Forget all the fuss about graphical user interfaces (all those of us that quite happily communicate with our computers via a keyboard and wince at the idea of an enormous layer of toddler oriented software that eats up the processing power and memory of the computer and hides what is really going on in the computer, already have) – the future is speech-oriented. That is the message delivered at the Siggraph ’89 show in Boston by the celebrated Nicholas Negroponte, director of the Massachusetts Institute of Technology Media Laboratory. Since Siggrah is a graphics event, he was sticking his head into the lion’s mouth, not only because attendees were presumably sold on silly little pictures of wastepaper baskets, but also because those that are not seduced by graphical user interfaces prefer to confine the remarks they pass to their computers to exhortations like Oh come on, get on with it, you stupid thing. But, as reported by Microbytes Daily, in the future, the primary means of communication with computers will be through speech, not through graphics, Negroponte asserted in his keynote address. He urged the audience be more concerned with how people communicate with computers instead of drawing teapots – a reference to the half solid/half-wireframe kettle in the Siggraph logo, which had been shown rendered with various surfaces in a presentation preceding Negroponte’s talk. Controversially, he claimed that a key reason speech-recognition capabilities have not reached the point where people can have a cosy chat with their computers is that people in the speech-recognition industry are fundamentally not interested in communication. Instead, they are interested in transcription, he suggested, challenging many people in this room… are not interested in communication either. As a result, the computer remains sensorily deprived. Another development in what Negroponte calls the sensory apparatus of computing will be vision capabilities that will enable the computer to receive and interpret visual clues from the user. As of now if we take our hands off the keyboard, the computer doesn’t know if we’re just pausing or leaving for the weekend, he noted – unless we switch it off, which should get the message through. Desktop metaphor doesn’t work Today the desktop metaphor is dominant in the human-computer interface, he agreed – Some people even have the temerity to go to court over it, Negroponte commented – but that’s going to change because the desktop metaphor doesn’t work. Any quantity of data and it starts to fall apart, he insisted. By the year 2000, he is satisfied that the interface will simulate efficient procedures that the user is familiar with. Finding a stored document, for example, won’t involve mousing around, it’ll be more like the current practice of calling one’s secretary through the speakerphone and asking her to please bring in a copy of the needed document. But instead of asking his secretary to find the document, he’ll verbally ask his computer. While the Luddites that believe that the keyboard has a lot of life left in it yet and simply yearn for ever more powerful macros and batch files that can be invoked with a single key and can be chained – anyone know how to put a return into an MS-DOS .BAT file? – his tirade must come as music to the ears of Wang Laboratories Inc – but will the market take the Negroponte fast-talking message to heart soon enough to save Wang’s bacon?
