Where My Noise Comes From

Someone sent me a link to Last.fm and I’ve been using it for a while now. It’s a small web radio player that learns what I like by actually paying attention to what I listen to. I start it with an artist or band, it plays something, and I either skip it or save it as a favorite. That’s the whole system. No complicated profiles, no playlists full of filler—just play, react, move on.

What surprises me is how fast it learns. After a couple hours the suggestions stop being random. After a week my own taste starts coming back at me—songs that feel inevitable, like of course that’s the next one. The player just watches what I keep and shapes itself around the pattern.

Most music services are trying to trick you somehow. They want your data, or they’re pushing whatever they got a deal on, or they’re turning you into a brand. Last.fm seems to want none of that. It’s just a player that watches what you’re playing and gets better at guessing what you want next. Anonymous. Free. No upsell.

I still don’t trust recommendation algorithms in general. There’s something creepy about being predicted, about your taste being read before you consciously know it. But this one doesn’t have that angle. It’s more like someone who actually paid attention to what I was listening to and made a suggestion based on that. The difference is subtle but it changes everything.

When I need music and don’t know what to put on, that’s where I go now.