Posted by Doug Weathers on July 28, 2003 at 16:18:26:
In Reply to: Concerned About Selection posted by David on July 28, 2003 at 08:23:40:
I tend to pick the poem that's the most grammatically correct. When words are semi-randomly thrown together, the shorter strings are much more likely to fit together correctly.
In other words, the more words the monkey types, the less likely it is he'll produce something I'll vote for.
How to fix? If it was me, I'd be tempted to meddle with evolution and declare non-viable any poem that is grammatical gibberish.
My excuse for this would be that the new poem-creature must be at least able to survive in its environment - the mind of an English-speaking reader - before it has a hope of reproducing. A non-grammatical string of English words can't live for long in MY head unless it's pretty short ("Think Different").
Survival of the fittest starts with survival. A mouse born with gills instead of lungs won't survive to reproduce - neither will a non-grammatical poem. Either the system must be able to evolve a grammar, or a grammar must be imposed.
That's my opinion, anyway.
Disclaimer: IANA Biologist - IAA Blowhard.
p.s. IS the system able to evolve a grammar?
: A couple of days ago somebody posted #3360 to the bulletin boards:
: the turned sword covered bodies
: guiding some wholesomeness in till
: densimetric evulsions round eye till
: invectively blood into berry good golden they
: yummy previsions devotions drew we the
: fight solidifying look do
: my shovel twists
: I ended up reading this several times, and thought it was one of the most promising developments yet. However I just looked at the database and its gone.
: It seems people are mostly choosing the shortest phrase that has some kind of semi-meaning. The population seems to increasingly consist of very short phrases (usually containing "sometime the life loved.")
: I'm starting to think it was a mistake to pick different snip points for each parent when new poems are produced. I programmed it this way specifically to allow poem length to change, and when poems started shortening I assumed they would eventually be recombined to form more complex poems, but the negative effect of short poems is that there is just a lot less genetic material in the gene pool and we have an increasingly homogenous population.
: Opinions?