More than a feeling.

Pretty much all the machine learning recommendation engines that emerged in the Netflix era were doomed to collapse under their own weight over time for non-mainstream users because the some limited number of mainstream modes dominate as most statistically "optimal" across the total user pool. These algorithms are best in the early days, when they're still exploring the content space for good novel fits but eventually get trapped into deep, boring grooves that work really well for tons of non-discriminating users with similar tastes.

Separately, in real commercial terms, they're all fundamentally poisoned by business model objectives of highlighting cheap content or servicing partnership/advertising deals, etc. And that problem also becomes more and more prominent as the companies running them grow and become more influential and as they need to squeeze harder and harder for revenue growth.

It was basically just a long, winding, wildly expensive road back to broadcast radio programming.

It was a good run for a while, but we're long due for a new model.

The sad thing is that before Spotify bought the Echo Nest[1], they had hosted some of the coolest discovery demos for non-mainstream (in my case ambient/IDM) where you would feed it a youtube video URL and it would make a really compelling radio playlist based off it. i found so many artists i still listen to today by just sticking a video in there in the morning and coming back to the tab when something incredible popped up.

When Spotify bought TEN i considered moving my listening over, but the radio button we ended up with in Spotify and Youtube Music are huge disappointments in comparison, so corporatist and flattened to 1.5 dimensions, I always wondered how the magic was lost.

Bandcamp's feed (especially once you trick the UI in to showing you how to follow tags) is usually interesting to leave running but limited in its own way by the artist pool lacking mainstream tentpoles to jump off of.

[1]: https://en.wikipedia.org/wiki/The_Echo_Nest

> over time for non-mainstream users because the some limited number of mainstream modes dominate as most statistically "optimal" across the total user pool

This isn’t true, YouTube recommendations when it chooses music are amazing (no idea if YouTube Music is good I mean the video site).

Spotify recs are intentionally recommending you things cheap to stream or that have been paid for. It’s not a raw rec engine and it’s not bad cos it’s collapsing under normies, YouTube is proof of that.

Absolutely, you're hitting the same conclusions I've reached. The algorithms are optimized for the lowest friction users that just replay the same music they like over and over again and accept whatever the popular music is. If you're a user that likes music discovery you're fighting against the system to get what you want.