Silicon Snake Oil: Second Thoughts on the Information Highway.

CLIFFORD Stoll's Silicon Snake Oil is essential reading for anyone who is devoted to the Internet, anyone who has ever gone anywhere near the Internet, or anyone who can spell `computer' without looking it up. It is entertaining, too, as essential books go. Stoll uses plain language and lots of examples to argue that computers and computer networks are ludicrously overvalued in modern America. The hype that surrounds them used to be funny, but no more. The computer culture, Stoll writes, `directly threatens precious parts of our society, including schools, libraries, and social institutions.'

The book is interesting not only for what it says but for what it is: a watershed. Many of the scientists, academics, and software architects who created the computer culture worship two different gods who don't get along. The god of technology knows machines and mathematics and science, doesn't care about politics, and inspires relentless ambition to achieve great things, get rich, or both. The god of unconventionality embodies the carrot-cake sensibilities of the establishment dropout; he is a liberal god, and fond of organic vegetables. The computer culture's divided loyalties account for the unmistakable yet hard-to-place oddness of its official organs -- Wired magazine, for example, or the electronic bulletin boards you find online. What is that strange aroma? Tuna fish Wellington? Cr - pes Suzette with salami?

It was inevitable that one day a computer wizard would arise from his keyboard with the medial carrot-cake centers of his brain (near the hypothalamus) in open revolt: `What am I doing this for? Computers aren't organic. This whole thing is a fraud.' Clifford Stoll isn't merely any old computer wizard, of course. He is a distinguished astronomer, author, and technology expert. But his book speaks for the wounded sensibilities of many of the thoughtful people who made this revolution, and it speaks with intelligence, distinction, and style.

`Life in the real world,' Stoll has discovered, `is far more interesting, far more important, far richer, than anything you'll ever find on a computer screen.' Log onto a computer network and you will encounter a `hollow world, devoid of warmth and human kindness,' a `weak substitute for dinner with my neighbors.' And the online world is `hardly a tolerant and accepting neighborhood' either.

It's hard to say what Stoll's overarching thesis is. Why have things gone so wrong? What fundamental problems or basic categories underlie the endless examples of computer-mediated stupidity he presents? How could intelligent and creative technologists have made such a mess? No response. Stoll has no patience for big theses or big pictures. His `conclusions' chapter covers one page and finishes with a rousing statement of principle: `I smell popcorn in the kitchen. I'm done meditating.'

The book's strength is in the details. Stoll has drawn up a plain-spoken indictment of computers and networks on perhaps 17,000 counts of being asinine. The indictment names names, gives examples, and is devastating.

On a Vermeer painting he can admire onscreen courtesy of a Microsoft CD-ROM disk: `a picture postcard shows more detail.' (Better color, too.) On the computer's many accomplishments in the field of graphics: `I've never seen a memorable computer-generated chart.' Of the computer's vaunted ability to deliver massive quantities of data: `Computers hide mistakes in logic while sanctifying information with an aura of truth.' On the design of computer hardware (for example, his modem with its row of lights marked CTS, DTR, CD): `I'm furious that engineers routinely use cryptic labels that mystify the technology and deny access to ordinary people.' (And to emphasize his contempt for the cretins who use these things, the typical modem engineer always locates the on - off switch amid the tangle of wires in back.) On the design of computer software (for example, the meretricious `new features' of the latest version of Microsoft's Word): `I'm happy to sacrifice functionality for simplicity, reliability, and especially, ease of learning.' Any software designer who fails to memorize that sentence by next Tuesday should have it laundry-marked by court order on his forehead.

Classroom computers, Stoll notes, are expensive, grow obsolete fast, are prone to be stolen, break down in ways teachers can't fix, compete with the teacher when they are deployed at each student's desk, and can't be seen clearly when they are not. `I'm surprised at how few scoff at these glitzy and ineffective toys.' The best paragraph in the whole book deals with this topic: `I see a parallel between the goals of Sesame Street and those of children's computing. Both are pervasive, expensive, and encourage children to sit still. Both display animated cartoons, gaudy numbers, and weird, random noises. . . . Both give the sensation that by merely watching a screen, you can acquire information without working and without discipline.' And by the way, computers `aren't necessary for most college studies' either. So there. Amen.

The indictment is set down in crunchy popcorn prose that is easy to take, pleasantly addictive, always informal, mildly monotonous. His editor shoulda told him to take out the gross colloquialisms 'n' awkward contractions (`Hey -- ', `Naw -- ', `It'd,' `it'll,' `who'll'). Stoll's habit of telling (often irrelevant) anecdotes about his friends at the drop of a hat is charming. He will do well as a grandfather.

Stoll shuffles casually from topic to topic, like a prophet in bedroom slippers who would be sharper, tighter, and better organized after a cup of coffee. Occasionally he gets carried away and the indictment goes overboard. Stoll scoffs at a St. John's College plan to teach a computer language in addition to Attic Greek. `Classical Greek lent itself to the promulgation of a rich culture, indeed, to Western civilization. Computer languages bring us doorbells that chime with 32 tunes, alt.sex.bestiality [an especially scintillating network bulletin board], and Tetris clones.' But as Stoll knows perfectly well, learning your first computer language is a serious business. To grasp what an algorithm is, how to express one, what makes one good and another bad, how programs get converted into working virtual machines -- these are deep and fascinating intellectual projects that can hold their own against Attic Greek any day.

When he talks about computer models Stoll focuses on his dissertation project, a model of Jupiter's atmosphere. It seems he is no longer sure of his conclusions; there is no way to be certain whether his numbers are right. He turns this uncertainty into a cautionary pronouncement about computer models in general. Granted, it is inconvenient to drop by Jupiter to check out your predictions. But the bug here lies in astronomy, not computer modeling. Stoll should be more forthright about the extent to which computers have revolutionized science, by allowing researchers to model physical reality in a vastly more powerful and sophisticated way than they ever could before.

But my only serious complaint is about Stoll's unwillingness to show us the big picture. I have argued (and still believe) that to understand the computer culture's dangerous potential, we need to recognize the dark feedback loop between the bad habits computers encourage and the bad habits we would cultivate anyway. It's our rotten luck that computers have made glitz, superficiality, and ignorance so convenient just as glitz, superficiality, and ignorance were coming into style in a big way on their own. Computers bring out the worst in us. That's my theory.

So, OK, the book isn't perfect, and it's not the last word. It is merely incisive, fascinating, and indispensable.