Some embodiments provide a program that creates a composite media presentation from several media clips. The program creates an audition set for a particular location in a composite display area for placing media clips to specify a description of the composite media presentation. The program adds to the audition set several different types of media clips that are candidates for placement in the composite display area at the particular location. The program iteratively selects a different media clip in the audition set as the only committed media clip in the audition set to include in the composite media presentation.