptive forces.

Another thought-provoking Richard Dawkins quote from the same source:

And it's also a smooth gradient, not a sudden leap from a flat plain in the phase space. Or rather it must be a smooth gradient in all those cases where evolution has actually happened. Maybe there are theoretical optima which cannot be reached because the climb is too precipitous. [quoted here]

It must be a smooth gradient.  Wait, how do we know it's a smooth gradient?  Because the smooth gradient is necessary for evolution to occur in those instances.  The peaks with no convenient reliable trail of breadcrumbs leading selection to it can only be theoretical optima.  Glad we got that cleared up.

Wait a minute!  Is he saying we don't need a multi-verse of 10^200 universes to reach these optima?  Why do some scientists invoke this ultimate deus ex machina at all?  Because of their lack of faith in selection.  But Dawkins is a true believer, and he staggers not at the promise of Darwin.

Schroeder complexity

Schroeder claims that bit strings formed from random coin tosses, most probably, will have zero information content zero.  I'm not sure how he can argue that unless he happens to known the receiving machine in question.

But, taken very generally, it is an interesting question to ask, I think: what is the minimum Kolmogorov complexity of a message that would be recognized as a valid proof of Godel's Incompleteness Theorem (in English)?  Once you figure out a number, call that number N.

What would Schroeder assess the average Shannon complexity of the shorter such proofs to be?  (Schroeder thinks exclusively in terms of Shannon complexity.)  Would it be more than N?  Does it make a difference that state change from not understanding the theorem to comprehending the truth of the proof results from a relatively small set of messages, even among the set of all proofs.

Now, what is the probability that of a randomly generated bitstring N bits long can be decrypted (or creatively recrypted) into one of these proofs using a decryption algorithm with Kolmogorov complexity less than N/4?

What, then, is the problem with claiming that specified complexity is a reliable empirical marker of intelligence? The problem isn't that establishing specified complexity assumes the form of an eliminative argument. Nor is the problem that specified complexity fails to identify a causal story. Instead, the problem is that specified complexity is supposed to miscarry by counterexample. In particular, the Darwinian mechanism is supposed to purchase specified complexity apart from a designing intelligence. But does it? [from here]

Gibberish

Gerald Schroeder at Aish.com

This impossibility of randomness producing order is not different from the attempt to produce Shakespeare or any meaningful string of letters more than a few words in length by a random letter generator. Gibberish is always the result. This is simply because the number of meaningless letter combinations vastly exceeds the number of meaningful combinations.

With life, such gibberish was and is lethal.

Piatelli-Palmarini on Socially Constructed Science

Revealing of many things -- about kneejerk anticreationism, about deference to neo-Darwinism, about cultural bias for or against various "research programmes" (in Lakatos' sense):

Suzan Mazur: Why has American science been slow to accept a reduced role for natural selection in evolution? Is it the physics that people just can’t grasp?

Massimo Piattelli-Palmarini: It’s not just American science, but rather Western science, though indeed France has, in this respect, a different story, not quite a noble one.

Some consider Darwinism to be quintessentially “Britannique” and they had Bergson suggesting a different approach to evolution, then the mathematician Rene’ Thom and his school, stressing the role of topological deep invariants. They may have come to anti-Darwinian conclusions for rather idiosyncratic reasons.

Anyway, even if we take the many, many biologists in many countries who have contributed to the new rich panorama we have today of non-selectionist biological mechanisms (including the masters of the Evo-Devo revolution), they are reluctant, in my opinion, to steer away from natural selection. They declare that the non-selectionist mechanisms they have discovered (and there are many, and very basic) essentially leave the neo-Darwinian paradigm only modified, not subverted.

I think that abandoning Darwinism (or explicitly relegating it where it belongs, in the refinement and tuning of existing forms) sounds anti-scientific. They fear that the tenants of intelligent design and the creationists (people I hate as much as they do) will rejoice and quote them as being on their side. They really fear that, so they are prudent, some in good faith, some for calculated fear of being cast out of the scientific community.

There are, however, also biologists who do not fear to declare, as Gregory C. Gibson (the William Neal Reynolds Distinguished Professor of Genetics, North Carolina State University) wrote in Science (2005), reviewing a book on robustness and evolvability:

“[this book] contributes significantly to the emerging view that natural selection is just one, and maybe not even the most fundamental, source of biological order”.
“Robustness must involve non-additive genetic interactions, but quantitative geneticists have for the better part of a century generally accepted that it is only the additive component of genetic variation that responds to selection. Consequently, we are faced with the observation that biological systems are pervasively robust but find it hard to explain exactly how they evolve to be that way”.
G. C. Gibson SYSTEMS BIOLOGY: The Origins of Stability. Science, 310 (5746), p. 237.
Prudent, but explicit.

And the distinguished evolutionary biologist Massimo Pigliucci, in an excellent book co-authored with the philosopher Jonathan Kaplan, writes:

“It is unwarranted to think that adaptation, diversification and evolution more generally are closely related phenomena that take place via the same mechanisms in the same populations […] Adaptation can, and verifiably does, take place without speciation, as does nonadaptive evolution more generally”.
Massimo Pigliucci & Jonathan Kaplan (2006), Making Sense of Evolution: The Conceptual Foundations of Evolutionary Biology. Chicago: The University of Chicago Press. (Chapter 9, Box 9.2).

There are other expressions of discontent with canonical neo-Darwinism, but, all in all, prudence prevails.

Explanatory Filter and Explanatory Blinders

Perception:  Even though the habit is to quibble/demur/deny/equivocate or change the subject, the problem most of the detractors have generally with Dembski's Explanatory Filter seems to be the choice between natural law explanations and interference by an intelligent agent.

What many of them seem to be saying is that it takes particular knowledge of a particular problem to know whether something bears signs of being a human artifact (based on specific facts about existing human artifacts) before we can know whether something is intelligently designed.  If that's true then their version of Occam's Razor (which is apparently the 4th or 5th Law of Thermodynamics) must require us to assume that some confluence of law-like behavior of atoms and "sheer dumb luck" interact to make what otherwise seems designed.  If it doesn't bear unique signs of human technology, it must be as purposeful as a snowflake.

I mean, this seems like it must be the case given the vitriol they inveigh against the Explanatory Filter.  The Filter is crap they say, and if you aren't convinced, I'm sure that Mark Perakh can start a Project Keith or Project Dave to change your mind.  Hey, just because they are not psychologists doesn't mean that scientists don't know about the Asch conformity experiments.

Wein and Perakh: Wein'ing about Dembski

The fact that this is on TalkOrigins makes me wonder about Wein, but still, he at least seems (at brief glance anyway) to try to clarify where he's coming from.

Perakh, where I got the reference to Wein, is more in the ad hominem business.  It's hard to take him at all seriously, with his disavowal of the term "Darwinism" which shows up in the evolutionary literature and Gould's essays (maybe they are "secular creationists"...) and his trafficking of consensus science -- which is noteworthy-- Perakh says that 1 in 142 scientists admit to doubting "the main tenets of modern evolutionary biology" which in context seems to be those elements of "the neo-Darwinian synthesis."

That many?!   I thought I was on the ground floor of a scientific revolution.  Don't worry, that virus has only infected 1% of the cells of your body.  You're fine!   No wonder they are desperate.  The NCSE has even diversified into climate politics.  They've had only 5 decades to get in on that.

I'll read Wolpert's disavowal of Dembski's interpretation of the NFL theorems again, but in writing his rather brief and politic statement, Wolpert succeeded at (a) saying that he was not trying to say what Dembski is saying nor meant his work to be taken as a challenge to the almighty force of natural selection and (b) noting that the "sit down and shut up" tactics of the Darwinist mob is actually a losing strategy at this point.

Maybe Wein is trying to do better than the mob.

But some things stand out to me already:

Wein quotes some researchers saying "averaging over all different fitness functions does not match the situation of black-box optimization in practice." That is hardly surprising. Solving any real problem (and any number of contrived problems) with stochastic/genetic methods generally involves parameterizing the problem such that gradients can be followed and that concerns remain separated enough to avoid -- or the algorithm is altered in order to not get stuck in local optima.

But real world problems often have some really nonlinear interactions between the parameters, complicating nonlinear responses to a single parameter. I could make a fitness function modeling the real world performance, or I could build into the fitness function some heuristics on the relationship between parameters. For parameters a and b, I might decide (this is pure fancy) that f '(sqrt(a ^2 + 2b^2) is an approximation of actual fitness F(a,b) that is likely to avoid getting stuck in local optima. For Wein, this is just making sure that the algorthm is properly designed to get the answer to the correct problem. Tbere is no concern of how this might actually be unnecessary to get a suboptimal answer to the correct problem but necessary to get an optimal answer to the problem.

"The trick in genetic algorithms is to find schemes that do this mapping from a binary bit-string to an engineering design efficiently and elegantly, rather than by brute-force..." In other words, applying engineering knowledge to understand the fitness landscape in advance, properly parameterizing the problem so that the phase space can be searched effectively, and building into the algorithm ways of dealing with obstacles.

"Deciding the best values for these parameters in a given application remains a black art, driven more by blind intuition and communal tradition than by sound engineering principles." Sounds like engineering in general to me. For Wein (quoting Wolpert), for more important than abstract generalizations of hill-climbing algorithms are the particular problems where you "start with the given [fitness function] f, determine certain salient features of it, and then construct a search algorithm, a, specifically tailored to match those features." This is what Schneider calls making the fitness function behave like the environment in natural selection -- "This is exactly how it works!" After all, we know that natural selection did it, the simulation is just demonstrating how.

Where are the solid mathematical foundations here? These guys need to get out the classics on population genetics and read them!

It could be (although I don't know) that the "crooked wire genetic antennas" Wein refers to might be a particular problem that lends itself to hill-climbing (to an apparently counter-intuitive solution) without a lot of teasing from the algorithm-designer. But Wein makes it pretty clear that you don't just rely on the fitness function to tell your algorithm everything it needs to know. You also need to "determine certain salient features of it, and then construct a search algorithm, a, specifically tailored to match those features."

Now, Wein (and this is a subject that deserves further exposition some time) claims that Dembski takes two bites from the cherry by offering that it is possible that the Darwinian fitness landscape-verse could be finely tuned for evolution. This seems unlikely given what we know of it. But are many of the biological design problems solved over the millenia the sorts of problems where you need to "determine certain salient features of it, and then construct a search algorithm, a, specifically tailored to match those features"?

Question: Is the landscape-multiverse well-suited to natural selection? Of course! After all, here we are...