“Specialisation is the enemy of innovation,” says a Wellington-born founding member of the Massachusetts Institute of Technology’s Media Lab.
The Media Lab is founded on the principle of mixing people from different disciplines, says Barry Vercoe, whose doctorate is in musical composition. He and his students at the lab are exploring various aspects of the digital storage, processing and expression of sound.
Vercoe made central contributions to the MPEG-4 audio compression standard and predicts, in an aside to an address at Media Lab South Pacific last week, that the recording industry will have even less hope of keeping a lid on online copying when compression techniques with a 100:1 ratio rather than MP3s 10:1 are routinely used.
But every one of the 35 faculty members of the lab has a radically different area of specialisation, he says.
“In an atmosphere like that, the only way a student can survive is to learn lateral thinking.”
Students come into the lab to pursue one course of research and may end up doing something completely different, or more likely exploring a related field influenced by someone else’s interests. “That clash [between disciplines] gives rise to innovation.”
He confirmed he would be talking to government representatives involved in the Growth and Innovation project. Industry NZ is a sponsor of his visit.
The MIT lab has a wealth of industrial sponsors that are free to observe the work and to commercialise any of the results on the basis of a royalty-free non-exclusive licence. If such a commercialisation looks like flying they will probably employ the research student concerned. Some companies, Vercoe suspects, have inveigled students into the lab specifically to report on anything promising that others are doing.
With such a ready means of finance and technology transfer “we don’t have to spend time writing proposals”, an activity which consumes too much of researchers’ time.
Vercoe gave an account of some of the broad range of projects the lab’s students had pursued. One line of research is to enable a computer to sense the emotional state of its user, through such devices as a pressure-sensitive mouse and chair, and modify its replies accordingly.
At present, it is impossible to relate to a computer as one would to a work colleague, he says; unless you have a colleague who barges into your office (or Office) unasked, second-guesses what you’re doing, offers unwanted advice, misunderstands your replies and finally agrees to leave “but insists on winking and doing a little dance first.” Ironically, during a subsequent demonstration, Microsoft’s Clippy put in an appearance.
A student in Vercoe’s own audio-influenced part of the lab has developed “spotlight sound”, a technique using an ultrasonic wave to “carry” audio frequencies in a narrow beam so only people in the beam and at a certain distance from the source can hear the sound. This has naturally been taken up by audio companies and by Daimler-Chrysler to install an exclusive sound source for each of four occupants of the same car.
Vercoe himself has developed technology allowing a musical accompaniment to follow a singer’s varying tempo and hence, he confesses, has played a crucial role in the latest generation of karaoke machines.