
A research group from the University of Tokyo has put forth a bold concept for the way we could interact with technology called "Invoked Computing." It's seriously outside-of-the-box, in more ways than one: The concept is that we would not learn how to use a technological device, but that device would instead observe us and learn to do what we wanted it to accomplish.
As a rather striking example of an application, Alexis Zerroug from the U. of T.'s Department of Creative Informatics shows how a banana (I swear I'm not making this up) could be used as a cell phone. The idea is that you pick up the banana and put it to your ear, the computer observes what you're doing and recognizes what you're trying to accomplish, then hooks up the phone call and projects the audio using parametric speakers so that it sounds like it's coming out of the banana.
The silliness of the banana aside, this raises the interesting point that a lot of our current technological objects are simply middlemen, go-betweens serving as an occasionally unwieldy bridge between us and a desired result. It reminds me of the design school maxim about how people want toast, not a toaster. Watch it for yourself and see what you think:
Comments
The larger the universe of things we could want to do, the greater the amount of effort necessary to identify our actual wishes. This is why, say, we spend so many years learning a language through which to express ourselves. Interaction with computing devices is similar - it is just another language. Now, the language will almost certainly change as the computing devices change, but the requirement to make a significant effort to express complex requirements will not suddenly disappear (unless we are willing to sacrifice the flexibility or the richness of what we do). Moreover, there is a certain conservation of effort principle at play - generally speaking, the greater the effort that must be spent to learn how to compress information (through abstract language, for instance), the the less the effort that must be spent transmitting information later. This is why people who speak the same language interact much more efficiently than people who do not.
aepxc - When I think about trying to get a system to understand my wants and needs, I think about cultural contexts. I also think about language, and how hard it can be to communicate with someone who doesn't speak your language.
We degenerate to ape-like miming of our wants and needs. I doubt those willing to spend money on the latest interactive hardware will want to gesture that they need to go to the bathroom.
So if this were invoked social computing, the banana would ring and your mom would be on the other end ;-)
Wild!