Last summer, I quit Spotify and wrote about it in a pretty big titleWhy am I quitting Spotify. “My reason is still reasonable: the software becomes clumsy, the ads are ruthless, and Sabrina Carpenter songs are also inevitable. I want to find a better music streaming service. I was unhappy to report that I rejoined a few weeks ago.
This algorithm made me get it. I don’t just mean get Me, the way the Tiktok algorithm pastes you onto the screen. Spotify’s algorithm makes me feel like an old friend makes me feel strange with me For the yacht rock Or continue to be fascinated French Touch Music from Medium. It took me months to dig out the well-known crates of Apple Music before realizing that Spotify has something that other streaming services don’t have: my 15 years of music listening habits and AI’s software can enhance them.
That’s why algorithms tend to be seen as villains these days. They are the technology behind Tiktok for your pages, which keeps feeding you Strange videos, you can’t stop watchingand Amazon’s recommendations It seems to know what prescription you are taking. At the same time, Facebook’s algorithm has At least ten yearsInstagram’s algorithm feed is Destroy the mental health of the entire generation. You can say that the meaning of the Spotify algorithm is quaint.
Spotify’s algorithm makes me feel like an old friend makes me feel strange with me For the yacht rock.
However, exiting and unquestionably Spotify made me realize something. Although algorithm feeds are the center of how you consume information, you have more control over how these algorithms affect your taste and behavior.
If the algorithm works for you – like Spotify did to me – don’t feel bad about obeying its easy and convenient product.
Music has always been important to me, and over the years it started to feel like I had to play Spotify to find songs I really like. When Spotify Launched in 2011This is basically a huge library of all music, but over the years it has introduced an increasing number of algorithmic suggestions and playlists that they promise to match my taste. Still need work to find good things.
This work now makes Spotify’s algorithm irreplaceable. It has my listening history for ten and a half years, and over the years I have learned its quirks and tinkered with it to suit my needs. I spent months trying to copy this experience on Apple Music, but it was surprised by its algorithmic efforts.
All music streaming algorithms are based on two basic principles: content-based filtering and collaborative filtering. Content-based filtering attempts to determine specific aspects of the song itself, including artists, genres, emotions, etc., to line up for the next song. Collaborative filtering is the suggestion made by others based on someone else’s listening to a song and what they listen to. If two people listen to the same five songs, then there is a good chance they both liked the song. It’s all math, and sometimes exceptions will make you happy.
“Some of the chance you get is a mistake and becomes a virtue.” Glenn McDonald’sSpotify’s pre-data Alchemist and Every noise once,tell me. “So you are surprised, sometimes these surprises are pleasant.”
Not only Spotify’s advice is often pleasant, as it has a lot of data about me. That’s Spotify 675 million people listen to historyits interests may overlap with me in countless different ways. Over the years, I have developed a range of habits that can help me hone these suggestions—such as making playlists, rejecting suggestions that I don’t like, exploring the artist’s catalogue, and even, most importantly, browsing other people’s playlists.
This is what I call listening. While clicking “Discover” once a week on Mondays, it’s easy to tilt and listen to the entire radio show and go to the next playlist, the more effort you put in, the more curated experience you have and the better the algorithm will be next time. At the very least, you will find a playlist that the algorithm has not created.
How to resist algorithm rules
Whether they like it or not, algorithm suggestions won’t appear anywhere. Companies like Spotify are like them because – when they work – algorithms make people fascinated by their products. Companies like Amazon like them because algorithmic advice enables them to guide people’s behavior. The right product advice can lead to someone buying something that they didn’t originally planned to buy. (We all did it.)
This status quo seems dystopian in many ways. Algorithm suggestions became popular decades ago when personalization felt convenient rather than creepy. Netflix should receive a lot of honors for this, because Created this concept It was a late 1990s for you to recommend customized movies. But by the early 2010s, It’s hard to tell the difference Between personalized suggestions and targeted advertising. Now, in fact everything you see online is personalized to some extent The New York Times’ front page arrive List of restaurants in your favorite food delivery app.
When you talk about music at Doordash’s Spotify or burritos restaurant, you may learn to live with it. “The stakes are a little higher when recommending products like Amazon and even higher when recommending content like Facebook.” Meredith Broussardprofessor of data journalism at New York University. “Because it is well known that false information and misinformation are very, very popular, but not good.”
A role algorithm designed to increase engagement plays a role in spreading misinformation is a long topic. For now, I’m just reiterating that you don’t have to lean back and let Facebook, Google, or X drown you the information generated by the algorithm. You can learn more about How these platforms use algorithms And guide them to your strengths.
If you’re tired of it Algorithms on X Give you right-wing propaganda, Try Brucekywhich allows you to select different algorithms for your feed. If Netflix or any other streaming service becomes obsolete, Try to cover up your opinion history Then start over. Spotify provides a list of detailed information How to recommend content about it And how to adjust. Amazon has a tool designed to Improve suggestions. (I tried all these things, including Amazon tools, which is very tedious, but still useful.)
Things get harder on big platforms like Google, Facebook, Instagram, and Tiktok, with their algorithms tending toward the spectrum of the black box. still, Know how algorithms work And playing an active role in making them better for your work can improve the experience on almost any platform. The algorithm is only responsible if you allow it.
In some cases, you might like it when the algorithm is in charge. Although I keep correcting it and guiding it, that’s how I usually feel about Spotify. This is also how I usually feel on Amazon, where I only try to buy the basics. Not long ago, I quit Instagram when I thought the algorithm was too responsible. If I get bored one day, I might try again.
A version of this story is also published in a user-friendly newsletter. Register here So you won’t miss the next one!
