A Need for the Dreaded Gatekeeper
Had we but world enough, and time,
This coyness, lady, were no crime….
But at my back I always hear
Time's winged chariot hurrying near…–Andrew Marvell, "To His Coy Mistress"
Regular readers of this magazine have surely concluded that the secret to understanding information technology lies in learning laws named after guys whose last names begin with M. So to the laws of Moore and Metcalfe, I would add a third and more fundamental one: Marvell's law.
Andrew Marvell was not a techie. He was a 17th-century English politician and poet. His greatest legacy may have been to help save his former boss, John Milton, from execution after Charles II restored the monarchy; without Marvell's maneuvering we might not have Paradise Lost. Milton's masterpiece is not, however, our concern here. Marvell's is. Its most famous lines appear above.
To read many a recent author, you would think that Time's winged chariot first appeared about a century ago, along with factories and railroads—that life was leisurely and unhurried until modern technology ruined its natural rhythms. A dip into the time-obsessed poetry of the 16th and 17th centuries quickly dispells that illusion. Today's technology may be speedier, but a sense of hurry is nothing new. Feeling the years slip away, Milton lamented the coming of his 23rd birthday.
Marvell's Law, then, is simple: Life is limited. Time is the truly scarce resource.
We do have more time than our ancestors: In Marvell's era, three score and ten was indeed a great age. (Marvell died at 57, Milton at 66.) Just as important, we have lengthened our days, lighting up the night. We have made convenience a marketable commodity. If, while writing this article, I am also running the washing machine, my effective life has been extended by the time it would take to do laundry by hand. To the consternation of convenience-denouncing social critics, I and millions of others are willing to pay well for such life extension.
But the chariot remains, ever hurrying. And, yes, Marvell's law really does tell us something about information technology. At least two centuries ago, probably longer, we passed the point where it was possible for educated people to master all their culture's sources of information. Yet only very recently—within my lifetime, really—1have we begun to absorb and accept the effects of that shift. You can still get the sympathetic attention of many intellectuals by decrying specialization and demanding a single, universal (and extremely narrow) canon of learning.
The flip side of information abundance is a huge sorting problem. Life is limited, so every new source of information must take time from an old one. We can ignore the new, or the old, or trade time from one to make time for the other. But we cannot have it all. And as information multiplies, the possible ways of combining it grow even faster; combinatorics makes exponential growth look positively lethargic.
Sometimes, of course, the game isn't exactly zero sum. A new information source arises that also provides faster access to what has gone before. Such innovations have enormous effects—and find hugely receptive markets. The explosion of the Web reflects this phenomenon. Whatever its value for commerce or entertainment, the Web functions most usefully as a research tool, the fastest way to dig up everything from government statistics and academic papers to the history of beach volleyball or contact lenses. But even as it cuts search time, it throws off more and more information. Eventually, the time we save in looking for things is dwarfed by the time it takes to absorb what we find.
The Net in all its manifestations is not, then, exempt from Marvell's law. Its usefulness is limited by our time and attention. Despite all the talk about "disintermediation" and "democratization," easy publication means that intermediaries are more valuable than ever. Unless we want to spend all our finite time sorting through information and none actually using it, we need specialized guides: editors, critics, synthesizers, scholarly referees, even travel or real-estate agents.
Good search engines and intelligent agents may someday make it easy to gather the information we know want. But we do not have infinite imaginations. And these technologies will not, at least for the foreseeable future, create combinations their users have not already specified (although inefficient searches may, like rambles through open-stack libraries, inadvertently turn up treasures). They will not ferret out meaning or exercise critical judgment. These tasks require the quirks and talents of human beings. Those quirks can create blind spots, which, as we shall see in my next column, are another product of Marvell's law.
But the dreaded "gatekeepers" are inevitable. The cyberdream that everyone will navigate everything for him- or herself is the ancient fantasy of omniscience, omnicompetence, and eternal life. It defies Marvell's law. We have not world enough, or time.
A somewhat shorter version of this column appeared in the June 2, 1997 Forbes ASAP. For further elaboration of these ideas, see her speech The Age of the Editor.
