While air pollution is a topic worthy of our attention, the data describing it is not the sexiest set of numbers on the block. But in the hands of NYC-based "programmer, artist, and joy evangelist" Brian Foo, otherwise known as the Data-Driven DJ, official air quality data takes on a whole new appeal.
For his projects, Foo combines data, algorithms, and borrowed sounds in an effort to create new ways to experience data beyond the written and visual forms we're accustomed to.
For Air Play, Foo generated a song in which three years of daily readings of fine particulate matter (PM2.5) in Beijing are set to music. The data was taken directly from the U.S Embassy Beijing Air Quality Monitor website's historical data files.
"The song becomes 'polluted' after many consecutive days of unhealthy readings rather than being affected by any single daily reading. Conversely, the song will 'clear up' after many consecutive days of good or moderate readings," Foo explains. "I decided to create the song in this way because it was more analogous to the impact of pollution on a Beijing resident's health over a longer period of time."
Foo did no actual composing for the song; it is entirely algorithmic and generated by a computer program that he wrote which takes the data and music samples – from Nine Inch Nails in this case – as input; the output is the music, which you can hear (and watch) below.
For all the dirt on Foo's process and tech details, visit Data-Driven DJ. His whole creative process is documented and published there – and his custom software is open-source. Stealing, extending, and remixing are inevitable, welcome, and encouraged, says he.