
Matt Gough Blog [closed]: zufall
Sunday, Jul 11 2004, 05:40
it seems that europe is the place to be at the moment, with impulsetanz and interaktionslabor i would rather be out there than over here. meanwhile i sit down and write, code and randomly get up and dance around the room testing what i have put to paper.
i'm expecting some questioning over the length of my development period for xdsn (currently 3.5 months), considering the manner in which i work i don't think it's a point of fault. but we will have to see. there is a body of thought that considers i should be developing existing notations, i think the strongest criticism will come from there.
whilst developing an implementation of artificial intelligence and artificial creativity that will produce 'pleasing' choreography. i've been wondering if i should take into account the way in which we watch movement - the tracking of a dancer in motion.
a model based on dance as a visual art (rather than physical) might throw up some interesting combinations. attach this model to real time eye tracking (the viewers eye) and more advanced interaction may be possible. with the inclusion of machine learning it might be possible to work out what particular style of 'choreography' an individual responds best to.
of course other responses are missing, but i think that visual recognition is the most important. as humans we see more often than closely engage, it is mostly from visual cues that we infer emotion, etc. i think this is something often overlooked.
the more time i spend developing models of dance composition the more appreciation i have of the choreographer that i love. and at the same time a deeper understanding of the works i dislike. yet it remains to be seem if i will embed 'agenda' into the systems i am developing over inspirational impulses.
as humans we make works for our own reasons, even if we do not understand them. perhaps it is the same for the essence of computation. or will such unknowing 'desire' be dependant on programming.