I’ve often wondered why user interfaces always seem to be always constrained to the two dimensions on your computer screen. Certainly in the business information world this is the case. On my way home today I heard an interview on NPR with Michael Sweet from the Berklee College of Music in Boston that got me thinking. Sweet was talking about composing music for video games, which rely heavily on sound feedback to set the mood for the player. When the player is about to enter the boss monster lair, the tension rises. When the player is walking through an African safari, a symphony plays a melodic tune. This is a new dimension to navigation that could be employed to good effect in information navigation scenarios.
When navigating through piles of information, it’s easy to get lost. Some web sites use cookie trails to help you get back to where you started from, but cookie trails are apparently not very popular with users. Another technique used by information architects is to ensure that there is always a short-cut to the main page so that when you follow a deep link from a search engine you can always head to the surface to establish a new starting point for your navigation. Would it be helpful for each zone within a website to have its own signature melody? It would have to be neither loud nor too distinctive, yet distinctive enough that a regular user could tell the difference when they navigate from the “bill payment” section to the “retirement savings” section of a banking website. When you get to the home page it could play a variation on the company’s jingo that they use for advertising. Or, if it’s a university website, they could play a variation on what the marching band plays at football games. Perhaps if you stayed in one section long enough the sound would taper off to nothing so it doesn’t become annoying.
On an intranet, a specific sound could be played out when the user is looking at a requirements document, and a different sound when the user opens a specification document or a test suite. The sound could be taylored to reflect important information about the document – for instance, if it is important to communicate the document size to the user when the document is opened, it could be at a lower pitch or played at a faster beat when the document is larger. In fact multiple such dimensions could be captured through various tweaks to the sound output. My brother, who is a violinist and conductor of our local symphony, could probably pick apart 6 or 7 different dimensions. An untrained human ear would probably have troubles differentiating between more than 2 or 3. But that’s enough to be useful.
When I worked for a speech recognition company in Cambridge, Massachusetts, we used little snippets of sound to help feed information back to the user so they would be able to more effectively navigate through a call center tree. They called these sound bites “earcons” (as opposed to visual “icons” – little speech recognition joke) and I believe they were pretty effective for power users. First time users found them somewhat irritating because they didn’t realize what they were for.
When I was doing my PhD thesis at Brookhaven National Laboratory, I always enjoyed listening to the accelerator “squawker”. It’s a sound that played each time a new pulse of heavy ions was sent through the accelerator. Each pulse lasted about 2 seconds, and when things were running smoothly it was a constant on… off… on… off throughout the day. Sometimes the accelerator missed a pulse, and the person responsible for monitoring the experiment would check the screens to see if there was a problem. After missing 5 or 6 squawks he or she would call the accelerator staff to see if the delay was going to be long. I can remember walking to the experiment trailer before dawn and hearing the squawker going, and I would know before I even walked in that things were going well. I don’t see why a business or an IT operations center couldn’t use a similar technique to help the operators monitor the facility. Higher sales volume could translate to a faster beat. Higher packet collision rates could translate to a higher pitch. The possibilities are endless! But as in all things IT, that pretty well goes without saying.

I agree that sound is a powerful and yet often neglected information tool in IT. When there’s something wrong, or even just a little off, on my motorcycle, I can tell just by the rhythm or the tone of the exhaust. How often do you recognize when something’s bothering a friend just by their voice on the phone?
However, using sound on web sites or intraweb sites would be problematic. Many companies don’t provide sound cards or speakers with the standard issue desktop machines. And then I also think of the few times that I’ve seen it used in IT, like the system beep that’s supposed to tell you when you’ve hooked something up backwards while assembling a desk top. Apparently I’m supposed to know that a BEEP bip BEEP means that the RAM isn’t seated correctly, and that bip BEEP bip means the power supply isn’t sufficient…
Good points. I think that IT workstations will start to ship with sound cards once they find a good use for them. There’s always after-market sound cards too, of course. And I’m thinking way beyond “BEEP”. How about text-to-speech “Network subsystem 24-0 is unavailable” for explicit warnings and error conditions… Computer and systems manufacturers need to start thinking more like Apple and less like IBM.
[...] one I touched on in an earlier blog, where I proposed that programs that monitor real time data use sound feedback to help grab our attention when something has changed in the data that is coming [...]