Edward Tufte, godfather of modern information visualization, speaks of successful interaction design as making the “information the interface”.

Lately, I’ve been doing research on how information and data visualization are beginning to impact interface design. There are loads of examples of great interactive visualizations, that’s not what I mean. I mean places where traditional interfaces are being transformed by the addition of techniques gleaned from data visualization.

In Alexander Galloway’s The Interface Effect, he defines, like I do, interface as an action-oriented processes. In fact, he says they are “autonomous zones of activity…processes that effect a result”. If we look at traditional interactive visualizations (like those made popular by the New York Times), we see some amazing and revealing stuff. However, it’s not an interface. The interface exists to help the users reveal the underlying data. What I’m becoming increasingly interested in is using visualizations to create interfaces that result in action.

To illustrate the difference, consider two examples. Modern music notation originated in the 12th and 13th centuries and was used for Gregorian Chants. Alternatives to it have been vying for dominance nearly since its inception (tablature (where how an instrument should be played is what is visualized), for instance, has been around since about the same time). In the mid-1980s, Stephen Malinowski introduced a new way to visualize music. In his program the Music Animation Machine, different instruments are given different colors. The length of note is indicated by the width of a rectangle and the pitch is (like traditional staff-based music notation) along the vertical axis. However, by visualizing the music in this way, a novice can understand the music better. Just watching videos of the software in action, one can “follow along” with a particular instruments much easier than if just listening to the music.

Here’s a YouTube clip of the Music Animation Machine in action.

This is an excellent example of “traditional” data visualization. Fast forward a few decades to Apple’s GarageBand. In GarageBand, the goal isn’t to inform the user, but to allow them to create music. To give them the ability to “effect a result”. In GarageBand, we see the same visualization (rectangles who’s width indicates length and who’s vertical position indicated pitch and who’s horizontal position indicated time), but it’s applied to creating.

As our interface design becomes increasingly more personal and abstracted away from real-world metaphors, I expect principles of data visualization to begin to seep into interaction design.

Next semester, in IMS261, we’ll be exploring more examples of this and looking more into how the information can become the interface.

You can leave a response, or trackback from your own site.