"Settlements on The Electronic Frontier" Transcript of a talk presented at IEEE CompCon '91 2-February-1991 Chip Morningstar American Information Exchange Corporation In recent years the growth and interconnection of the various major computer networks (e.g., the Internet, Usenet, BITnet, etc.) and online information services (e.g., CompuServe, Genie, MCI Mail, etc.) has reached the point where many of us no longer think of them as independent entities at all. Instead, we collectively refer to them simply as "the Net". This agglomeration of computers and communications pathways has not developed according to any particular unified plan. Instead, the growth has been ad hoc, driven by the irregularities of user demand, opportunistic alliances of system managers, and voluntary efforts by programmers and other technical people to enable things to interconnect. The result is rather motley; some find it unaesthetic and many find it confusing. However, the system as a whole shows a remarkable vitality and robustness that transcends that of any of its particular components. The absence of centralized control has enabled experimentation with a wide variety of communications protocols, interconnection topologies, data transport technologies, addressing schemes, administrative procedures, and so on. More importantly, this lack of uniform regulation has permitted the Net's user community a striking degree of freedom of expression. Many of us view this freedom as one of the benefits we hope computer networking will provide to our entire civilization. Unfortunately, the diversity of protocols, addressing schemes, administrative organizations, and the like that characterizes the Net is increasingly becoming a burden rather than a benefit. As people come to rely more and more on the interconnectivity that the Net provides, the bewildering complexity of it all is increasingly viewed simply as an obstacle to effective use of the system. The explosive growth of the Net has brought a growing number of users who view it as a mere tool rather than an object of interest in its own right. These users are less willing to tolerate some of the oddities that we old hands have come to think of as normal. In addition, many messy and complex legal issues regarding life in this electronic environment have been left dangling because, until recently, the technicality of this world has kept it below the threshold of perception of legislators, regulators, mass media, political interest groups, law enforcement agencies, and other powers that be. However, as the technology has continued to evolve it has become more pervasive and more important in the world beyond the academic and technical communities. Consequently, questions of rights, responsibilities, power, and authority have started to become genuine political issues. As the Net's significance in our society grows, there will be calls for increased government involvement in the organization and regulation of the system. Indeed, the clamor for state intervention has already begun and can only increase in volume. The government is already a major player in the system, both as an extensive user and as the sponsor of important parts of it. While one may debate the merits and drawbacks of government sponsorship of, for example, the Internet, to the extent that such support is a matter of historical fact, the government arguably has legitimate interests in its various roles as both user and owner of parts of the system. Those responsible for overseeing such interests will, of course, argue for them. However, there are other motivations for bringing the government into this which cause me greater concern. History teaches us that calls for government action often originate with those who see government power as a tool to attain ends that they cannot attain by other means. These ends may be political, commercial, technical or aesthetic. They may be good or evil or sensible or foolish. However, they all share the common property that attaining them requires (or at least appears to require, in the mind of the person advocating the regulation) that some person or group of people be compelled to do something that they wouldn't otherwise do if left to their own devices. There are many reasons why people might not do what is desired. It may be that it is not in those peoples' economic self- interest (e.g., paying taxes); it may be that they have moral or religious objections (e.g., having their children learn the theory of evolution); it may be that their action in the absence of any guiding constraints would be random (e.g., what side of the street to drive on); it may simply be that they don't know what to do (e.g., building codes). In all cases, however, the government possesses means that ordinary citizens do not to coerce people into doing particular things. In most societies, and certainly in ours, the power of the government is an almost irresistible force. Civilized societies have long recognized that such power is extremely dangerous to nearly everyone, and have therefor evolved traditions and institutions for containing and controlling it. These include democracy, the principle of the rule of law, constitutional separation of powers, bills of rights, and so on. Nevertheless, such power remains a seductive lure to anyone with a cause, be it public spirited or venally self-interested. Consequently, nearly every new development, technological, economic, cultural, or artistic, results in a turf war between the various interests who want to define the scope and nature of the relationship between the government and that which is new. Some of these interests simply seek their own advantage, while others act in what they believe to be the public interest. Inevitably there is a battle over the interpretation of the existing principles of containment and control of government power. Some wish to strengthen these principles in light of the new development, others wish to weaken them, while still others simply wish to preserve the status quo. I believe we are in the early stages of just such a turf war with respect to the Net and to electronic media in general. Right now the situation is characterized largely by confusion. The various sides (and certainly there are more than two) have not yet sorted themselves out into definable camps and, given the pace of technological development, may never do so. It is not clear what all the different interests want and it is arguable whether the course of action advocated by any particular party will actually get them something like what they think they want even if they had their way. Consequently, I would like to set aside my strong feelings on some of these issues for a moment and present what I hope will be a helpful metaphor for sorting the situation out in our own minds. Having made things a little clearer, I can then go on to present what I hope will be a coherent argument for a particular position on these matters. Even if you do not agree with my position, I may have at least contributed a helpful intellectual tool for considering these questions. The class of problems which interests me here relates to how we think about the nature of the electronic community itself. Two schools of thought contend for dominance in this area. The first maintains that computer networks are simply extensions of existing institutions (e.g., the press) and therefor subject to the same constitutional protections and moral and ethical principles that govern traditional media. I believe this is the position taken, for example, by the Electronic Frontier Foundation. Oddly, this position can accurately be called both liberal and conservative. It is liberal in the sense that the principles that its proponents wish to apply in the electronic realm are liberal principles. It is conservative in the sense that it draws on tradition for its sense of direction and denies that new technology invalidates fundamental ethical standards. However, both labels already carry so many emotional and ideological associations in most people's minds that I hesitate to use either of them. The second school of thought maintains that these new technologies are different in kind from anything that has come before and therefor that the old rules do not apply. Instead, they maintain, a new set of rules must be drafted, based on the unique characteristics of the electronic medium. Notable support for this school comes from two groups that in most other respects are antithetical to each other. The first group consists of the utopians and revolutionaries who wish to create in the electronic realm a new (and presumably better) world, entirely free of the mistakes and problems of the present one. I met a number of these people last year at the First International Conference on Cyberspace. The second group includes certain authoritarian establishment members who view existing constraints on their powers (e.g., constitutional limitations on state authority) as annoying restrictions that merely hinder them in attaining what they believe to be worthy goals. The latter group would like to see these constraints confined to as limited a sphere as possible; consequently, they readily adopt the position that electronic media are "different" and therefor not subject to existing protections. One might include in the latter group a number of the law enforcement officials whose actions last year incited the formation of the EFF. Both of these groups really subscribe to the same belief, namely that the electronic realm represents an opportunity to fashion an order whose form they can design and control. Adopting a term from F. A. Hayek, I will refer to this position as "constructivism", the belief that one can construct a society according to a definite plan. I suppose then that I can call the first position I outlined, the both- liberal-and-conservative-position-with-no-good-label, "non- constructivist." While I believe that the non-constructivists are on the side of the angels, there is some merit to the constructivist's argument that what we have here is a new situation demanding new rules. While there are many parallels that can be drawn between the electronic environment and the traditional one, there are also many distinctive differences which cannot be dismissed lightly. Indeed, it seems clear that the rules necessary in the electronic world will be different; the issue is whether they will reflect continuity with our present legal and cultural traditions or whether government regulators and system designers can arbitrarily decree an entirely new model de novo. In a nutshell, the dispute between these two schools of thought comes down to the question of whether the constructivists are to have their way. Obviously, the non-constructivists would say no and the constructivists yes. Discussions of this issue always seem to center on the question of who is to determine the shape of this new world and whether their vision of it is good or bad. I don't want to go into the particulars of any given vision. However much argument over particular visions may make for stimulating conversation, I think framing the debate in this form overlooks a more profound problem with the constructivist position. Indeed, the presence of this problem is the basis for my categorization of the constructivist camp, the reason why I place the revolutionary and the FBI man in the same group. At the heart of a constructivist point of view is the belief that through design or regulation it is possible to make the final form of the system conform to the designer or regulator's vision. For systems involving large numbers of human beings and tremendous technical complexity, I think this belief is profoundly mistaken. I could point to half a century or more of failed attempts at centralized economic planning in the eastern bloc and throughout the third world, or to the problems and enormous cost overruns associated with gigantic technological ventures such as the space shuttle or the B-2 bomber. However, my perspective on this issue is much more personal. From the summer of 1985 to the winter of 1988 I worked for Lucasfilm Ltd. as the project leader and principle designer of Lucasfilm's Habitat project. How many of you here know anything about Habitat or have even heard of it? That's pretty much what I thought. Habitat was (and is-- later incarnations of it are alive and well) a many-user virtual online environment, a make-believe world that people could enter using their home computers connected via telephones and packet nets to a centralized host running on a commercial online information service. It was intended as both an entertainment and as a sociological laboratory. Habitat presented each of its users with a real-time animated view into an online simulated world in which they could communicate, play games, go on adventures, fall in love, get married, get divorced, start businesses, found religions, wage wars, protest against them, and experiment with self-government. (I intended to show some slides of the latest version of Habitat. We asked our associates in Japan for some slides but they sent us these things that look like slides on steroids, about 3" on a side -- different media standards there, I guess. Unfortunately, I don't have time to go into much detail. There is a paper that my principle cohort in this undertaking and I wrote that I can send to anyone who wants to know more; talk to me afterwards.) It was a small but more or less complete world, with hundreds and later thousands of inhabitants. And I, along with my coworkers, was God. We were faced with the challenge of creating and managing this world. In a strict technical sense, the control we had was total. We were omniscient and omnipotent. We could create, destroy and rearrange things at will, and we could peek in anywhere we liked to see what was happening. However, our real control was far more tenuous. Again and again we found that activities that we had planned based on often unconscious assumptions about user behavior had completely unexpected outcomes (when they were not simply outright failures). The more people we involved in something, the less in control we were. We could influence things, we could set up interesting situations, we could provide opportunities for things to happen, but we could not predict nor dictate the outcome. Social engineering is, at best, an inexact science, even in proto-cyberspaces. Propelled by these experiences, we shifted into a style of operations in which we let the users themselves drive the direction of the design. This proved far more effective. Instead of trying to push the community in the direction we thought it should go, an exercise rather like herding mice, we tried to observe what people were doing and aid them in it. We became facilitators rather than take-it-or-leave-it gods. As the experts on how the system worked, we could often suggest new activities for people to try or ways of doing things that people might not have thought of. In this way we were able to have considerable influence on the system's development in spite of the fact that we didn't really hold the steering wheel -- more influence, in fact, than we had had when we were operating under the delusion that we controlled everything. This brings me to the title of my talk. I hear a lot of the rhetoric associated with electronic media speaking in terms of "creating a new world." My experiences lead me to believe that this is a poor choice of metaphors. Instead, I would prefer to speak of "settling a new world." Although the electronic environment is, to be sure, the creation of people, it is not the creation of any single person or group. While each part of it is designed and organized by somebody, the totality is not. It is, in Hayek's words, "the product of human action but not of human design." The metaphor of creation implies a creator, which leads you to the constructivist view I described earlier. The metaphor of settlement implies a diversity of actors, some individuals, some groups, some large organizations, each with its own aims and its own local knowledge. As the settlements grow, their boundaries brush up against one another. The various participants encounter each other and interact, and evolve conventions for getting along together peaceably. Consider the settlement of North America by Europeans and the growth of our nation. Nobody "created" or "designed" the United States of America in the sense that we now know it. It was formed in a piecemeal fashion and evolved the complex structure we see today as it grew. This growth did not, however, take place in a vacuum. The various settlers drew on a variety of intellectual, legal and cultural traditions in the formation of their new nation. This, of course, is the non- constructivist model. It is also, I believe, a better model for the actual way the Net has grown up until now. Now, what do these two metaphors, these two models, imply when we think about the theme of this session, "Protections FOR the future"? A couple of things strike me as significant. First, while I believe the constructivist vision is, in a practical sense, impossible, it is nonetheless pervasive. It is a natural frame of mind for someone with an engineering background, and our field, being largely technical, has a lot of such people! It is also a predominant tradition, for reasons I haven't the time to go into here, among much of the community of people who work in the areas of public policy and government. A great deal of fuss has been made, and rightly so, over the government's recent ham-handed handling of a number of cases involving what are popularly and mistakenly called "hackers". These cases, such as the raid on Steve Jackson Games, involved truly astonishing abuses in the persecution of individuals who, in the final analysis, were not only entirely innocent of any wrongdoing but were also, in many cases, not even plausible candidates for suspicion. However, as outrageous as some of these cases are, they do not cause me enormous worry for the long run. The reason is that their very outrageousness makes them highly visible and tends to provoke a strong and determined counter-reaction, as, indeed, they have. I am far more concerned about the more subtle danger posed by such constructivist visions as Senator Albert Gore's so-called "Data Highway" initiative, which, in some of its variants, proposes to take the U.S.Government into the high-bandwidth data transport business in a big big way. If you can imagine our long-haul communications arteries being run by the same people who brought you the U.S. Post Office, you can see why I might be a little uneasy. Another frightening constructivist vision, one which seems to descend at times into Kafka-esque black comedy, is the International Standards Organization's Open System Interconnect protocol suite. For those of you unfamiliar with it, this is an attempt to draft the ultimate and complete set of communications protocol standards for the world from a standing start. The result is an unwieldy and excessively complex specification that seems nearly impossible to implement. As someone on the Net (I think it was Henry Spencer) said recently of the OSI documents, "read them? I can't even lift them!" Given the weight that ISO standards carry in much of the world, I sometime fear the foolishness of the ISO more than I fear the excesses of the FBI! To protect ourselves in the future, we need to be on guard against attempts to tear down our motley systems that nevertheless work, if somewhat less than ideally, with a singular, grandiose, all encompassing total replacement, which, its designers assure us, will make everything just right. Complex systems tend to need a lot of debugging. While we need to constantly make an effort to improve things, I think we are far better off if we proceed in a piecemeal and pluralistic fashion. Second, we should think about the intellectual traditions from which our nascent international telecommunity is arising. I earlier spoke of the growth of the United States as being organic and unplanned. However, it was organized in part by a loose set of guiding principles, such as the English liberal tradition embodied our Constitution. Many other nations grew just as organically from other traditions, with the result being, basically, a mess. For example, I think much of Latin America falls into this category. This suggests that one's starting point is an important element in one's success (or lack thereof). Fortunately, I think our classical liberal tradition, embodying such notions as the rule of law, individual rights, and private property, form a good foundation not only for our nation but for our new electronic worlds as well. Please don't throw away institutions which are known to work well. If you think you can design something better, by all means try. One of the virtues of the electronic environment is the degree to which pluralism and experimentation are so eminently practical. Just try to avoid demanding that everyone else conform to your own vision of the One True Path. If we can try many things in parallel, time will tell us what works and what does not.