Science & Technology

Hooray for Risk


For its second conference on "Cyberspace and the American Dream," the Progress & Freedom Foundation decided to match the medium to the message. It not only brought various digerati to Aspen to discuss cyberissues. It also established a Web page and lined up co-sponsors, including Reason, to moderate on-line discussions and record and post the Aspen proceedings.

Thus I found myself amid a tangle of wires in the "digitally enabled" section of a large hotel ballroom, experiencing real life in the information age: frozen screens, dysfunctional passwords, on-line services that didn't answer, and lots of help from my digitally enabled friends.

I'm not complaining; we did eventually get the system to work—most of the time. In fact, the experience was pretty typical. From the Altair to the Pentium to Windows95, the digital "revolution" (a favorite word at Aspen) has come with bugs.

And that's truly remarkable. The American legal and regulatory system rarely tolerates imperfection, even in brand new inventions. In most of American law, as Peter Huber notes in Liability, "It is the innovative and unfamiliar that is most likely to be condemned." Old, familiar risks are okay; new products with unfamiliar risks often get wiped out in court—or at least denounced by "consumer advocates." As a result, our political—regulatory system is biased in favor of stasis, against the dynamic, the new, the unknown.

This static bias seeps into our popular culture. You can even find it in Vogue , a magazine that owes its existence to ever-changing fashions. The June 1994 issue featured an attack on one of my favorite new technologies: disposable, extended-wear contact lenses, a tremendous boon to hopelessly myopic residents of earthquake country. (The only thing worse than being jolted awake is being jolted awake blind.) The risk of infections, medical columnist Dr. Isadore Rosenfeld informed readers, means that wearing the lenses overnight is "not safe."

"Removing the lenses at night takes no more than a minute or two, and, after all, you don't need them while you're sleeping (unless you're having nearsighted dreams)."

Or unless you sometimes wake up. Which is exactly why I still wear them—despite having had one of those very painful, very scary eye infections.

Rosenfeld's flip tone captures a common attitude toward other people's risk-taking. He presents the medical evidence that sleeping in contact lenses is hazardous and declares the benefits trivial, a matter of mere convenience. He implies that people who sleep in their lenses are ignorant or lazy. He suggests that safety is an absolute. And he assumes that he has all the relevant information needed to determine whether I, someone he has never met, should sleep in my contact lenses.

It is a short step from such attitudes to product bans. Just ask the manufacturers—or the would-be buyers—of silicone gel breast implants.

Information technology, by contrast, has spread rapidly largely because third parties haven't seized the power to decide what risks users can take. At first, the new technology was too obscure for scrutiny; now it's too ubiquitous, too obviously valuable. Information technology is a model of what you can get if you accept the risks of imperfect products: As Dave Cell warned in CompuServe Magazine, "First-generation hardware is expensive and never works right. The first buyers are the last beta-testers."

That risk is as old as the personal computer: Many people who sent away for the first PC, the home-assembled Altair, couldn't get it to work at all, thanks to its notoriously unreliable memory boards. Yet who in retrospect would tell those earliest adopters not to waste their money?

Classic early adopters—the people who buy the fastest new computer at prices that look exorbitant six months later—seem irrational to bargain hunters, especially if they can't give a "good reason" for their impatience. But the risk takers who buy the early editions not only finance the development of new and improved ones. By serving as the "last beta-testers," they also provide crucial information about what needs to be corrected. The product that comes out six months later has generally eliminated the old bugs, though it may (and probably does) have new ones.

By accepting risks, information-technology customers have created a dynamic culture that values innovation over certainty, speed over perfectio, and progress over all. Machines don't even stay current long enough to establish a track record for Consumer Reports, says Garr Di Salvo, senior project leader for electronics at Consumers Union: "A model might have a life cycle of maybe six months from the time it appears in the store to the time it's on the discount shelf."

Tolerating a little imperfection can take us all a long, long way. Something to keep in mind next time your screen freezes and the guy at CompuServe says, "Yeah, sometimes that just happens."

A slightly shorter version of this article first appeared in the December 4, 1995 issue of Forbes ASAP.