Acerca del autor

John O. McGinnis is the George C. Dix Professor of Constitutional Law at Northwestern University.

De la contraportada

"This is an outstanding book with a timely argument. McGinnis makes the important point that information is accelerating and democratic governance needs to evolve in response to rapid changes in information technology and other scientific fields. The breadth of his analysis and the keen insights he provides at many levels of the problem are impressive."--Darrell M. West, author ofDigital Government: Technology and Public Sector Performance

"McGinnis discusses the challenges and opportunities for governance created by the rapid advance of technology, and analyzes these issues in a manner that is new and distinct.Accelerating Democracy tackles an important subject that has not been properly addressed in the literature to date."--Glenn H. Reynolds, University of Tennessee

De la solapa interior

"This is an outstanding book with a timely argument. McGinnis makes the important point that information is accelerating and democratic governance needs to evolve in response to rapid changes in information technology and other scientific fields. The breadth of his analysis and the keen insights he provides at many levels of the problem are impressive."--Darrell M. West, author ofDigital Government: Technology and Public Sector Performance

"McGinnis discusses the challenges and opportunities for governance created by the rapid advance of technology, and analyzes these issues in a manner that is new and distinct.Accelerating Democracy tackles an important subject that has not been properly addressed in the literature to date."--Glenn H. Reynolds, University of Tennessee

Fragmento. © Reproducción autorizada. Todos los derechos reservados.

Accelerating Democracy

PRINCETON UNIVERSITY PRESS

Contents

INTRODUCTION.....................................................................................1CHAPTER ONE The Ever Expanding Domain of Computation............................................9CHAPTER TWO Democracy, Consequences, and Social Knowledge.......................................25CHAPTER THREE Experimenting with Democracy......................................................40CHAPTER FOUR Unleashing Prediction Markets......................................................60CHAPTER FIVE Distributing Information through Dispersed Media and Campaigns.....................77CHAPTER SIX Accelerating AI.....................................................................94CHAPTER SEVEN Regulation in an Age of Technological Acceleration................................109CHAPTER EIGHT Bias and Democracy................................................................121CHAPTER NINE De-biasing Democracy...............................................................138CONCLUSION The Past and Future of Information Politics..........................................149Acknowledgments..................................................................................161Appendix.........................................................................................163Notes............................................................................................165Index............................................................................................203

Chapter One

The computer is now the fundamental machine of our age. Continuing exponential increases in computing power both propel the potentially cascading benefits and catastrophic threats that demand better governance and create the tools for better governance. The computer is the force behind most material technological advances as more and more fields from biotechnology to energy are brought within the domain of its digital power. Advances in these material technologies will generate many benefits but also may create new dangers such as novel kinds of pollution and weapons.

The rapid rise of computers likely reflects technological acceleration, a process by which technological change has moved faster and faster over the course of human history. Thus, the growing power of computation will increase the pace of change, potentially generating social turbulence and instability.

Fortunately, computational advances are also driving advances in information technology, from the growth and deepening of the Internet, to the burgeoning power of empirical methods, to the increasing capability of artificial intelligence. The key to improving governance is to bring politics within the domain of such information technology. Only a politics that exploits the latest fruits of the computational revolution can manage the disruption that this revolution is bringing to the social world.

The Exponential Rise of Computers

Computational capacity is growing at an exponential pace. Moore's law, named after Gordon Moore, one of the founders of Intel, is the observation that the number of transistors that can be fitted onto a computer chip doubles every eighteen months to two years. This forty-year-old prediction has correctly foretold that every aspect of the digital world—from computational calculation power to computer memory—is growing at a similarly exponential rate.

Recent studies confirm that exponential growth in computational power is still on track. In an article in early 2011 two researchers calculated that the "computing capacity of information," which they define as the "communication of information through space and time guided by an algorithm," grew by approximately 58 percent a year, very close to the eighteen-month doubling posited by Moore's law. The temporal communication aspect of computation, such as broadband capacity, has grown at 28 percent per year, doubling in approximately thirty-four months. The spatial capacity for storage has grown at 23 percent per year, with a doubling time of approximately forty months.

While Moore's law is the best known perspective on the exponential growth of computing, Bell's law may better capture its social effect. Bell's law posits that computational increases of a hundredfold create new classes of computers approximately each decade. Each class of machines becomes dramatically smaller in size but with as much or greater functionality as the class it displaces. The new class then becomes a new nexus by which humans exploit computational power in everyday life. In the 1960s mainframes were a primary locus of computing. In the 1970s so-called minicomputers began their run, only to be succeeded in the 1980s by PCs. In the 1990s PCs were joined by laptops, and in the 2000s "smartphones" appeared. Networks of sensors operating together are just now beginning to collect data on such matters as traffic flows, and the ubiquitous computing of interconnected sensors will soon be upon us. By the 2020s, computers will be small enough to be routinely introduced as medical devices in the body, enabling closer interaction between humans and computational machines. Already, devices like electronic tattoos are being placed on the skin for monitoring.

The declining cost of computation fundamentally alters its social role. Exponential growth in computational capacity is moving computation rapidly to what technology theorists call a stage 4 technology. In stage 1 a technology is very valuable and used only sparingly and controlled generally by elites. In stage 2 the technology is still expensive but cheap enough that ordinary people can use it. In stage 3 it becomes cheaper still and is integrated into every part of daily life. In stage 4 the technology becomes extremely cheap, and its use is so pervasive that it almost goes unnoticed. Computers have rapidly moved through these stages. Mainframe computers marked the first stage, personal computers the second stage, and we are now well into the third stage with mobile computing. Soon we will be in an era of ubiquitous computing when computers become so cheap they are everywhere. Computation has moved through these stages in less than a century. In comparison, paper, which began with papyrus in Egypt, took more than two millennia to become a ubiquitous presence of human life.

It is difficult to overstate the power of exponential growth. As economist Robert Lucas once said, once you start thinking about exponential growth, it is hard to think about anything else. The computational power in a cell phone today is a thousand times greater and a million times less expensive than all the computing power housed at MIT in 1965. The computing power of computers thirty years from now is likely to prove a million times more powerful than computers of today.

For years, the imminent death of Moore's law has been foretold, but the relentless exponential progress of computation has continued nonetheless. Intel—the computer chip–making company that has a substantial interest in accurately telling software makers what to expect—projects that Moore's law will continue at least until 2029. It is, of course, the case that the increase in computer power based on silicon chips will not continue indefinitely, because even with recent improvements in techniques, such as 3-D chips, this kind of computing process will bump up against physical limits. But in his important book The Singularity Is Near the technologist Ray Kurzweil shows that Moore's law is actually part of a more general exponential growth in computation that has been gaining force for over a hundred years. Integrated circuits replaced transistors, which previously replaced vacuum tubes, which in their time had replaced electromechanical methods of computation. Through all of these changes in the mechanisms of computation, computational power relentlessly increased at an exponential rate. This perspective suggests that other modes of development—from carbon nanotechnology, to optical computing, to quantum computing—are likely to continue exponential growth, even when silicon-based computing reaches its physical limits. Although exponential growth cannot go on forever, its end is not yet in sight.

Focusing only on the exponential increase in hardware capability actually understates the acceleration of computational capacity in two ways. Computational capacity advances with progress in software as well as progress in hardware. A study considering improvements in a benchmark computer task over a fifteen-year period showed that computer speed had improved by one thousand times through improvements in hardware capacity. But computer speed had also been increased by approximately forty-three thousand times through improvements in software algorithms. Like many creative human endeavors, progress in software alternates between breakthroughs and periods of consolidation where gains are less spectacular. Improvements in software may also speed some tasks more than others, but in general it is a force multiplier for the gains of Moore's law.

Gains in connectivity may also increase the effective power of computation. Computers are beginning to interconnect among themselves and with human intelligence, most notably through the Internet. A concrete result of this interconnection is that more people in areas like India and China, previously remote from the core areas of innovation in the West, can now come online and contribute to the growth of science and technology. The results of the faster and greater collaboration made possible across long distances are reflected in exponential growth in the volume of scientific knowledge. This knowledge creates the platform for further, faster innovation.

It might be thought that, unlike the exponential increases in computer hardware and some kinds of software, the Internet represents a onetime gain that will be exhausted when everyone goes online. But in reality the current state of the Internet represents only the beginning of deeper connectivity among machines and among humans through machine connections. First, the amount of time spent connected electronically continues to grow as connectivity is made easier through mobile devices. Second, the smaller classes of computers predicted by Bell's law will connect more and more of the physical world to the Internet, making this vast machine more sensate. Soon afterward still smaller classes of computers may also make more organic connections between the human mind and machine intelligence, rendering the idea of a global brain more than a metaphor. These interconnections are likely to further accelerate innovation. More rapid innovation requires more social interaction. The denser the web of computational connections becomes, the better the innovations that are incubated there.

Science fiction writer Neal Stephenson has recently argued that our interconnectivity and informational capacity might actually decrease innovation. He fears that greater opportunities for researching ideas and talking to others may discourage risk taking, because potential innovators will learn of past failures and future pitfalls more easily. But blind risk taking is unlikely to be as helpful as informed risk taking. Inventors will see what mistakes were made in the past and adapt their ideas in response, making success more likely.

The culture of improvement that has been ongoing in the West for the last thousand years also underscores the power of computational advances. Once invented, almost all technologies become more efficacious as they are developed over time. Given the many routes to modifying an invention, many minds can work simultaneously for its amelioration. Thus, combining multiple linear improvements has often been able to generate overall exponential improvement in many kinds of mechanisms and machines.

The exponential growth of computers, however, generates broader and more powerful waves of technological change than previous inventions. Almost everything in the world can be improved by adding an independent source of computational power. That is why computational improvement has a greater social effect than improvements in previous technologies.

Computation-Driven Technological Change

The exponential increase in computing power is driving fundamental innovation in other technological fields as they are brought within the domain of information technology. Information can be digitized and then easily manipulated, permitting the analysis and simulation of the real world without as much need for physical experiments. As a result, many fields are making enormous progress and are poised to create substantial social change. Medicine, nanotechnology, and solar energy provide excellent examples of such computation-driven breakthroughs.

Medicine

Medicine is undergoing a computational revolution. The cost of sequencing a genome—a bounty of digital information about the particularities of individuals—is falling faster than the rate of Moore's law. This progress heralds a whole new discipline of genomic medicine that will help find causes of diseases and personalize cures to make them more individually effective. Of course, it is not the case that all, or even most, diseases are wholly dictated by our genes. For instance, in recent years scientists have studied how environmental factors can affect even hereditary traits, as when cigarette smoking accelerates puberty by affecting gene expression. But such factors are now intensively studied as well. And improvements in data collection and measurement made possible by the ongoing computational revolution can help us better quantify those environmental influences. This research has the potential to create a kind of personal preventive medicine, where people would be advised to avoid the risks most dangerous to them.

Medical research will also be aided by the digitization of medical records, which will enable researchers to match genetics and personal environments with life medical histories, and by the Internet, which will allow for ease of access to such material, suitably redacted of identifying personal information. Computers are becoming smaller, with the result that they will be introduced more frequently into the body as early warning systems of acute disease, as monitors of chronic conditions, and as maintenance mechanisms of well-being. Thus, computational advances will lead to many different kind of innovations that will transform a central aspect of our lives—in this case, medical care.

These developments in medicine will provide benefits and yet produce social disruption—all characteristic of progress in the coming decades. As information is made more quantifiable, precise, and accessible in digital form, it will be more easily shared among researchers and become a platform for innovation, providing benefits to the public. But even beneficial improvements will create social distortions that will need to be addressed by more nimble and accurate policy changes. For instance, a rapid rise in longevity may increase the pressure on state pension schemes, forcing the reweaving of the safety nets of the modern welfare state. The ability to manipulate the genome may also create substantial risks by making it easier to create biological weapons of mass destruction. Faster transportation and globalization more generally will provide more vectors to spread engineered germs or viruses throughout the world.

Nanotechnology

More than fifty years ago, the famous physicist Richard Feynman gave a talk titled "There Is Plenty of Room at the Bottom" in which he suggested there was much progress to be made in molding and operating matter at microscopic levels. Nanotechnology—an entirely new field of research—was born soon afterward. Nanotechnology involves the use and manipulation of objects between 1 and 100 nanometers in scale, essentially at the atomic and molecular level. Nanomachines are being created to engage in such manipulation.

The fundamental dynamic of this field is also being driven by increasing power in computation. Nanoscale devices and dynamics are sufficiently small that computers are progressively better able to model and predict their behavior. This modeling allows researchers to experiment with new ideas without having to endure the time and expense of creating physical models. Computers are creating virtual worlds to speed the improvements in our world.

On the agenda of this science is the creation of machines that will self-assemble. Growing computational capacity is also at the heart of self-assembly. Parts must become programmable if they are to replicate and transform themselves and must recognize their place and their geographic relation to other programmable parts. If the promise of nanotechnology is realized, such self-assembly will lead to very low cost production of all kinds of industrial and household goods.

Nanotechnology will be progressively deployed in 3-D printing, another digital innovation that is already in action. Here a three-dimensional printer directed by a computer program prints objects by adding materials to a virtual mold, similar to how a two-dimensional printer prints words by adding ink to a page. This kind of computer driven manufacturing is already taking advantage of the connectivity of the Internet; companies specializing in 3-D printing can rapidly test crowd-sourced prototypes of a wide variety of products. Nanotechnology will also have a wide variety of uses in medicine, as doctors will be able to make changes to the body at the cellular level without invasive surgery and 3-D printers will be able to reproduce body parts.

But nanotechnology also brings potential costs and risks. Humans may experience some forms of nanoparticles as deadly pollution. More dramatically, self-assembly may permit nanomachines, because of mistake or malevolence, to replicate without end, threatening to envelop the whole world.

Energy

Computational improvement is also creating faster improvement in the alternatives to fossil fuels. For instance, solar energy is moving along an exponential path, improving at about 7 percent a year. If this trend continues, solar power is projected to become competitive with fossil fuels sometime in the relatively near future and perhaps half their cost by the 2030s. The improvements also depend on computational technology that permits the rapid design of more efficient solar cells and the production of such cells at lower cost.

One method of creating solar power deploys mirrors to concentrate sunlight so that intense heat is generated. Large mirrors are very expensive to make and maintain, but one company, eSolar, has created a system of small mirrors whose positions can be monitored and changed by computer chips in each mirror, making for both cheaper fabrication of mirrors and yet more efficient concentration of sunlight. This invention provides another example of how computer chips will add value to all kinds of products in the coming years.

Such innovations also complicate collective decision making, because our decisions about how stringently to regulate today may depend on our assessments of the technology of tomorrow. If we are confident that solar energy will be available soon, government might need to intervene less to make fossil fuels more environmentally friendly. In short, the future shape of technology becomes increasingly relevant to our current decisions, because the speed of technological change is accelerating.

(Continues...)

Excerpted from Accelerating Democracyby John O. McGinnis Copyright © 2013 by Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.