(Spotify's) Fresh Finds takes a central component of The Echo Nest's original methodology—its web content crawler and natural language processing technology—to mine music blogs and reviews from sites like Pitchfork and NME and figure out which artists are starting to generate buzz, but don't yet have the listenership to show for it. Using natural language processing, the system analyzes the text of these editorial sources to try and understand the sentiment around new artists. For instance, a blogger might write that a band's "new EP blends an early '90s throwback grunge sound with mid-'80s-style synthesizers and production—and it's the best thing to come out of Detroit in years." If this imaginary act goes on tour and writers in Brooklyn dole out praise of their own, the bots will pick up on it. It helps address an issue some people have voiced early on with Apple Music, that its selections aren't adventurous and it tends to recommend things you already like rather than things you might like.
I have the feeling that Apple Music is closer to getting curation right than Spotify. People respond to recommendations when there's a personal aspect … like when it's a mixtape from a friend (or someone you admire), or a recommendation from that blog writer whose taste is so spot on, or that guy at the counter in the hip record store who is always handing you cool 12" singles. Apple Music's apparent understanding of this might be in part because they have publicly hired tech-savvy musicians to oversee these things, while Spotify seem to be bringing on music-savvy techies.
Are Apple Music's playlists a bit obvious? Sometimes … but I was recently surprised by a dance-oriented playlist focusing on Factory Records that contained songs I'd never heard before (and I thought I was a Factory completist), and a space-rock playlist compiled by a musician I hadn't heard of which turned me on to a few other new artists. Apple Music's playlist recommendations can get a bit uncanny (in a good way) once it gets to 'know' your taste.
Spotify's idea of intensively data-sourced curation is intriguing, and I am sure they are utilizing some amazing innovations bordering on artificial intelligence to try to make it work. But a playlist delivered weekly under the same headline -- that the recipient knows is auto-generated -- is easy to ignore. And the discovery-bot will inevitably get it wrong a few times, throwing in curve-ball songs that are completely outside of the listener's taste-zone. I don't know about you, but something like this is only allowed a few times to get it wrong before I'm not interested.
That said, the emphasis on discovery that the streaming services are embracing makes me hopeful. If Spotify's system does start turning people on to emerging, self-released artists then that's an amazing thing. Likewise, it would be nice to see Apple Music's playlists include more emerging artists. I think having regular (monthly?) playlists from notable tastemakers -- music bloggers, cutting edge musicians, and even non-music types like fiction writers and film directors -- would add to the 'personally curated' touch and increase the chances of discovery.
Apple Music also needs to improve their Pandora-like 'radio' function. In my experience there is zero amount of potential discovery going on there. When I tell Apple Music I want a station that sounds like The Slits, that doesn't mean I want to hear The Slits every other song, and it certainly doesn't mean I want to hear "The Killing Moon" for the dozenth time. Pandora has this problem, too, though not as pronounced as with Apple Music. Maybe this is where Spotify can find an advantage?