[Ichneumon]

Anyone who has programmed for a living knows that budget constraints and managerial incompetence often result in the reuse of bad code, and the accumulation of dead code. Looks like a designer to me. More likely a team of Microsoft programmers.

I know you're just being puckish, but since creationists often make a similar lame excuse for why the DNA of different species would have similarities, I'd like to point out for the lurkers that the specific *kinds* of characteristic similarities and differences found in DNA point unmistakably to evolutionary common ancestry, and *not* to "common design".

Furthermore, since we're using the programming analogy, it should also be pointed out that no one could possibly mistake the results of "evolutionary programming" (like genetic algorithms, etc., whereby evolution is harnessed to produce program code without direct human intervention or programming) for the results of a program written directly by a programmer (i.e. "designer"), even one which incorporated a lot of "code re-use" or cut-and-paste from other projects.

The results of the two methods of producing programs are *vastly* different in character and structure, and any programmer could tell at a glance whether a particular program was actually written by a human, or "grown/evolved" via genetic algorithms. And the same goes for DNA -- it looks exactly like the results of an evolutionary process, and not at all like the results of a "design team".

[js1138]

Puckish. Now there's a word I haven't heard in a long, long time.

[anguish]

and any programmer could tell at a glance whether a particular program was actually written by a human, or "grown/evolved" via genetic algorithms

Or rather, any programmer could tell if it was written by a normal human, or one of the following:

• Evil code-obfuscater
• Genetic algorithms
• Raving lunatic programmer with enormous luck

[bobdsmith]

Furthermore, since we're using the programming analogy, it should also be pointed out that no one could possibly mistake the results of "evolutionary programming" (like genetic algorithms, etc., whereby evolution is harnessed to produce program code without direct human intervention or programming)

Just a minor clarification, which is largely irrelevant and you didn't really make any mistake, but I like to hear myself type..

Genetic Algorithms don't produce program code. An individual in a genetic algorithm is represented as a string of "genes" that encode its phenotype. The result of a GA is the most optimal string representation found. So GA's are closest to natural evolution, which is why I find them more interesting than Genetic Programming, which does use computer programs as a representation. So in Genetic Prgoramming an individual is represented as a prgram and the actual program is mutated and evaluated for the selection process. In this case the output will be the most optimal program found.

GP's are not GA's - they are both different applications of evolutionary computation.

[ml1954]

The results of the two methods of producing programs are *vastly* different in character and structure, and any programmer could tell at a glance whether a particular program was actually written by a human, or "grown/evolved" via genetic algorithms.

And one can often tell at a glance that 'a particular program was actually written by' an idiot.