Having an all-you-can-eat music service like Apple Music is fairly incompatible with how music is traditionally released. Releasing singles and then eventually full albums and then sometimes deluxe albums and then even remix albums causing a lot of confusion. I end up with tons of duplicates in my library which then skews the “randomness” of shuffle. Here’s the scenario:
- Hear a new song that you like and add it to you library. This song comes from the single.
- Listen to “radio” streams or curated playlists and hear the same song. That song is being sourced from album. You add it to your library again because you don’t really know if it’s in your library yet or not.
- Listen to your library in shuffle and keep hearing the same songs over and over again.
- Realize you’ve added five versions of the same song – several are identical versions.
The multiple versions of songs also creates confusion when individual albums are pulled from the service by the studio for whatever reason. I’m assuming most reasons relate to contract negotiations. I also hate when songs are pulled and then are no longer playable then we they do get added back they’re not always associated with what’s in your library.
I’m not sure what the answer is to solve this. Maybe a machine learning algorithm to help reduce the duplicates by predicting matches better (“this song is already in your library”)?