stackless python revisited
i've been interested in stackless python for quite a while now, but basically i've been too chicken to really dive in. so today i went and had another look and decided to just switch my system's python installation to stackless. so far, there are no compatibility problems with working code, just as they claim. so, i see no reason not to switch for good -- all it can do is ADD functionality.
been playing with it a little, and i think it's the solution to some horrible problems i was facing with trying to write a new schnolgo module. performance problems will be fixed and streaming will essentially be implemented automatically. this is an exciting prospect because it should make it actually possible to:
1. "lexically" parse musical expressions in real time (!!) [i'm pretty pleased with how the interface for parsing musical expressions is turning out, real time functionality or not]
2. write spontaneous improvising methods based on parsed patterns in real time (that is, real time musical improvisation that is context-aware)
both things will need more code to get up and running in any useful way, but the underlying framework will be there and kicking ass if all goes well.
i've also been devoting a bit of thought (again) to getting together a gui for schnolgo. trying to conceive of something that will be usable while maintaining schnolgo's flexibility is not easy. i think a lot of thought will need to go into the design phase.
that's allz i got. word.
been playing with it a little, and i think it's the solution to some horrible problems i was facing with trying to write a new schnolgo module. performance problems will be fixed and streaming will essentially be implemented automatically. this is an exciting prospect because it should make it actually possible to:
1. "lexically" parse musical expressions in real time (!!) [i'm pretty pleased with how the interface for parsing musical expressions is turning out, real time functionality or not]
2. write spontaneous improvising methods based on parsed patterns in real time (that is, real time musical improvisation that is context-aware)
both things will need more code to get up and running in any useful way, but the underlying framework will be there and kicking ass if all goes well.
i've also been devoting a bit of thought (again) to getting together a gui for schnolgo. trying to conceive of something that will be usable while maintaining schnolgo's flexibility is not easy. i think a lot of thought will need to go into the design phase.
that's allz i got. word.