I spent some time this morning transcribing the motives that make up Terry Riley’s In C into TinyNotation so that I could use them in the creation of a script which creates scores instances of hypothetical performances of the work. I have this crazy idea that repeatedly doing this on a large enough scale could generate data enough to facilitate a formalist statistical analysis of this seminal aleatoric work.
Most of the computation time seems to go into into showing the MusicXML score. For scores with parts numbering less than four, this is passable. Better to render to Lilypond, if you can. However, this creates an unwieldy image or EPS file that is next to useless for print purposes. Anyway, the point here isn’t so much the printed score as end product, it’s the creation of a symbolic representation of the score which can be then further reduced and analysed with the help of music21 and/or common sense.