High frequency listening & algo rythmic training

October 22, 2016 11:00 AM

In April, 2016, music taste profiler Ajay Kalia compiled data on listeners from Spotify and on the popularity of the artists they listened to from the music intelligience and data platform, Echo Nest. The data sets indicated that teens listen almost exclusively to the most popular artists and suggested that as we age into college and out to about age 25, our music tastes evolve rapidly as we discover more and more new, but less “popular” music. After 25, the “un-popularity” of the artists we listen to continues to expand outward, but by age 33, listeners are in a rut. 

Kalia hypothesizes that most listeners over 33 stop discovering new music. Despite many popular online outlets like Fox, The Guardian, Reddit, and NME running the catchy clickable headline as ‘people stop listening to new music,’ his data does not show this. Rather, the data indicates that older music listeners stay in the same “popularity zone. If I listen to artists on Spotify with between 1,000 and 2,000 followers I may be staying in the same popularity zone, but there are thousands of artists at this popularity zone, so I have not necessarily stopped discovering new music. The study was furthered by using Spotify and Echo Nest data. Echo Nest, which was acquired by Spotify in Q1, 2014, compiles mountains of metadata about every song and user. Its algorithms power the suggestions we see, the songs that are added to our online radio stations, and the recommended songs it sends us. 

In addition, it gathers information on how mainstream an artist is; how often you listen to artists that then become popular (discovery), how fresh, how diverse, and how buzzworthy the music is that you listen to. 

These algos then nudge people to “discover” new artists that are similar to ones they already listen to. So, these algo-driven platforms encourage new music discovery while placing a high percentage of that music in the same musical genres and popularity (and comfort) zones of the listener. They dig you deeper into the same hole rather than directing you to a new place to dig. In short, people on Spotify (and Pandora, and other platforms) listen to more music similar to the last piece of music they just listened to. A more accurate conclusion of the research is that music listeners are far more likely to discover new artists within the music genres they appreciate, than they are to experiment with new music genres. 

So why not dig a new hole? Search for me in Spotify (jbvannatta) and follow what I am listening to. This one is on you.

Dinosaur jr.   

Give A Glimpse Of What Yer Not

Give a Glimpse opens with just 1.5 seconds of an amp fuzzing out, ringing in a new song from an old band. Dinosaur Jr. has gone through iterations with and without bassist Lou Barlow (Sebadoh, The Folk Implosion) and drummer Murph (The Lemonheads). This iteration sees the original trio back together and buzzing through eleven melodic guitar-driven songs. Gone are the keyboards that added textures under Mascis’ guitar on previous albums. The more “classic” sounding lineup allows the guitars to pop, the drums to roll, and the bass to drive the songs forward. Murph’s drumming is solid throughout, especially the tight rolls and fills that lend urgency to songs between guitar parts. Closer, Left/Right is a solid tune that relies heavily on bass for the melody and a sense of urgency, but it is only when Mascis arrives with his guitar solo that the song starts to fly. After 3:51, you are left wanting just a bit more -more of Mascis’ biting guitar. Give a Glimpse’s songs are melodic and are every bit as good as Dinosaur Jr. was in the late ‘80s. While the sound is not “fresh,” it is familiar – distorted guitars that avoid ringing power chords in favor of repeated runs, arpeggios and enough feedback to fuel your inner angst. Better put the kids to sleep, get in the car and turn it up loud.

About the Author