Big Music Meets DataBy Nanther Thangarajah
September 11, 2015
Admit it, we're all data collectors.
We know how many minutes (or hours) our daily commutes take, on average, allowing for a standard deviation of ±1.3% because the subway system can't seem to stick to a schedule. Unscientifically collected, or not, that's data. It's all around us and we've been using it to look back and look forward longer than we care to admit.
Technology, being what it is, has changed the way we collect and interpret this data. Take Polygraph, a site that aspires to "data-driven storytelling." Their "Most Timeless Songs of All-Time" article uses data from Spotify to measure the popularity of music. They go so far as to assert that Blackstreet's "No Diggity" is timeless—using data, of course.
They look at which songs from the 90’s were the most played in 2014 on Spotify and, using a horizontal scale, plot the songs based on popularity with embedded audio samples. I was on this chart a while, just reminiscing and maybe regretting some of my music preferences at the time, just a little bit (I make no apologies for the Spice Girls, though).
Data is nothing without an effective way to present it and reveal the story it wants to tell you. Sometimes there are many stories. And sometimes, there's only one, and it might not even have a happy ending. Case in point, if you scroll further down to look at the other charts and read the accompanying copy, you too will agree that past popularity of any song or artist is no predictor of future popularity. The data doesn't lie.
I love how the varied data is presented and interpreted, and how you can have some control about the parameters, like with the "Present-day Popularity of Five Decades of Music" chart. You can see the whole list, or simply select a particular decade, artist or song. I mean, why isn't the Spice Girls' "Too Much" ranked higher?!?
The charts aren't the only things great about this site. They've embedded audio samples within the copy to provide as full an audio picture as possible—styles, tastes and popularity. The site demands exploring and presents new insight at every turn, uh, click. Finally, they use data from 2013's hits to predict longevity of hits from that year. Time will tell, I suppose.
Music means so much to so many of us, at a level impossible to quantify. But for the data nerds amongst us, sites like Polygraph give us a whole new way to appreciate it.