Beyond Calculation

By 2047 almost all information will be in cyberspace—including a large percentage of knowledge and creative works. All information about physical objects, including humans, buildings, processes, and organizations, will be online. This trend is both desirable and inevitable. Cyberspace will provide the basis for wonderful new ways to inform, entertain, and educate people. The information and the corresponding systems will streamline commerce but will also provide new levels of personal service, health care, and automation. The most significant benefit will be a breakthrough in our ability to communicate remotely with one another using all our senses.

In 2047, Internet is everywhere. In seventy-four years it has grown from an idea to interconnect heterogeneous packet-communication networks to a world-wide, ubiquitous communication web joining people, businesses, institutions, appliances, and all forms of electronic equipment in a common communication framework. Like electrical power, it is assumed to be available whenever and wherever needed. The little vignette below suggests some of what I believe will be commonplace.

The first million was easy. Computers have improved by a factor of millions in price, performance, capacity, and capability in their first fifty years.

When the idea to write about the next fifty years of computing first entered my head, I wrote it off as utterly preposterous: what sane scientist purports to be able to see so far into the future? But then I realized that in a way that is precisely what educators do all the time: when designing our courses, we do dare to decide what to teach and what to ignore, and we do this for the benefit of students, many of whom will still be active forty to fifty years from now. Clearly, some vision of the next half century of computing science is operational. To this I should add that it is all right if the crystal ball is too foggy to show much detail. Thirty-five years ago, for instance, I had no inkling of how closely program design and proof design would come together, and in such a detailed sense my life has been full of surprises. At the same time, these surprises were developments I had been waiting for, because I knew that programming had to be made amenable to some sort of mathematical treatment long before I knew what type of mathematics that would turn out to be. In other words, when building sand castles on the beach, we can ignore the waves but should watch the tide.

It requires a brave person, or else a fool, to attempt a detailed, and accurate prediction for the next half century of computing. But without a vision of the future we will tend to wander like the proverbial drunken sailor and will not successfully attack many of the important problems facing us. As impossible as accuracy may be, it is nevertheless necessary to make some predictions. Of necessity, this is more of an essay than a technical report.

The important waves of technological change are those that fundamentally alter the place of technology in our lives. What matters is not technology itself, but its relationship to us.

In the “Sim” series of computer games (SimCity, SimLife, SimAnt, SimHealth), a player is asked to build a community or an ecosystem or to design a public policy. The goal in each case is to make a successful whole from complex, interrelated parts. Tim is thirteen, and among his friends, the Sim games are the subject of long conversations about what they call “Sim secrets.” “Every kid knows,” Tim confides, “that hitting Shift-FI will get you a couple of thousand dollars in SimCity.” But Tim knows that the Sim secrets have their limits. They are little tricks, but they are not what the game is about. The game is about making choices and getting feedback. Tim talks easily about the tradeoffs in SimCity between zoning restrictions and economic development, pollution controls and housing starts.

A common prediction among technologists—and a common fear among the general population—is that computers and robots will come to mimic and even surpass people. No way. Computers and people work according to very different principles. One is discrete, obeying Boolean logic; and deterministic, yielding precise, repeatable results. The other is nondiscrete, following a complex, history-dependent mode of operation, yielding approximate, variable results. One is carefully designed according to well-determined goals and following systematic principles. The other evolves through a unique process that is affected by a wide range of variables, severely path dependent, fundamentally kludgy, difficult to predict, and difficult to emulate. The result—biological computation—is complex, parallel, multimodal (e.g., ionic, electrical, and chemical).

We understand half the mind fairly well and the other half barely at all—which is surprising given the suggestive facts that are all around us, scattered like shells on the beach at low tide, so plentiful it is hard to avoid crunching them underfoot. Most cognitive scientists do largely succeed, though, in proposing theories that simply ignore the nonanalytic, non-problem-solving, non-goal-directed aspect of thought. It would be absurd for me to claim that my work (or that of the small number of others who are interested in these problems) accounts for all the neglected facts. It doesn’t, but at least it acknowledges that they exist and need explaining.

Imagine a world where smart machines do all the work—a world in which man no longer lives by the sweat of his brow. Intelligent robots, freely available to all, provide all the economic benefits of slavery without any of its moral and ethical drawbacks. Want a new home? Just ask a robotic architect to design your dream house, a crew of construction robots to build it. Want to travel to a faraway place? A robot taxi can take you to the airport, where a robotically piloted aircraft can whisk you to your destination. Ready for dinner? A few words to your robotic chef and the food will be prepared just as you wish, with inventive touches ensuring that no meal is just like a previous one—unless you want it that way, of course.

When asked to project fifty years ahead, a scientist is in a bit of a quandary. It is easy to indulge in wishful thinking or promote favorite current projects and proposals, but it is a daunting task to anticipate what will actually come to pass in a time span that is eons long in our modern, accelerated age. If fifty years ago, when the ACM was founded, biologists had been asked to predict the next fifty years of biology, it would have taken amazing prescience to anticipate the science of molecular biology. Or for that matter, only a few years before the initiation of the ACM even those with the most insight about computing would have been completely unable to foresee today’s world of pervasive workstations, mobile communicators, and gigabit networking.

With personal computers essentially ubiquitous, an emerging broadband WWW, the $500 Internet interface, DRAM selling for less than 50 microcents per bit, 300 mm wafer silicon heading to 0.1 micron and smaller, virtual reality moving in all directions, and other such levers, the computer industry is changing more swiftly at this time than ever before. Indeed, the key players have changed at a dizzying pace.

Everyone in my office uses a computer. I never travel without one, and my three- and five-year-old grandchildren can play games for hours on their parents’ laptops. Not only do many people have access to computers, but given the Internet, we now have access to millions of people to whom we did not have access before. Fifty years ago, we could not have predicted this. I am therefore hesitant even to speculate about specific information-technology (IT) products and services that will be invented throughout the next fifty years. In this paper, however, I will discuss some things I am confident will be central in the direction of computer and software design over the next fifty years. This confidence comes from an observation about the history and evolution of disciplines and industries dominant today. If we look at the history of many disciplines, we can distinguish certain crucial moments in which the uncovering of a set of simple, stable recurrences opens the door to a sudden rush of innovation, providing orientation to the field for some time to come. Perhaps the most obvious example is the field of chemistry. Here, the articulation of atomic structure and the creation of the periodic table created the foundation of our modern discipline, which has consequently led to the rise of several of our most powerful industries. None of these resulting events could have been predicted in detail or with complete accuracy. But once the periodic table existed, it was clear that it would be a secure foundation for design and fruitful exploration for many years to come. Another example is the field of genetics. Certain aspects of heredity were well understood long before DNA was identified as the source of genetic inheritance and its chemical structure determined. But once Watson and Crick discovered the simple, recurrent structure of base pairs and their helical arrangement, suddenly all the variations of species could be understood in terms of the myriad combinations of this simple set of elements. And with the more recent development of chemical means for manipulating DNA, the biotechnology industry has emerged. Again, none of this could have been envisioned clearly at the time the structure of DNA was discovered. But it was clear that for anyone interested in heredity and genetics, DNA was the only game in town and that it is likely to stay that way for some time to come.

The phrase “information warfare” has recently reached the popular press. While the phrase may suggest hostile actions between two warring nations against one another’s military information systems, hostile actions are limited neither to the military nor to nations at war. They may be undertaken by small groups of people or even individuals, and they may target civilian systems. The ideas of nation states and geographic boundaries are social artifacts that do not necessarily apply in the new world of networked computing. Neither are the responsibility and authority for protection clearly defined. As with physical systems, individuals must assume responsibility for their own protection but must also relinquish certain powers to properly authorized groups for greater protection. Professionals have a civic responsibility to warn the general population of the risks and to help them cope with those risks.

Taken as a whole, the practices of many companies today reflect a sea change in the relationship between business and government. Companies are increasingly relying upon outsourcing, strategic alliances, acquisitions and mergers, relocation of plant and equipment, and aggressive money management to compete in the global marketplace. Although these practices and not entirely new, they have only recently become “normal” management instruments.

The laws of thermodynamics tell us that in a closed system, order tends to be replaced by disorder and randomness. But approximately three billion years ago a chemical structure appeared on the earth with a tendency to organize itself into increasingly complex forms. This structure, made mostly out of carbon and hydrogen, is called DNA. Of course, the tendency of DNA toward increasing order is not a thermodynamic anomaly, because the earth is not a closed system, receiving as it does a steady rain of energy from the sun. Nevertheless, the behavior of DNA is unusual enough that we give a name—“life”—to its complex forms, and devote a science—biology—to their study.

We rarely choose to think of it this way, but presence consumes resources and costs money. Typically it costs us more (in hotel charges or office rents, for example) to be present in places where many people would like to be than it does to be present in places where few people want to be. And it costs us time and effort to get to places to meet people, conduct transactions, and see performances. Being in the right place, at the right time, can be expensive.

Research always brings to mind the paradigm of an absent-minded scientist playing aimlessly with his pencil (and computer) seeking some universal truth. It is a paradigm that gave us, over the centuries, great progress in science, and it resulted in relative prosperity for mankind. The basis for this paradigm was and still is that the researcher is highly intelligent, self-motivated, and rather undisciplined. Can we go to the next millennium with this paradigm? In a society where everything is measured, quantified, and rationalized, how can researchers continue to live in an idealistic world? We already see the signs of trouble. Research budgets and research institutes are under pressure in both the public and private sectors. This is the time to revise our position, especially in the fast-moving area of information technology.

At the close of the twentieth century, higher education is facing a series of strong, sometimes contradictory pressures that will transform the two major missions of the university—teaching and research. On the teaching side, these pressures will be resolved by a new distinction between knowledge and information, between “knowing how” and “knowing about.” This change will be accompanied by a strong alignment of graduate educational offerings with the needs and interests of working professionals, with a special emphasis on certifying competence in selected areas. This distinction will also foster a new commitment to offering broader perspectives that enable people to deal with complexity and uncertainty, act with wisdom, build powerful social relationships, and practice the skills of entrepreneurship. Digital media and Internet communications will transform learning practices from the sequential classroom curriculum to nonlinear hyperlearning environments. A new kind of teacher will emerge—the teacher who is a course manager and a coach rather than an information transmitter.