| 1 |
| 1 |
| 1 |
| 1 |
 |
| 1 |
 |
|
|
|
| From Contract Professional, May/June 1998 Art and Utility . . . . . . . . . . . . . By David Brittan Machine Beauty: Elegance and the Heart of Technology, by David Gelernter (Basic Books, $20). WHEN DAVID GELERNTER—computer visionary, Yale professor, and Unabomber target—promises to explain how advances in computing are driven by a quest for beauty, he ought to grab the attention of hardware and software designers everywhere. Who wouldn't like to know how to write more beautiful code or build more beautiful machines? Who doesn't want to be at the heart of a grand historic movement to beautify our world? The trouble with Machine Beauty is that it does not deliver on its promise. It's an ambitious book, to be sure, and an amiable one. Gelernter proposes that beauty, which he defines as "a happy marriage of simplicity and power," underlies not only artistic achievements but technological ones as well. The Hoover Dam, Henry Dreyfus's stark 1937 telephone, the Macintosh desktop, and the Java programming language all display simplicity and power in a way that sets off a "resonant hum" in the brain of the beholder. And so does any technology worth building. Unhappily, Gelernter complains, we live in a society that regards beauty as wimpy and effeminate. So real men resist beautiful technologies and embrace ugly ones. "We tolerate junk cheerfully—in the form of commercial software, for example—that hurts our productivity and adds nuisance to our lives," he writes. How do we escape this Philistine existence? First, by opening our hearts to innovative computer interfaces—specifically Lifestreams, an ostensibly beautiful product Gelernter has developed with Yale students, and the promoting of which absorbs some 20 percent of his slim, 144-page text. And second, by sending programmers to art school. "We ought to start teaching Velasquez, Degas, and Matisse to young technologists right now on an emergency basis," he admonishes. A Lovely Idea Beauty as "the driving force behind technology and science" is a lovely idea. Somebody should write a book about it. But to do so would require, at a minimum, a disciplined argument, a welter of telling examples, and several hundred dense pages. A convincing conception of beauty also would come in handy. In Machine Beauty, these ingredients are nowhere to be found. Gelernter's first mistake, probably, is to define beauty. Philosophers have wrestled over that one for millennia with little agreement. His second mistake is to define beauty carelessly. "A happy marriage of simplicity and power" can certainly be present in great art. It also sounds like a reasonable goal for software development. But you have to wonder if that's all there is to beauty. Elaborating, Gelernter explains that power means "the ability to accomplish a wide range of tasks, get a lot done." This clause would rankle the likes of Immanuel Kant and Arthur Schopenhauer, who argued that beauty is something contemplated as an end in itself, not as a vehicle for achieving some practical purpose. A definition of beauty that stresses utility is suspect from the start. Gelernter's definition also seems to deny beauty its sensual, its literally "aesthetic," dimension, which demands that an object be appreciated (or, in the case of literature, imagined) in a way that stimulates the senses. The power and simplicity of a computer program, much as it satisfies the intellect, does not exactly jangle the nerve endings. To be charitable, you might presume that Gelernter has in mind only some specialized form of beauty, a kind that adheres to machines but not to art or nature. There you would be wrong. He assures the reader that when he calls machines beautiful, he means "in the real, literal way that a painting or symphony or rose can be beautiful." But is Machine Beauty really about beauty? The validity of the book's pronouncements on our cultural failings, and of its proposed solutions, hangs on the answer to that question. Utilitarian Elegance To judge from his own usage, Gelernter appears to be talking not so much about beauty as elegance, a term he uses interchangeably with beauty. Elegance, of course, has an aesthetic sense: graceful style. But it also has a technical sense: "scientific precision, neatness, and simplicity," as Merriam Webster's puts it. The latter definition of elegance is remarkably close to Gelernter's own "power and simplicity." And this is exactly how he applies the term. In this utilitarian sense, elegance has little to do with aesthetics and much to do with efficiency—providing the greatest output from a given input, or perhaps the most bang for the buck. So if the gist of Gelernter's complaint is that consumers and producers tolerate inefficient designs, then it may not be necessary to invoke aesthetics to explain what moves or doesn't move at CompUSA. To Gelernter, the triumph of the "ugly" DOS machine over the "beautiful" Macintosh is proof that "machine beauty bothers us." He implies, somewhat fantastically, that the Apple machine has fizzled because it has been marketed more to women, who appreciate beauty, than to men, who don't. But if what we're really talking about is a battle over efficiency, then consumer behavior can be explained by simple economics. People buy the PC not because they are uncultured swine. They buy it because they believe that the PC, always priced lower than the equivalent Mac, delivers more output for the resources (labor and money) they put into it. An argument based on efficiency also clears up a mystery that baffles Gelernter: why consumers who supposedly rejected the Mac interface because of its "beauty" would suddenly flock to an almost identical PC interface in the form of Windows 3.0. Simply put, the price was right. Likewise, programmers, contrary to Gelernter, don't have to be beauty-impaired to turn out inefficient (or "inelegant") software. People may cluck over the swelling of Microsoft Word from 27,000 lines of code in its first release to 2 million lines in 6.0, and to the addition of hundreds of dubious new features. But since computer memories have also grown 100-fold, demand for more efficient word processors is low. So it's likely that programmers create inelegant code for the same reason fast-food restaurants serve lousy hamburgers: because they can. The same cannot be said of authors who create inelegant books. The constraints of Gelernter's chosen genre, the "slender volume," are such that not a word can be wasted without marring the argument. There is no room for lengthy sales pitches whose only apparent justification is to show that "there are alternatives" to the Macintosh desktop. No room for tangential discussions of radio physics. No room, in so small a space, for the mind to wander. Should programmers study art? Yes. Everybody should. But as Machine Beauty unintentionally shows, art is more easily studied than practiced. db |