Elegance and the Heart of Technology
By David Gelernter
Basic Books 166pp $20
Americans love technology, but they feel a little queasy about beauty. That's a real problem, according to David Gelernter, the Yale University computer science professor and author of Machine Beauty. "Insisting that beauty is at the heart of science and technology is like ordering wine at lunch, or tacking ruffles to your office furniture--it takes a serious proposition and makes it frilly and frivolous," he writes.
But Gelernter does insist that beauty can and should inform the way we conceive of and build machines. If the author had his way, all computer scientists and software engineers would be rushed off to crash courses in drawing, design, and art history. Give every computer scientist a grounding in Michelangelo and Matisse, Gelernter believes, and software and hardware would be more beautiful and more functional. Thus, aesthetics of the sort endorsed by such modern architects as Mies van der Rohe and Philip Johnson are a main focus of this disorderly but fascinating treatise--a book that also takes up such wide-ranging topics as the future of computing and the shortfalls of American taste, training, and education.
"Machine beauty" entails an inspired mating of "simplicity and power," the author writes, citing as backup some sublime technological feats: the Hoover Dam, the Empire State Building, and the 1938 J3 steam locomotive. Then he turns to one of the most ubiquitous machines of the Modern Age, the computer, which started out as a classic example of sophisticated simplicity: It harnessed the power of zeroes and ones to perform an almost infinite number of calculations and tasks. Gelernter's history of computers is interesting and accessible without being simplistic.
One modern computer achieved a high degree of machine beauty: the original Apple Macintosh. Employing technology developed at Xerox Corp.'s Palo Alto Research Center, the Mac's interface was designed to look and act like a virtual office, with file folders and trash cans and an area in which you could work simultaneously in several different program "windows." It was straightforward and manageable, a happy contrast to rival Microsoft Corp.'s user-hostile DOS. Remember the sorry days when a user command might result only in a computer response of "Abort, retry, fail?" Or the terrifying "Fatal error"?
Yet even before it switched to Windows, Microsoft's ugly system overwhelmed Apple's machine beauty. How Microsoft managed to pull off this feat is hardly a new story--price, advertising, IBM market dominance, and the clones were all involved--but Gelernter believes there's a perplexing aesthetic angle involved here, too. Microsoft's Windows is almost a carbon copy of Apple's system. Did beauty ultimately triumph? Yes and no, Gelernter thinks, asserting that "it is possible" that "right up to the moment people fell in love courtesy of Microsoft with Apple-style beauty, they were repelled by it." One factor behind this "beauty paradox," he feels, is a kind of sexual discrimination: Some Americans, particularly those in business, didn't like Macintoshes because their very simplicity meant they were too girly or "sissy." The IBM PC, on the other hand, was, in the words of one technology columnist, "a man's computer designed by men for men."
Other factors as well may discourage an appreciation of machine beauty. Their very success has convinced many people that science and technology, as opposed to the arts, "are the only intellectual attainments that count." Then, too, there are America's schools and colleges, which have ceased to offer a well-rounded education.
Gelernter complains that software makers are always adding useless bells and whistles, including clocks that announce they are on "U.S./Eastern" time, as if the user didn't know where he was. And since the chances that software will have bugs increase with its complexity, complicated software is often neither beautiful nor user-friendly. As an alternative, Gelernter promotes a project he and his graduate students have been developing called Lifestreams, which would provide a multimedia archive of a person's life that one can tap into just about anytime, anywhere.
Nor is Gelernter pleased with current computer hardware. "Almost every office in the country centers on an electronic Model T," he says, calling current desktops "graceless, lumpy objects." As substitutes, he offers his own sketches of computers "made of warm orange wood, cherry or mahogany."
Gelernter is an engaging and amusing writer, even when he turns his hand to such a potentially dry topic as computer science. Unfortunately, there are times when his narrative wanders aimlessly. At other moments, the book seems to cry out for a red-pencil-wielding editor. In a section attacking the notion that science is merely a social construct and not "a strong, hard, objective business," Gelernter writes that "any child...can see that the wavelength of blue light is the same in New York and Nigeria...that sine waves were shaped like sine waves yesterday afternoon and ten centuries ago." Any child? Any reader can see that such matters aren't child's play.
Despite its quirkiness and perambulations, Machine Beauty is an interesting, informative, and quick read. Who knows, Gelernter just might persuade some designer to put a little more Matisse in his next mouse--and we would all be the better for that.