I listen to a LOT of old music from 1910s to present. There seems to be a cyclic nature to complexity in pop music. From most lines repeated, to none, and back. Sure technology has always been a huge driving factor, but human nature stays the same.
Because analysis I've seen points to reduced complexity over the years, not to cycles or "ups and downs". There were always repetitive songs (lyrics and music wise), but they get increasingly larger percentage of top music. Can check the top 20 of any decade between 1910 and 2020 to compare.
Or check the article's "Repetition of Popular Music, by Year" plot (it covers from the 60s to today).
>Sure technology has always been a huge driving factor, but human nature stays the same.
Human culture doesn't stay the same though. Nor all aspects of cultural change are cyclical.
Another type of repetition in pop music that may not be surfaced with this analysis is the structural variation in the lyrics. Maybe it's because pop songs that are entirely 1-3 syllable lines personally irk me.
Your definition of EDM does not match mine. Who are those modern equivalent of Fatboy Slim or Chemical Brothers that have 7 songs in the top 10? Or do you mean more like Aphex Twin?
If by EDM you mean music produced on a computer, that's even less of a genre than rock is.
I feel like there are many definitions of EDM out there. GP probably means electronic music in general - to me EDM is a genre coming out of the US since maybe 2006, heavily influenced by pop, French house and UK dubstep - basically a watered down and more palatable version of underground styles of EU electronic music. But we can ask 10 people and get probably a dozen different definitions.
Searching for dance music, it seems to bring similar results as EDM usually, I haven't heard the term "dance music" though. I thought it might include maybe things like hispanic dance or otherwise, but it too seems to be mostly EDM.
Also I haven't heard of "electronic music" that much and it has different connotation in my head. Maybe I have been brainwashed by the US term, although I didn't even know it was a US term. I'm from Europe, but of course US brainwash has reached me, no question about the size of the radius.
However to me the term EDM definitely has a certain meaning and I think it's generally for most, as when I search for EDM I get exactly what I'm looking or feeling for most of the time.
Nonetheless even knowing I might be brainwashed it feels special in terms of the feeling.
My definition of EDM is electronic music. You know, the ones where vocals are more just another (optional) instrument than the main thing driving a song.
But if you want something like Fatboy Slim or Chemical Brothers from '10s (similar level of complexity vs mainstream appeal): Duck Sauce, ZHU, Rufus Du Sol, Tchami, Peggy Gou, maybe some Disclosure. If I'm not limited to just house music, then add Flume and Kaytranada to the mix.
As someone that grew up on this music in '10s, I'm sure I'm gonna view these artists with the same sense of nostalgia a couple of decades from now.
I would disagree with that to an extent for the distribution below "I excluded the 20 most repetitive songs in the dataset from the above chart. When you see what it looks like with them, you'll understand why". I had both distributions on screen so I could compare them, and was staring at them for a good while thinking I was missing something obvious. The animation that actually revealed the difference didn't start until I'd given up and scrolled the diagram most of the way off the screen. Had I scrolled a little faster to the next section (or possibly used page down/space), I would have missed the animation entirely.
I didn't see the animation and was just as confused. I finally shook it off, but it made me more skeptical for the whole article, because I couldn't align with judgment of the author.
Only now seeing your comment, I realised that there was an animation and now things make sense yes.
I think I had scroll at the image + the paragraph below and once I read the paragraph I decided to scroll quite quickly and either to completely new page or I was already focused on the table, because that did seem interesting information.
Thinking back now, it's weird how this one thing made me view the whole thing more skeptically, only to realise it was a miscommunication in a way.
The "Repetition of Popular Music, by Year" graph doesn't work for me. First it's empty, When I scroll past the graph and back, it seems to show data sometimes, but only when it's not in the viewport. As soon as I scroll it into view, it undraws the data and is blank again. I don't even have any blockers installed on this browser.
Beautiful for me = less shitty JavaScript that prevents me from seeing the graphs.
It's also more globally marketable. If you only have words like "hey baby, oh oh oh" then pretty much anyone in the world will be able to understand it, but "that's me in the corner losing my religion" has a high barrier to understanding for non natives.
> I suppose it's more appealing to our TikTok-conditioned minds.
Tik Tok directly influences the charts as well as video views are counted as streams. As a result, songs that are fun or funny to dance to for 20 seconds can quickly end up in the Top 10. Tik Tok also has had a significant indirect influence recently with songs and artists that go viral there finding mainstream success afterward, e.g. Lil Nas X, Doja Cat, and Dreams by Fleetwood Mac.
It's really confusing (and unnecessary) to keep switching between two kinds of percentages: compression ratio and size reduction. At least one of the illustrations seems to be labelled size reduction, but is showing compression ratios.
Maybe they mean 'reduced size %' == 'compression ratio' when they say size reduction, in which case why use both terms?
If we're talking about the top-100, what most people listen to, it's getting more repeatitive lyrics, and more simplified and crude music wise, and with less variety in sound.
Well, the top 100 is the literal lowest common denominator. It's music for people who don't like music, who make up the majority of the population. It's expected that it's shit.
Yeah, like the top 10 programming languages are the lowest common denominator. It's basically programming for the people who don't like programming. The rest of us use lisps like a real programmer.
Programming languages are not commissioned and designed by mega conglomerates for maximum likeability and to sell conference tickets, tshirts and merchandise.
We're getting there, some of the languages that at least been commissioned and designed by mega corporations, albeit for fuzzy reasons:
Go (Google), Swift (Apple), Kotlin (JetBrains), TypeScript (Microsoft) and Dart (Google)
One could of course argue that these were created to solve problem X or for reason Y instead of just raising the profile of the company who created it.
>Well, the top 100 is the literal lowest common denominator.
It's also the widest and more representative index of the tastes of each era/generation.
That's despite whether there are "true Scotsman/real music lover" listening to 15th century lute music or modern free jazz albums (or the "really good obscure EDM few listen to"). Of course there are those too. That's beside the point.
>It's music for people who don't like music, who make up the majority of the population. It's expected that it's shit.
And yet it was excellent in the past, like the 60s and 70s, and far more complex lyrically and musicallly (outliers aside).
But it's also the reasons why you are listening to the music.
E.g. if I'm working I don't want to focus on the lyrics. And e.g. if I go for a run, I also want to dive deep into my own thoughts instead of thinking or interpreting the song.
I think there's just maybe more activities now where you listen music as an addition rather than the main thing to do, so you want it to be a pleasant and stimulating background vibe of the sort.
In the past, you didn't heave easy to carry on ear or headphones, but now it's so much easier to just listen to everything while doing other activities.
The music of today is built on the music of yesterday and so on and so forth. The top 100 of 1970 has very little variety when compared to the top 100 of 2023. There’s 5 decades of genres that have developed since 1970. Electronic music hadn’t even entered the mainstream! Go and listen to the top 100 of 1970 and then the top 100 of 2023.
You can argue that music in the 1970s was better / unique / less repetitive but to say it was more varied is demonstrably false.
>There’s 5 decades of genres that have developed since 1970
Which is neither here, nor there, because those "5 decades of genres" are not represented in the current top 100 for the most part. Formulaic, cookie cutter, 2024 songwriting and production is.
Music distribution has changed so radically in the last 30 years that I'm not sure it's meaningful to make historical comparisons. The Billboard Hot 100 was once based purely on physical sales, but now incorporates online streaming and radio play.
Charts have always been a fairly poor indicator of what people actually want to listen to because of the huge incentives record companies have to manipulate them, but they're a particularly poor indicator now. The snake is eating its own tail - streaming and radio algorithms heavily rely on chart placement to build playlists, but those charts are determined largely by streaming and radio playlists.
Wait a nanosecond. It is assumed that you do not do this. Besides, these are the top ten, it means what the $population hears the most, especially the young ones: these two entities are sometimes not quite good. Right now a song is playing in the loudspeakers with the following lyrics: Dreaming of enhanced feelings ... words without loss of meaning ... known limits shall give away ... fresh that won't decay ... breathing craft and will. Circa 2005. And of course it is a pop song, even if it never reached the top. Catchy melody, good instrumentation, 4/4, verse verse chorus etc. I know a song about tectonic plates by a very popular guitarist and songwriter, I could summarise it using my mind algorithm: we live in an ocean of fire ¿and how about you? He also wrote a song about what it means to take your pets for a walk. The algorithm's summary could be: why ... walk ... dog.
Back in the 1980s I hated Phil Collins' "Take Me Home" it seemed to be just those words, the title of the song, but I guess I am remembering it worse than it was.
> The songs that reached the top 10 were, on average, more repetitive than the rest in every year from 1960 to 2015!
So looks like they're filling a need. Which makes sense to me; singing along is the most accessible way of participating in making music, and repetitive lyrics make that even easier. As someone who doesn't play an instrument (and likes doing bad karaoke), I definitely appreciate lyrics that are easy to pick up.
52 comments
[ 3.2 ms ] story [ 113 ms ] threadBecause analysis I've seen points to reduced complexity over the years, not to cycles or "ups and downs". There were always repetitive songs (lyrics and music wise), but they get increasingly larger percentage of top music. Can check the top 20 of any decade between 1910 and 2020 to compare.
Or check the article's "Repetition of Popular Music, by Year" plot (it covers from the 60s to today).
>Sure technology has always been a huge driving factor, but human nature stays the same.
Human culture doesn't stay the same though. Nor all aspects of cultural change are cyclical.
Every decade since the '80s has some type of EDM song taking the number one spot, but '10s has like 7 in the top 10.
If by EDM you mean music produced on a computer, that's even less of a genre than rock is.
For dancing purposes, lyrics as in meaning matter less, so it's about what goes with the flow to make you move.
EDM is about pleasant, energetic vibing, moving and energy.
Also I haven't heard of "electronic music" that much and it has different connotation in my head. Maybe I have been brainwashed by the US term, although I didn't even know it was a US term. I'm from Europe, but of course US brainwash has reached me, no question about the size of the radius.
However to me the term EDM definitely has a certain meaning and I think it's generally for most, as when I search for EDM I get exactly what I'm looking or feeling for most of the time.
Nonetheless even knowing I might be brainwashed it feels special in terms of the feeling.
But if you want something like Fatboy Slim or Chemical Brothers from '10s (similar level of complexity vs mainstream appeal): Duck Sauce, ZHU, Rufus Du Sol, Tchami, Peggy Gou, maybe some Disclosure. If I'm not limited to just house music, then add Flume and Kaytranada to the mix.
As someone that grew up on this music in '10s, I'm sure I'm gonna view these artists with the same sense of nostalgia a couple of decades from now.
It indeed appears that we're increasingly leaning towards more repetition, I suppose it's more appealing to our TikTok-conditioned minds.
Absolutely. The visualizations in that song text really add value.
I wish I could use these kinds of visualizations on my programming language benchmarks analysis posts :D.
(Knowing JavaScript and HTML/SVG beforehand certainly won't hurt.)
I would disagree with that to an extent for the distribution below "I excluded the 20 most repetitive songs in the dataset from the above chart. When you see what it looks like with them, you'll understand why". I had both distributions on screen so I could compare them, and was staring at them for a good while thinking I was missing something obvious. The animation that actually revealed the difference didn't start until I'd given up and scrolled the diagram most of the way off the screen. Had I scrolled a little faster to the next section (or possibly used page down/space), I would have missed the animation entirely.
Only now seeing your comment, I realised that there was an animation and now things make sense yes.
I think I had scroll at the image + the paragraph below and once I read the paragraph I decided to scroll quite quickly and either to completely new page or I was already focused on the table, because that did seem interesting information.
Thinking back now, it's weird how this one thing made me view the whole thing more skeptically, only to realise it was a miscommunication in a way.
The "Repetition of Popular Music, by Year" graph doesn't work for me. First it's empty, When I scroll past the graph and back, it seems to show data sometimes, but only when it's not in the viewport. As soon as I scroll it into view, it undraws the data and is blank again. I don't even have any blockers installed on this browser.
Beautiful for me = less shitty JavaScript that prevents me from seeing the graphs.
Tik Tok directly influences the charts as well as video views are counted as streams. As a result, songs that are fun or funny to dance to for 20 seconds can quickly end up in the Top 10. Tik Tok also has had a significant indirect influence recently with songs and artists that go viral there finding mainstream success afterward, e.g. Lil Nas X, Doja Cat, and Dreams by Fleetwood Mac.
Maybe they mean 'reduced size %' == 'compression ratio' when they say size reduction, in which case why use both terms?
Unclear what has happened in the 7 years since.
Microphone check as I choose my route
I'm playing on the road, I've got no fear
The sound from my mouth is a rap you hear
No valley too deep, no mountain too high
Reach the top, touch the sky
They tried to diss me 'cause I sell out
I'm making techno and I am proud
[0] https://en.wikipedia.org/wiki/The_Complexity_of_Songs
At least not yet.
Go (Google), Swift (Apple), Kotlin (JetBrains), TypeScript (Microsoft) and Dart (Google)
One could of course argue that these were created to solve problem X or for reason Y instead of just raising the profile of the company who created it.
It's also the widest and more representative index of the tastes of each era/generation.
That's despite whether there are "true Scotsman/real music lover" listening to 15th century lute music or modern free jazz albums (or the "really good obscure EDM few listen to"). Of course there are those too. That's beside the point.
>It's music for people who don't like music, who make up the majority of the population. It's expected that it's shit.
And yet it was excellent in the past, like the 60s and 70s, and far more complex lyrically and musicallly (outliers aside).
E.g. if I'm working I don't want to focus on the lyrics. And e.g. if I go for a run, I also want to dive deep into my own thoughts instead of thinking or interpreting the song.
I think there's just maybe more activities now where you listen music as an addition rather than the main thing to do, so you want it to be a pleasant and stimulating background vibe of the sort.
In the past, you didn't heave easy to carry on ear or headphones, but now it's so much easier to just listen to everything while doing other activities.
https://www.billboard.com/charts/hot-100/
Compare to the musical and tonal variety of:
https://en.wikipedia.org/wiki/Billboard_Year-End_Hot_100_sin...
Also: https://www.reuters.com/article/2012/07/26/science-music-idU...
You can argue that music in the 1970s was better / unique / less repetitive but to say it was more varied is demonstrably false.
Which is neither here, nor there, because those "5 decades of genres" are not represented in the current top 100 for the most part. Formulaic, cookie cutter, 2024 songwriting and production is.
Charts have always been a fairly poor indicator of what people actually want to listen to because of the huge incentives record companies have to manipulate them, but they're a particularly poor indicator now. The snake is eating its own tail - streaming and radio algorithms heavily rely on chart placement to build playlists, but those charts are determined largely by streaming and radio playlists.
No tell me no, no, no, no, woah, woah, woah, woah
Oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh
Baby, come gimme your lo-lo-lo-lo-lo-lo-lo-lo-woah-woah-woah-woah-woah
You got me like, "woah-woah-woah-woah-woah-woah-woah-woah-woah"
Shawty come gimme your lo-lo-lo-lo-lo-lo-lo-lo-woah-woah-woah-woah-woah, mhmm
A
https://www.youtube.com/watch?v=fikUELeU-S4
So looks like they're filling a need. Which makes sense to me; singing along is the most accessible way of participating in making music, and repetitive lyrics make that even easier. As someone who doesn't play an instrument (and likes doing bad karaoke), I definitely appreciate lyrics that are easy to pick up.