Mastering Inner Space: Telecommunications Technology and Geography in the 20th Century Ronald F. Abler Executive Director, Association of American Geographers International Geographical Congress Seoul, Korea 16 August 2000 Outer and Inner Spaces Most people born after 1960 remember exactly where they were at 1:56:15 a.m. Greenwich Mean Time on 21 July 1969, when Neil A. Armstrong stepped onto the surface of the moon. I would be very much surprised if anyone here today remembers where he or she was on 24 May 1844, when Samuel Finley Breese Morse, one of the many tinkerers and scientists trying to develop a device capable of transmitting information over distance, demonstrated a workable prototype telegraph by sending the message "What hath God wrought?" a distance of 40 miles from Baltimore, Maryland to Washington DC. The Morse telegraph's practicality resided more in its binary code than in its transmission technology. It had been preceded by a series of techniques for transmitting information over distance ranging from signal fires capable of sending a single, simple message ( "the Armada is coming!") through various kinds of optical and electric telegraphs that permitted the transmission of complex message, albeit slowly. The development of a workable telegraph in 1844 and of a commercially viable telephone 32 years later sparked a sustained industrial boom unprecedented in human history, especially in geographical terms. Telegraph circuits spread over the United States and the world at a pace unmatched by any previous innovation. The electrification of human communication that began with the telegraph was reinforced by the telephone. Within two years of Bell's innovation hundreds of telephone companies had been founded under licenses issued by what was eventually to become the American Telephone and Telegraph Company, or AT&T. Out of AT&T's laboratories a century later came many of the building blocks and conceptual models of the digital computer that is the foundation of today's information and internet society. We are all familiar with the ways the outer space programs of the United States and other countries have fired the imaginations of people in all walks of life throughout the world. Boys and girls now dream of growing up to be astronauts in much the same way youngsters dreamed of becoming railroad engineers a century ago. We tend often to think of the future in terms of the science and technology of extra-terrestrial space, as exemplified in the popularity of programs such as Star Trek. Much of the art we associate with the future, for example Robert T. McCall's mural at the Epcot Center in Orlando, Florida, depicts the human past and future as inextricably linked to outer space. Today we associate high technology and great opportunity for the future with the programs begun in the 1960s to explore outer space and with the digital revolution brought on by computers and the internet in the 1980s and 1990s. Had someone of my generation come to maturity between 1850 and 1900 rather than 1950 to 2000, he or she would have viewed telegraphy and telephony as the most advanced technologies of the day, as the critical and exciting keys to an alluring future. The technologies employed to link and unite the inner spaces of worlds well known in the second half of the nineteenth century are analogous in many ways to the technologies of outer and inner spaces that have been sources of wonder to those of us alive during the second half of the twentieth century. Moreover, the geography of telegraphy and telephony in the nineteenth century set the stage for the geography of intercommunications technologies that produced today's information society. In the same way, I believe, developments in telecommunications over the last half century help us identify some of the questions that will occupy the attention of geographers interested in communications technologies over the next few decades. I will review briefly the geography of intercommunications in both the nineteenth and the twentieth centuries, to see what might be inferred about a geographical agenda for the information society on the threshold of the twenty-first century. Telecommunications Challenges and Solutions Telecommunications, as the word implies, are communications conducted over distance. My definition of "telecommunications" includes the specific techniques of mail, telegraphy, telephone, and the multiple forms of data and information transmission that use today's internet. For simplicity's sake I will focus here almost exclusively on telephony, bearing in mind that the challenges, solutions, and principles evident in telephony are virtually identical in any other telecommunications medium. The challenges to easy, inexpensive communications over distance inhere geographically in the need to move messages across distance, the need to route or direct messages to their respective destinations, and the need to perform both tasks at costs that are reasonable within times that users deem acceptable. More broadly, communication volumes must be sufficient to yield revenues that will cover the costs of operations and yield profits acceptable to entrepreneurs attempting to provide such services. More succinctly, the primary telecommunications problems-which are inherently geographic-are transmission, switching, and establishing a threshold of use that generates self-sustaining growth. A (Geographical) Paradox As an aside, I should note a distinctive quality of interconnecting networks that Thomas Falk and I labeled the "utility-penetration paradox." The dilemma faced by entrepreneurs who hope to establish intercommunications networks can be illustrated by the unlikely example of a person owning the only telephone in the world. He or she might take selfish pride and pleasure in having something no one else has, much as the private owner of a valuable painting might relish exclusive access to it. As a means of communication, however, the world's only telephone would have zero utility. Connect that telephone to another, however and it acquires value; make it possible to connect it to 10,000 others, and its utility rises sharply. The capacity of today's telephones to reach almost any other telephone on earth makes them so indispensable that many of us are never without them. For an entrepreneur trying to start a new interconnected network, the fundamental challenge is selling the first two terminals. Beyond the first pair lies the problem of selling enough additional terminals to build a viable network. A consequence of the paradox distinctive to interconnecting networks is that few have truly diffused geographically. Such networks have historically been established by governments or by entrepreneurs with access to capital sufficient to connect the threshold number of users needed to make the medium minimally viable. Transmission Returning to network fundamentals, the raw problem of moving information across distance yielded to a variety of transmission media. Mail has been carried by almost every means of transportation ever invented. In telegraphy and telephony, single strands of wire gave way to multiple wires bundled into cables, which in turn were superseded by such higher capacity channels as coaxial cable, directional microwave transmission, and fiber optic links. Each successive transmission medium could move more information among places, beginning with a single conversation on a pair of wires. Bundled wire cables could transmit the equivalent of hundreds of simultaneous conversations. The coaxial cables and microwave technologies developed simultaneously in the 1930s carried tens of thousands of conversations or their information equivalents. Fiber optic cables now carry the equivalents of millions of simultaneous conversations or even more, depending on the kind of multiplexing employed. I should note here that for well over a century, every innovation in transmission technology and capacity has been heralded as providing virtually unlimited communications capacity for the future. I note further that every forecast of unlimited capacity has been woefully shortsighted. Unlimited transmission capacity (or bandwidth) is a chimera and will remain so. Like highways, transmission media stimulate the very traffic that chokes them in short order, sparking yet another quest for a medium with greater capacity. Switching Stringing a pair of wires from each telephone to every other telephone is one polar geometry for providing access to any randomly selected pair of instruments. Beyond a network of three or four nodes, that solution becomes impractical. Even a directly connected network of 100 nodes would require 4,950 pairs of wires to be strung across the landscape, as well as some means of terminating that many wires at each terminal. Alternatively, one could connect every telephone in a network to a single central node and accomplish the interconnections there and only there. Again, even in small networks, that solution quickly becomes unwieldy and inefficient. The switching capacity required at the single interchange would be enormous, and the wiring required would be inefficient owing to the fundamental geographical fact that most requests for interconnection are local. It's not clear that telephone exchanges were part of Bell's original thinking, but they were established as early as 1878 and have been a critical component of electronic information technologies ever since. Originally staffed by young men, attending switchboards quickly became women 's work. Foreshadowing the software developers and computer scientists who have taken fiendish delight in tormenting neophytes in this century, the arrogant male operators of the late 1870s and early 1880s treated customers rudely, even profanely. By 1882 male telephone operators were history, superseded by the fairer, gentler sex possessed of a more helpful nature. Male or female, human operators switched calls by making manual connections between the pairs of wires leading to telephones with flexible cords. Being labor intensive, manual switching was an early target for automation. Electro-mechanical switches began to replace human operators about the time of World War I, and subsequent innovations largely eliminated human operators by 1970. The geography of switching offers a fascinating topic that is too complex to be pursued here. Suffice it to note that the tradeoff between constructing long transmission lines to a few switching centers or alternatively, constructing short transmission lines to many centers, governed the density and location of such centers in the past and that tradeoff will continue to influence the geometry of telecommunications networks in the future. Organization, Vision, and Agency Natural networks happen in response to basic natural forces and constraints. Telecommunications networks do not happen; they are the deliberate products of goal-oriented human activity. From the outset of telecommunications, centrally-directed coordination and planning were prerequisite to achieving communications over distance. Postal systems have been government monopolies from the Persian empire onward. In most of the rest of the world, infant telephone systems were conceived or kidnaped in infancy by national postal services, with generally negative consequences. In Anglo America, the failure of the United States Congress to exercise its rights to Morse's telegraph patents left telegraph and telephone in the private sector. Inventing dispersed operating systems was a distinctly new geographic challenge in the mid-nineteenth century, one that arose simultaneously in railroading and telecommunications in their shared infancy. As they co-evolved, the private and public networks inventing organizational and management techniques often traded ideas and individuals, progressively building geographic concepts and methods of that we take largely for granted today. Academic geographers have become devoted considerable attention to national and multinational corporations in recent decades. The roots of those organizations and indeed of the global economy lie in the transportation and telecommunications networks invented in the nineteenth century. "Make no little plans," enjoined Ralph Waldo Emerson, "They have no power to stir men's souls." The handful of key individuals who sketched out the broad outlines of telecommunications in the nineteenth and twentieth centuries certainly subscribed to Emerson's dictum. But why? What factors of personality, upbringing, education, and experience explain their abilities to articulate visions of future telecommunications networks that came to harness the work of thousands and millions of individuals, over a century of more or less progressive effort, to yield today's information society? No firm explanations have emerged from my study of telecommunications technologies, but some of the possible causes certainly intrigue. Born in 1791, Samuel Finley Breese Morse was the son of Jedidiah Morse, a New England cleric known to many of us as the author of The American Gazetteer published in 1797. It was the first American geography. The American Gazetteer and its accompanying map were widely reprinted and used during the early nineteenth century. It's difficult not to believe that his father's fascination with geography on continental and global scales influenced Morse and contributed to his devotion to developing the network that would come to move information across and among continents well before his death in 1872. Morse first made his living as a painter and photographer and founded the National Academy of Design. In 1832, he conceived the idea of an electromagnetic telegraph while returning from Europe. Morse's telegraph offered little improvement over the dozens of others extant at the time, until his development of an efficient code. Though known subsequently as Morse code, it was most likely devised by Alfred Vail, the son of one of Morse's financial backers. Vail senior was an industrialist, the owner and operator of the Morristown, New Jersey Speedwell Iron Works, where Morse built his original, three-mile experimental line. When the United States Congress declined to purchase Morse's patents he and his backers faced the problem noted above, that of building and managing what must by its very nature be a dispersed but coordinated geographical organization. Morse and Alfred Vail turned in 1845 to Amos Kendall (1789-1869), United States Postmaster General from 1835 to 1840. Kendall became the architect and builder of what was to become Western Union, the largest North American telegraph system. History and geography repeated themselves in 1885 when Gardiner Hubbard (Alexander Graham Bell's chief financial backer) hired Theodore Newton Vail to be General Manager of the infant American Telephone and Telegraph Company. Theodore Vail was a cousin of the aforementioned Alfred Vail, and had been a telegrapher in his youth. He subsequently joined the U.S. Postal Service. By 1885 he was General Superintendent of the Railway Mail Service, an operation that had by then become continental in scale. Vail became the architect and builder of AT&T. The vision of integrated, national service he carried from the postal service to AT&T guided its subsequent evolution right through the forced breakup of the Bell system in 1988. Within these broad outlines lie some fascinating details of how Samuel Morse, Alfred Vail, Amos Kendall, Alexander Graham Bell, Gardiner Hubbard, and Theodore Vail came to realize that the networks they were attempting to build must be viewed as integrated, coordinated systems of continental and even global scale. Planners and craftsmen less far-seeing than they focused on local or regional aspects of telecommunications services. Kendall and Theodore Vail in particular quickly realized that their networks must be seen as whole, as truly integrated systems in which each and every component was designed to function seamlessly with each and every other element over continental scales. Also hidden in this sketch, but well documented elsewher e, are the financial maneuvers in which almost everyone involved engaged. We are wont to think of today's dot com booms and busts as something new under the entrepreneurial and financial sun. They have numerous precedents in the stock sales and manipulations, hostile takeovers, mergers, and lawsuits that were commonplace after the U.S. Civil War in the 1860s. Everything that has happened in the United States stock markets in the last ten years would be familiar to Theodore Vail-in the two months after he joined the Bell system he raised more than $450,000 in capital for the new enterprise, mostly among personal friends. Outer and Inner Spaces Permit me to stretch the motif of outer and inner spaces somewhat further as a way of highlighting some points to which I believe we must attend as we try to tease out the agenda for geographical research and teaching on the information society. Today in contrast to the past, it is especially important to distinguish the inner space of telecommunications technologies from the outer space of the way customers use them. We will also think more clearly about the information society if we attend to the differences between the perspectives internal to geography and those of interest to geography's customers, the audiences we wish to reach in our research and teaching. Technology Producers and users invariably view telecommunications technologies differently. Producers deal with the inner spaces of their technologies, the infrastructure and operations behind the scenes where the work of telecommunications is accomplished. Consumers see the outside in the forms of the terminals via which they access technologies, the services rendered to them, and the prices they pay for using media. Accordingly, producers and consumers deal with different geographies. Traditional pricing for non-local telephone services was based solely on the distance between two telephones without regard for the substantial differences in the actual cost of providing interconnections. The network cost of linking two telephones located in small villages 1,000 kilometers apart might be fifty times greater than the cost of connecting two telephones separated by the identical distance if both telephones were situated in large cities served by bulk transmission links. Average cost pricing-charging an identical distance-based fee for both calls-shielded customers from the inexorable laws of economic geography. Those providing service could not ignore that internal geography, which had to be dealt with when planning facilities and setting prices. A general inattention to inner and outer geographies is compounded in modern telecommunications by the invisibility of components that were once obvious to customers as well as to producers. Wires and cables have gone underground. Radio signals are invisible and the antennae they use are innocuous in most places, though pitched battles are now being fought in some locales over attempts to place mobile phone towers in particularly scenic or sacred locations. My students once enjoyed field trips to telephone exchanges because of the sights, smells, and sounds associated with electro-mechanical switching. All they would see in one of today's telephone exchanges would be a large computer that doesn't even hum. Regardless of their invisibility, geographers hoping to understand telecommunications technologies will learn far more if they analyze both producer and consumer geographies. They are, have been, and will continue to be different geographies. Producers and consumers can agree, for example, that the perfect telecommunications medium would be instantaneous and ubiquitous. Indeed the entire history of telecommunications has been a quest to make intercommunication instantaneous and ubiquitous. Telephony and its offshoots have come perilously (some might say maddeningly) close to achieving those two goals. For consumers, a third desideratum in a perfect medium is that it be costless. Understandably, producers do not share that view. The real costs of many telecommunications services have dropped sharply since 1960 making them close to costless for many consumers. All the same, most customers would like to see them reduced to lower levels, and prices will remain a fundamental point of contention between producers and consumers. Geography Geographers should always be careful to specify whose geography they address in their discourse. Many intriguing geographies exist outside the usual ken of academic or even practicing geographers. An example of a producer's geography will make my point. The individual who in 1953 mapped out the sectional center/ZIP Code system for transporting mail in the United States used a Rand McNally Road Atlas and a dime store compass to plot the locations of sectional centers and the territories they would serve. He was a career postal employee entirely innocent of formal training in geography, yet he knew his business and was a superb geographer. The proof of his geographical pudding is Federal Express, which modeled its highly successful collection and distribution network on the postal service's basic framework. Dozens if not hundreds and thousands of similar empirical geographies exist, and people contend with them day after day, whether or not certified geographers pay any heed. There's no question that the world needs badly the insights only academic and professional geographers can provide. Yet it's healthy to be reminded that the worlds we ignore get along well without us. Where we risk sowing serious confusion is in failing to distinguish sharply the geographies of the space outside academic geography from the inner space of the discipline we pursue. When non-geographers come to us, they usually seek insights into the geographical problems they face. Too often, geographers tend to respond with ideas and methods more relevant to internal, abstract debates about how to conduct academic geography. In biblical terms, non-geographers come to us seeking bread and we give them stones. We should serve ourselves and our societies more effectively if we attended more to the geographies of the spaces outside our disciplinary boundaries, and if we spent less time pursuing questions of interest only to other geographers. Especially in our teaching of undergraduates, geographers would be farther ahead in the long term if they restrained their natural and even admirable wish to recruit apostles to our religion and focused more on the most trenchant things we have to say about the empirical world. Whither Geography in the Information Society? If we could first know where we are and whither we are tending, we would better know what to do and how to do it. --Abraham Lincoln The geographies of the spaces external to academic geography are ineluctable. They will abide and people will continue to occupy them with our without our tender attentions. Some professional geographers may be dismayed by the success some non-geographers have achieved in organizing space. I prefer to think that the geographic complexity of the world and its persistence, even in the information age, are the best guarantors of job security we could hope to enjoy. Given that neither the world's geography nor academic geography is likely to wither away, an assessment of what we know seems prerequisite to deciding what to do. What We Know We know that geography matters and that it is forever. Friedrich Ratzel characterized the technologies of telecommunication as raumbewältiger (conquerors of space). The popular and geographical literature of the last several decades has been heavily larded with both facile and learned forecasts of the imminent demise of distance and space. Distance has not yet been annihilated, nor will it be. Localities have not succumbed to homogenization under the impacts of telecommunications and transportation, nor will they. Even when customers or users are shielded from the internal geographies that service providers cannot escape, system planners must still grapple with them and geographers may have some useful things to say about them. Moreover, a slow but seeming inexorable trend toward the application of marginal-cost pricing for telecommunications services suggests that fewer and fewer users of telecommunications services will remain shielded from internal system geographies in the future. We know the basic principles that have historically governed the geography and geometry of interconnecting networks and that will continue to do so in the future. Tradeoffs between transmission costs and switching costs govern the layout of telecommunications networks, as well as the geography of electric power networks, interstate highways, and air transportation systems, among others. In telecommunications, transmission (or linehaul) has historically been cheap and switching dear, although both have fluctuated in absolute terms as well as in relation to each other. Their relative costs at any time shape the increments that are added to network edges or laid atop networks that are already combinations of the tradeoffs extant during previous eras of network development. We know the desiderata for the telecommunications networks of the future. They will ideally offer instantaneous access to any location, from any location, at minimal cost. Consumers will continue to try to drive the costs toward zero. Service providers will continue to insist that they need revenues sufficient to cover costs and a reasonable profit, and if not prevented from doing so by competitors, will charge whatever the market will bear. Keeping the desiderata of a perfect telecommunications network in mind as the debates between consumers and providers (and increasingly irrelevant regulators) play out will help greatly in gaining insights into what seem to be arcane issues. The issues may indeed seem arcane, but they are rooted in rock bottom fundamentals of the economic and social geography of telecommunications. We know that geographers possess in GIS one of the most powerful tools extant for mapping and analyzing the geography of telecommunications systems. Geographic information systems were originally developed to help manage geographically dispersed natural resources. No phenomena are more dispersed than global telecommunications networks. There exists, in fact, a professional society devoted to the applications of GIS to networks and especially telecommunications networks now named GITA (The Geospatial Information and Technology Association), formerly known as AM/FM (for Automated Mapping/Facilities Management) International. As GIS continues to evolved into GISc (geographic information science), geographers should find themselves in an excellent position to play key roles in analyses of information society hardware and software, and in the uses people make of that infrastructure. We know that geographers have a head start in addressing the human dimensions of information societies. The International Geographical Union established a Study Group on the Geography of Communications and Telecommunications in 1984. That group's productive work has continued though the last sixteen years and it will continue into the future. If any national groups exist that focus on telecommunications research and teaching I am unaware of them, either in geography or in related social sciences such as anthropology, economics, history, political science, or sociology. Geographers specializing in telecommunications topics may indeed be few in number, but we form a larger nucleus and are better organized than those in any other specialty. What to Do Identifying definitively the specifics of an agenda for geography with respect to the information society is a task that demands a breadth and depth of knowledge far beyond that I have been able to bring to this lecture. I believe IGU commissions and study groups would make valuable contributions to our thinking and planning within the Union were they to devote time at each of their meetings and symposia to identifying trends in their respective specialties and opportunities that seem particularly promising. With the caveat that I have not consulted widely, I suggest that research over the next several years focus on two dimensions of access. Those of us privileged to use telecommunications intensively and regularly often come to believe that access to those technologies is universal or nearly so. Though I suspect that more progress has been made more rapidly in providing general access to the internet than to any prior telecommunications innovation (over fifty percent of United States households now have internet access), many individuals and groups remain unconnected. The geography of access is what might be expected: high rates of participation in wealthy countries and very low rates of penetration in most of Africa and Latin America, and in much of Asia. If, as I believe, the distribution of talent and genius is largely random among the human population, leaving large numbers of people disconnected from the nöosphere telecommunications and the internet have created leaves us all poorer than we need be. Finding ways to make basic access more universal is a grand challenge, worthy of the best thinking we can muster. It is also one in which the insights into the nature of connecting networks geographers can provide and the GIS tools that geographers command may be particularly helpful. If one of the major challenges geographers might address is hypoaccess, I believe one of our grandest opportunities lies in hyperaccess. Geographically, the most exciting recent development in areas of the world that enjoy advanced telecommunications services is the near geographical universality of telecommunications services and the rapid progress toward ubiquitous computing. Imagine being able to assemble and integrate every bit of information that exists about a place and being able to access and use it at that place. That capacity is rapidly developing as computers, GIS, GPS (global positioning systems), the internet, and wireless telecommunications become more integrated. That capacity will enable geographers to used archived information to enhance the reality of places, whereas stored information has traditionally served as a substitute for places. I despise the terms "revolution" and "revolutionary." They have been grossly misused in connection with communications technologies. Yet mobile communications based on wireless telephony have produced broad regions in which access to intercommunications services is rapidly becoming wholly independent of location, and in which the two-way capabilities of unified telecommunications and computing technologies could reverse patterns of thought and practice that have prevailed in geography and related specialties for millennia. The increasing ability to take information about places to the places themselves rather than assembling it in laboratories or offices far removed from sites of investigation strikes me as offering great potential for improving the way geographers think. Historically, stretching back to the beginnings of our craft, geographers have enjoyed field work but done most of their analysis at a remove from the places and spaces they study. Mobile computing can be used to eliminate the distinction between the outer space of empirical problems and the inner spaces of geographic information and analysis. I rank that potential and its consequences at the top of my personal agenda for such time as I am able to devote to research on telecommunications in the years to come.