Since we only have tested prototypes of the environment we haven't used the environment for weighty musical applications yet. However, we would like the speculate about some of the possibilities it offers.
First. Sounds can be rendered in real-time from within the composition environment. No intermediate steps between composition and synthesis need to be taken. We hope that this will make working with sound in composition more efficient. More important, we hope that this will bring forth new ideas for the problem of the control of sound synthesis. As said in the introduction, sound descriptions, and thus the control of sound synthesis, has to become part of the composition environment to fully exploit the richness of computer generated sounds. The proposed environment makes an attempt to provide this integration.
Second. We showed how synthesis processes can be created and manipulated from within the Lisp environment. The inverse is also possible: synthesis processes ask the evaluation of Lisp expressions. Since synthesis processes cannot create new objects for fear of waking up the garbage collector, the evaluation is conferred to another thread that runs in lower priority. We call it the evaluation thread. We still have to explore the musical applications of this possibility. We imagine a synthesis process that analyses incoming sound. The analysis data is subsequently passed on to a compositional process expressed and evaluated in Lisp. The results from the compositional algorithm are then used to influence the synthesis.