“…Let us start with evolved computer music systems, such as objectoriented systems (Pope, 1988), even if, as we have seen, they cannot be considered truly intelligent: KYMA (Scaletti, 1987), (Fagarazzi, 1990-this issue), Ios (Roads, 1983), PLA (Schottstaedt, 1983), FORMES (Rodet & Cointe, 1984) are the most significant examples. Even if they do not allow inference mechanisms, object-oriented systems include features of significant interest from an AI viewpoint, such as conceptual, object-oriented representations of music, and 'good' mechanisms such as inheritance and information encapsulation» The purpose of these computer music systems is to assist composer's activities.…”