An interesting short analysis of how Napster's Japan's subscription-based offering is doing:
http://www.japaninc.com/mmw110
It is amusing to see how often Music services boast the number of available tracks in their Subscription library, when, oftentimes, this sheer number of tracks really is of little interest to consumers assessing the attractiveness of a Mobile Music subscription service offering. These services need to look at Relevance -- I'd rather have access to a smaller sample of current or recent music of a particular genre than an impressive library of millions of tracks of past songs at my disposal. Consumers probably already have those songs in their computer or CD collection.
I'm not sure how concentrated the US Music market is (I'm pretty sure that we don't have a genre like J-Pop that encompasses 80% of the market), but the lessons learned in the more advanced Japanese mobile music market should undoubtedly be applied here when subscription services negotiate the makeup (not just the size) of their library with the labels.
