In the area of Music Information Retrieval (MIR), great technical progress has been made since this discipline started to mature in the late 1990s. Yet, despite the almost universal interest in music, MIR technology is not that widely used. There seems to be a mismatch between the assumptions researchers make about the users’ music information needs, and the actual behaviour of a public that to begin with may not even treat music as information. Therefore, the emphasis of MIR research should be more on the emotional, social and aesthetic meaning of music to regular, untrained people. MIR applications could greatly benefit from using the results of recent research into the spontaneously-developed musical competence of untrained listeners.