Before I continue, I want to make clear that I am certainly not an expert on Aristotelian metaphysics. In addition, obviously what I say about Catholics and Protestants are broad generalizations that may not reflect the actual state of affairs.

Catholics believe in Transubstantiation. Protestants do not. Catholics believe that God supernaturally acts in the Eucharist even though nothing empirically seems to change. Protestants do not understand or do not believe the Catholic’s view on the Eucharist. When they see bread and wine, they see bread and wine.

Point is that Catholics have no problem with evolution and God supernaturally creating, providing, and sustaining the universe. They do not need empirical evidence to suggest that God had designed this or that.

“the same random selection” -> “the same as random selection”

Conservation of Information, as described in various papers on the Evolutionary Informatics Lab’s publication page, constitutes such a law governing information and is directly pertinent to establishing the insufficiency of conventional material mechanisms for generating biological information AND the need for information sources not reducible to such mechanisms (which includes characterizing the flow of information among them).

CoI, as discussed in the EIL papers, seems to say something like the following:

Given a search S that involves a function ƒ : X → Y (which may be a fitness function, a probability distribution, etc.), if ƒ is chosen randomly from Y^X (or a permutationally closed subset of Y^X), then S will on average perform the same random selection.

This certainly illustrates the limitations of randomly choosing ƒ, but how does it show “the insufficiency of conventional material mechanisms for generating biological information AND the need for information sources not reducible to such mechanisms”?

I’m an MSEE with extensive experience in some of the information theoretic topics discussed on UD.

Then I’m sure you knew that so-called “conservation of information” is just the data processing inequality of information theory. See chapter 2 of Cover and Thomas.

If Sooner is maintaining that a piece of software might generate information, I can only laugh. If he is not, he might want to rephrase and clear up his pronouncements.

Computational search is automated observation of a physical system. Typically, a search program observes the responses of a function subprogram to inputs. The search program yields information. It does not create information.

Dembski and Marks refer over and over to Leon Brillouin’s “brilliant insight” that computers do not create information. You can read the relevant portions of Chapter 9 (?) of Science and Information Theory at Google Books. Brillouin’s mid-1950′s perspective of computing was that scientists use their laboratory instruments to make observations, and then feed the information they gain into the computer for processing. He saw the processing as translation from one language to another, with possible loss, and without gain, of information. Information theory has made considerable advances in the past 50+ years, and the data processing inequality gives a more general understanding of what he was driving at.

It’s important to understand that Brillouin did not contemplate the scenario of one software module making observations of another. His comments are not relevant to computational search.

I realize that I should have put the point slightly sharper than I did in my last message. There are almost guaranteed to be search procedures which function better than random for the problems we are interested in. For purposes of NFL, these search procedures would have to have worse than random performance on other problems – problems which we are not interested in, perhpas cannot even state in a universe as small as ours.

Because our universe, on our size and time scale, is relatively smooth we do wind up asking relatively compact questions – look how far we got with assumptions like continuous and everywhere differentiable! We’re so proud of our ability to model the weather with partial differential equations, but most problems look more like multidimensional TV static.

The kind of computational search we address is one in which the the search algorithm supplies inputs to a “black box” and observes outputs. In practice, what’s inside the box is usually a function written by a human. The search practitioner can read the source code of the function and gain knowledge, possibly exploitable in search, that is inaccessible to a black-box search algorithm. The practitioner gains the knowledge by observation, not creation of information.

It’s also possible for a computer program to read the source code for a function and attempt to establish exploitable properties of the function (e.g., differentiability). No one has established whether humans perform analyses that are not Turing-computable. That’s a philosophical issue, and it will not be resolved by taking the logarithm of a relativized performance measure.

I knew I could count on you to get what I was saying, and that’s why I addressed my earlier comment to you. Anyone with any depth of understanding of computational search would see that I had tossed off some substantive remarks. Who can say whether Dembski could not see or did not care to acknowledge?

I also find it to be very productive, stimulating multiple empirically testable hypotheses, which in turn provide multiple opportunities to figure out how the natural world works.

And, I find the theory of evolution intriguing; it stimulates my curiosity and my desire to get out and look at the world around me, rather than simply sit around and think about it in the abstract. It’s that overwhelming sense of curiosity that drives most scientists to “hold…the mirror up to nature”.

Usefulness, productivity, and curiosity yes; “belief” no. Indeed, to be a good scientist means to maintain an attitude of radical skepticism, especially about one’s chosen field of study. T. H. Huxley said it best:

“Science has taught to me the opposite lesson. She warns me to be careful how I adopt a view which jumps with my preconceptions, and to require stronger evidence for such belief than for one to which I was previously hostile.

My business is to teach my aspirations to conform themselves to fact, not to try and make facts harmonise with my aspirations.

Science seems to me to teach in the highest and strongest manner the great truth which is embodied in the Christian conception of entire surrender to the will of God. Sit down before fact as a little child, be prepared to give up every preconceived notion, follow humbly wherever and to whatever abysses nature leads, or you shall learn nothing. I have only begun to learn content and peace of mind since I have resolved at all risks to do this.”

(see http://aleph0.clarku.edu/huxley/letters/60.html)

“What is the underlying theology, anyway?”

I’m unclear on this whole subject: is O’Leary asking what the “underlying theology” is for evolutionary biology? If so, the answer is, of course, none. Evolutionary biology is not a theological enterprise, it’s an empirical science. Yes, some of the founders of the “modern evolutionary synthesis” (e.g. R. A. Fisher, Sewall Wright, and Theodosius Dobzhansky, among others) were theists, as are some prominent evolutionary biologists today, but they all have taken care to keep their theology out of their biology.

Perhaps O’Leary is asking a rhetorical question about the “underlying theology” of people like Richard Dawkins and P. S. Myers. Forgive me, but I thought they were atheists. Have I misunderstood the proper use of the common Greek prefix “a”? Does it’s use here mean the opposite of what it usually means?

But perhaps O’Leary is inquiring about the “theology” that underlies ID. I thought the answer was again none. Am I wrong?