07 February 2007

More Fun and Games with Evolution

Finally I found a simulation of the ID/creationist model of random variation and selection. It's extremely simple, but if you're patient, you can select and revert your way to some positive changes in the "genome". The key in my view is to start with random character strings so you aren't distracted by the initial state. Basically, one selects (saves) any positive change, and then continues mutating. If nothing better comes along, you revert to your selected case, and then continue mutating again. The effect is to simulate a larger population with only the single individual character string. This is being discussed at Pharyngula.

Also being discussed there is a rebuttal simulation that is pretty interesting, and much more sophisticated. It actually shows how irreducible complexity can evolve. If you follow the discussion of it, you'll learn that most cases of irreducible complexity result from a decrease in the fitness of the individual. One commentator gives a useful analogy with an arch that becomes irreducibly complex only when the scafolding used to build it is removed. There is a net reduction in robustness at that point, and then the arch has many parts that can no longer be removed without its falling. Pretty cool!

It's a bit disappointing watching the evolutionists and creationists chewing on each other, usually on separate blogs. The level of discussion has sunken to a new low since the Dover decision, I suppose. Still, the evolvers are not the least bit gracious in victory, nor do the creators give any quarter. Perhaps I'm just being wishy washy, but I would really like to see them compare the above simulations and the standard weasel program and argue the points to a polite agreement to disagree, rather than engaging in endless hurling of sophomoric epithets and insults.

No comments: