Creation Bits

I think its amusing how much evolutionists think that genetic algorithsm are their salvation. The funny thing about genetic algorithms is that they _rely_ on a stable, semantic base for their operation. Creationists don't have any problems with non-deterministic operations happening within populations or even individual organisms. It's that these operations have specific boundaries past which the organism or population falls into error catastrophe. You can't pass gradually from one system to another without hitting error catastrophe.

I always hear "but genetic algorithms produce information". This is one of the dumbest comments I have ever heard, and it pains me that it comes from people who actually program computers! Genetic algorithms are explicitly designed, and include both changing and non-changing parts. It's the non-changing parts that are most important and make the algorithm useful at all!. Evolutionists would have a point if they could point me to a system where all parts of the algorithm change, and have the program, through any sort of non-codal selection (i.e. the selectors cannot simply view the current coding -- that's looking ahead), and the system fundamentally changes what its doing without going through error catastrophe. It simply cannot be done.

Avida is an interesting concept, but it actually shows the weakness of Darwinism. First of all, Avida is incapable of error catastrophe. Second of all, most of the functions of Avida are NOT susceptible to the genetic algorithm. While in evolutionary theory, all parts of the organism are susceptible to evolutionary change, in Avida only a VERY SHORT part of it is. The non-existance of error catastrophe should be enough to disqualify Avida anyway, but even more in order to get it to produce even the smallest, tiniest algorithm, not only to you have to provide HUGE incentives for the algorithm, you have to provide HUGE incentives for ALL of the operations leading up to the algorithm. This is basically forcing a path. It is itself a form of programming. In addition, it is nothing like what happens on earth -- the "benefits" from "beneficial evolution" are not as large, and small deviations are not as costly.

I know someone will say, "but that's what we think happened -- the earth (the environment) programmed the genes". First of all, its funny that to salvage anything, they simply move the design argument to the earth. The earth contains the design is what they are actually arguing, whether they think so or not. So, they are denying creation by explicitly affirming theistic evolution. You see, in order for the environment to serve as a sufficient program, it has to be specifically designed to get you to specific stages! In the Avida example, ALL of the intermediates were PRE-PROGRAMMED into the environment. When the environment did not contain justification for ALL the intermediates PLANNED OUT (remember, evolution requires NO PLANNING), evolution simply did not occur. Programming depends on a plan. If the plan is not in the algorithm, it is in the environment, which would be simply another embodiment for the algorithm. But without design neither one works. However, we know that the plan was not encoded in the environment based on the fact that the environment does not work in the way needed to form drastic semantic change. Therefore, it is reasonable to conclude that the design lies in the organism, or at least that is one of the locations where design is present.

For justification of this, when experiments on Avida were carried out using conditions similar to those in real life, nothing managed to evolve, even given the all of the unrealistically-favorable pieces inherent in the Avida system. For details see here and here.

A funnier reply is here.

The actual Avida software can be found here. I haven't personally used it yet, but plan to shortly. I tried awhile ago and couldn't get it to compile on my Linux box. It has since moved sites, so hopefully they've updated the platform to work with modern compilers.