Do-While Jones wrote:
> It would be interesting to measure the time between drum beats played > by a professional drummer, to see how much variance there is in the
Perhaps the variation adds something to the music. Perhaps it detracts.
Music is a social activity. Attempts to produce music by machine
might do so by modeling an ensemble as independent agents, with each
agent having its own idea about variations in the beat, tempo, and
emphasis, listening and conversing musically with the other voices.
Perhaps real-time participation with human musicians could be added.
That would be a great variation on the Turing test, can we tell which
instruments in a jam session are being played by machine, and which by
humans? Of course, Ada would be a splendid language for these