Data Audiation
Music and Data
I recently attended an awesome carillon concert from the top of UC Berkeley’s Campanile. Then again, every carillon concert is an awesome one. I’m only a little biased.
But this concert was unique from all the other carillon performances I’ve seen — and all other musical performances I’ve experienced — in that the music was crowd-sourced.
Here’s how it worked: For the duration of the concert, members of the audience could access the Polar Tide: Pacific Rim web page on their smart mobile devices, and tap on any of the 7 colored circles in the middle panel of the web page.
Each tap triggers an animation in both the top and middle panels. A unique tone plays simultaneously with the animation. I’ve yet to understand how the pitch of the tone relates to the animation. All I know is that the note, the height of the animated bar in the middle panel, and the height of the colored circle in the top panel, somehow correlates to the level of the tides at certain coastal locations along the rim of the Pacific Ocean. These locations are illustrated in the bottom panel.
As audience members tap on the circles and play the different tones, data for each circle tapped is collected and translated to musical notes by a computer program. The notes are then displayed in real time on a computer screen in front of the carillonist, who then sight-reads the notes and plays them on the carillon.
Get Nicole Fronda’s stories in your inbox
Join Medium for free to get updates from this writer.
It wasn’t the most melodic of musical pieces, but it was an incredible experience to witness this live experiment of data audiation.
Data Audiation is the creation and study of the audial representation of data.
Okay, I kind of made up that term by changing a single word in the Wikipedia definition of Data Visualization.
Data visualization is becoming increasingly popular in understanding big data and the stories data can tell. It’s clean, it’s beautiful, it’s much more interesting than an Excel graph. It’s an art as much as it is a science. There are now jobs and companies devoted to taking data and turning it into pictures or interactive graphics that are digestible, aesthetically pleasing, and clear in the information it’s meant to convey.
But why hasn’t data audiation become a big thing yet? The people behind Polar Tide are musicians as much as they are data scientists. And undoubtedly there are many other data scientists with musical backgrounds who have designed similar experiments with the intent of using data to tell a story in much the same way an infographic would.
So here are my ideas — a list of pros and cons — of data audiation and making data a musical experience.
Cons
1) May not be a pleasant listen.
Data doesn’t often make sense. It would take a considerable effort to turn a large data set into a catchy melody. That’s not the point of course, but wouldn’t you rather listen to Mozart?
2) Only exists in time and not space.
With an infographic, you can take in all the information at once, stare at it for as long as you like. Trying to understand data by listening will certainly require multiple replays of the piece.
3) Still requires some visual representation.
As with Polar Tide: Pacific Rim, there would be no way of understanding what each of the notes meant without looking at the animations.
Pros
1) It’s still a unique and beautiful way of experiencing data that isn’t boring numbers on a spreadsheet.
2) Provides a different medium of expression and storytelling for data scientists who are more musically artistic than visually artistic. And think of all the music obsessed data geeks who would load their playlists with data music. I’d probably be one of them.
3) Provides another appropriate medium for data sets expressed as a function of time. Just think, halfway through a piece, the key changes, or a new motif is introduced, or a group of instruments takes over the melody, representing maybe a drastic change in the pattern of data after a certain point in time.
Since the Polar Tide concert I’ve been trying to find other examples of data translated into music. So please send some my way if you find any! I just find it a cool new way of appreciating data and composing music.







