Live data from Hacker News

Show HN: Selecta – Tune Your Own Spotify Recommendation Algorithm

selecta.ciaran.co.za

31–32 of 32 posts

Re: Show HN: Selecta – Tune Your Own Spotify Recommendation Algorithm

#32
post #19
post #8

Interesting! Too late for me; Spotify's declining recommendation capability was one of several issues that allowed me to cancel my subscription. I really wish streaming services would just let us give conscious direction instead of trying to hack our brains. They're not as good at as they think.

Did you end up switching to a different platform? If so, which one?

I continue to use YouTube, which is less dynamic (meaning, you can end up on an actually or nearly deterministic playlist starting at a given song, rather than on "random" songs), but really surfaces some incredible stuff. It does take more work to break out of a mold sometimes, but I tend to find more interesting things when I do. Plus every song is on there.

I also began using Tidal and I'm unsure if that's going to be a long-term thing for me. It's good so far but I find I'm not using it as much; I think it will take effort to get my library into its algorithm before it becomes really useful.

One problem with Spotify is it has a very poor memory. It would only focus on things I had recently listened to and almost never accounted for my full library of "liked" songs/albums when giving suggestions (so something like Pink Floyd, despite having many liked songs, would absolutely never appear on a generated playlist). I remember Pandora was even worse at this; I don't know why. Anyway it's too soon yet to tell if Tidal will do the same, but I don't have high hopes. Even YT tends to do this. It's extremely frustrating.

Post reply on HN