Algorithmic Entropy: Difference between revisions

From FizzWiki
Jump to navigation Jump to search
← Older edit
Lattaio Meimnagaan JS (talk | contribs)
93 edits
VisualWikitext
No edit summary
mNo edit summary
 
(One intermediate revision by one other user not shown)
Line 1: Line 1:
'''Algorithmic Entropy''' is the common phenomenon of media recommendation/suggestions/autoplay/mix algorithms undergoing drift into lower common denominator media/music overtime, largely due to the way that recommendation algorithms handle "risk taking" (being weighted towards being careful not to rec you something you might not like) and the way "shared audience" is a core vector in the decision process. Algorithmic Entropy can be particularly annoying if you are a musician that just found a cool track and wish to find more stuff like it, such as being a [[Torus of Fizzosophy|Fizz]] musician and letting your playlist autoplay after Fizz vol. 1 or [[Telefuzz]] vol.  1, and before you know it you're listening to popular Vulfpeck songs from 2019.  
Have you ever found a cool song or artist whilst using a music streaming service e.i Spotify, YoutubeMusic, Tidal etc. and then hit the autoplay or mix feature to find more stuff like it, only to within seven or so songs been served a stream of [[Fizz vs. Jazz|barely sonically related]], increasingly popular by view count tracks? What you are experiencing, typically frustratingly, is '''algorithmic entropy.''' This phenomenon is the result of recommendation algorithms tending to opt for low risk vectors, such as '''shared audience''', so as to lessen the probability that you will '''dislike''' a track being automatically recommended to you. This ironically for a '''based''' music head (such as a fizz musician) tends to result in a rapid '''de-based-ening ('''or at least '''de-sonically-comparible-ening)''' of your incoming music. You find '''blowfish snorkeling,''' and before you know it you're getting jump-scared by '''Dean Town at Madison Square Garden''' or '''Night in Tenusia (in 13/8)''' for the hundred millionth time.


"Dude the algorithmic entropy on Youtube Music is so hilariously bad, I was just listening to some cool girl indie electronica music and after only like five songs it was just a sea of 6-2-5-1-strummed-on-a-guitar-with-a-soft-voice-core"
What the solution for this problem is is unknown at present time, other than just hoping that the algorithms improve overtime and that streaming services start letting you tailor what the incoming recs are opting for (and doing so without also using the tech in a way that facilitates the '''[[The Great Slop Reset|Great Slop Reset]]).'''


"yeah it makes sense you got algorithmic entropy after that song since you are listening for the genius production while most people just came for the vocals"
"dude the algorithmic entropy on YouTubeMusic is insane. I found this really cool video game music inspired EDM track and within 8 tracks it was just playing harmonic minor festival slop"
 
"Yeah that's not just algorithmic entropy. That indie girl song you autoplayed from is just a really unique [[Utility Monster Track|utility monster track]]"  


=== The Slay Queen Problem ===
One of the most brutal examples of algorithmic entropy is commonly experienced by fans of "elevated" dance/synth/electro/club/DnB pop music, which many musicians love. Many of these artists, such as Pink Pantheress or charliexcx, have an undeniable "slay queen" aesthetic to them. It's great. It's part of their vibe and aesthetic and it serves their style well, and it makes total sense that the girlies and the gays love these artists, but this slay queen aesthetic is shared among many female pop artists of greatly varying subgenres e.i Pink Pantheress, charliexcx, Doja Cat, Zara Larsson, Upsahl, Ariana Grande, Sabrina Carpenter, Chappell Roan, etc, which results in all of them being put into one shared basked of listeners who derive their music taste from that shared slay culture. This results in particularly harsh algorithmic drift for listeners that were hoping to discover more songs that share sonic and structural vectors rather than a shared cultural aesthetic.
[[Category:Music Theory]]
[[Category:Music Theory]]

Latest revision as of 05:44, 16 July 2026

Have you ever found a cool song or artist whilst using a music streaming service e.i Spotify, YoutubeMusic, Tidal etc. and then hit the autoplay or mix feature to find more stuff like it, only to within seven or so songs been served a stream of barely sonically related, increasingly popular by view count tracks? What you are experiencing, typically frustratingly, is algorithmic entropy. This phenomenon is the result of recommendation algorithms tending to opt for low risk vectors, such as shared audience, so as to lessen the probability that you will dislike a track being automatically recommended to you. This ironically for a based music head (such as a fizz musician) tends to result in a rapid de-based-ening (or at least de-sonically-comparible-ening) of your incoming music. You find blowfish snorkeling, and before you know it you're getting jump-scared by Dean Town at Madison Square Garden or Night in Tenusia (in 13/8) for the hundred millionth time.

What the solution for this problem is is unknown at present time, other than just hoping that the algorithms improve overtime and that streaming services start letting you tailor what the incoming recs are opting for (and doing so without also using the tech in a way that facilitates the Great Slop Reset).

"dude the algorithmic entropy on YouTubeMusic is insane. I found this really cool video game music inspired EDM track and within 8 tracks it was just playing harmonic minor festival slop"

"Yeah that's not just algorithmic entropy. That indie girl song you autoplayed from is just a really unique utility monster track"

Retrieved from "http://fizzwiki.com/index.php?title=Algorithmic_Entropy&oldid=1456"