Help me find the music I like!


Apologies to all my readers – it’s been way too long since I posted here. I’ve been frantically busy and on the road, but these are no excuses. I resolve to do better, and have no shortage of interesting stuff to write after my recent travels, so here we go…

“Collaborative filtering” – systems that allow us to collaborate with others to find what is relevant to us in a world awash with information – will rapidly become central to our lives, whether or not this is visible to us. One of the best single implementations I’ve seen is Last.FM, a personalized online radio station. It builds a profile of your preferences based on your nomination of your favorite artists, albums, tracks, and music labels, as well as what you choose to listen to. When Last.FM is playing on your desktop, you can either let it run if you like what it’s playing, or if you don’t like the song you can press skip to go immediately to the next one, or let the system know you love or hate a particular song if you wish. As it builds an increasingly accurate profile of what you like and don’t like, it can identify other individuals with similar musical taste to you, and play you what they like. In this way you both hear what the music you like, and get to hear new music you like that you wouldn’t have discovered otherwise. There’s an article on Last.FM on Wired News that got it a lot of attention at the time.