Data Can Be Fun and Musical

This is a guest post from Richie Jay.

I was talking to a friend last night about data. Specifically about how modern technology makes it so much easier for us (and, for better or worse, other people and companies) to collect tremendous quantities of data about each of us.

Most of that data is boring. Or necessarily private in the hyper-connected world (birth date, social security number, income, credit report, etc). But some of it can be quite fun. And very revealing if shared.

In Summer 2004, in the early days of iTunes and Internet 2.0, I started using a little software plug-in called Audioscrobbler (later renamed to Last.fm and purchased by a giant media conglomerate), whose sole purpose was to keep track of the music that I listened to (to 'scrobble' it, in Audioscrobbler parlance), and to use that information to provide music recommendations. Initially, it only worked with iTunes, but later it incorporated many streaming sites and portable players (although Pandora, Songza and some other sites that I frequently use are not supported or require creative work-arounds).

Long story short, 8 years later I'm just a few songs short of 17,000 individual pieces of data about my listening habits. This is actually only a partial sample of my total music consumption in 8+ years (misses some ways I listen to music at home, online, and on portable devices, plus concerts, radio, PA systems, parties, friends' houses, and other places I hear music; occasionally has data errors, like double-counts or missing 'scrobbles'), but probably a pretty good representation of my overall musical preferences.

Obviously, my tastes have changed over the years, and I also can't be blamed for the capricious whims of the oft-used shuffle feature, so this list does not necessarily represent a present-day ranking of my favorite or most-listened-to artists. Furthermore, any time you share this much data about yourself, there's bound to be something 'embarrassing' in there (perhaps, in this case, a total lack of certain genres high on the list, even though I aspire to broaden my musical predilections, and like to think of myself as musically diverse and adventurous).

Beyond that, because the technology now exists (and is often used) to correlate various sets of data, this data set begs a few additional questions: What do my music listening habits say about me? What other information can be gleaned from this data? Can it be used to determine any number of seemingly unrelated (but most likely highly-correlated) things like my age, gender, race, political leanings, religion, education level, income, profession, home state, favorite sports team, favorite ice cream flavor, hobbies, boxers vs. briefs, cats vs. dogs, and/or blood type?

Without further ado, here is the list of the Top 50 Most Listened to Artists by me, Richie Jay, since August 2004, according to Audioscrobbler / Last.FM:

RANK          ARTIST

1		Barenaked Ladies
2		Fountains of Wayne
3		Phish
4		Nickel Creek
5		Chris Thile
6		Ben Folds
7		Sufjan Stevens
8		Ben Folds Five
9		The Decemberists
10		Sean Watkins
11		Self
12		The Beatles
13		Rufus Wainwright
14		Air
15		Counting Crows
16		Fiona Apple
17		Richie Jay
18		The Avett Brothers
19		of Montreal
20		Brad Paisley
21		Fluid Ounces
22		George Winston
23		Glen Phillips
24		Aimee Mann
25		Crooked Still
26		Woo
26		Mutual Admiration Society
28		The Polyphonic Spree
29		Jack Johnson
30		Maroon 5
31		Jamiroquai
32		The Actual Tigers
33		Wilco
34		Beck
34		Radiohead
36		Garth Brooks
36		Gnarls Barkley
38		They Might Be Giants
39		Jack White
40		Billy Joel
41		Eels
41		Me First and the Gimme Gimmes
43		Florence + the Machine
44		Phantom Planet
44		Optiganally Yours
44		Foster the People
47		The Duhks
48		Coldplay
49		The Shins
50		Works Progress Administration