Big fat music collections with millions of songs seems like a common thing nowadays. I don’t know about you, but I have got the feeling that I am missing out on a lot of great music. Once all the music download sites (like magnatune) sprang up, the number of songs you could access increased.
Then came the music recommendation sites (like Last.fm). There are many recommendation sites I have come across, but none of them are complete. They somehow don’t allow me to discover new music. Their algorithms are either text-based classification, user-centric (comments etc), “expert” classified or a mix of all three. There is one problem with these methods and that is they don’t actually present the complete picture. You need to know about what you are missing out on more than what you like. Expand your boundaries so to say.
Thats where this comes in :
This was one of the papers at the International Society of Music Information Retrieval Conference (ISMIR 2009) Carles F. Julia, Sergi Jorda , Music Technology Group, Universitat Pompeu Fabra, Barcelona, Spain called “Songexplorer : A Tabletop Application for Exploring Large Collections of Songs”.
It presents the massive music collection as a map with similar songs closer together. This way you get the bigger picture first. Nothing is hidden. Brilliant idea. Its obviously not much use from a personal PoV but if they can port it onto a desktop or a more common hardware I am pretty sure this will catch on.