[kryten123]

Thought I'd share this article proving what has long been suspected...that modern music is all the same...only louder. The article contains the link to the original techie Nature report for those interested.
 
http://www.gizmag.com/pop-music-trends/23535/

 

[Penarin]

Interesting read, thanks.

 

[grapefruit]

How could science possibly prove something so subjective?  
 
The english language only uses 26 letters over and over again, I guess gizmag would say all books are the same also. 
 
I honestly hate social sciences, they are not sciences, they apply quantities to abstract concepts, they abuse statistics test, and pack it all up into jargon so it sounds legit.  
 
I really dont know why people blast pop music, it's enjoyable.  Sure its tasteless, but it has its value.

 

[qusp]

this is really quite ridiculous.

 

[mikeaj]

Just another dumb website/newspaper/magazine/whatever (doesn't really matter which, for these things) reporting on a recent publication and trolling for hits with a sensationalist title that is probably not at all implied by the actual paper.
 
Oh wait, this thread is over a week old.  I missed it the first time then.

 

[Clarkmc2]

How could science possibly prove something so subjective?  

Not defending the source, it seems suspect in its too simplistic modeling of the problem. But to answer your question, modifying Pandora's algorithm might be a possible place to start.

 

[TMRaven]

This just in.  Scientists prove modern pop music to sound like Dada poetry performances-- only louder.

 

[mikeaj]

Quote:

Not defending the source, it seems suspect in its too simplistic modeling of the problem. But to answer your question, modifying Pandora's algorithm might be a possible place to start.

 
What's simplistic about the modeling?  If you're going to do analysis on large data sets, you need real algorithms that can actually classify and quantify things.
 
If the goal is to make outlandish claims (what the website reporting did, just in the title), you need a lot deeper analysis, maybe value judgments.  If you're investigating trends in music over time (what I'm presuming is the scope of the paper), then those metrics seem pretty appropriate to me.  What else would you propose?  For example, if there are fewer pitch transitions on average, that does say something meaningful about the music.  That's also not necessarily a good or bad thing.  There's a lot of pop music full of gratuitous key signature changes.

 

[PeterCasta]

This is a most recent discovery: Scientists prove hard rock is the same as classical music!!!! Lol

 

[Clarkmc2]

P

What's simplistic about the modeling?  If you're going to do analysis on large data sets, you need real algorithms that can actually classify and quantify things.

All I was trying to say was that the conclusion(s) of the study were sweeping given how few metrics were studied. I have been at this playing and listening thing a long time and have heard musicians who could create an opus on a C cord through rhythmic and dynamic modulation. And works of great complexity that were going nowhere worth hearing. The Pandora algorithm posesses a lot of metrics and actually works, at least for me. If it were to be harnessed for research I would give it a lot more weight than this study. Not to mention that it was created by music experts and not sociologist types, a criticism mentioned earlier I concur with.

I don't consult physicists when I need an engineer or an auto mechanic. If the researchers were experts in music, I will stick to there being too few metrics.

 

[JadeEast]

If anyone is concerned with the kind of data that this is drawn from, the original article in Nature has a link to the "Million Song Dataset" the info was dredged from. Here is a link to the Metadata that is included in each entry: 
 
http://labrosa.ee.columbia.edu/millionsong/pages/example-track-description
 

 

[mikeaj]

Quote:

P
All I was trying to say was that the conclusion(s) of the study were sweeping given how few metrics were studied. I have been at this playing and listening thing a long time and have heard musicians who could create an opus on a C cord through rhythmic and dynamic modulation. And works of great complexity that were going nowhere worth hearing. The Pandora algorithm posesses a lot of metrics and actually works, at least for me. If it were to be harnessed for research I would give it a lot more weight than this study. Not to mention that it was created by music experts and not sociologist types, a criticism mentioned earlier I concur with.
I don't consult physicists when I need an engineer or an auto mechanic. If the researchers were experts in music, I will stick to there being too few metrics.

 
Are experts in music technology, data analysis, etc. not suitable for this?  Would you rather ask musicians for an empirical analysis of musical trends over time?  
 
Just to be clear, these are the conclusions of the actual study (not of the reporting, which is decidedly more sensationalist; the authors are not responsible for any kind of suggestion that pop music is all the same):
 

Beyond the specific outcomes discussed above, we now focus on the evolution of musical discourse. Much of the gathered evidence points towards an important degree of conventionalism, in the sense of blockage or no-evolution, in the creation and production of contemporary western popular music. Thus, from a global perspective, popular music would have no clear trends and show no considerable changes in more than fifty years. Pitch codeword frequencies are found to be always under the same underlying pattern: a power law with the same exponent and fitting parameters. Moreover, frequency-based rankings of pitch codewords are practically identical, and several of the network metrics for pitch, timbre, and loudness remain immutable. Frequency distributions for timbre and loudness also fall under a universal pattern: a power law and a reversed log-normal distribution, respectively. However, these distributions' parameters do substantially change with years. In addition, some metrics for pitch networks clearly show a progression. Thus, beyond the global perspective, we observe a number of trends in the evolution of contemporary popular music. These point towards less variety in pitch transitions, towards a consistent homogenization of the timbral palette, and towards louder and, in the end, potentially poorer volume dynamics.  [emphasis added]

 
I don't see how any of that is unfounded based on the data or overly simplistic.  There is a lot more to music than the characteristics analyzed, but that doesn't mean that this kind of analysis is meaningless.  Anyhow, look at the methodology for yourself.  I mean, this isn't an amazing paper, but I'm not understanding all the criticism.
 
The part about how old music could be altered with modern sensibilities and sound modern was just a suggestion or conjecture, not really a conclusion. And that conjecture is somewhat true in many cases, just in my opinion.
 
 
Somebody else referenced social sciences.  Authors are in research groups that broadly cover areas like artificial intelligence, music technology, machine learning, mathematics...Are these considered as sociologist types?

 

[grapefruit]

If it's not STEM it's social science. And about the homogenienity of the timbral pallete, there are plenty of songs with only one timbral variety, yet they are considered some great pieces (take fur elise or chopin's nocturnes for example). At the end, music is very very very subjective, and it's not up to scientists to define what it is and what its not. 
 
Pop music is a fixture of modern society, and people largely like it. For those who cannot believe or understand that, I'm sorry, deal with it.

 

[anetode]

Measuring the Evolution of Contemporary Western Popular Music

we prove important changes or trends related to the restriction of pitch transitions, the homogenization of the timbral palette, and the growing loudness levels

The growing loudness levels and to a lesser extent the restriction of pitch transitions could be at least partially explained by the loudness war. I'm interested in the purported homogenization of the timbral palette in a time period which saw the birth of electronic music. Then again, rock's guitars and drums are surely ubiquitous.

Looking over the data, it could be that the 60s spike in timbres could be related to some shortcoming of the compilation of the dataset.

How could science possibly prove something so subjective?  

The english language only uses 26 letters over and over again, I guess gizmag would say all books are the same also. 

I honestly hate social sciences, they are not sciences, they apply quantities to abstract concepts, they abuse statistics test, and pack it all up into jargon so it sounds legit.  

I really dont know why people blast pop music, it's enjoyable.  Sure its tasteless, but it has its value.

If it's not STEM it's social science. And about the homogenienity of the timbral pallete, there are plenty of songs with only one timbral variety, yet they are considered some great pieces (take fur elise or chopin's nocturnes for example). At the end, music is very very very subjective, and it's not up to scientists to define what it is and what its not. 

Pop music is a fixture of modern society, and people largely like it. For those who cannot believe or understand that, I'm sorry, deal with it.

Did a social scientist kill your puppy or something? Relax, this study has very little to do with deciphering subjective opinions.

 

[OJNeg]

Quote:

If it's not STEM it's social science. And about the homogenienity of the timbral pallete, there are plenty of songs with only one timbral variety, yet they are considered some great pieces (take fur elise or chopin's nocturnes for example). At the end, music is very very very subjective, and it's not up to scientists to define what it is and what its not. 
 
Pop music is a fixture of modern society, and people largely like it. For those who cannot believe or understand that, I'm sorry, deal with it.

 
I come from an engineering background myself, but I don't know why you're so quick to discredit social sciences. It's not like their writing anything in stone; that's not what these studies are about. A study doesn't have to be 100% conclusive to be credible. Someone had an idea, did some research, and came to a conclusion. In a few years, someone might do another study and disprove it. Or reaffirm it. Science is not about absolutes, it's about throwing things at a wall and seeing what sticks.