Is there in fact, as is so often claimed, an “information explosion”? Why do we think so? What does it signify? Certainly the press has had no doubts for some years that information is exploding. A single issue of the Sunday New York Times a decade or more ago yielded the following two items:
Will a full week of shorter trading hours bring happiness to brokerage firms whose back offices are jammed with paperwork?
George A. Miller, a professor of psychology at Harvard, warned that by 2000, the limit of man’s mind to absorb information may be reached. “We may already be nearing some kind of limit for many of the less gifted among us, and those still able to handle the present level of complexity are in ever increasing demand.”
This is simply a tiny sample from a very much larger number of items— the first two that came to hand. The first conjures up a fascinating picture of the Stock Exchange slowly submerging under a tide of paper; the second promises prosperity to “those still able to handle the present level of complexity”—and I assume that means all of my readers. How valid are the predictions of the impending Flood? To answer those questions, we must sort out the elements of stability and the elements of change in human affairs.
Change—extremely rapid change—there certainly is along technological and economic dimensions. We know that technology is advancing with great speed. We know that it is beginning to make technically possible for the first time in human history the elimination of acute poverty (provided that we give adequate attention to problems of distribution as well as total production). We know that it is even providing means, if not yet the will, for combating the over-rapid growth of population—the most serious threat to the prospect of banishing poverty.
But (there is always a “but” at this point in the argument), if we measure the world by the values and goals of our species, we have good reason to doubt whether it is changing very much at all. We must not suppose that with the progress of technology, or even with the progress of our economies, mankind will become deliriously happy. For human aspirations have a way of adjusting to opportunities. We must not expect that technological progress will produce Utopia; it is reasonable to hope that it can bring relief to acute hardship and acute pain.
1. Evidences for the Revolution
With these cautions and reservations, let us examine the changes that are taking place in information production and processing. Forty years ago, at a meeting of the Operations Research Society of America, Allen Newell and I made some very specific ten-year predictions. I wish I could tell you that each one had been exactly fulfilled. The last of them (a computer as world chess champion) is still not quite there—but very close. But rather than either defend or explain away these past predictions. I will simnlv make a few general comments about them.
We did not guess correctly how research efforts would be allocated to particular areas or the relative difficulties of specific problems. Hence, although chess-playing by computer has made enormous progress, it is, in forty years, just reaching our ten-year target. On the other side of the account, fundamental understanding of natural language, including speech recognition and understanding, the construction of high-level computer languages, automatic design by computers, visual pattern recognition, and robotics, among others, have advanced more rapidly than we would have dared to predict in 1957.
Hence, in the light of the actual progress, there is no reason to revise our basic thesis: that electronic computers are general-purpose information- processing devices; that we will continue to learn step by step to do with them any kind of thinking that people can do; that, with the help of computer simulation techniques, we are learning how human beings learn and think and how to help them to learn and think better.
My discussion in this chapter of the consequences for organizations of the computer and communication network culture has been strongly influenced by my having lived in that culture for many years, with e-mail since 1972, and with a campus-wide network since 1985.71
1. Attending to the Information That Is There
Given the rate at which the technology of information processing is pro- gressing, why won’t there be an information explosion? The mountain climber, Mallory, when asked why he wanted to climb Everest, gave his famous reply: “Because it is there.” Not all of us accept that reply for ourselves. Not all of us aspire to climb Everest or would look forward to the prospect with any relish or sense of purpose.
Now it is possible to be just as skeptical about processing information as about climbing mountains. Specifically, information doesn’t have to be processed just because it is there. The telephone doesn’t have to be answered just because it is ringing; or the newspaper read just because it was tossed on our doorstep. Information is sometimes ignored at our own peril, but we are more often guilty of the opposite error—of supposing that all would be well “if we just had more information,” a pathetically naive belief in the technological fix. The following is an old example, but one that remains quite relevant today:
The U.S. State Department, drowning in a river of words estimated at 15 million a month to and from 278 diplomatic outposts around the world, contracted, a number of years ago, for a $3.5 million combination of computers, high-speed printers, and other electronic devices. These were aimed at eliminating transmission bottlenecks in the system, especially during crises that bring in torrents of cabled messages from world trouble spots. With the new system, computers could absorb cable messages at 1,200 lines a minute. The old teletypes could receive messages at only 100 words a minute.
Leave aside the fact that the technology mentioned in this example is already, thirty years later, buried with the dinosaurs, and that the flow of cable messages has undoubtedly increased by several orders of magnitude. What is most instructive about the example is that the new system was designed without anyone asking about the capacity of the human users to process the greatly accelerated flow of information. The sorcerer’s apprentice is at large. Who will read the flood of words that the new enlarged communications channels will deliver? The bottleneck is no longer the capacity of the electronic channels but the capacity of the human users.
2. Attending Selectively
We cannot save ourselves from drowning in information by installing faster printing devices. Lack of information is not the typical problem in our decision processes (although lack of the right information sometimes is). The world is constantly drenching us with information through eyes and ears millions of bits per second. According to the best evidence, we can handle only about fifty. The limit is not information but our capacity to attend to it.
Saturation with information is no new thing. The movements of the stars, visible to us throughout the tens of thousands of years of our history, contain all the information that is needed for Newton’s laws of motion or the law of gravitation. The information was there all along. What was lacking, until a few hundred years ago, was the basis for selecting the tiny fraction of it that could be used to establish powerful generalizations.
If we cannot avoid living in a world that drenches us with information— whether made by ourselves or nature—still, we can and must select for our processing the information that is likely to be useful to us and ignore the rest. Our scientific and technological knowledge, our decisionmaking and information-processing systems should permit us to absorb information very selectively, extracting from it just the parts we want.
In the same vein, most of the contemporary concern about the infor- mation explosion in science is misconceived, because it is based on an presses it. A generation or two ago, for example, organic chemistry was a mass of particulars only weakly organized by known theoretical generalizations. Today, although knowledge of organic chemistry has grown vastly, the principles of quantum mechanics provide powerful organizing means for that knowledge. As a result, it is undoubtedly easier today to gain a mastery of organic chemistry adequate for doing significant original work than it was in an earlier era when very much less was known.
The example I have chosen is not an isolated one. In the scientific endeavor, “knowing” has always meant “knowing parsimoniously.” The information that nature presents to us is unimaginably redundant. When we find the right way to summarize and characterize that information— when we find the pattern hidden in it—its vast bulk compresses into succinct laws, each one enormously informative. Herein lies the real significance of today’s information revolution. Information and the processing of information are themselves for the first time becoming the objects of systematic scientific investigation. We are laying the foundations for a science of information processing that we can expect will greatly increase our effectiveness in handling the information around us.
Thus, at a time when we are acquiring devices that will transmit, store, and process symbols at unprecedented rates and volumes, the most important change is not the growth of these devices but the growth of a science that helps us understand how information can be transmitted, how it can be organized for storage and retrieval, how it can be used (and how it is used) in thinking, in problem-solving, in decision-making. This growing understanding of information processing returns to us the decision of whether information must overflow and we must drown in it.
A major task ahead is to design effective information-processing systems for making decisions in business and in government. It is important that we talk about designing information-processing systems and not just designing computers and electronic networks. The design must encompass far more than the computer hardware and software; it must handle with equal care the information-processing characteristics and capabilities of the human members of organizations who constitute the other half of the systems.
For generations to come, although organizations will have many mechanized components, their most numerous and crucial elements will continue to be people. Their effectiveness in handling problems will depend as heavily on the effectiveness of the thinking, problem-solving, and decision- making that people do as upon the operation of the computers and their programs.
Source: Simon Herbert A. (1997), Administrative Behavior, Free Press; Subsequent edition.