- - - - - - - - - -

T A B L E__T A L K

Will Netscape's decision to make its source code public improve its chances of beating IE once and for all? Join the discussion in Table Talk's Digital Culture area

- - - - - - - - - -

R E C E N T L Y

Flicks from the underground
By David Hudson
Berlin transforms subway tunnels into movie screens
(01/27/98)

21st Challenge
By Charlie Varon and Jim Rosenau
Haiku error messages.
Plus: Results of Challenge No. 3.
(01/26/98)

Let's Get This Straight
By Scott Rosenberg
Microsoft spins, Netscape liberates, Wired ... synergizes?
(01/23/98)

Let's Get This Straight
By Scott Rosenberg
When it comes to new media, the New Yorker remains hopelessly out of the loop
(01/22/98)

The mayor of cyberville
By Mary Elizabeth Williams
A conversation with Echo founder Stacy Horn
(01/21/98)

- - - - - - - - - -

BROWSE THE
21ST ARCHIVES

- - - - - - - - - -

David Gelernter

argues not only

that it can, but

also that it

must be so.

MACHINE BEAUTY: ELEGANCE AND THE HEART OF TECHNOLOGY
BY DAVID GELERNTER_|_BASIC BOOKS_|_166 PAGES

BY SCOTT ROSENBERG | David Gelernter is a computer scientist who believes, as he told the New York Times a couple of years ago, that software engineers ought to study Keats -- not only for their own enrichment but in order to become better software engineers. In his new book, "Machine Beauty: Elegance and the Heart of Technology," he expounds more fully his view that "great technology is beautiful technology" -- that aesthetics are no mere frosting on the technological cake but rather the basic ingredients, the very flour and sugar and egg that give it substance and value.

Most of us are brought up to view science as a cold, objective realm of verifiable truth -- and art as a kingdom of subjectivity where personal expression and private responses hold sway. While one corps of scholars today leads a charge on the bastions of scientific objectivity, suggesting that scientists and their work are as heavily governed by the circumstances of their times and places as the rest of us, Gelernter takes precisely the opposite tack. In his view, the core principles of beauty -- which, in the field of technology, he defines as "a happy marriage of simplicity and power" -- are as independent of fashion and prejudice as the laws that govern the physical universe. While notions of beauty may not be utterly immutable, they are rooted in essential and persistent objective realities of human perception. Or, as Keats once put it in "Ode on a Grecian Urn," "Beauty is truth, truth beauty."

One does not have to take Gelernter's side in this eternal dispute to appreciate the value of his point of view as it applies to the digital world today. At its best, "Machine Beauty" aims delightful and deserved barbs at the orthodoxies of today's technology design -- particularly, the "feature bloat" that afflicts successful software products as they "mature" through successive upgrades. Gelernter maintains that beauty is a "truth and rightness meter" that is "crucial to software" as "the ultimate defense against complexity" -- and, particularly, against the tendency of software designers to overload their abstract creations with too many awkwardly integrated capabilities that no one really needs. He also suggests that we often resist beauty in technology because we think that "elegance is sissy" -- that beauty is feminine and technology ought to be masculine.

For a good half of "Machine Beauty" -- a slender polemical volume that follows hard on the heels of "Drawing Life," Gelernter's account of surviving the Unabomber's attack -- the author sustains an impassioned, engaging argument: Beauty, he holds, is not only a desirable trait for new technologies but actually a good predictor of what technologies will prosper in the market. Before you can say, "Then why didn't the Macintosh conquer the world?" he's got a novel, trenchant answer, based not on a business analysis but on a kind of psychological profile. The DOS-based personal computer had always been "a man's computer designed by men for men" (the words, quoted by Gelernter, are from a 1984 column by PC Magazine pundit John Dvorak). And so Microsoft was able to beat Apple with Apple's own ideas -- Windows being a minor variation on the Mac desktop -- because Microsoft was "equipped in Nixon-to-China fashion to make elegance legitimate in a way Apple never was."

For examples of beautiful technologies, Gelernter looks not only at today's software world but at computing history, where he praises the pioneering programming language Algol 60, and at older technologies like the grandly Gothic radio consoles of the 1930s. In stubbornly refusing to draw arbitrary lines between aesthetics and engineering, or between art and science, "Machine Beauty" is in the good company of other recent books in this field, like Steven Johnson's "Interface Culture." Programmers talk about "elegance" all the time; today most of us are willing to apprehend beauty wherever it occurs -- in a museum or on a computer screen, in the words of a poem or the patterns of a fractal design.

Unfortunately, Gelernter loses steam as "Machine Beauty" looks into the future of elegant computing. The bulk of the book's second half is devoted to a summary of Gelernter's own Lifestreams project -- an ingenious scheme for organizing all of one's personal files, documents and digital information in chronological "streams." From this fascinating but still fuzzily speculative research outline, Gelernter hopscotches to a rant about the conformist mediocrity afflicting the physical design of today's personal computers; then the book peters out via a peroration on the exquisite glories of 1930s radios. In its own grab-bag design, alas, "Machine Beauty" fails to live up to the standards of elegant simplicity it extols.

Gelernter is also prone to interrupting his arguments with outbursts of ire against the forces of cultural relativism that he sees threatening truth and beauty at every turn. His cantankerousness, which can be brusquely amusing in the technological arena ("You wish software like this came with a rear end so you could kick it"), loses its charm as he rails against "the prostitution of scholarship to politics" or thunders that the current educational system ought to be bulldozed flat. (It's unclear whether that prescription includes Yale, where Gelernter teaches.)

If anyone has the right to be bitter about the more rotten fruits of the academic left, it is surely Gelernter, who lost a hand to the bomb built by Harvard mathematician-turned-Luddite-killer Theodore Kaczynski. But these tirades mar the surface of "Machine Beauty" just as inevitably as the pointless features tacked on to the latest-model software suite wreck its elegance. For sheer focus and exuberance, Gelernter's 1991 book "Mirror Worlds" provides a far better case study of the real-world power of technological beauty in action.
SALON | Jan. 28, 1998