

As one might expect, the rock zone (red) separates metal (gray) and pop(green). Reggae and ska (pink) exist in a little data peninsulaequidistant from pop and rock. You can search for bands on the map to see where they fall, or enter any Last.fm user name to see where their favorite artists are on the map.
There’s a great deal of data represented here and no ability to zoom, making the map more a curiosity than a useful tool for music discovery. But it sure is pretty.
http://sixdegrees.hu/last.fm/interactive_map.html
http://www.wired.com/listening_post/2008/11/interactive-map/

He proposes incorporating the passion index into a recommender filter. Make Genius Playlists smarter still. I’m guessing it will happen.
But more, in this week following the $1.92 million jury award in the retrial of Jammie Thomas-Rasset for the illegal downloading of 24 songs (see here, here, here and here), I’m remembering how the big media companies fought the VCR. Then, in testimony before the House Judiciary Committee, Jack Valenti famously said:
The k-means clustering algorithm is a straightforward technique that attempts to find a classification of the vectors, putting them in clusters of users that are similar in their musical preferences. Their definition to ending up in the same cluster, is that they are all closest to their cluster's centre point (with respect to Euclidian distance). When the number of clusters is set to 5, we get a clear separation of sub populations in Last.fm. Below is a depiction of the clusters, where each colour denotes a cluster. It is clear that the clustering algorithm found "indie", "rock" and "metal" to be three significant sub populations of Last.fm users.
Clustering of data into 5 clusters (principal components 1 and 2)
One of the trickier questions I've been trying to visualize is how long pop songs are staying on the charts relative to the past. Are they staying on the charts longer than in the past?
In the chart below, I plotted the total number of weeks charted for all 23,924 songs that appeared on the Billboard Hot 100 from 1957 to earlier this year. (In other words, a little dot on the "60" line means there was a song released that week that stayed on the Hot 100 chart for 60 weeks.)

See the heavy dropoff on the 20th week starting in 1991? In an attempt to increase diversity and promote newer artists and songs, Billboard changed their methodology, removing tracks that had been on the Hot 100 for twenty consecutive weeks and slipped below the 50th position. These songs, called "recurrents," were then moved to their own chart in 1991, the Hot 100 Recurrent.
Unfortunately, this shift makes it much harder to compare the last 15 years to the decades before it. In the chart below, I've isolated the effect by only showing songs that reached the top 50.

A couple interesting observations... Looking at the very bottom of the chart, you can see that in the last couple years, it's become very common for a single to appear in the Top 50 and fall out of the Hot 100 within four weeks. Prior to the mid-1990s, this almost never happened.
Also, songs are staying in the Top 50 for far longer than they used to. Unfortunately, I don't have any actual sales numbers to compare to, so it's hard to say if these 30-70 week singles are massive megahits eclipsing the #1 singles of the past, or if it's because the record industry is producing fewer hits than before.
Diversity
Did Billboard's methodology changes in 1991 make the charts more diverse, like they hoped? By looking at the total number of unique songs that have charted yearly, it's clear their changes did nothing to slow the decline.

According to Billboard, the late 1960s were the peak of musical diversity in popular music, with 743 different songs appearing on the 1966 Billboard Top 100. It's fallen consistently since, hitting an all-time low in 2002 with only 295 songs. Since then, it's improved only slightly, with 351 unique songs appearing on last year's Top 100.
One Hit Wonders
I've always thought the 1970s were the decade of the one-hit wonder, but now I have the data to see for sure.

http://waxy.org/2008/05/the_whitburn_project_onehit_wonders_and_pop_longevity/


What about blogs? It’s not unlike Pirate Bay. Everything’s free.
EO: But it’s not like Pirate Bay. You totally get me wrong. Because when someone has a blog while saying, “Here is the music I like,” that’s nice because she’s talking about us and putting one song up, not the entire album. It’s kind of cool. I feel happy when I see that. With Pirate Bay, there is no announcement, there is nothing; you just search for what you want, instantly you get it for free, the entire thing. It’s not selected, it’s not personal, it’s just, “Here it is.”
So what I was saying in the MySpace blog was that if someone should give it away for free, it should be us. If we decide to put our album on our MySpace page the for free, that should be our decision. So we are the ones who are noble. We are the ones who are Robin Hood. “Hey, it’s for free! It’s fine! You get it from us because we love you, and it’s free!” But someone else does that.


(It's often claimed that live revenues are only/mostly benefitting so-called 'heritage acts'. Unfortunately, the data doesn't shed any light on this because live revenues are not broken down by type of act, gig size or ticket price.)..
It's interesting too that, overall, industry revenues have grown in the period - though admittedly not by much - which arguably adds strength to the notion that, when the BPI releases its annual report claiming how much 'the music industry' has suffered from the growth in illegal file-sharing, what it perhaps should be saying is how much the record labels have suffered.
http://www.boingboing.net/2009/11/13/labels-may-be-losing.html
(It's often claimed that live revenues are only/mostly benefitting so-called 'heritage acts'. Unfortunately, the data doesn't shed any light on this because live revenues are not broken down by type of act, gig size or ticket price.)..

Researchers used a set of ratings from the LAUNCHcast team at Yahoo! to map artists, connecting those who were rated similarly by a large number of users.
“In the central-right portion of the image is a ‘diffuse’ set of points with some strong connections,” creator David Gleich said. “This region represents ‘mainstream’ music and includes many popular artists. Immediately adjacent is a set of points that represents ‘indie’ music. Thus, ‘indie’ music is not as independent as some might like to think.”
It looks like you're new here. If you want to get involved, click one of these buttons!