Ahhhh, algorithms. It seems like that word follows us around everywhere these days. Of course we hear the word all the time, but what does it REALLY mean?
If you are a Spotify user, you are familiar with the ‘Discover Weekly’ playlists created every week. These playlists of course are curated by the almighty algorithms, and it got me thinking.
What is the music algorithm? How does it exactly work? And how can it be harmful to certain artists when apps work like this?
LET’S GET TECHNICAL
The algorithm starts analyzing your music taste once you get about 30 seconds into whatever you are listening to. Once it collects that data, it will report it back to bots that survey and work on playlists catered to you.
According to an article in Analytics India Mag. :
“Spotify’s algorithm is always finding new ways to understand the kind of music you listen to – from the songs that are always on repeat to the favorite genre that you can’t let go.”
The organization of the Spotify homepage is done with an AI system called Bandits for Recommendations as Treatments, or BaRT. This system is the reason your searches are limited; it helps connect you with artists based on your likes and listens in the past.
Analytics India Mag. also says, “… if one has been on Spotify for some time, it starts recommending songs based on previous listening activities. …BaRT also drops in fresh music that it thinks users will like, to keep them out of the same listening loop.”
This is an amazing way to expose listeners to new artists as well as new songs. The Spotify system has learned your tastes as well as it could.
HOW CAN THIS BE HURTFUL?
Disclaimer: Research done is not limited to Spotify as a streaming platform.
In an article written on the Grammy website, they explain how streaming platforms are hurtful to many female artists. “Music streaming platforms … not only do the popular playlists favor male artists, [but] research [also] shows their algorithms do as well, consequently promoting a continued gender imbalance.”
Unless a female artist is streamed to a global level repeatedly, the chance of seeing a female artist in the top 5 songs is low. This data comes from Pop! Inquirer.
Gender inequality is rampant. It can be seen in tons of the genre playlists on the Spotify platform. The Grammy article points out that while there is a multitude of female rappers that are dominating the scene, the ‘RapCaviar’ playlist will still hold predominantly male artists.
The article says:
As Spotify explained, they choose music experts and cultural ambassadors to curate many of their playlists. As people continue to make gender-skewed playlists, algorithms will follow their lead. In Mokoena’s words; “That’s not just a coincidence – algorithms reflect the culture in which they’re developed.”
This is where it can harm. Limiting female artists in their playlists, narrows the ability to be recognized. This recognition could later lead to performances in festivals and such.
“Inevitably, Spotify charts and data from other streaming services; and social networks play into which artists get booked at festivals, which are slowly starting to address gender inequality.”
SO WHAT CAN WE DO?
In the end, we listen to the music we listen to because we like it. The algorithms only help us branch out to connect with new artists or songs that we may enjoy. The way we consume music is shaped a lot by our cultural biases and what the majority streams.
To make a change, we have to start advocating more for female artists and celebrating them as performers and artists. No different to how we have always done for our favorite male artists.
Interesting topic! The algorithm bias is an interesting issue; it isn’t necessarily that biases against women are programmed into the algorithm, but rather the algorithm learns those biases from users. There are lots of consequences to “algorithmication”, and amplifying the biases of people who use the algorithm is an issue we’ll have to deal with if we want to have a more equitable music industry.