I'm more responding in general to the thread, but I felt impelled to post this message after reading this comment:
Parsimonious_Razor writes:
Most of it relied on arguing that a computer simulation is not biology and the outcome was "predetermined" by the core of the program.
We tend to forget the tasks we complete and remember those left unfinished, and so I still remember the last time I discussed this topic. It was with Fred Williams, and his last post on the topic was
Message 246. Fred first objected that evolution simulation programs included the outcomes within the code, and later that
genetic algorithm programs do not emulate evolution. We were making progress by discussing actual programs (see
Fixation of genetics - program) when, can you believe it, Fred disappeared!
About Avida, if memory serves correctly, this effort has been going on for a long time. After all the subsequent efforts that have attempted to simulate evolution in ways more analogous to actual biology, I was surprised that the Discover article didn't give much space to any of the other efforts beyond Avida. It's no knock on Avida, but it began when computers were slower and memory was a scarce commodity, and so because it evolves computer programs instead of simulated organisms it is less directly applicable because analogs must be drawn between, for example, what constitutes metabolism in a computer program versus an actual biological organism. This has no impact on its use as a research vehicle, but as evidence for Creationists that evolution can evolve new and novel structures and mechanisms it must be admitted that it is less than ideal. The additional layer of abstraction makes the task of comprehension much more difficult for those already opposed to the concept of evolution.
--Percy