This project needs no introduction.
Watch a sunset. You will need to set aside a couple of hours or more to do this properly. You should watch from the moment at which the sun hits the horizon, to the point at which all trace of its colour has gone from the sky.
Now try to imagine how a sunset would sound. What are the main features? How do the elements of a sunset translate into sound? And, most importantly, how do the feelings the sunset produced in you translate into sound? How do you convey the impression created? Make written or aural notes to help you if necessary.
Now, create a piece that reproduces as nearly as possible what you have imagined. Observe and note the gaps between what you have imagined and the finished result. Have other people listen to the piece, without telling them what the piece is about. Do they get it? Try again with a title ('Sunset' is the obvious one), or more information. Does that change things?
Finally, get hold of some video footage of a sunset and attempt the same exercise but with the addition of the visual imagery. What has changed? Is the imagination liberated or constrained by the visuals? Can anything be done to improve the imaginative experience?
Mapping decisions are significant here. The computer-controlled version is straightforward to execute, but how can the results be made musically interesting? Giving the user control may require some kind of training, if they are to complete the puzzle successfully. Is it the solution, the original, or the processes involved in solving the puzzle, that is being mapped? If this is to be performed by the composer, to what extent (if at all) can the algorithm be used to reflect compositional intentions?