Information in the Genome

Many times I read that there is no evidence that evolutionary mechanisms have added information to a genome. Recently this claim was reproduced on a blog post titled Entropy, Evolution or De-Evolution.

The entire post is filled with misunderstandings and misrepresentations. I wanted to focu on one of the claims reproduced there. The claim is attributed to Bruce Malone, though no further information is given concerning it. Malone apparently has a Bachelors in Chemical Engineering, though no relevant expertise regarding evolutionary biology.

The quote is as follows:

No experiment has ever shown how useful functioning information can be added to the DNA molecule by random changes.

I have tried to track down the souirce of the quote so I can see it in context, but have not found it as yet.

Here, Bruce Malone claims that DNA changes don't result in any additional "useful functional information". No definition of what this might be is given. In the absence of any definition, we can look to the field of Information Theory.

As I understand it, by any measure of information as understood within Information Theory, taking 2 identical strands of DNA, and then changing one of them results in an increase in information. It doesn't matter if this is a change of a single base pair, and it doesn't matter if this change would result in a beneficial, detrimental or neutral result in the organism.

How about "complex specified complexity" or "specified complexity", a term coined by Intelligent Design proponent William Dembski. He defines it as a property of organisms that cannot have arisen through natural processes, and that must have arisen through the actions of an intelligent agent. However, it's been shown that evolutionary mechanisms can produce "information" in this sense, and that repeated application of these mechanisms can produce sufficient information to cross Dembski's claimed "Universal Probability Bound". Dembski's ideas have been found to be without merit. As a brief example, researchers in the field of Artificial Life, where computer simulations are used to study evolutionary mechanisms, all ostrs of novelties and increased complexity are observed which should be impossible were Dembski's claims correct.

What about "functional information", as proposed by Stephen Meyer. Meyer defines this as the production of linear sequence-specific arrangements of characters. He states that it is our uniform experience that this sort of information only arises with the actions of an intelligent agent. There's no doubt that we, as the main intelligent agents we are aware of, can create this sort of information. However, Meyer's leap from this to the conclusion that an intelligent agent is necessary for these sorts of strings of symbols (as we could view DNA) is not valid. In fact, as was mentioned above when discussing Dembski's "specificed complexity", evolutionary mechanisms can and have generated this information.

So that would mean that Malone must be using "useful functional information" to refer to something else. Whatever it is, I can guess that it's either a useless concept, or the claim that evolutionary mechanisms cannot account for it is false.