TELEGEOGRAPHY 1993 In Cooperation with the International Institute of Communications (HC) GREGORY C. STAPLE EDITOR GLOBAL TELECOMMUNICATIONS TRAFFIC STATISTICS & COMMENTARY TELEGEOGRAPHY 1993_ In cooperation with the International Institute of Communications (IIC) GREGORY C. STAPLE EDITOR GLOBAL TELECOMMUNICATIONS TRAFFIC STATISTICS & COMMENTARY Copyright © TeleGeography, Inc. 1993, except as stated. All rights reserved. No part of this report may be reproduced, stored in an information retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without written consent from the publisher. Printed in the United States of America THIS IS COPY NO. DO NOT REPRODUCE telegeography ’tel-~-j~-iig-ref~ n (1990) abbrv, of telecommunications geography [fr. Gk t~le, far off, at a distance and L. communicatus, pp. of cornmunicare to impart fr. Gk geO (earth) graphein, (to write)] 1. a new branch of geography that maps the global pattern of telephone traffic and other electronic communication flows; 2. the telecommuni- cation artefacts (radio antennas, terminals, signs) on a site; 3. the balance of telecommu- nications power in one country or region vis-a-vis another (cf. geopolitics, archaic). © TeleGeography, Inc. 1993 The Front Cover The age of the compass has passed. The mariner’s sextant and rhumb line have given way to a network of global positioning satellites, allowing anyone with a transceiver to instantly learn their latitude and longitude. Today we navigate in new electronic worlds generated by telephones and computers. We orient ourselves with a cybercompass: a computer "mouse" or telephone keypad. We dial directions and embark with the push of a button. We "click" on places, real and imaginary. And we measure the distance by how much it costs or by how long it takes for someone to answer our E-mail. Illustration: Alexandra M. Rehak, Gregory C. Staple and Marian Kester Coombs © TeleGeography, Inc. 1993 All rights reserved. -i- TeleGeographsr 1993 is published in cooperation with the IIC, an independent educational and policy research organization with members in more than 70 countries. The IIC focuses on telecommunications and broadcasting issues on a world-wide basis. Institute publications include a bimonthly magazine, .!..n.termedia, and a range of topical briefing papers. Previous reports in this series: TeleGeograohv 1992 The Global Telecommunications Traffic Report 1991 The Global Telecommunications Traffic Boom (1990) Global Telecommunications Traffic Flows and Market Structures (1989) For copies of prior reports (1989 to 1992) and other IIC publications, contact: International Institute of Communications Tavistock House South, Tavistock Square, London WC1H 9LF Tel: 071-388-0671/Fax: 071-380-0623/Telex: 24578 IICLDN G For copies of TeleGeography 1993 and editorial queries, contact: TeleGeography Suite 1000 1150 Connecticut Avenue, N.W. Washington, D.C. 20036 Tel: (202) 467-0017 Fax: (202) 467-5915 E-mail: 4579195@mcimail.com First Printing October 1993 Second Printing November 1993 ii TeleGeography 1993 We are indebted to the numerous carriers, government departments and regulatory organizations from around the world who responded to our informational requests. Special thanks are due as well to the dedicated people at these organizations and at the International Telecommu- nications Union (ITU) in Geneva who set aside the time necessary to ensure that the data reported here are as current as possible. We also wish to thank the Commission of the European Communities (CEC), Directorate General XIII, for underwriting our research on European telecommunica- tions and for permitting us to publish certain Commis- sion data on traffic to and from member states. The CEC’s independent effort to improve the scope of statistics on international telecommunications flows is ever welcome. The new maps in this year’s report and the broader national coverage were also supported, in part, by special publication grants from MCI Communications Corporation and Stentor Resource Centre Inc. TeleGeography 1993 reflects a team effort. Alexandra M. Rehak served as the assistant editor. Word processing and desk-top publishing help were provided by Barbara Frank. Bennett Moe at Maryland CartoGraphics (Columbia, Maryland) contributed his computer graphic skills to the map pages. We have tried to ensure the accuracy of the statistics presented here by relying upon primary sources whenever possible. But, in a project of this magnitude, some errors are bound to occur. We invite readers to bring any mistakes or inaccuracies to our attention so that they are not carried forward to future editions. Gregory C. Staple Editor .oo lU TeleGeography 1993 For four consecutive years, this almanac of global telecommunications statistics has highlighted new areas of the industry and uncovered emerging trends. This year, TeleGeography 1993 enters the world of computer networking with the first publication of route-by-route international traffic for the Internet -- a network of computer networks. Electronic communications -- in all its forms -- has blurred the borders of countries and continents. Nowhere is this more obvious than in the gathering and disseminating of information via computer. As we move deeper into the decade, computers and connect time become less expensive. More and more, communication across time zones is facilitated through E-mail messaging and on-line conferences or bulletin boards. And, with the interconnection of computers on a global scale, has come an explosion of new users who have entered the world of "mind and modem." Users and suppliers of telecommunications services are all looking for better ways to provide broader access to more data in more places than what is currently available. The maps, charts and tables provided on the following pages are both interesting and beneficial tools for analyzing the dynamic growth in telecommunications worldwide, and for understanding why the convergence of computer and telecommunications technology is accelerating. They also point clearly to the growing need for seamless communications that make geographical borders transparent. The information contained in TeleGeography 1993 is invaluable to researchers, government agencies and regulators, industry watchers, as well as service providers. We are both pleased to have underwritten this, the fifth edition of TeleGeography, and to recommend the power of its information to you. Eugene Eidenberg Executive Vice President, MCI C.W. Scott President and CEO Stentor Resource Center, Inc. iv TeleGeography 1993 : "-.:.-.-).:. ... :.......~:: ..... i.i:’ ’:""?:i::.:i-.."-..-: .. Preface Index of Maps, Tables, Text Boxes and Figures I. Introduction II. Review Articles and Notes Computer Networks "Who’s ’On Line’ With Whom: The New Global Internetworking Business From Computer Bulletin Boards To National Infostructure" by Anthony M. Rutkowski and others "Where In The World Is carmensandiego @ mci.com?" by Alexandra Rehak "The Ends of the Matrix" by John Quarterman "Life Is Better On The Net ... Oh Really?" by John Perry Barlow Telephone Networks "Numbering The World’s Phones: From Geography To Services" by Stewart Fist "Automated Call Back Services: An Update" by Daniel Briere and Mark Langner "Global Satellite Phones and Cross-Border Cellular Services: New Challenge For Incumbent Carriers" by Robert Frieden "Telegeography And The Explosion of Place" by Gregory C. Staple page vii 1 18 22 25 26 38 40 49 -V- TeleGeography 1993 III. IV. Maps and Tables A. TeleMaps B. Inter-Continental Transmission Facilities C. The World’s Largest International Carriers, Markets and Routes D. International Telecommunication Traffic Statistics: Countries and Routes 2. 3. 4. The Americas Europe The Middle East and Africa Asia and the Pacific Methodology and Sources Page 57 69 72 77 90 117 122 138 3. 4. 5. Ti~e TeleGeography Blue Pages A Quick Reference Section National Telecommunication Indicators (a) Population, GNP, Area, Telephone Lines (b) Mobile Phones, Fax Terminals, Intemet Host Computers (c) Outgoing Traffic, Per Capita Traffic, Balance of Traffic Index to National Telephone Codes and North American Area Codes Map of World Telephone Codes By Country and Capital City Map of World Time Zones Map of Political Geography vi TeleGeography 1993 MAPS A. Bo Co Cartograms Page 1. PetroPower 58 2. TelePower 58 3. People Power 59 4. GeoPower 59 Traffic Maps 1. Growth of International Calling (1990-1992) (a) The Americas and Europe (b) Africa & Asia-Pacific Countries 2. Balance of Telecommunications Traffic (1992) (a) The Americas and Europe (b) Africa & Asia-Pacific Countries TeleContinents 1. United States 2. Germany 3. Japan 60 61 62 63 64 65 66 STATISTICAL TABLES A. Inter-Continental Transmission Facilities Capacity and Cost, Per Voice Path, Of Selected Trans-Oceanic Cables (1956-2000) 2 Capacity, In Voice Paths, Of Trans-Oceanic Cable Systems (1986 2000) Capacity, In Voice Paths, Of Tmns-Oceanic Satellite Systems (1986 2000) 69 70 71 vii TeleGeography 1993 TABLES (Cont.) Bo Co The Largest International Carriers, Markets and Routes 4a 4b 4c 5 6 The Traffic Base Of The Top 20 International Carriers (1991 and 1992) The Traffic Base Of The Second 20 International Carriers (1991 and 1992) The Traffic Base Of The Top 20 International Carriers (1988-1992) Market Shares Of Competing International Carriers: U.S., U.K., Japan, Korea, New Zealand, Chile, Philippines and Sweden (1985- 1993) International Routes With The Largest Volume of Telecommunication Traffic (1992) International Telecommunication Traffic: and Routes Countries The Americas 7 8 9 10 11 12 13 14a 14b 14c 15 16 Argentina Brazil Canada Chile Colombia Mexico Peru United States Outgoing Traffic United States Incoming Traffic United States Annual Growth (1991-1992) Uruguay Venezuela Page 72 73 74 75 76 78 79 80 81 82 83 84 85 86 87 88 89 Vlll TeleGeography 1993 TABLES (Com.) Page The European Communi _ty (EC) 17 Belgium 91 18 Denmark 92 19 France 93 20 Germany 94 21 Greece 95 22 Ireland 96 23 Italy 97 24 Luxembourg 98 25 Netherlands 99 26 Portugal 100 27 Spain 101 28 United Kingdom 102 29 Cross-Border Traffic Between The Twelve Nations Of The European Communities (EC): (a) 1992 Statistical Matrix (12 x 12) (b) 1991 Statistical Matrix (12 x 12) 103 104 Europe (non-EC) 30 Austria 106 31 Finland 107 32 Iceland 108 33 Norway 109 34 Sweden 110 35 Switzerland 111 36 Czechoslovakia 113 37 Hungary 114 38 Poland 115 39 Russia 116 ix TeleGeography 1993 TABLES (Cont.) Page Middle East and Africa 40 Israel 118 41 Saudi Arabia 119 42 South Africa 120 43 Turkey 121 Asia and Pacific 44 Australia 123 45 Bangladesh 124 46 China 125 47 Hong Kong 126 48 India 127 49 Indonesia 128 50 Japan 129 51 Republic of Korea 130 52 Malaysia 131 53 New Zealand 132 54 Philippines 133 55 Singapore 134 56 Sri Lanka 135 57 Taiwan 136 58 Thailand 137 -X- TeleGeography 1993 MISCELLANEOUS TABLES, FIGURES AND TEXT BOXES The Market For International Calls: Traffic Volume (1986-1995) Origin of International Telecom Traffic (1992) Who’ s "On Line" With The U.S. Monthly Internet Traffic To and From NSF Net (June 1993) Who’ s "On Line" With Europe? Monthly Traffic To and From CERN (March 1993) Internet Global Traffic Growth: Increase In Internet Networks and Traffic Volumes From And To NSF Net (Sept. 1991 to June 1993) How Big Is The Matrix? Traveler’s Tales ... The Net Cyberspace The Metaverse Telephone Numbering Zones Of The World Ex-U.S.S.R. Telephone Codes The Future of The North American Numbering Plan (NANP) The Information In A Telephone Number Dialing Past the Receptionist What’s been left out of the numbering debate? Using Hubbing In A Call Back Arrangement Estimated Size of Call Back Market (in Accounts) Inmarsat’s Current Mobile Satellite Service How Satellite Phone Calls May Be Routed Major Satellite Phone Ventures International Cellular Roaming: Settling Accounts Mobile Satellite Services: Settling Accounts Page 1 2 12 14 16 20 20 24 24 27 29 30 33 35 39 39 41 42 44 45 46 xi TeleGeography 1993 Introduction The report contains a wealth of other information: Introduction TeleGeography 1993 marks the fifth anniversary of this report. Details on earlier editions appear at p. ii before the acknowledgements. The inaugural edition concentrated on foreign telephone traffic from a handful of industrialized nations. This year’s report provides route- by-route data on international telephone circuit traffic for over 50 countries and offers comparative statistics on as many carriers. ~7~o TeleGeography nowo~i charts approximately 80- 95 % of the 5~ international calls traversing the world’s public telephone 3s- networks. For 1993, 3o we estimate that the 2~- world-wide market for ~o- international calls will total 47.5 to 48.5 billion minutes (see o- Figure 1) with each minute generating approximately $1 in carrier revenues, on average. -, A quick reference section -- the TeleGeography Blue Pages -- lists basi~ communications indicators for more than 50 countries (e.g., telephone lines, fax machines, cellular phones, international traffic balances). In addition, the Blue Pages, at the back of the report, include reference maps showing national telephone dialing codes, time zones and political boundaries. Figure 1 THE MARKET FOR INTERNATIONAL CALLS Traffic Volume (1986-1995) Updated tables have been compiled on the transmission capacity and cost of the principal trans- oceanic cable and satellite systems to the year 2000. Current market share data are provided for carriers in countries with competitive markets. 1986 1987 1988 1989 1990 1991 1992 1993 1994 © TeleGeography, Inc. 1993 New telemaps have been prepared to highlight country-by-country differences in international traffic 1995 growth and national traffic balances. The map section also includes cartograms on the world’s major telecontinents. TeleGeography 1993 also presents comprehensive statistics on the volume of trans-national data traffic (file transfers, electronic mail, database queries) handled by the world’s largest public computer network -- the Internet. Outbound and inbound Internet traffic statistics and growth rates are provided for the top 25 U.S. routes; similar data are stated for Europe. With this new database, TeleGeograph¥ provides a tool for mapping the basic contours of computer-to-computer as well as phone- to-phone communications throughout the world. The country-by-country statistics in this report show that the telegeography of the early 1990s is still largely a reflection of the traffic among industrialized countries. See Figure 2. But, the number of telephones in developing countries is rising. (From 1995 to 2000 China is expected to add more telephones each year than the U.S. or any Western European nation.) Likewise, the old rules regarding market entry, traffic routing and pricing in the industrialized world are beginning to bend as a new generation of competitive carriers comes of age. 1 TeleGeography 1993 Introduction These changes are already shifting the world’s major telecontinents closer to Asia and the newly industrialized countries astride the Equator (Malaysia, Indonesia, Taiwan). And, within a decade, the resulting traffic patterns are likely to birth a new telegeography with its own distinctive topography. As with past issues, several essays offer readers a perspective on the TeleGeography database. This year’s articles address global computer networks; telephone numbering plans; and international mobile telephony (satellite phones and cellular roaming). However, before turning to these subjects, some general points bear mentioning. A. Interpreting The N-tubers First, the international traffic statistics for 1992 confirm that economic growth continued to slow last year, not only in the English speaking countries, but worldwide. Global traffic growth in 1992 was approximately 12-13%; down from 13-14% in 1991 and 15-20% in the late 1980s. Further, the data in TeleGeo~ show that, in rich and poor countries alike, cross-border Minutes of TelecommunicationTraffic (MITT) are a fairly reliable indicator of basic economic trends and the differences in regional growth rates. For example, from 1990 to 1992, international MiTT grew approximately 11% in Western Europe as compared to 14% in the Americas and 25 % in Asia. Second, in the late 1980s, a buoyant world economy allowed the major carriers to "grow down" as competition was introduced because the size of their markets was expanding more rapidly than new carriers could win market share. The 1992 statistics show that this situation has changed. In the U.S., the U.K. and Japan, the volume of international traffic added by new carriers was, for the first time, greater than the volume of traffic added by the incumbent carrier (i.e., MCI added more traffic than AT&T; Mercury added more traffic than BT; IDC and ITJ together added more traffic than KDD). This suggests that head-to-head competition for international market share will be the reality for most carriers in the years ahead as markets grow more slowly. It also implies that the half dozen new international carriers which will cut their teeth in the _early 1990’s in Sweden, Australia, Malaysia, Indonesia, Israel and Chile -- may have a harder time than first generation. Third, although the macro picture painted by global MiTT is reasonably clear, some caution must be taken in interpreting year-on-year changes in MiTT on particular routes. One reason is that TeleGeography only measures MiTT on the Public Switched Telephone Network (PSTN); the traffic data exclude international private (leased) lines which are not metered by carriers. The share of international traffic carried on leased lines probably accounts for 10-20% of total traffic on routes with substantial business traffic (e.g., the U.K.-Japan; Germany-Netherlands). No more precise figures are ORIGIN OF INTERNATIONAL TELECOM TRAFFIC (1992) N. Amodca © TeleGeography, Inc. 1993 2 TeleGeography 1993 .... : ..... :." :. :: i::iilI :i .. Introduction generally available. However, as countries begin to relax the limitations on reselling leased line capacity to the general public for providing switched services, the share of total telephone traffic carried by private lines is likely to increase and, along with it, the need for more accurate measurements. The rapid growth of Country Direct services since 1990 also is affecting the route-by-route statistics reported in TeleGeography. The main reason is that, since 1991, the only traffic statistics available to TeleGeography for the U.S. count Country Direct traffic as originating where it is billed (that is, in the U.S.) rather than where the call is placed (outside the U.S.). Country Direct traffic billed by U.S. carriers is thus considered as outbound not inbound U.S. traffic. In contrast, carriers in the rest of the world have generally reported traffic data to TeleGeography based upon the "flow of traffic" (where the traffic originates) not on the "flow of funds" (where or how the carrier is compensated). The introduction of Country Direct calling plans thus has: tended to (a) inflate the volume of outbound MiTT reported from the U.S. on some routes; and (b) led to some double-counting of outbound calls because a Country Direct call to the U.S. may be reported as "outbound" traffic where it originates and terminates. The net effect of Country Direct traffic in country statistics can not be fully determined, however, without taking into account the Country Direct calling plans of non-U.S, carriers as well. We estimate that the proportion of Country Direct calls is now at least 15-20% on major U.S. routes and 5% to 10% on routes served by European and Japanese carriers offering similar programs. Regulators in the U.S. have been asked (but have not yet agreed) to compile traffic data from carriers based on the direction of the call as well as the billing point. Similar demands are likely to be made elsewhere as trade negotiators and industry groups seek a more accurate picture of the innovative payment mechanisms which are driving the over $50 billion market for international telephone services. Global Computer Networks: Mapl~ing The Matrix Today’s telegeography is being shaped by computer networks as much as by telephony. The installed base of computers -- approximately 140 million worldwide and rising -- is rapidly being linked. Moreover, telephones and computers are becoming more alike as computers take on fax boards and modems, and telephones gain display screens and enhanced keypads. Mapping who is "on line" with whom around the world thus becomes a more and more important part of any telegeographic enterprise. The statistics necessary to begin this survey are slowly becoming available. Although substantial transborder data traffic is routed via private networks run by airline, banking and other industries, basic metrics do exist for the traffic on the largest public networks. Anthony M. Rutkowski, Vice President of the Internet Society, begins our survey of global data highways. The survey is supplemented by John Quarterman, the founding Editor of Matrix News in Austin, Texas. Much of what we know about who is "on line" to whom in the world is based upon traffic flows over the Internet -- a loose affiliation of computer networks with common access protocols -- which now connects over 11 million terminals in over 60 countries. It is the world’s largest and most rapidly growing data network. The Internet traditionally has served the educational and scientific communities. Today it is fast becoming a common network for commercial and household users as well. Indeed, much of the Internet’s recent growth can be attributed to the rising number of commercial gateways. These gateways make it increasingly possible for E-mail users anywhere in the world to correspond even though they subscribe to different computer networks. See p. 18. 3 TeleGeography 1993 A number of the Internet’s more creative users have begun to distribute faxes and video clips; a weekly radio talk show is also available via the Internet. In a very few years, these trends may transform the Internet from a meeting place for global bitheads and other assorted digiterati to a popular medium which competes with the public telephone network. Rutkowski’s article is followed by a special contribution from John Quarterman. For those readers who find it helpful to keep one eye on their old wall maps while exploring cyberspace, Quarterman’s personal travel log, "The Ends Of The Matrix", is the place to begin. The electronic frontier has proven to be a fertile terrain for fiction as well. William Gibson’s trilogy (Neuromancer, Count Zero, Mona Lisa Overdrive) and Marge Piercy’s He, She and It are perhaps the best known. One allure of this new world is obvious. Literature traditionally has taken the landscape as a given, even if the action takes place on another planet or galaxy. But in cyberspace, geography is simply a computer artefact (software) and thus malleable. In this Alice in Wonderland world, the imaginary is the real; the real can be imagined; and the imagined can be coded. That is why some of the leading cartographers of this new space are as likely to be novelists and computer programmers as social scientists and statisticians. C. Telephone Numbering Plans Telephone dialing codes are the border guards of the information age. They determine what can get in and out of any locale. They mark a nation’s sovereignty as much as its flag or its currency and are often better known. Without a code of one’s own, as the would be telestates of Central and Eastern Europe quickly discovered, one’s telephone traffic must be routed via another country’s network. At best, this is inconvenient; at worst, it can lead to a national security nightmare. The International Telecommunications Union (ITU), the Geneva-based U.N. standards body, has now assigned telephone country codes to the Baltic States; Introduction Azerbaijan and Moldova are also slated to receive individual country codes which will separate them from the rest of the former U.S.S.R. But, just as we are learning these new codes, it seems that the traditional, locale based numbering scheme is about to slip its geographical moorings. In the second major article, Stewart Fist, an Australian journalist, explains why. For much of the last century, telephone numbers have offered a coded map of the world with a fairly transparent public key. Any (listed) telephone number begins with the country code; it is followed by the code for the regional or city exchange and then the code for the local exchange and subscriber line. But, as Fist advises, "smart" switches, competing carriers, freephone ("800") services and the next generation of personal (wireless) communicators are forcing governments to revise the current numbering scheme. The debate is particularly intense where competing carriers have emerged because control of a nation’s numbering plan can convey a significant market advantage. This has led to growing political pressure from new carriers to ensure that national numbering organizations -- the Information Age equivalent of boundary line commissions -- operate at arms-length from the dominant carrier. As yet, the numbering debate has focused largely on traditional concerns. However, computer technology not only allows a departure from the existing zone- based numbering system; it might also allow telephone numbers to provide much greater choice to users. For example, a dialing suffix might be used to direct a call to a fax or data terminal on the same line or to select a lower grade of service (bandwidth) with a commensurate reduction in price. Paradoxically, although telephone numbers allow us to work as nomads, they also root us. We say "my" telephone number because, though we do not own it, the number is tied to our phone, which we (often) do own. More importantly, we associate our telephone number with the private space which we create when 4 TeleGeography 1993 Introduction we use the phone. Telegeography suggests that the rapid expansion of international communications has made our sense of place and of community whether it is generated electronically or anchored in concrete ever more important. (See p. 49.) Future telephone numbering systems that do not take account of this may not be long lived. D. Industry Structure The telegeography of any age is a function of regulation as much as technology. Telecommunication cables and satellites need know no boundaries; governments are defined by the boundaries they keep. The tension between telecommunications technology and national sovereignty has sharpened since the mid 1980s as the pressure for market liberalization -- licensing competing carriers -- has spread. In the 1990s, the vast pools of national telecommunications traffic, once a country’s patrimony as much as its forests or its mines, have become the subject of fierce multinational bidding. The articles section in TeleGeograph¥ 1992 discussed how the growth of Light Carriers -- carriers which resell, repackage or reprogram the services of facilities-based or Heavy Carriers -- has begun to change the paradigm for international service. The Heavy Carrier, which typically has a monopoly or oligopoly franchise, provides service by connecting its half-circuits with the matching cable or satellite facilities of another monopolist. With no protected home market and few facilities of its own, the Light Carrier seeks to provide end-to- end service on a global basis. Because Light Carriers are driven by software rather than hardware, they have come into their own as the basic building blocks for international service -- electronic switches and long distance transmission capacity -- become more and more abundant. In the U.S., for example, there are now 19 facilities-based international carriers and over 150 resellers, including several automated call back companies. Yet, as AT&T recently advised the U.S. Government, there is still substantial excess international transmission capacity. "In fact," said AT&T, "currently installed or planned ... submarine cable facilities ... could accommodate a tenfold increase in traffic demand." And that situation is likely to continue. The undersea cables which will come into service in the 1995-2000 period will have a capacity of 5 Gigabits per second (Gbps) or more (i.e., enough to handle at least 300,000 simultaneous phone calls). Yet, according to AT&T, wavelength division multiplexing (WDM) can potentially put "up to 1000 channels of 1-10 Gbps each on a single fiber"! The transition from a supply constrained to a demand driven world populated by a mix of Heavy and Light carriers is prompting a number of regulatory conflicts. One of the most bitter involves automated call-back or "boomerang box" companies, which offer reverse billing of international calls from high tariff countries. Most of these companies are based in the U.S. and the Federal Communications Commission (FCC) in Washington, D.C. has been asked to declare the call set-up practices of these companies illegal. Several South American countries have already done so, although at this writing the FCC has yet to act. Despite this legal cloud, a new report on the call back business by the TeleChoice consultancy, (see p. 38) suggests that these novel service providers will be hard to dislodge. Our third major article explores another area of controversy between established carriers and new entrants: cross-border wireless telephony. Rob Frieden, a Professor at Penn State University and a former counsel to several entrepreneurial carriers, focuses on two businesses: satellite phones and trans-national cellular roaming. Frieden notes that while mobile-to-mobile calls now account for a very small percentage of long distance traffic and international mobile calls are a rarity, change is on the horizon. Several major cellular radio companies (e.g., McCaw) are now allied with or owned by leading 5 international telephone companies (e.g., AT&T) and at least two global satellite phone ventures (Iridium and Inmarsat’s Project 21) seem likely to be launched. These developments, argue Frieden, are putting new pressure on the regime which wireline carriers now use to settle international accounts. The current accounting rate regime for international telephony has been widely criticized for sustaining excessive toll charges. This regime also is at odds with the mechanisms being developed to compensate carriers for providing cross border "roaming" service to cellular subscribers. In these circumstances, Frieden suggests that the debate over trans-national mobile services may well point the way toward alternative settlement and access charge regimes for the international telephone industry as a whole. Introduction E. TeleGeography 1994 Our plans for TeleGeography 1994 are already on the drawing board. We began by mapping the world’s telephone and data flows -- tWO projects which we will continue to refine. Next year, we plan to turn our attention to the developing market for global V-mail (video mail) and teleconferencing. We also want to get back to the basics by taking a closer look at the major demand drivers for international communications -- economic, social and cultural -- and how the most successful carriers have tailored their services in response. As always though, we welcome the assistance of our readers and invite your comments and suggestions. Gregory C. Staple September 1993 Washington, D.C. -6 i ARTICLES AND COMMENTARY TeleGeography 1993 WHO’S "ON LINE" WITH WHOM" THE NEW GLOBAL INTERNETWORKING BUSINESS FROM COMPUTER BULLETIN BOARDS TO NATIONAL INFOSTRUCTURE By Anthony M. Rutkowski and others Computers have been talking to other computers for more than a generation. For much of this time, however, there were three separate networking words. One world was mainly occupied by corporate users who needed high capacity, dedicated networks to handle the day-to-day demands of processing very large volumes of financial transactions, customer orders, inventory records and the like at geographically dispersed sites. A second networking world was created by the non- profit sector -- government, universities and research institutes. In contrast to the corporate world, these networks prided themselves on being fairly open and non-structured, allowing people and computers to build a practical set of internetworking tools to accommodate their varied interests. The personal computer boom in the 1980s led to yet a third networking world as millions of new users came "on line." Although the Personal Computer (PC) based networks overlapped in large part with the non- profit nets already in place, they were also serviced by a new breed of commercial vendors, including Compuserve, Delphi and thousands of smaller computer Bulletin Boards Services (BBSs). Anthony M. Rutkowski, Vice President of the Internet Society and Director of Technology for Sprint, was the principal author. Daniel P. Dern (ddern@world. std.com), author of The Internet Guide For New Users (McGraw-Hill, 1993) and Gregory C. Staple (gstaple@mcimail.com), a communications lawyer at Koteen & Naftalin, Washington, D.C. also contributed. © TeleGeography, By the early 1990s, these three networking worlds had begun to converge, making the world of computer networking at once more plural and more alike. At the center of this convergence is a new global internetworking business. One computer world is being linked with another and new electronic gateways and services are being opened to a wider and wider public. In the process, the specialized research and PC driven networks of yesterday are all becoming a vital part of tomorrow’s national infostructure. The Past As Prologue Not very long ago, computer networks were based on large central mainframes, with remote "dumb" terminals connected by basic public data networks. This universe became a little more varied when minicomputers and supercomputers arrived. But the same basic networking model was retained; a central machine was surrounded by a cluster of computer terminals. And on the telecommunications side, most carriers designed their products to support this simple world of monopoly providers and remote computer processing. However, the computer world did not stand still. PCs and work stations began to very quickly shrink mainframes down to desktop size. Local Area Networks (LANs) and a vast array of computer operating systems and applications also entered the market. Some of these small but powerful computers were designed to communicate and share resources simultaneously with large numbers of other computers using "client-server" models. The slogan, "The Computer is the Network," was born and along with it the business of internetworking. The basic internetworking technology was largely Inc. 1993 8 TeleGeography 1993 :. :. devised in the early 1970s by researchers working for the U.S. Department of Defense Advanced Projects Research Agency (DARPA). The networking "glue" to accomplish this integration was a set of computer protocols known as TCP/IP (Transmission Control Protocol/Internet Protocol). These protocols provided standards for diverse computers to talk to one another in a fast and efficient manner. It didn’t take long before companies trying to integrate their own computer networks began using the TCP/IP technology and it diffused quickly into the private sector. In a matter of years, the collection of DARPA based networks called the "Internet" grew from a handful of computers to several thousand computers in hundreds of locations. Then, in 1985, in what is sometimes referred to as the Second Great U.S. Communications Divestiture (the first being AT&T’s separation from its local operating companies), DARPA kicked its child out of the house to find civilian and commercial step parents. Many stepped forward. Soon government, scientific and academic organizations throughout the U.S. and Europe all began to help grow the Internet. At about the same time, various large companies also began using the ICP/IP technology in their dedicated enterprise networks. A few thousand interconnected computers became a few hundred thousand by the end of the 1980’s. Defining Internets and "The Internet" The distinction between internets and the network of internets called the Internet is a source of endless confusion. The term "internet" (lower case "i") refers to any network using "open" computer protocols that allow dissimilar kinds of computer systems to communicate with each other. For all practical purposes, the Internet Protocol (IP) is the predominant way of accomplishing this, although there are many smaller scale implementations using proprietary and public domain protocols, including some proposed under the name Open Systems Interconnection (OSI). The Internet Protocol is best envisioned as a kind of universal overlay network that rides on top of the basic telecommunication network. It makes little difference if the underlying network is telephone wires, radio or fibre optics cable. So long as the transmission medium moves electrical signals from one point to another, it Mapping The Matrix can carry the IP protocol. Similarly, for IP communications, it makes little difference whether the computers are mainframes, supercomputers, minis, Apple Macintoshes or IBM compatible PCs. This incredible ability to link together directly any kind of computer or process running on a computer with that of any other computer or process anywhere represents the enormous value of the Internet Protocol and accounts for its wide implementation. Most of the newer enterprise networks being constructed by corporations today are IP internets. IP networking has long been provided as part of every AT&T UNIX computer operating system and has recently become a standard feature of most other computer operating system now made, including the popular Windows for PCs. At September 1993, there were more than 50,000 internets registered for unique network addresses. This base was growing by more than 500 new internets per week. The Internet consists of about one-third of these internets. This IP network of networks is also richly endowed with gateways to countless other specialized networks, especially those for electronic mail. This larger agglomeration, which includes Bitnets, Fidonets, UUCP Nets, MiniTel based nets, and virtually every public messaging system in the world, has been annointed "The Matrix" by John Quarterman, a popular analyst in the field. The Matrix encompasses an estimated 20 million users. (See p. 20 below.) The very notion of such a massive unbounded communications medium may be difficult to comprehend. Most networks -- for buses, airlines, even telephony -- are reasonably discrete. But, in some ways, the Internet is similar to the more familiar radio spectrum which is also an unbounded and loose collection of millions of users all sharing a common medium. The Internet Is Not Just For Electronic Mail One of the most common myths about the Internet is that it merely provides an inexpensive way for academics to exchange e-mail. In fact, rather a small fraction of the traffic -- less than 20 percent -- is -9 TeleGeography 1993 ~’ ’: ..... e-mail. The Internet currently supports more than 1000 different kinds of computer services, although only about 400 are defined, registered services, and only a few dozen are in widespread use. The rest are largely experimental or provide specialized gateways between different kinds of computer networks. Approximately 50 percent of the Internet traffic is file transfers. The Internet allows extremely fast and easy movement of files, including the "binary" files that constitute the software code used by computer programs. Not surprisingly, some of the Internet sites with the heaviest global traffic are those used for distributing new versions of UNIX software. A partial listing of the accessible information on the Internet includes: Real-time access to databases, indexes and interactive programs. These include over two million public-domain and shareware programs (which can be down-loaded to the user’s own computer), plus documents, images and other files at over 2,000 archive sites; Over 2,000 "Gopher" servers, providing access to over 1 million files, services and resources, structured by topic, organization or location. (Gopher is a customized search and data retrieval program.); Several hundred full-text-searchable databases accessed through WAIS (Wide-Area Information System), a search engine developed originally by researchers at the Boston based Thinking Machines Corp.; Over 700 on-line library catalogs with a mix of record entry, abstract and full-text items; Government archives, thousand of special- interest BBSs, document repositories and information services; Remote-log on access to supercomputer centers and to leading on-line information services such as Dialog, Dow Jones, and MEAD Data; and ¯ Usenet News newsgroups (from 3,000 to Mapping The Matrix 5,000 topical discussions open to anyone, and an unknowable number of private, by- invitation-only, discussions). Other heavily used capabilities include directory services, remote log-on, remote mounting of computer drives, network management and dozens of new browsing and knowledge tools. One of the best known examples of the latter are "Knowbots®’’ (for knowledge robots) that search the information resources of the Internet at the behest of their human masters. (The concept is the brainchild of the "father" of the Internet and Internet Society President, Vinton G. Cerf.) Finding Your Way In The Internet The most rapidly growing new knowledge service on the Internet, Gopher, is named for the mascot of the University of Minnesota. This research tool, although only two years old, has spread to thousands of sites worldwide and its traffic is growing at an annual rate exceeding 1000 percent. Gopher is best envisioned as an information kiosk where one can use a joystick or "mouse" to browse by selecting geographic-markers or information key words. Its interface is so basic that even young children can use it. Numerous private companies, libraries, museums and government agencies have created Gopher kiosks to provide public information, white page directories, product information and sales brochures. Some entrepreneurial publishers even provide abstracts of their books and the ability to instantly place an order vi___~a these kiosks. New multimedia services which permit audio and video multicasting also are beginning to appear. The Internet Society, among others, has used these services for the global conferences needed to establish Internet standards. Currently more than 500 sites in 20 countries participate interactively in Internet standards- making conferences via two "tv channels" which are multicast worldwide. As new multimedia standards for Internet mail come into widespread use, the distribution of images, sounds, complex graphics and texts with non-Roman character sets is expected to grow rapidly. Some popular uses of the Internet entwine several 10 TeleGeography 1993 capabilities under such headings as "networked collaboration" or "collaboratories." These services enable researchers or specialists in a given field to form virtual committees to pursue common interests. Another Internet capability is a network management service for highly distributed networks with many different kinds of computers and devices. Actually a collection of tools, the Internet Simple Network Management Protocol (SNMP) has been adopted by both the computer networking and telecommunications worlds. Some of SNMP’s more unusual implementations include the control of remote stereo systems, toasters and soft drink vending machines. Economics of the Internet A second important but frequently misunderstood facet of the Internet is the network’s economics. It is not uncommon to hear remarks about the Internet being "free." The Internet is NOT free. However, because of its architecture, technology and accounting practices, the costs are low and often shielded from some end users. The Internet consists of 17,000 networks (at August 1993) and the costs for connectivity and operation are collectively borne by all those networks. A great many of these networks are institutional enterprise networks, operated as part of the normal business of companies, academic institutions and government agencies. There is also a "no settlements" policy; networks nearer to the end user purchase capacity from a higher level backbone network on a flat rate, not a volume sensitive basis. Lastly, the connectionless technology employed by the Internet allows extremely efficient sharing of transport and switching facilities. Fee structures typically range from a low of $20 per month for basic service by dial-up modem to $2000 per month for service via a 1.5 Megabit leased line connection. Global carriers such as Sprint allow resellers to market the services and offer a multitude of different access options including free phone ("800") numbers and X.25 data services. Prices usually tend to be higher outside the U.S. due to the lack of competition in providing circuit capacity. Commercialization of the Internet Most of the registered internets today already are commercial entities. And most of the growth today of interuets and the Internet is dramatically commercial.- The recent announcements by all of the major telephone carriers in the U.S. to provide Internet services underscores this trend. The rapidly rising demand for services by major business customers is another sign of the Internet’s changing character. In 1994, the trend will doubtless accelerate as the U.S. government withdraws subsidies from its core backbone network, the National Science Foundation (NSF) Net. Some lament the transition to a more public global and commercial environment. Yet, because the Internet now encompasses so many diverse uses, users, and technologies world wide -- mirroring society -- it is likely that a rich mixture of networking cultures will continue to coexist. Analyzing the Tables and Charts The tables in this article and the charts which follow display some of the major relationships and trends for internetworking. The Internet community frequently refers to this information as metrics. The Internet world of metrics has both positive and negative features. One constant problem with Internet metrics is completeness. The Internet’s architecture, involving 17,000 or more networks, is ever growing. The good news is that much of the traffic proceeds through a relatively few network backbone computers which keep very detailed and comprehensive data on traffic flows and use. The Internet Society, in Reston, Virginia, consistently publishes different renditions of the data as it becomes available. Figures 1 to 5 in this article reflect the metrics that are seen at the U.S. NSF Net backbone and the largest European backbone at CERN (European Center For Nuclear Research) near Geneva, Switzerland. The Internet is a highly disaggregated world of distributed networks whose backbones route traffic for all the networks which are made known to them. For example, many commercial internets which use only commercial Internet back bones (e.g., PSI, ANS (Cont. on p. 16) -11- TeleGeography 1993 Mapping The Matrix Figure 1 Who’s "On-Line" With The U.S.? Monthly Internet Traffic To and From the Largest U.S. Correspondent Network in Megabytes (June 1993) NOTES: N.Amm~ca Europe Mid. East Africa Asia Counby Networks Inbound Outbound Inbound Outbound Megabytes Megabytes % of Total % of Total Un~tad States 5,576 6.057,255 5.445,963 53% 47% Canada 429 116,703 277,418 30% 70% Mexico 34 5,892 52.274 10% 90% TOTAL 6.039 6,179.850- 5,775.656 52=/= 48% Brawl 52 1.873 9,592 16% 84% ChLOe 14 3,351 7,484 31% 69% Argentina 1 62 3,065 2% 98% Venezuela 5 510 1,890 21% 79% Costa Rica 2 33 1.098 3% 97% Ecuador 28 126 599 17% 83% TOTAL 102 5,954 23.728 20% 80% United Kingdom 420 69.838 169,383 29% 71% Gem~ny 443 47,511 112,331 30% 70% France 453 89,776 80,677 53% 47% Netherlands 131 52,166 48,870 52% 48% Italy 169 13,475 47,699 22% 78% Switzerland 87 40,880 45,218 47% 53% Sweden 87 32,202 39.341 45% 55% Finland 102 50,763 27,171 65% 35% Noway 52 15.026 23.745 40% 60% Aust~a 84 9,052 23,120 28% 72% Spain 39 2,433 23,058 10% 90% De~mark 8 7,232 13,613 35% 65% Belgium !4 1,225 7,196 15% 85% Ireta~ 24 802 4,445 15=/= 85% Czach Repub~c 51 1,076 4,243 20% 80=/= 11 640 3,706 15% 85% Po~ 35 999 3,586 22% 78% Tudmy 9 489 3,226 13% 87% Hungary 20 479 3,177 13% 67% Poland 45 844 3,061 22% 78% Slovalda 13 489 1,453 25% 75% Iceland 13 377 1,413 21% 79% Luxembou~ 4 40 278 13% 87% TOTAL 2,353 439.335 693,264 39% 61% Israel 48 6,737 17,454 28% 72% Kuwait 1 12 330 4% 96% TOTAL 49 6,750 17.764 28% 72% South Africa 42 1.656 9,723 15% 85% Tunisia 1 2 5 30% 70% TOTAL 43 1,658 9,728 15% 85% Australia 189 59,348 93,818 39% 61% Japan 257 19.475 41.825 32% 68% Taiwan 78 33,222 40.593 45% 55% Singapore 16 3.417 19,019 15=/= 85% South Korea 35 4,322 17,737 20% 80% Hong Ko~g 8 1,682 13,988 11% 89% New Zealand 50 2,031 7,526 21% 79% India 3 1,375 1,499 48% 52% Thailand 13 258 883 23% 77% Malaysia 3 87 423 17% 83% TOTAL 653 125.216 237.311 35% 65% The table is based on traffic to and from the U.S. National Science Foundation (NSF) Network, presently the largest Internet backbone. It serves as a global traffic exchange mechamsm among the 9,239 networks =ndicated. The basic data is available on the network server: nis.nsf.net. Some smaller country traffic is not shown, but the networks and traffic are counted in the regional totals. © TeleGeography, Inc. 1993 12 TeleGeography 1993 :i.:iii:.i’::i: :ii::i:.:..i....:..,:.~.._. ’,.....,:.. ...... Mapping The Matrix Who’s "On-Line" With The U.S. lnternet Traffic to and From The Largest U.S. Correspondent in Megabytes (June 1993) USA Canada Mexico Brazil Chile Argentina Venezuela Costa Rica Ecuador United Kingdom France Germany Netherlands Switzerland Finland Sweden Italy Norway Austria Spain Denmark Belgium Ireland Czech Republic Greece Portugal Turkey Hungary Poland Slovenia Slovakia Iceland Luxembourg Israel Kuwait South Africa Tunisia Australia Taiwan Japan Singapore South Korea Hong Kong New Zealand India Thailand Malaysia Total traffic transiting to and from the NSFNet backbone (Megabytes) 50,000 100,000 150,000 200,000 250,000 300,000 350,000 _ I Inbound traffic to NSFNet I I I I Outbound traffic from NSFNet ,,,,~ r r How Big Is a Megabyte? 250 pages or 400 messages 250,000 pages 1,000 MBytes 1 MByte © TeleGeography, Inc. 1993 Figure 2 400,000 11,500,000 13 TeleGeography 1993 Mapping The Matrix Figure 3 Who’s "On-Line" With Europe? Monthly Traffic To and From The Largest European Correspondent (CERN) in Megabytes (March 1993) N.~rnedca Europe M.East Asia Country Inbound Outbound Inbound Outbound Megabytes Megabytes % of Total % of Total United States 195,000 104,000 65% 35% Canada 4,200 6,700 39% 61% Mexico 500 300 63% 38% TOTAL 199,700 111,000 64% 36% Switzerland 77,400 63,200 55% 45% Germany 41,400 59,300 41% 59% Italy 20,400 79,800 20% 80% Netherlands 38,300 44,800 46% 54% France 27,500 30,300 48% 52% Austria 17.200 35,400 33% 67% United Kingdom 13,600 14,600 48% 52% Sweden 5,400 7,300 43% 57% Spain 1.800 10,100 15% 85% Finland 7,600 3,400 69% 31% Hungary 1,400 4,600 23% 77% Czech Republic 1,100 5,100 18% 82% Norway 1,900 3,900 33% 67% Poland 2,100 3,100 40% 60% Denmark 700 1 °800 28% 72% TOTAL 257,800 366,700 41% 59% Israel 0 9,800 0% 100% Australia 4,300 3.000 59% 41% Taiwan 1,200 1,300 48% 52% Japan 1,000 1,500 40% 60% TOTAL 6,500 5,800 53% 47% NOTES: The CERN (the European Center for Nuclear Research) is presently the largest European traffic exchange point for the Internet. It serves as a global traffic exchange mechanism for even more networks than the U.S. NSFNet. This data was collected on a one-time basis and is derived from J.M. Jouanigot O. Martin, and J. Yu, "IP Traffic Measurements and Analysis at CERN," Proceedings of the International Networking Conference (INET ’93) San Francisco, CA 17-20 August 1993. The data is also available on the network server: mordor.standford.edu. Traffic from smaller countries is not shown. © TeleGeography, Inc. 1993 14 TeleGeography 1993 Mapping The Matrix Who’s "On-Line" With Europe? Internet Traffic to and From The Largest European Correspondent (CERN)in Megabytes (March 1993) Figure 4 Total traffic transiting to and from the CERN Exchange (Megabytes) 50,000 100,000 150,000 200,000 250,000 300,000 United States Canada Mexico Switzerland Germany Italy Netherlands France Austria United Kingdom Sweden Spain Finland Hungary Czech Republic Norway Poland Denmark Israel Australia Taiwan Japan Outbound traffic~l from NSFNet J How Big Is a Megabyte? 250 pages or 400 messages 1 MByte 250,000 pages I 1,000 MBytes © TeleGeography, Inc. 1993 15 TeleGeography 1993 Mapping The Matrix .. ’ .... .. +.. :.:.: ii.’..~:i:ii.’. -.-" "~::!.i. ,’ Internet Global Traffic Growth Increase In Internet Networks and Traffic Volumes From and To the NSF Net Backbone in Megabytes (Sept 1991 to June 1993) © TeleGeography, Inc. 1993 Figure 5 Country United States Canada United Kingdom Germany Australia France Mexico Netherlands Italy Switzerland Japan Taiwan Sweden Finland Norway Austria Spain Singapore South Korea Israel Hong Kong Denmark South Africa Brazil New Zealand Chile Networks Sap/1991 Jun/1993 1,758 5,571 144 429 44 420 144 443 96 189 95 453 5 34 52 131 30 169 35 87 73 257 0 78 25 87 12 102 8 52 14 84 7 39 2 16 9 35 16 48 1 8 3 81 0 42 8 52 15 50 1 14 Ave. Annual Growth 124% 113%1 488% 119% 55% 215% 331% 87% 265% 85% 144% 142% 429% 314% 286%1 261%1 400% 165% 114% 400% 95% 314% 133% 743%! Inbound Traffic Ave. Sepl1991 1,034,428 47,953 12.602 19,256 22,601 8,757 2,366 10,716 5,549 9,427 4,616 0 8,926 4,217 4,571 3,904 932 2,420 3,942 5,569 256 2,821 0 232 1,536 67 Jun/1993 Annual Growth 6,057,250 277% 116,703 82% 69,838 260% 47,511 84% 59,348 93% 89,776 529% 5,892 85% 52,166 221% 13,475 82% 40,880 191% 19,475 184% 33,222 32,202 149% 50,763 631% 15,926 142% 9,052 75% 2,433 92% 3,417 24% 4,322 6% 6,737 12% 1,682 319% 7,232 89% 1,656 1,873 405% 2,031 18% 3,351 2,805% Outbound Traffic Ave. Sap/1991 1,108,022 34,103 4,177 8,131 11,000 5,495 1,521 5,397 4,772 7,715 2,239 0 7,386 9,880 2,510 1,030 419 325 613 1,859 28 1,074 0 43 539 52 Jun/1993 Annual Growth 5,445,958 224% 277,418 408% 169,383 2,260% 112,331 732% 93,818 430% 80,677 782% 52,274 1,907% 48,870 460% 47,699 514% 45,218 278% 41,825 1,011% 40,593 39,341 247°/= 27,171 100% 23,745 484% 23,120 1,225% 23,058 3,088% 19,019 3,284% 17,737 1,597% 17,454 479% 13,988 28,279% 13,613 667% 9,723 9,592 12,578% 7,526 741% 7,484 8,100% CO+RE, Sprint Link, InfoNet, PipEx, SwipNet, TIPNet) may route little or none of their traffic over the government supported NSF Net and CERN backbones. However, based on the melxics informally reported by other networks, these statistics are known to represent the major portion of global internetworking traffic. Several other observations can be made about the metrics charted here. Although the number of networks and the traffic for the U.S. greatly exceeds that for any other country, connectivity outside North America is growing. Like other telecommunications streams, there is a simple but strong correlation to GNP, with some exceptions. The Nordic nations, for example, proportionately have far more Internet connectivity than either their size or GNP would indicate. Similarly, many developing countries have significant and rapidly growing connectivity. This speaks to the ability of internetworking to permit such countries to compensate for an otherwise underdeveloped infrastructure. 16 TeleGeography 1993 The traffic balance metrics (in/out statistics) in the tables also bear examination. They suggest that Internet traffic has a significant correlation with the availability of accessible and valuable information products. Because so much more information is openly accessible in the U.S., it’s Internet hosts seem to be more heavily queried. More traffic thus typically flows out of the U.S. than into it, but not because more people in the U.S. are calling out, as in the telephone world. In contrast to Figures 1 5, the charts following this article focus on the global distribution of hardware or computer hosts serving the Internet and other networks. Taken together this information on facilities and flows provides a reasonably good overview of the global Matrix. Mapping The Matrix Internetworking In The 1990s The Internet traffic charted in TeleGeographv offers one doorway into the internetworking business of the 1990s. As the number of internets and users continues to double each year, the Internet and internetworking will very quickly become a central part of every nation’s infostructure. Anyone who has worked "on line" or dialed into a computer BBS for a social chat, or to receive assistance with a technical problem, knows that connectivity is often it’s own reward. The more people and resources that are reachable through a medium, the more valuable it becomes, attracting still more people and resources. The internetworking businesses of the 1990s are already taking advantage of this phenomenon to make the net a vital tool for virtually every kind of institutional, governmental, professional, social, or personal activity -- domestically and internationally. These twin dimensions -- the business and the social ones -- will continue to shape the evolution of the world’s internets as they evolve into the next public telephone network. END 17 TeleGeography 1993 Mapping The Matrix ¯ ’ :’ ¯ ....... :~:~’ ...... : -::i: Where In The World Is carmensandiego @ mci.com Eric Raymond, Editor, The New Hacker’s Dictionary (MIT Press, Cambridge, MA, 1991). Messages sent over computer networks, such as the Internet, are known as electronic mail, or e-mail. This method of communication has become so popular that the w~rd e-mail is now also used as a verb (cf. definition above), making the same grammatical crossover that the noun "telephone" made at the end of the 19th century. As with telephoning, to e-mall someone you need an address which directs your message through the network to the appropriate destination. Computer addresses, such as carmensandiego@mcicom, have long baffled the uninitiated. The addresses seem to belong to a world of their own -- a virtual space without geographical landmarks or obvious directions about how to get there. By comparison, the country and city codes in most telephone numbers give callers a sense of place, although these physical handholds may soon become a relic of computer technology too. (See "Numbering the World’s Phones" below.) What do computer network addresses actually mean and how do you find them? INTERNET E-MAIL Internet e-mail addresses are fairly straightforward. If both the sender and the recipient are connected to the Internet (a likely scenario for over half of the 20-million plus e-mailers in the world), is the standard address format which is used. Internet addresses consist of two parts: a username and a hostname. Suppose, for example, we want to send e-mall to the following address (which is fictitious): carmensandiego@telegeography, com carmensandiego is the username or local part, and refers to the mailbox, user i.d., or login name of the recipient. These names may be personal, corporate, or even based on a random assignment by a network manager. Once the e-mall message reaches the host computer, the computer uses the username to figure out where to "deliver" it. telegeography.com is the host part of the address. It identifies the host computer where the account is located using a standard Internet format. Each host computer on the Internet is assigned a unique 32-bit numerical address. To make life easier for human users, this numerical address is converted to an assigned domain name through the Domain Name System (DNS). The DNS refers both to the method of name-assignment and to the worldwide database of Internet names and addresses. DNS names are organized in a hierarchical fashion, moving from more specific to more general domains as one reads from left to right. The lowest level of the hostname (’telegeography’ in our example) comes first: it refers to a specific host computer. Next, separated by a period, there may be the name of a higher network or of the organization the computer belongs to. For example, if our hostname were telegeography.iic, corn, iic would refer to the sub-network which telegeography belonged to. The rightmost part of the name (. com in our example) refers to the high-level domain under which our organization is categorized within the DNS. These high-level organizational domains were created along with the DNS; they include COM (commercial organizations), EDU (educational), GOV (governmental), MIL (military) and ORG(miscellaneous organizations). For Internet host computers outside of the U.S., two-letter country codes are commonly used for the highest-level domains (e.g., ’uk’ for the United Kingdom, or ’ch’ for Switzerland). cont’d. © TeleGeography, Inc. 1993 18 Mapping The Matrix ~Where In The World Cont’d. The confusing aspect of computer network addresses is that, unlike telephone numbers, they do not necessarily indicate geographical location. A name points to a numerical address associated with a specific computer; the computer itself may be anywhere in the world. Just because a name is part of the domain ’uk’ doesn’t mean that the computer is located in the U.K.; it may only be registered there, and like an oil tanker registered in Liberia, could be berthed halfway around the Internet world. E-MAIL OUTSI])E THE INTERNET There are a number of computer networks which are not part of the Internet but which can still exchange e-mai! with Internet networks. This is done through ’email gateway’ computers, which translate the different e-mail protocols when mail is sent from one network to another (for a discussion of this international system of networks connected through gateways, see "How Big Is The Matrix" below.) E-mail addresses for users of these other networks (known as outernets) may need to provide more information than the standard Internet addresses. Some outernet computers have DNS names; in this case, the DNS database will automatically f’md the right gateway for the e-mail to go through. MCIMail (userid@mcimail.com), America Online (user@aol.com), and Compuserve (userid@compuserve.com) are examples. For other outernet networks, the e-mail sender may need to specify the gateway computer. For example, to send e- mail to someone on the UUCP or BITNET networks, you may need an address in the following format: username %hostname@gateway-hostname The e-mail is sent to gateway-hostname, which then delivers it to the specified user on the hostname computer. 1NTERNET ADDRESS "BOOKS" What if you don’t know the name of the account or computer you want to reach? Unfortunately, no general Internet directory exists, yet. However, a number of Internet functions and ’white pages’ type directories are helpful. Some of them are: Finger: This command, for computers with a UNIX operating system, allows you to find a user if you know the host computer. Whois: This is both the name of a white pages directory and the command you need to access it. It contains over 70,000 entries and can look up people by last name or by a piece of the last name. X.500 Directory Services: This service is a collection of directories, having an agreed X.500 protocol, with each member group being responsible for keeping the directory of its lower organization up-to-date. It is most easily accessed by logging on to an e-mall computer which runs the fred query application. X.500 may be the wave of the future for Internet directory searches, but currently many users are still not listed. For more information, see Tracy Laquey and Jeanne C. Ryer, The Internet Companion: A Beginner’s Guide to Global Networking (Addison-Wesley, Reading, MA., 1993) and Ed Krol, The Whole Internet User’s Guide & Catalog (O’Reilly & Associates, Inc. Sebastopol, CA, 1992) A.R. ® TeleGeography, Inc. 1993 !9 TeleGeography 1993 Mapping The Matrix ¯ "? ,::" ’ %.." i "~.:i:. ,~ ¯ ".’~ ~’ ’ _ ~, .:’..-.- ,’~ ’ -" ,:: ....,..: How Big Is The Matrix? The Matrix may be def’med as all computer networks worldwide that exchange electronic mail. The four global networks that carry most of the public traffic through the Matrix are Intemet, BITNET, FidoNet and UUCP. There are also many large enterprise networks -- private networks dedicated to a particular company (Exxon, Mitsubishi, Citicorp) or industry (e.g., SWIFT, which serves the banking industry; SITA which serves the airline industry). The table above shows the current size (mid-1993) of these networks. The data were compiled by Matrix Information and Directory Services, Inc. (MIDS); a/k/a mids@tic.com. A graphic portrayal of the interrelationship between these networks appears on the opposite page. Glossary: The Internet is a network of networks, mostly leased line but also dialup, operating from 2400-bits per second (bps) to 100 Megabits per second, which utilizes the TCP/IP (Transmission Control Protocol/Internet Protocol) protocols. The key TCP/IP protocol is the Internet Protocol which provides a uniform address space and permits routing across diverse networks from one computer to another. The Transmission Control Protocol uses IP to provide reliable delivery of bitstreams in order and without loss or duplication from a process on one machine to a process on another machine. B1TNET (Because Its Time Network) is the mostly IBM mainframe and Digital VMS, mostly 9600 bps leased line, mostly academic and research network that uses the NJE (Network Job Entry) protocols. The size of BITNET is defined by the BITEARN node list, which gazettes hosts for EARN (European Academic and Research Network), GULFNET (Arabian Gulf countries), NetNorth (Canada), BITNET (United States), and other NJE networks. FidoNet is the mostly DOS, mostly dialup, network of PCs used mainly by hobbyists. It uses FidoNet protocols and its size is defined by the FidoNet node list. cont’d 20 TeleGeography 1993 ~i".,-i:-’ ’ i:’ cont ’d ... How Big Is The Matrix? Glossar,€ Cont’d Copyright 1993 Matrax, Inc.. Austin, UUCP (Unix-to-Unix Copy Protocol) is the store-and-forward network of mostly UNIX, mostly dialup machines. The users are a diverse group and the size is partly defined by the UUCP node list. Each of the four networks described above supports other applications in addition to mail. One application carded across all four of them in USENET (Users Network) news, which provides conferencing services allowing people to post and read messages on numerous machines worldwide. USENET is often referred to as a network, but it isn’t really, since there is no USENET transport protocol. USENET news is carried over whatever underlying mechanisms are available on a carrier network. The Internet has numerous applications that are not provided by these other three networks, including remote login (TELNET), interactive file transfer (FTP), distributed file indexing (archie), remote document indexing (WAIS), remote menu lookup (Gopher), and remote text searching (World Wide Web-WWW). If you can FTP to is.intemic.net, you are on the Internet; if you can’t, you’re not. If you can send mail to is.internic.net, you are in the Matrix; if you can’t you’re not. If you are on the Internet, you are in the Matrix. But, remember if you are in the Matrix, you may not be in the Internet. Adapted, with permission, from "Network Sizes" Matrix News, Vol. 3, No. 8, August 1993 ’The Net’s-. a public utilit7~’ Malkah"said ’Commtmities, multis, towns, even individuals subscribe,’.. .... .-:: ... ’Yourtowndoesn’t subscn’be?’ Avram asked Nili ineredulouslyi::.:-. ...:.. Nili s~gged, ’We know a tot abaut:-some tlaings, tess ab~izt .otl~rs,: ...:it doesn’t seem as. if.-i-, i! :" .-" 5"5-’" your Net tins inade you -"tinlversaily ~w~se. .::.. -%.~ .... .... :: .. .... .., ~Tt~,y. we~.aiI.:a...tittte::shockext.:--It was a truism ..tha~:everyone. wad:On ~the"Het.,...a!thou.g!9..., child m~ght, grow..upm the Gl0p~ work for a gang.0r setl-:l~.to.d:multi~ die ..o.f.:tm~ Of ~,e ~’ii~i!.~...: :::::?:~: ..... plaguestI~t swept-the Glop eue~ ..y~ir, and neve~ once :pttig reto., access th.ei.N6t:.. btili Jn whatwas urtive~sally considered, the ceatt~d artifact o.t~:eontem~ry ¢ .~ttU~s~’il ’".-..:.- ... : .. ’ .-.- i: From .He. She [~ by Marge l~ercy (Baltantin.e"Books, bL Y,, 1991)" ..... .~ -!!~ 21 TeleGeography 1993 Mapping The Matrix ,,::. :. THE ENDS OF THE MATRIX By John S. Quarterman Where are the geographical ends of the networked world? The question is interesting, partly because the answer keeps changing. Here’s a mid-1993 report. If you know more, please let me know. SOUTH In 1990 the U.S. Palmer, Siple, and South Pole stations were all connected to the Matrix -- the worldwide web of computer network which exchange electronic mail (E-mail). The stations used a satellite link to a NASA machine in Florida. But the Siple station is usually closed now, making the South Pole station the most southerly networked place in the world, at 90 degrees south. Amundsen-Scott South Pole Station does not have a permanent link and does not use the usual protocols of the four main networks: the Internet, UUCP, FidoNet and BITNET. But E- mail still goes to and from the Matrix. The "AQ" designation for Antarctica was registered as an Internet address under the Domain Name System (DNS) in March 1992 by the New Zealand Department of Scientific and Industrial Research (DSIR). DSIR has a permanent satellite link to Scott Base on Ross Island. They run IP (Internet Protocol) over that link, and it is routed with the Internet. But apparently it isn’t configured for international access; I can’t reach it from here. This is apparently because they reserve their bandwidth for scientific voice and data traffic. However, two kilometers away is the U.S. McMurdo base. McMurdo was first connected to the Internet in John S. Quarterman, jsq@tic.com, edits Matrix News and is the author of the pioneering survey The Matrix: Computer Networks and Conferencing Systems Worldwide (Burlington, Mass, Digital Press, 1990). An earlier version of this article appeared as "The Edges of the Matrix," in volume 3, No. 3 of Matrix News (March 1993). Contact: mids@tic.com, voice: 1-512-451-7602, fax: 1-512-450-1436. about February 1992. Apparently there was some problem with 200 mph winds during the Antarctic winter, but McMurdo has been up solidly since at least December 1992 (Antarctic summer). That’s about 78S 166E. It is rumored that Australia has a link to their Antarctic base (Casey station?) to Hobart, Tasmania, but it is apparently not routed with the rest of the Internet. Other countries are also considering Intemet connections from Antarctica, and some may be up already. There are no FidoNet, UUCP, or BITNET sites in Antarctica that I know. The southernmost networked place not in Antarctica appears to be Hobart, Tasmania, which has UUCP, FidoNet, and Internet connections. There is no BITNET in Australia or New Zealand. NORTH The northernmost networked place in the world is harder to discover because there are so many contenders. Most northerly appears to be a node on the NASA Science Network (part of the Internet) in Thule (Qaanaaq), Greenland, at 77 40N 69 0W. There are a few towns farther north than that, in Greenland, on Ellesmere Island, in Svalbard, and in Franz Josef’s Land, but they don’t seem to be networked. The second most northerly place on the matrix appears to be a PC clone running the Waffle Protocol on the North Slope of Alaska in Atqasuk: 70 28 10N 157 23 45W. This node is not north of the North Pole, but it is north of North Pole, a town in Alaska. This host is really with the UUCP net, and has the associated Internet domain name, coldbox.cojones.com. It also uses the WWIVnet protocols. According to the owner of coldbox, Atqasuk has 220 Eskimos, 15 Caucasians, and no roads. Other than Atqasuk, the northernmost con ’t. next page Copyright ® 1993 Matrix Information and Directory Services, Inc. (MIDS) 22 TeleGeography 1993 Mapping The Matrix con ’t. from prior page networked place in the United States is apparently Fairbanks, Alaska. The most northerly Canadian hosts appear to be in the Northwest Territories. The most northerly one is a UUCP node in Rankin Inlet, 62 48 41N 92 6 57W. Next most northerly networked in the Northwest Territories is Yellowknife, with both FidoNet and UUCP, at 62 27 43N 114 26 04W. The most northerly networked Yukon town appears to be Whitehorse, at 60 43N 135 03W, with a FidoNet node and a UUCP node. Egedsminde, Greenland is a small island south of the larger island of Disko, off the west coast of Greenland at about 68 50N 52 45W: there is a FidoNet node there. The Danes named the town after Hans Egede, a missionary, but the Inuit call it Ausiait. Nuuk is also on FidoNet, at 64 10N 51 35W. Iceland is farther north, and Kopasker appears to be the most northerly site in Iceland, with both Internet and UUCP connections, at 66 20 30N 16 29 30W. Reykjavik is only a few degrees farther south, and has UUCP, FidoNet, BITNET, and Internet connections, at 64 08 35N 21 57 35W. Other northerly nodes on the matrix include: Norway -- Tromsoe, with Internet and FidoNet connections, at 69 40N 18 50E. This is also the site of the most northerly networked university in the world. Sweden -- Malmberget, on FidoNet, at 67 10N 20 40E, followed by Lulea, with FidoNet, UUCP, and Internet connections at 65 35 22N 22 10 00E. Finland -- Tornio, on FidoNet, at 65 50N 24 12E, followed by Oulu, with Internet, BITNET, and FidoNet connections, at 65 13 00N 25 19 00E. In Russia, Murmansk doesn’t seem to be networked, but Sverodvinsk and Archangelsk are, around 64 34N 39 50E, on FidoNet. Then there is Komi, on UUCP, at 63 33 00N 53 38 00E; Petrozavodsk, Karelia, also on UUCP, at 61 47N 34 20E; Ivanovo, on FidoNet, at 60 32N 36 22E; and eventually St. Petersburg, on most networks, at 59 55N 30 15E. WEST Westernmost doesn’t mean much on a round planet, but the networked place farthest west in longitude from Greenwich is Hawaii, with Honolulu at more than 157 degrees west. Exactly which town on which island is the westernmost networked is unclear EAST Easternmost from Greenwich is Fiji, where the Department of Mathematics of the University of the South Pacific in Suva, at 18 6S 178 30E, has a UUCP link to New Zealand, such as Wellington, which is connected to most networks except BITNET, at 41 17 25S 174 46 07E. Christchurch is also far east, followed by Port-Villa, Vanuatu, on the RIO UUCP network, at 17 44S 168 19E. Then there is Scott Station (NZ), Antarctica, on the Internet; Noumea, New Caledonia, on RIO; and McMurdo Station (US), Antarctica, on the Internet, all between 166 and 167 degrees east. ENDS Gold Coast, Queensland, is probably the easternmost networked place in Australia, on FidoNet, at 28 04S 153 25E. Russia’s easternmost networked outpost appears to be Magadan, north of the Sea of Okhotsk, and at the end of the road from the west, with both FidoNet, and UUCP, at 59 34N 150 48E. There appears to be nothing on the Kamchatka Peninsula, or farther east. Sakhalin Island is on the map, with UUCP. Guam’s easternmost networked point is on FidoNet, at 13 29 02N 144 48 23E. The easternmost networked place in Japan is Kushiro, Hokaido on the Internet, at 42 58N 144 23E. Sapporo follows soon after, with Internet and UUCP connections. Acknowledgements Particular thanks to Rick Broadband (north), Bob Faulhaber (south), Eartmut Haberland (north), Bryan Lockwood north), Marty Lyons (south), and Peter Salus (north). Copyright © 1993 Matrix Information and Directory Services, Inc. (MIDS) 23 TeleGeography 1993 Mapping The Matrix 24 TeleGeography 1993 Mapping The Matrix LIFE IS BETTER ON THE NET ... OH REALLY? By Johll Perry Barlow 1. It allows you to meet interesting people from all over the world. 2. No matter where your body may be, your mind can always be found online. 3. You can commute to work at the speed of light, earning big dollars at home. 4. It provides access to all the world’s written knowledge with a few keystrokes. 5. You can conduct passionate long-distance love affairs in e-mail. 6. The Net eliminates such barriers as race, gender, attractiveness, or social grace. 7. It helps keep you and your teen-age kids off the streets. ......... : -.....-- .; .:’:-" ~:-!" --:-i "’i-.- ’ .. "~;:-::~?;-]’i’." .: :( -.’: .’ ;’ i~;’, :’.. 8. It’s interactive and therefore less addictive than television. 9. You’ll rediscover the community lacking in your physical world. 10. No matter how obscure your interests, you can always find someone who wants to talk. 11. Terabytes of software are available online, either free or dirt cheap. John Perry Barlow is the co-founder of the Electronic Fronder Foundation (EFF). His article is reprinted, with permission, from the May/June 1993 issue of Wired magazine, 544 Second St., San Francisco, CA 94107; Fax: 415-904-0669; Voice: 415-904-0660. E-maih subscriptions @ wired.com © Wired USA Ltd. 1993 25 TeleGeography 1993 Global Numbering GLOBAL TELEPHONE NUMBERING PLANS- FROM GEOGRAPHY TO SERVICES By Stewart Fist With the possible exception of minting a new currency, few changes are likely to disrupt a nation’s social and economic fabric so much as launching a different telephone numbering plan. Each particular number change may be only a moth-hole sized annoyance but they come in plague proportions. Many countries nevertheless are facing the prospect of new national numbering plans. And changes are also on the horizon internationally. By the decade’s end, the current system of International Direct Dialing (IDD) for country codes -- "44" for the U.K., "61" for Australia and so on -- may not connect callers to many of the world’s telephones. There are various reasons: Non-geographical Services. New radio based services are making some forms of numbering identification quite independent from geographically- based wireline services. Digital phones based on the GSM (Global System for Mobile communications) have already been designed to ’roam’ across national borders in Europe. LEO (Low Earth Orbiting) satellite phone systems will extend their reach globally. Tram-national Services. New telephone services are being designed for organizations and corporations which operate on a regional basis. These services need to be recognized at the global or regional level in a numbering plan rather than solely at the national level. Intelligent Network Services. Call- diversion and ’follow-me’ services have begun to migrate from the company switchboard to national networks. Soon they may be available on a global level via the proposed Universal Personal Telecommunications (UPT) service -- a one number one person scheme. A user’s UPT number is completely divorced from the geographically based number of his or her local residence. Political Changes. Finally, the break-up of the Soviet Republic and the disintegration of many of the states in Eastern Europe has put pressure on the global numbering plan for the release of new country codes. Back~ound To understand the stresses on the current global numbering plan and the options for tomorrow, some background is helpful. The current regional or zone system for telephone country codes dates from 1964. The codes adopted then were based on early 1960’s projections of telephone population densities and population growth. The numbering plan also assumed a reasonably stable political and technological world. Stewart Fist, an independent technical journalist based in Lindfield, New South Wales, is a regular contributor to Australian Communications. But "these assumptions are no longer valid," notes Dr. Theodore Irmer, the Director of the U.N. telephone standards body which overseas global numbering plans. "We face increasing problems in allocating international country codes. [And] we © TeleGeography, Inc. 1993 26 TeleGeography 1993 Global Numbering shouldn’t underestimate the high political importance that countries attribute to their international code. For them, it’s a symbol of their independence and self-confidence." These changes to the old world order are revitalizing activity in the Telecommunications Standards Section (TSS) of the International Telecommunication Union (ITU) which Dr. Irmer heads. Formerly known as the CCITT (International Telegraph and Telephone Consultative Committee), the Geneva based TSS has allocated telephone numbering issues for the next generation of phones and services to Study Group 2 (SG-2). It is this obscure technical committee which will be the battleground during the 1990s for the world’s future telephone codes. CCITT Recommendations The current 1964 international numbering plan for the world reflects a CCITT Recommendation known as E.164, itself an update of an older scheme, known as E. 163. The foundation of these schemes- is a ’country code’ (one to three digits long) followed by a national number which is set by the country’s telephone authorities. The style of the country code varies from continent to continent. But essentially it consists of a single- digit World-Zone code followed by a one or two digit national identifier. In North America, however, the Zone Code is followed directly by regional area codes which have no national significance. See Figure 1. (A global map of Telephone Numbering Zones of the World Figure 1 North America Africa ~-Europe South and Central America South Paclhc Former Soviet Union Far East M~ddle East and Southeast Asia 27 TeleGeography 1993 Global Numbering dialing codes for countries and capital cities appears in the TeleGeography Blue Pages at the back of this report.) There are also quite separate numbering plans currently in use for non-voice services. For example, CCITT Recommendation F.69 is the calling code standard for telex. Recommendation X.121 is used for international data (mainly X.25 packet-switching); it consists of a four digit Data Network Identification Code (DNIC) followed by a Network Terminal Number (NTN) which doesn’t have any geographical significance. SG-2 is attempting to prepare a general set of numbering principles so that data networks could use either X. 121 or E. 164. As yet, however, there is no consensus on how best to do this. Network Intelligence One thing on which Study Group 2’s members do agree is that E.164 must lose some of its geographical rigidity. The technical ground work for change has been laid over the last two decades by the progressive introduction of network intelligence. In 1964, the world’s telephone system, especially at the international level, was largely bound by hard-wired mechanical switching. A strict numbering hierarchy thus was generally required to pass a call from geographic zones to countries, regions and local exchanges. Today, however, international exchanges can call upon all the flexibility and customization possibilities provided by computerization and software control. With common channel signaling systems, such as SS7, a switch can search an internal ’look-up’ table to determine the routing for a particular numbering scheme. Thus, each country (or carrier) can be identified as a single entity, not subject to hierarchical (regional or zone) switching. These technological changes have made possible the first major departure from the current global numbering plan. Until now, as noted above, country codes have been strictly allocated to World Zones with the same numerical prefix. For example, any country code beginning with a "3" or "4" is in Europe. See Figure 1. Those headed by a "5" are in Central and South America; a "6" indicates the Oceanic and Australian region. "Most people aren’t even aware that this zone differentiation exists," says Paul Reptis, an engineer with Telstra, Australia’s international carrier. "They will know the country code. But they won’t appreciate the regional significance of this code." A future recommendation from SG-2 will be to abandon this unnecessary zonal rigidity and, in the process, a large number of new ’country codes’ will be made available for future use. Problems arise, of course, in attempting to ’retro-fit’ the world’s telephone networks with new numbering plans. Unless the nations concerned all benefit, some countries will be loath to make a change. Moreover, changing international country codes is a costly and complex business. Historically, this antipathy to change has been most troublesome for the technology leaders. In North America, for instance, Zone 1 telephone codes are limited to the U.S., Canada and miscellaneous Caribbean nations. Further, the zone has been divided up with little geographic or national significance and the zone code "1" is not incorporated in the three digit sub-regional code, thus increasing the length of the international dialing codes for callers from other zones. For example, the Zone Area Code for Minneapolis, Minnesota, is "1-612" while to reach Ottawa, Canada it is "1-613". There is no country-code separation here at all -- just a zone area code. The Caribbean is even more confusing; the "1-809" sequence is shared by both Jamaica and Puerto Rico, while "1-808" will get you Hawaii. As a consequence, the caller often has no indication of distance or geographical proximity from the North American Numbering Plan (NANP). 28 TeleGeography 1993 ."7" ",, :: ::-:,;?:!!: :.:..:: ........ Global Numbering Ex U.S.S.R. Telephone Codes Fi~zure 2 LATVIA 371-2 ESTONIA 372-2 LITHUANIA 370-2~ /~Tallinn B E LA R U S ...~.~,,_~-Viini u s 7-01 72 ~nsk ) UKRAINE ’, Moscow 7-04~ ~’~ ¯ Kidv i~Chisina~, MOk~OVA 373 RUSSIA 7-095 KYRGYZSTAN 7-3312 "Dushanbe TAJIKISTAN 7-3772 Map shows diahng codes for co~.lz~try al)d caDtal city Balhc states have the same code for the country ancl caDtal ctty. ~Future code Except for the remnants of the ex-Soviet Union, where numbering chaos still reigns (See Figure 2), the rest of the world has followed a pattern of using either two- or three-digits to identify the country (which includes a single zone-identification digit at the head). Europe is subdivided into two zones. But, again, this was done in an ad hoc fashion over time, with no geographical significance. Who Controls National Numbering Plans? local telephone directories, control over national telephone numbering plans can be a useful weapon indeed. In Australia, for instance, Australian Telecom has allocated to itself large blocks of easy-to-remember numbers. The 018 service prefix, which identifies all analogue cellular phone services, now has an inquiry number of 018 018 018 for Telecom’s own MobileNet service, which their competitors claim provides MobileNet with considerable advantage. Under the existing CCITT Recommendations, each country is responsible for the allocation of national codes within its own borders and some zones have their own numbering groups. Generally, these groups are dominated by the old monopoly telephone companies which has made numbering a key issue in the battle between old and new telephone carriers. Especially when the older carrier owns and produces In the U.S., Bellcore, once part of AT&T and now owned by the seven Regional Bell Operating Companies (RBOCs), has long had responsibility for administering the North American Numbering Plan (NANP) for Zone 1. See Figure 3. In recent years this has led to a vigorous fight between MCI, Sprint and the RBOCs over Bellcore’s alleged favoritism. "It’s a question of ... control", says Peter Guggina, 29 TeleGeography 1993 Global Numbering Figure 3 The Future of The North American Numbering Plan (NANP) Telephone numbers in the North American Numbering Plan (NANP) generally consist of 10 digits in the format NWX- NXX-XXXX, where X is any digit 0-9; N is any digit 2-9; and W is 0 or 1. The first three digits are referred to as the Numbering Plan Area (NPA) or area code. The next three digits are the central office code (COC). And the last four digits are the station number. NPA Code CO code SN Key ." N = 2 to 9 NPA = Numbering plan area W = 0 or 1 SN = Station or Subscriber number X = 0to9 CO = Central Office The NANP is nmning out of numbers. Hence, in 1995, interchangeable NPAs will be introduced using the NXX format. This will increase the number of available NPAs by 640, representing a fivefold increase in numbering resources. (There are 160 numbers (8x2x10) in the format NWX. By going from NWX to NXX, the number of potential NPAs is increased from 160 to 800 (8xl0xl0). After subtracting the eight Nll abbreviated dialing codes -- see below -- the number of available NPAs is increased from 152 to 792.) The post-1995 numbering plan in North America will thus be organized as follows: N00 Codes Nll Codes (short codes) Other NPA Codes 200 Unallocated 300 Unallocated 400 Unallocated 500 Personal Communications Services 600 Non geographical services in Canada 700 Carrier specific services 800 Freephone 900 Information services 211 Unallocated 311 Unallocated 411 Directory Information 511 Unallocated 611 Repair service 711 Unallocated 811 Local business office 911 Emergency services NOX N1X N2X N3X N4X N5X N6X N7X N8X N9X For geographical based services; mainly public telephony but also paging, cellular mobile, premium rate. Reserved for both geographic and non-geographic services. Also one set of 80 NPA codes reserved for expansion of NANP to 11 digits. For non-geographic services; e.g. mobile, satellite, Universal Personal Telecommunications (UPT). Key: NANP = North American Numbering Plan NPA = Numbering Plan Area N = 2 to 9 X = 0 to 9 Source." FCC and BellCore © TeleGeography, Inc. 1993 30 TeleGeography 1993 the Director of Technical Standards Management for MCI. "We believe the numbering administrator should maintain a neutral posture; Bellcore shouldn’t be free to apply their own judgement, or control issues, when there is a lack of consensus. Simply put an administrator cannot be owned by any entity that has a vested interest in the administration of numbering resources." The conflicts between the RBOCs and other U.S. carriers are likely to sharpen in the months ahead as competition for mobile services intensifies. (AT&T’s $12.6 billion acquisition of McCaw Communications is a case in point.) Significantly, in August 1993, the U.S. Federal Communications Commission (FCC) asked Bellcore to halt the assignment of "500" number area codes for Personal Communication Services (PCS) pending further public comments. Shortly thereafter, Bellcore’s CEO, George H. Heilmeier, surprised many observers by advising the FCC that Bellcore would relinquish its historic role as the NANP Administrator. Heilmeier urged the FCC to find a successor quickly. Alternatives suggested include the National Telecommunications Information Administration (NTIA) in the U.S. Department of Commerce as well as various government-industry councils. The European Commission (EC) is also attempting to play a role in regional numbering issues. It has already taken steps to harmonize access and emergency codes across Europe. Further, the EC has tentatively decided to set up a European Numbering Office (ENO) to supervise the region. An EC draft Directive on Open Network Provision (ONP) says that, as a first step, control of national numbering plans should pass from the monopoly Post Telegraph and Telephone (PTT) administrations to national regulatory authorities. First Come, First Served The general pattern of number allocation based on World Zones may appear to be logical. Yet, it hasn’t always worked well in the past. In general, Global Numbering i iiiii.:. ..-~::~:.~::-::’:~::::-" ". ::.::-iiiiii.iiiii..-.::!:::: ’. ’ i!:i ... the countries that developed international telecommunications facilities early and had the greater populations quickly took the two-digit country codes within their zone. Later, as it became obvious that more than ten country-codes would needed in each zone, the few remaining unallocated proposed two-digit codes were reserved for three- digit use. This was very much a stop-gap measure. Thus, France and Spain ("33" and "34") were early IDD users in Europe’s Zone 3, while "35" was eventually reserved and subdivided into Portugal ("351 "), Ireland ("353"), Malta ("356") and Finland ("358"). This practice has been widely followed in most world zones (except, of course Zone 1). However, with the break-up of Eastern Europe, the few remaining two-digit codes weren’t enough for all the new countries that have emerged. Fortunately, the unification of Germany released the old East German ("37") country code for subdivision and reallocation. A similar process will also be followed with the fragmentation of the old Yugoslavian "38" codes, into "38x" for Croatia, "38y" for Serbia, and, possibly "38z" for Bosnia Hertzegovnia, if it still exists. The Far East developed international telephone connections at a stage when it was already apparent that three digit codes were essential and so these were allocated in Zone 8 from an early stage. As a consequence, there are still forty ’country-codes’ available in this region. This has permitted the existing plan to be extended beyond its original strictly geographical use, as discussed below. From Country Codes to Service Codes Because the old telephone zone system currently has little value for major carriers, and because it has little practical significance for telephone users, why not abandon it altogether? After all, the system was always inefficient; some zones obviously had too few numbers to allocate (without forcing major revisions to the existing codes), while others had more than they could use. 31 TeleGeography 1993 Global Numbering "The pressure on country codes is not coming from countries, although it may appear to be that, with the break up of Yugoslavia and other Eastern European countries," says AT&T’s Robert Madden. "This is just an aberration which is noticeable at this time. The real demand being placed on country codes is for service-identification rather than country- identification." "The only ’service’ to which country-codes are now allocated is the Inmarsat ship-to-shore satellite telephony services," explains Paul Reptis. "But it is a sort-of geographic code; you’ve got to find out where the ship is -- which ocean it is in." Inmarsat was given four Far Eastern (Zone 8) codes. The "871" number addresses the East Atlantic; "874" is for Atlantic West; and "872" and "873" are for the Pacific and Indian Oceans respectively. Further allocations will later be used to split both the Pacific and the Indian oceans into twin zones as well. Inmarsat’s minor divergence from the old zone system is now being extended further by a current proposal for regional and global freephone numbering. The idea is to use the Far East series from "800" to "809" for an International Numbering Scheme (INS). (Don’t confuse your INS’s! The same acronym is used for both the International Numbering Scheme and Intelligent Network Services; one depends on the other.) David Lewin of London consultancy, Ovum, points out that phone number sequences serve purposes other than just dialling. They are also important to give the customer an indication of distance, time- differences, costs, and the type of service they are accessing. These functions need to be considered when designing number systems. The widespread use of the code "800" (or something similar) for freephones is a case in point; geography isn’t the only factor to be taken into account. See Figure 4. The International Numbering Scheme proposal backed by Lewin and a group of European numbering experts would make "800" into a global freephone service prefix, with free connection for the customer from any part of the world. This would be important to those who "leave home without their American Express" card or have it stolen. Not coincidentally, it would also foster global access to the "Country Direct" and "Country Beyond" calling plans of major carriers. It may be fortuitous, therefore, that this "80x" number series has remained vacant in the Far East allocation plan. To provide for a slightly more geographically-restricted range -- ideal, say for Car Rental firms, and the like -- the INS proposal suggests that the codes "801" to "809" would be available for freephones within existing World Zones, using the standard zone code as the third digit. These prefixes would be used for premium rate services and directory inquiries also. Under the foregoing plan, "803" and "804" would thus be available for pan-European use. This would enable EC wide businesses to advertise a singe "803 xxx xxxx" number in all these countries. These pan-’zone’ numbers would exist in parallel to existing national numbers for the same services. The "801" series would become a pan-North American sequence, and the "809" freephone or directory numbers wold be used by pan-Middle East services or corporations. Not everyone is convinced Lewin’s approach is a good idea, however. "[It] is the thin edge of the wedge. If the Study Group agrees to this proposal, where do we stop?" asks one member of SG-2. "The answer is that you can’t. Which suggest that there will always be a shortage of country codes in the foreseeable future." Personal Phone Numbers A further demand for more international codes will come from personal communications, often loosely described as Universal Personal Telecommunications (UPT). The term UPT seems to cover every concept from being branded-at-birth with a telephone 32 TeleGeography 1993 number to plans for globally-extended hardware- based cellular networks using pocket communicators. CCITT Recommendation E.168, a recent update to E.164, calls for the reservation of one ’country- code’ as a global service identifier for UPT. And Global Numbering this is the most likely course to be followed. When calls are placed to a personal number, the network will route them to wherever that person happens to be by using an internal network database. This database must be checked whenever a caller dials-a The Information In A Telephone Number Figure 4 ¯ Distance: ¯ Choice of Service: ¯ Choice of Carder: ¯ Choice of Terminal: ¯ Service Quality: Called Party Identification: Through the use of readily identifiable zone, country and area-codes the calling party has some appreciation of the distance and likely cost associated with any given call. Number prefixes at a national level, such as "800" in the U.S., tell the caller that this is a "free" (receiver paid) service. However there is no universal standard for these access identifiers; other prefixes are used in other countries. User groups have therefore called for harmonization of the various prefixes needed for access codes to premium services (’900" in the U.S.), international directory services assistance, and for emergency services. Access codes are also used to carry other embedded information. For instance, recorded information services may be split between those suitable for general-family ("0891" in the U.K.) and adult- only ("0898" in the U.K.; "976" in the U.S.). Sometimes numbers do identify a carrier network. In the U.K., the "0500" prefix is used to select a destination on the Mercury network. Numbers can be used to identify mobile and satellite phones, telex machines, etc. In the case of mobile terminals, these numbers also often carry the implication that any unanticipated costs (due to distance) will be absorbed by the receiver, rather than the originator of the call. Some carriers use a special access code for higher quality lines, such as those used for international facsimile transmission. However, in the future, users may be able to select from a range of different service qualities for voice as well as data services. For instance, an access code could enable users to chose a 4 or 8 kbit/s international voice circuit at a cut-rate, and suffer the slight loss of quality over the standard 64 kbit/s circuit. The Universal Personal Telecommunication (UPT) number scheme is aimed at assigning one number per person, irrespective of location or the subscriber’s business or private life. Directory listings for wire-line numbers, however, generally let the caller know whether she is ringing a residence or a business and hence the social role of the person being called. UPT may change all that. © TeleGeography, Inc. 1993 SoFo 33 TeleGeography 1993 Global Numbering UPT number because the network needs to translate the dialed number into a real number. Because UPT will be starting from ground zero with no pre-existing applications, the idea of having a single universal prefix throughout the world carries obvious advantages -- especially since the UPT number will have no geographical significance. According to AT&T’s Madden, the UPT recommendations "are now coming into the final stages of development, and there are early trial applications -- precursors of UPT. But ... the viability of the idea depends on the customers. Will there be a strong market for these kinds of services? Early indications say, ’Yes’ there is a market for it, and the optimists say that ultimately all numbers will be personal numbers." But, says Madden, "it won’t happen in my lifetime!" UPT and Number Portability The UPT concept concentrates on mobility -- the ability to have people call you anywhere at any time, plus the ability to call out from anywhere and have the cost of the call billed to a standard account. One example of this service is AT&T’s $7 a month EasyReach "700" number. It is advertised as a ’lifetime telephone number’. No matter where you are in the U.S., your long-distance calls, through AT&T, will follow you. (The AT&T access code must be dialed before the "700" number.) EasyReach gives its subscribers the ability to re- program their particular directory-reference through a Touch-tone phone using a PIN (Personal Identification Number) security code to divert calls. Users can also filter calls by giving callers a special PIN-number. Number portability is a quite different concept. ’Portability’ here means something less transient than mobility; it is the ability to maintain the same number if you change carriers, or shift to a new home, or even change countries. As discussed above, in most networks, numbers are still largely geographical and they contain carrier-required information which determines whether the charging is to be local, long-distance or international. Intelligent Networks can overcome this problem but only at the cost of complexity. Hence, because number portability implies a loss of carrier and geographical identification, many technical hurdles must be overcome before portability can be implemented. Likewise, because UPT services personalize fixed networks (they rent people a number rather than a phone line) they also raise large technical hurdles. The network must be able to immediately recognize that a UPT call requires intelligent processing and route the signal to the appropriate database which holds the user-profile data. (Ideally, this database will need to be a shared national facility, independent of any of the national carriers.) The signal is then returned to the exchange from which it originated so that the call can be set up. Number portability, UPT and other new services (e.g., virtual private network offerings) thus are all heavily dependent on the database and processing power of intelligent networks. This has created a paradox; network intelligence provides many of the answers but at the same time it also creates new and more complex demands. For as switching becomes wholly computerized, it creates an open invitation for network developers to ’push the envelop’ of numbering possibilities beyond geography. Conclusion The international telephone network has been described as the world’s largest machine. This metaphor underscores the interworking of the network’s many parts and it is what makes telecommunications so standards-driven. As voice, text and video services are integrated in the 1990s, the size of this machine will double and then possibly quadruple over the rest of the decade. A further doubling of network terminals may occur as we progressively break away from wireline delivery, allowing mobile radio and satellite connections across borders and continents. 34 TeleGeography 1993 Global Numbering Dialing Past The Receptionist Figure 5 What’s been left out of the numbering debate The debate over new telephone numbering plans seems to have overlooked the possibility for using standard codes to reach an extension line within an office or a designated terminal (e.g., fax rather than phone). Yet, agreed international standards for extension service dialing could be a boon to users; it might also make more efficient use of existing access lines. With many large companies, users can already direct-dial the main switchboard and be connected to an extension or direct- dial the extension itself. This is possible because the carrier treats the company switch as if it were a sub-exchange and allocates to it a range of numbers (usually the last two or three digits of the main number) for extensions. Some call- diversion facilities and voice-mail systems will then allow one to dial an extra digit to divert a call to a mail-box, a secretary or back to the switchboard. ISDN (Integrated Services Digital Network) offerings in some countries have a similar extension dialing capability. With a basic-rate ISDN plan for homes or small businesses, eight numbers can be allocated to each service. The last digit in the series is then used by a network termination box (at the customer’s premises) to divert an incoming call to one of eight extensions. But what is needed for normal telephone services is a standardized set of post-dial suffixes which any caller knows. The caller shouldn’t need to consult a special directory or know the extension digits from past calls. A telephone call could then be picked-up instantly by a small intelligent switch (before the bell is rung) at the customer’s premises. The switch would interpret any further dial tones as commands to connect a designated extension. If no further dial tones are received in, say, two seconds, the switch would ring the bell and revert to normal phone operations. It has been suggested that the standardized set of suffixes should be: Nos. 1 to 4 for extension phones, in a hierarchical order of importance. (In a small business, Ext. 1 would be the manager; at home it might be the family-room phone). As an alternative, these numbers could be used to generate distinct ringing tones which identify which person is being called. No. 5 reserved for user-nominated devices and two-digit suffixes. No. 6 reserved for home/security control systems (with two-digit suffixes). No. 7 for direct connection to a computer or local area network. No. 8 for direct connection to an audio/data recording device. No. 9 for direct connection to a facsimile machine. No. 0 used to recall the switch board in larger businesses and to cancel previous suffix dialing in small businesses. The foregoing numbering proposals are not considered by the main telephone standards body to be part of network dialing. This leaves them in limbo. An international forum with a greater emphasis on customer requirements might see things differently. S.F. ©TeleGeography, Inc. 1993 35 TeleGeography 1993 Global Numbering All of this makes having a fool-proof, intemational way of addressing the myriad communications terminals in the world -- from smart card operated mobile terminals to company laser printers to the home phone -- many times more complex. The proliferation of communications terminals also makes solving the numbering problem ever more urgent. The current global numbering standards (E.163, E.164 and E.168) were devised by monopoly national carriers and implemented through simple bilateral connection agreements. Fortunately, most national-boundary changes have been accommodated by this early numbering structure. But service competition is now putting more pressure on numbering authorities. So too is the recent expansion of regional carriers into trans-national telecommunications corporations, with a mix of international virtual network services. To date, solutions to these numbering problems have largely been seen from the technical and/or marketing perspectives of carriers. Customers interests have primarily been considered only when new or more profitable services can be provided (e.g., pay-per-call offerings). As the impact of the number debates rises, however, perhaps, it is time for a change. And it raises the problem of whether to assign numbers to people in their personal or business/employment capacity. This decision has important social as well as economic consequences. If separate numbers are used to identify people in their separate roles (as parent or manager or girlfriend), then we will need many more numbers than are now available. And better directory services to sort through them. At root, however, the major problem with new numbering plans is not technological but social. Technologies may change overnight and can be introduced to a network in months or years. But it takes decades to implement number changes at the pace that the society can absorb. The cost of even incremental changes (a new city or regional code) can be enormous, as London and Tokyo have recently learned. Hence, it is probably safe to say that the steady shift away from geographically-based numbering plans is likely to be a painful and contentious process for years to come. It is essential that users be involved. END What appears to be needed in the numbering field, as with many other standards matters, is a consumer body capable of grappling with these deeply technical issues and able to state the consumer interests with conviction. One place to start might be electronic directory services. They are an integral part of numbering. They provide alternatives to numbering portability problems and can be the source of messaging and call-forwarding information. Yet, major carriers have been reluctant to provide cheap, accurate electronic directory services because paper directories are such a highly profitable "milch-cow". Consumer input is also needed regarding plans for UPT. The one person-one number UPT approach depends on intelligent network databases as well. 36 TeleGeography 1993 Global Numbering For Further Reading: CCITT Blue Book, Volume II, Fascicle 11.2, Section 5 Numbering Plan of the International Telephone Service [Recommendations E. 160 E. 167] (International Telecommunication Union, Geneva, 1989). "Questions Allocated To Study Group 2 (Network Operation) For The 1993-1996 Study Period," COM 2-l-E, World Telecommunication Standardization Conference, Helsinki, March 1993. "North American Numbering Plan Administrator’s Proposal On The Future Of Numbering In World Zone 1," A. Gaechter, Bell Core, January 2, 1992. Notice of Inquiry, Administration of the North American Numbering Plan, Common Carrier Docket No. 92-237, Federal Communications Commission (FCC), Washington, D.C., released October 29, 1992. "Australia’s Telephone Numbering Plan," Australia Office of Telecommunications (Austel), Canberra, April 1993. Numbering: Choices For the Future, Consultative Document, Office of Telecommunications (Oftel) London, June, 1993, "Telecommunications Numbering Policy For Europe, A Study For the Commission of the European Communities," 16 January, 1992, Ovum, Ltd., London. 37 TeleGeography 1993 Call Back Services :i~:::..:.: :..:ii!:.’.: A~tomated Call-Back Services: An Update by Daniel Briere and Mark Lagner Since the U.S. long distance telecommunications market was opened to competition in the 1970s, resale of telecom services has been a fact of life. Various types of resale offerings have flourished -- facilities-based private line resale, switchless resale, international private line resale, and now, "call back services." All of these resale services try to take advantage of differences in the way carriers price their services for various market segments. Call back services, for example, generate profit from the disparity in international call prices around the world. The Call back provider offers customers in high-priced countries access to switches in low-cost countries. These switches take a caller’s request for an outbound international call originating in a high priced country and provide the caller with a dial tone in the low-cost country. Who are the main players? The call back mzrket is becoming crowded. Most call back providers operate out of the U.S. Market leaders include companies such as Viatel and IDT of New York, TeleNational of Omaha, NE, and Telegroup of Fairfield, IA. Each of these companies has their own approach to the market; all except IDT provide some form of value added service capabilities in conjunction with the inexpensive call back arrangement. The number of players in the market will continue to grow rapidly as new providers discover the call back market and established players recruit alternative marketing channels. What is the technology? Call back can be provided through any type of switching equipment, including PBXs and central office switches. However, most call back providers use "programmable" switches manufactured by companies such as Summa~ Harris or NACT. Such switches have a flexible architecture allowing for the integration of billing and value-added services, e.g., voice mail platforms, which call back providers offer as part of their services. A basic programmable switch costs approximately $45,000; a fully outfitted one may cost $200,000 or more. For individual companies which would like to set up their own call back arrangement, small "black box" systems, provided by companies such as Logotronix, provide the call back capabilities for one or two lines for approximately $1,000. What will happen in the future? Resale almost never occurs as planned by carriers or regulators, thanks to the entrepreneurial spirit and staying power of resellers. These companies are adept at looking for the easiest way around obstacles and the next available opportunity in the market. The history of the U.S. market suggests that resellers do not easily go away. Some governments and carriers have tried to stop call back resale by making it illegal. Call back providers have not responded by going out of business; they have concentrated their energies on working around the language of the law and new service platforms. Some hide behind "value added" services; others alter the product slightly so the definition used in the laws no longer specifically applies. The bottom line is that call back will not go away because a particular carrier or government wishes it so; the business will only wind down when the conditions which created the profitable call arbitrage -- high international accounting and eseller Switch ’ TA~41 Fm Automated Call Back System Priwte Netwo~ FsctllW Customer accesses private line network. Customer Is provided ©TeleC~oice, Inc. 1993 with U.S. dial tone and places calls as ff in the U.S. 38 TeleGeography 1993 :- Call Back Services collection rates and dissatisfied local customers -- begin to disappear. The fact that so many customers are willing to move so quickly to totally unknown providers should be a shock to carriers. Based on an internal study, one South American carrier has made estimates that more than 5% of its international voice traffic will be carried by resellers by the end of 1993. The primary reason: The huge discounts offered by the competing services. Even at low rates, the profit margins can be substantial. The next big profit booster for the call back providers will be when international toll-free ("1-800") rates drop to the same range for outbound IDD rates. As this occurs in the U.S., there will be even more room for resellers to lower rates in response to competitive challenges by foreign carriers. Is there no end to the price competition? Yes, but only when prices approach costs. The problem for most countries where call back is common is that they do not want to lower their rates. The result will be that calling streams will be diverted to resale, and if not call back, then something else. Resellers in the U.S. have proven that "where there is a will, there is a way" -- a phrase which nicely captures the attitude of most call back entrepreneurs. Using hubbing in a call back arrangement. 1902 © TeleChoice, lnc. 1993 Adapted from the TeleChoice Report on International Call Back Services by Daniel Briere and MarkLangner. For additional information contact: TeleChoice, Inc., 15 Bloomfield Ave., Suite 3, Verona, NJ 07044, Voice: (201) 239-0700; Fax: (201) 239-4443. 39 Satellite Phones ’iii::’ili: .i Satellite Phones and Cross-Border Cellular Service: New Challenges For Incumbent Carriers by Rob Frieden Mobile communications is one of the largest new markets for incumbent telecommunication carriers as well as their competitors. The global base of cellular radio subscribers is fast approaching 30 million; it was already 23 million in January 1993, up 43% from January 1992. Moreover, cellular telephony is becoming as important in developing countries as in industrialized ones, with nations such as Thailand and the Philippines having higher ratios of wireless to wireline phones than many OECD countries. As the base of wireless subscribers mounts, some industry observers have begun to predict that the distinction between fixed and mobile communications will begin to disappear in the mid- 1990s. Well over 90% of cellular radio calls today originate or terminate on the wireline network. End- to-end wireless calls -- cellular handset to cellular handset -- are still the exception. International cellular service (cross-border roaming aside) is even rarer, with the only large scale service now available through Inmarsat’s global satellite system. See Figure 1. Over the next decade, however, this situation could change significantly. A range of new trans-national mobile services will come on line and a growing volume of traffic will be routed outside the wireline Public Switched Telephone Network ("PSTN"). These new wireless services will "by-pass" traditional facilities and toll revenue division arrangements by using terrestrial and satellite systems as well as new interconnection arrangements between and among wireline and wireless systems. AT&T’s recent $12.6 billion offer to acquire McCaw Communications, Inc., the largest U.S. cellular telephone carrier, evidences the stakes involved and the desire of wireline carriers to vie for the wireless market. Incumbent carriers are reacting with a mix of fear and skepticism. There is fear that insurgent mobile telecommunication carriers will be able to migrate traffic and revenue to their own networks. Yet, there is skepticism that these "by-passers" will be able to carry large enough volumes of predominately wireless traffic at a cost which is significantly below that of the terrestrial network. The remainder of this article looks briefly at: (a) current proposals for end-to-end international mobile services; and (b) the ways in which mobile carriers will likely settle accounts between themselves and with terrestrial carriers. A concluding section comments on the impact that cross-border mobile service arrangements may have on the traditional interconnection and settlement practices of terrestrial carriers. A New Service Paradigm For Mobile Communications Rob Frieden is an Associate Professor in the School of Communications at Penn State University. He was formerly Deputy Director, International Relations, for Motorola’s Iridium satellite venture and also has served on the legal staff of PTAT, the first private trans-Atlantic fiber optic cable. Since its inception in the early 1980s, most wireline carriers have provided cellular radio service as an adjunct to their existing services. Although cellular radio required a network of new base station facilities, the core infrastructure was already in place, because the vast majority of traffic either © TeleGeography, Inc. 1993 40 TeleGeography 1993 Satellite Phones originated or terminated on wireline facilities. Until recently, therefore, most wireline carriers saw mobile radio primarily as a secondary transmission vehicle. It would add, not subtract, traffic from the wireline PSTN, which was the core facility. This meant that because terrestrial carriers continued to own "bottleneck" wireline distribution facilities, they could negotiate access terms with the mobile services industry from a position of strength. Consumers had a similarly limited view of cellular telephony. For most, the service was simply a convenience a way to avoid pay phones and to avoid "dead time" when driving to and from work or making service calls. The average subscriber did not think in terms of bypassing the conventional telephone network or using mobile facilities to avoid long distance toll charges. (Indeed, there is substantial evidence that, even in competitive markets such as the U.S., most cellular subscribers don’t even know the identity of their long distance carrier). Mobile radio fundamentally was a local service, limited by the range of transmitters, _operating on a few channels and without broad geographical coverage. This comfortable world is now changing nationally and internationally. In Europe and North America, continent-wide cellular radio systems will soon Pae~ Ocaan region mobile satellite Two.way €omnlutd~Uon between a telex subscriber on land and a mobile satellite temdnal Two.way €~mmmJ~Uml between 2 mobile satelUte Source." Adapted From Singapore Telecom A message is sent j lo anobher mobge lemdnal whidl can be carded in a briefcase 4i TeleGeography 1993 Satellite Phones Figure 2 How Satellite Phone Calls May Be Routed Motorola’s Ori~nai Iridium Plan A fleet of sophisticated satellites in low-Earth orbit pick up mobile phone calls from anywhere on Earth and could relay them from satellite to satellite, bypassing long- distance companies. Competing Satellite Plans The satellites’ only role is to link mobile phones to regional ground stations. Calls are switched and routed between ground stations over long distance phone lines. Far fewer satellites are needed, especially if unpopulated areas and oceans are not covered, and the satellites can be much simpler in design. SOURCE: Office of Technology Assessment, 1993, based on Motorola, Loral, from The New York Times, August 26, 1992, p. D5. become a reality. It appears that the level of demand and entrepreneurial interest will support multiple operators: there will be a mix of conventional mobile radio and regional geostationary satellite systems. Later these systems will be joined by micro- and pico-cell Personal Communication Networks ("PCN") plus a new series of low earth orbiting, satellite delivered global phone services. Taken together these systems may give mobile telecommunications the potential to compete head-to- head with terrestrial networks in terms of coverage, price and versatility. Satellite-Delivered Mobile Telecommunications At least a dozen Low Earth Orbiting (LEO) and Middle Earth Orbiting (MEO) satellite systems for mobile telephony have been announced. The best known include Iridium, a consortium led by Motorola; TRW’s Odyssey; the Loral/Qualcomm Globalstar; and Inmarsat’s Project 21. With some variation, each of these ventures aims to provide ubiquitous, wireless, digital coverage to pocket-sized telephones via a fleet of satellites. See Figures 2 and 3. 42 Satellite Phones For example, Iridium plans to build a new mobile service from scratch by investing $3.4 billion in a constellation of 66 new LEOs. These satellites will operate in circular, polar orbits at an altitude of 42.5 nautical miles. In August 1993, Iridium announced that it had obtained binding investor commitments for an initial equity offering of $800 million. Investors in Iridium include: BCE Mobile of Canada; Muidiri Investments BVI, Ltd., a Venezuelan consortium; Raytheon; Sprint, the third largest U.S. long distance carrier; Mawarid Group, a diversified Saudi Arabian industrial group; Khnmichev Enterprise, a Russian aerospace engineering organization and manufacturer of the Proton rocket; China’s Great Wall Industry Corp.; United Communication Industry Co., Ltd., a major cellular and paging operator in Thailand; and a consortium of 18 Japanese companies led by Kyocera and Daini Denden (DDI), Japan’s second largest domestic long distance carrier. In contrast, Inmarsat’s Project 21 vision anticipates incremental service evolution; service would start with a fourth generation of high-powered GSO (Geostationary Orbiting) satellites in the late 1990s, possibly augmented by MEO satellites and terrestrial cellular systems. Inmarsat recently introduced a suitcase size portable (Standard M) terminal designed to provide mobile phone and data services via the cooperative’s existing GSO satellites. Inmarsat plans to introduce a hand held Standard-P terminal to provide global satellite phone service in conjunction with future generations of satellites. Proponents of Mobile Satellite Systems (MSS) recognize the need to integrate their services with available terrestrial options. MSS users are likely to pay $3.00 or more per minute. Thus, users will want to have a transceiver that can access lower priced cellular radio or Personal Communication Service (PCS) where available. Additionally, the deployment of dual-mode (terrestrial and satellite) terminals means that existing terrestrial operators might consider MSS a new profit center, rather than yet another competitor and potential bypasser. Revenue Settlements and Facilities Access As mobile telecommunication carriers generate larger long distance and international traffic volumes, they will more aggressively seek alternatives to conventional traffic routing and toll revenue division arrangements. AT&T’s proposal to acquire McCaw Communications has focused the industry’s attention on this issue in the U.S. Although the traffic routing and settlements impact of the AT&T-McCaw deal probably will not become clear for sometime (even assuming approval by the FCC and antitrust authorities, the deal will not close until 1994), it has highlighted the need for established carriers to rethink their role in linking mobile systems together and the charges connected therewith. Otherwise they may be by-passed by a mix of mobile carriers and their wireline allies. If that happens, a new settlements process for long- haul mobile traffic may arise which involves the daisy chain interconnection of local facilities or a new type of settlements arrangement between mobile carriers and resellers. Today wireline carriers do not compensate wireless carriers when they terminate calls originating on wireline networks; the charge is passed through to the mobile service subscriber. This is also true for international calls; the wireline carrier, not the cellular radio carrier, receives the lucrative settlement payment from the foreign carrier which originates a call. As the subscriber base of cellular companies rises, however, the ability of wireline carriers to "fence out" their cellular correspondents from this international revenue stream is likely to be less and less defensible. Further, mobile telecommunication operators and their subscribers now pay a separate and arguably over-priced access fee for originating and terminating local calls on the PSTN. Long distance charges (domestic and international) are simply added to that. Thus, as long distance call volumes rise, mobile operators have an incentive to bypass these access charges via leased lines that have lower access charges and are exempt from the application 43 TeleGeography 1993 Satellite Phones Major Satellite Phone Ventures Figure 3 Name of System Project 21 Globalstar Odyssey Al’ie$ American Mobile Satellite Corporation (AMSC) Description of Venture A developing concept for Inmarsat’s 4th satellite generation to serve handheld land mobile terminals without burden to the cooperative’s maritime service mission. Cost: $1 billion or more. A constellation of 66 low earth orbiting ("LEO") satellites providing ubiquitous service; design includes inter-satellite links to reduce ground segment complexity. Cost: $3.4 billion. 24-28 LEO satellites for U.S. or global coverage; technical design includes code division multiple access. Cost: $748-829 million. A constellation of 12 middle earth orbiting (MEO) satellites in 3 planes. Cost: $1.3 billion. 48 polar orbiting LEO satellites in 4 planes. Cost: $292 million. The company plans to provide mobile telephone, paging, data communication and geo-locational services via two dedicated GSO satellites. Prior to the launch of its ftrst satellite, scheduled for late 1994, AMSC will provide service using capacity leased from I.marsat. AMSC has a U.S. domestic monopoly and will serve the North American market. Owners lnmarsat, a cooperative of over 70 primarily government owned carriers based in London, U.K. Motorola, Inc., Shaumburg, IL and a consortium of investors including BCE Mobil; Sprint; Khnmichev Enterprises (Russia); DDI (Japan); Great Wall Industry Corp. (China) and United Communication Industry Co. (Thailand). Loral Aerospace, Inc., New York, NY; Qualcomm, Inc., San Diego, CA. TRW, Inc. Redondo Beach, CA. Constellation Communications, Inc. (Microsat Launch Systems, Inc., Herndon, VA); Defense Systems, Inc., McLean, VA and private investors. Major shareholders in AMSC are: Hughes Communications, Inc. (Los Angeles, CA); McCaw Cellular Communications (Kirkland, WA); and Mobile Telecommunications Technologies Corp. (Jackson, MS). 44 TeleGeography 1993 Satellite Phones .-." -: .. ."i:.. ,:’:5~- ~-’ :..ii~?i ,,: ...... ~:" :’ ¯ -:’-"- ’ "’.,,i’.,~ : ’~ .-: .. ’ ’ International Cellular Roaming: Settling Accounts Figure 4 Today international cellular roaming (e.g., using a Venezuelan cellular phone in the U.S.) follows two primary settlement models: one for frequent settlements (typically in adjacent countries); and the other to handle infrequent roaming, where operators are likely to negotiate on an ad hoc basis. Frequent roaming typically requires standard billing and record keeping protocols; clearing houses to collect, process and format billing records; and a central bank to receive and disburse funds. All of these measures have been put in place to settle revenue for cellular roaming traffic between the United States, Canada and Mexico. These systems piggy-back on the systems for settling U.S. domestic roaming traffic. In the U.S., Cybernet Corporation, a wholly owned subsidiary of the Cellular Telecommunications Industry Association (CTIA), provides member carriers with standard protocols and billing formats for the exchange of roamer billing data. Two major clearing houses collect billing data and submit the financial position of member carriers to a designated bank for receipt or payment of funds. To support these arrangements, U.S., Canadian and Mexican carriers have executed agreements and implemented the necessary software programming for real-time "backhauling" of cross-border roamer traffic to the cellular user’s home carrier. Each operator has the capacity to meter roamer traffic and apply the agreed upon toll rate. Roaming of U.S. cellular units in other countries has not yet generated significant traffic streams, primarily because a number of different technical standards and rampant toll fraud. As yet, therefore, U.S. operators have not agreed to extend the geographical scope of the North America clearing houses arrangements beyond Mexico and Canada. Similarly, the promise of pan-European roaming has yet to reach fruition, although users can roam across the Scandinavian nations and implementation of North American style clearing house procedures is contemplated soon. Limited one-way international roaming opportunities do exist in South America. For example, BellSouth International provides subscribers in its five Latin American franchises (Guadalajara, Mexico; Caracas, Venezuela; Santiago, Chile; Buenos Aires, Argentina; Montevideo, Uruguay) the opportunity to roam in North America. While the Latin American systems operate on the same frequencies and use the same analog technology as most United States systems, the roaming subscriber must physically bring in the radio for activation of a secondary telephone number. The Latin American cellular operator (the "home carrier") must notify BellSouth (the "serving carrier") of the number provided from a dedicated pool of BellSouth Mobility numbers. In effect, the foreign roamer is treated by the U.S. network as a domestic roamer, able to use the widespread service opportunities available to BellSouth subscribers. The roamer incurs regular domestic roaming charges that are first posted with BellSouth Mobility and subsequently forwarded to the foreign cellular operator for customer billing and collection. United States cellular customers do not have reciprocal roaming opportunities in South American, because of the absence of a clearing house that can validate calls and prevent toll fraud. Absent a roaming agreement, customers can make international calls using a credit card but at a rate well above typical roaming charges. R.F. 45 TeleGeography 1993 Satellite Phones Mobile Satellite Services: Settling Accounts Figure 5 Future access options for global satellite phone services are likely to include one or more of the following: 1) Facilities-Based Access Changes -- Cost-based interconnection to the Public Switched Telephone Network (PSTN) at an agreed international gateway on terms and conditions similar to those paid by other inter-exchange carriers; 2) Accounting Rate Based Changes -- Access to the domestic network at agreed international gateways based on the accounting rates negotiated between wireline carriers. 3) End-to-End Billing -- A sum of the tariffs method under which each carrier responsible for a portion of the routing is paid its tariff rate, which is passed through to the subscriber by the originating carrier. 4) Sender Keep All -- An agreement among carriers who hand-off and receive roughly the same traffic volumes to dispense with any revenue division or billing arrangement, opting instead to collect and keep all charges received from the originating caller, with a share paid to the satellite carrier. 5) Flat Rate or "Commission" Billing -- The satellite carrier compensates the terminating or originating terrestrial carrier based on a fixed charge per minute regardless of where the call is picked up or delivered. of accounting rate settlements. Ad hoc arrangements between adjacent or regional mobile carriers are also likely to emerge. Significantly, in the U.S., the international settlements policy governing terrestrial carriers has not been applied to cross-border cellular radio traffic thus far. See Figure 4. For example, United States subscribers using a Mexican carrier to route a call back to the United States pay a "roaming fee" which is not necessarily commensurate with the wireline accounting rate between AT&T and Tel Mex; nor is the cellular roaming necessarily divided equally between the Mexican and U.S. cellular operators as in the case for terrestrial accounting rates. Ad hoc arrangements outside the existing accounting rate region apparently also exist for cellular roaming in Europe. The prospect for wider utilization of mobile telecommunications facilities for routing international traffic thus is beginning to generate real concern among some carriers. The revenue impact of these arrangements is one issue. Another is the legality of ad hoc cross-border mobile settlements in lieu of the conventional accounting rate regime. This is also why regulatory hurdles (e.g., interconnection terms) rather than technology may be the most troublesome barriers faced by new satellite phone providers. See Figure 5. Beyond that, with "smart switches" and flexible routing arrangements, much of the international mobile traffic requiring interconnection can be aggregated in such a way as to make it difficult for the wireline carrier to identify the origin of the call or to require an accounting rate settlement. In most 46 TeleGeography 1993 Satellite Phones cases, the wireline carriers tasked with terminating traffic into the PSTN typically have no knowledge of where a call originates. Billing is done by the sending carrier. And, while automatic number identification (ANI) and out-of-band signaling technology (such as SS7) may be in place in the future, it remains uncertain whether foreign wireline carriers will be able to demand the availability of such information as a precondition to the carriage of mobile (or other) inbound foreign traffic. Without such information, incumbent carriers may receive foreign-originated traffic, but treat it as local. When this occurs, the terminating carrier misses an accounting rate settlement or at least some degree of financial compensation beyond what it would charge for conventional, often non-metered, local exchange service. Worst of all perhaps, the terminating carrier has no reliable means of calculating the volume of the traffic which should have generated inbound settlements. wireline settlements and when a growing percentage of traffic never accesses the wireline PSTN. A "hands off" regulatory policy might well accelerate the demise of the existing accounting rate regime -- a regime which has been notably slow to flow through technology-based cost reductions to users. Yet, if change happens piecemeal, some mobile traffic will incur excessive charges to access the PSTN while other, probably larger, traffic flows are likely to avoid access payments altogether. In these circumstances, regulatory intervention may be inevitable. END In sum, over time, cross-border mobile radio services, the permeability of regional private line networks, transborder satellite footprints, resale of leased lines and a host of other factors seems likely to tip future settlements policies in favor of the mobile carriers and their customers. Conclusion Mobile telecommunications has evolved to a point where it can no longer be considered a minor aspect of the telephone world dominated by wireline carriers. Consumer demand for wireless services, and increasing use of mobile technologies for cross- border calls, will strengthen the importance of new wireless carriers and force incumbent carriers to rethink the terms and conditions for access to the wireline network. Further, at some point, incumbent carriers and regulators will have to confront the consequences of two increasingly independent systems for traffic routing and toll settlements. Mobile systems present financial challenges to the status quo both when they route traffic into the PSTN at rates lower than 47 TeleGeography 1993 For Further Reading Satellite Phones Douglas M. Hursey, "Inter-American Roaming," Wireless ’93 Cellular Telecommunication Industry Association (CTIA) Convention, Dallas, TX, March 2, 1993. Olof Lundberg, "Global Personal Communication: A Satellite Close At Hand." Notes For A Presentation To The Pacific Telecommunications Conference PTC 93, Honolulu, Hawaii, 21-23 January 1993. Robert Frieden, "IntemationaI Toll Revenue Division, Tackling the Inequities and Inefficiencies," Telecommunication Policy, Vol. 17, No. 3, April 1993. Gregory C. Staple, "Winning The Global Telecoms Market: The Old Service Paradigm and The Next One," Gregory C. Staple in G. Staple, Ed. TeleGeoCraphy .!992 (International Institute of Communications, London, 1992) The 1992 World Administrative Radio Conference: Technology and Policy Implication,% U.S. Congress, Office of Technology Assessment, OTA-TCT-549, (U.S. Govt. Printing Office, Washington, D.C. 1993) 48 The Explosion of Place TELEGEOGRAPHY AND THE EXPLOSION OF PLACE Why The Network That Is Bringing The Worm Together Is Pulling It Apart By Gregory C. Staple If you often feel that the world simultaneously is getting bigger and smaller, coming together and falling apart, you are in good company. In the last year this theme has become common in novels and news magazines alike. "All the countries were big, single blobs of color, from one side to the other", muses the young heroine in William Gibson’s latest 21st century tale, Virtual Light, as she looks at a set of vintage maps. "There’d been countries big as anything: Canada, USSR, Brazil. Now there were lots of little ones where those had been .... Even California had all been one big state, once." Henry Louis Gates, the chair of Harvard’s Afro- American Studies Department, wrote in a similar vein in The Economist’s 150th anniversary issue (11 September 1993): "We live in confusing times .... [W]hat communism joined, nationalism has been all too eager to pull asunder. In the industrialized West, though ... never in this century has ... supranational unification seemed closer to realization." "Today, both the forces of fragmentation and the forces of consolidation are abundantly on display," continued Gates. "Just which side is history on, anyway?" Patrick Glynn, another American scholar, recently framed the issue this way in Commentary magazine (July 1993): "On every continent, in almost every major nation and in almost every walk of life, the overriding political reality today is that of increasing social separatism and fragmentation...by ethnic group, race, religion and even (to a less dramatic extent ...gender or sexual orientation)." This clash, observed Glynn, "between ethnic and other types of particularism on the one hand, and, what might be called democratic universalism," on the other hand, seems to be replacing the Left-Right and class divisions of the past century; it is becoming "the new dialectic of a new age." But, why is this th__ee "new dialectic" of our time? What is driving the process? Why is the world being divided even as it is being unified? How is it that an increasingly cosmopolitan economic order seems to progress in step with an ever greater insistence upon local identity? Is this phenomenon a passing one, limited to the end of the Cold War? Or are there underlying factors, operating on a wider scale, which will make the tension between localism and globalism, between self-determination and economic integration, between nationalism and trans- national (human) rights, a much more permanent fixture on the world stage. Any answer to these questions will inevitably be a partial one. But surely the novel role played by information technology and, more specifically, the global telecommunications network -- the largest and most complex machine the world has ever known -- helps to solve this intellectual puzzle. There are two main reasons. First, the network, as we now know it, did not exist a generation ago. What was once, circa 1960, a disparate group of poorly connected national telephone systems has become an integrated web of telecommunications and computer terminals implicating half of humanity. Thirty years ago, North America and Europe accounted for over 85% of the world’s 150 million telephone lines. Fax was virtually unheard of and computer networking was an oxymoron. The sole trans-Atlantic telephone cable could handle but 89 calls and a handful of tropospheric radio links were the only way for the human voice to cross the Pacific. By contrast, the global network now comprises over 600 million telephone lines and more than 1.2 billion terminals in approximately 190 countries. Today’s telephone subscriber can expect to spend a year of their life on the phone (two years, if they live in North America). The cable and satellite network across the Atlantic can now handle almost a million s,.’multaneous calls. And, in 1993, almost every aspect of people’s work and play touches the network at least once. Second, the global telecommunications network has a uniquely contradictory dynamic. Though the network actually has fewer and fewer moving parts (silicon chips are the order of the day), in the social and economic sphere it works as if powered by a giant array of mechanical pistons. The downstroke of the piston creates a geographical implosion, compressing space and time, and © TeleGeography, Inc. 1993 49 The Explosion of Place bringing economies together. But with the piston’s upstroke comes a geographical explosion, fragmenting shared experiences and values, and creating a multiplicity of new places to be cultivated by partisans of one vision or another. Up and down the pistons go, in country after country, day in and day out, three hundred and sixty five days a year. And with each cycle, as each new telephone and computer and facsimile connection is made, the piston simultaneously compresses and explodes the everyday geography of its users. Most chronicles of the communications revolution have focused on the first half of the piston’s cycle -- on the downstroke, the geographical implosion. But in many ways, it is the upstroke -- the powerstroke -- and the geographical explosion of place accompanying it, which is more telling and to which this essay draws attention. Whose side is the network on? Both sides. Everyone’s side and no one’s. The network is at once universal and parochial. It affords people great intimacy and yet can make a virtue out of their separateness. It is able to unify and to divide; to open new markets even as it razes old ones; to knit together a diaspora across the world while factionalizing a tightly knit community elsewhere. It is at once specialized and non-discriminatory; public and private; empowering politicians and pornographers, international conservation groups and local mining companies. It respects neither geography nor sovereignty and thus provides a common vehicle for both the centripetal (economic) and centrifugal (social and political) impulses of the day. The network’s routine contraction and expansion of place, and the hopes and fears engendered by these changes, have important implications for the telecommunications industry as well as for public policy. The network is creating a vast new electronic landscape, even as it devalues the physical locations many people have known for generations. Not surprisingly, this process has triggered contrary responses. As George Gilder has said, "In an age when men can inscribe new words on grains of sand [silicon chips], particular territories are fast losing significance." It is people and ideas that matter, not places and things. Yet, "The network works as if powered by a giant array of mechanical pistons. The downstroke of the piston creates a geographical implosion, compressing space and time, and bringing economies together. But with the piston’s upstroke comes a geographical explosion, fragmenting shared experiences and values, and creating a multiplicity of new places to be cultivated by partisans of one vision or another.... Global telecommunications has not led to the end of geography so much as to the rebirth of place." "blood and soil" are still paramount in many parts of the world. And there is a strong, often desperate impulse to defend the land of one’s birthright (by force of arms, if necessary) lest this sacred space be polluted by the outside world or washed into an electronic no-man’s-land. Religious fundamentalists are often at the forefront of this reaction. For example, Iranian customs officials recently seized scores of satellite TV dishes smuggled across the border from Turkey. Private possession of satellite antennas, which can receive various foreign TV programs, is forbidden in Iran. At the same time, however, there are growing popular demands to expand and secure the myriad new electronic spaces the network has opened. The network brings new worlds into people’s homes. And here, the operative words in many countries are "privacy" and "property rights," community values be damned. In sum, the global telecommunications network has not led to the end of geography so much as to the rebirth of place. To better understand how we have reached *,his geographical juncture, a brief historical aside is useful. The Rural Bias Telecommunications was born into an overwhelmingly rural society. At the time of the telegraph’s inventi_on in the 1840’s at least two-thirds of Europe’s people lived in the countryside. Similarly, of America’s 20 million inhabitants, only 1 in 7 lived in cities; rural populations predominated in much of Asia and Africa as well. The frontier, be it West of the Mississippi or East of Aden, was remote. On land, news traveled at horse speed and intercontinental connections depended on the vagaries of wind and sail. In this environment it is little wonder that the telegraph and later the telephone (1876) were hailed as tools of unification. The telegraph would end the tyranny of space. And with trans-oceanic cables, America would no longer be isolated from Europe, nor Europe from her colonies. There would be instantaneous communications from Vancouver to Bombay. Trade and commerce would 50 TeleGeography 1993 The Explosion of Place ... ’ ...... ::i::iiiii:i?’ flourish and military surprises would become a relic of the past. Of course, in the ensuing century, economic and political integration proceeded fitfully, at best. Much of the Western world lived through wars and depressions on a scale which led to an isolationist sentiment among many. But, in the 1960s, following the launch of the Russian satellite, Sputnik, the popular vision of telecommunications as the great integrator was given new life. Marshall McLuhan, the Canadian communications theorist, was its best known spokesman. In his book, The Gutenberg Galaxy (1962), and elsewhere, McLuhan suggested that the widespread availability of satellite communications would create a "global village". We would all be neighbors on the same "electronic common." It may also be worth recalling the words of Arthur C. Clarke, the intellectual father of the communication satellite, at the 1971 signing ceremony establishing Intelsat, the multinational co-op which owns the world’s principal telecommunication satellites. "What the railroads and telegraph did [in the U.S.] a century ago," said Clarke, "jets and communications satellites are doing now to all the world .... For today, gentlemen, [there were apparently no women signators] whether you intend it or not -- whether you wish it or not -- [y]ou have just signed the first draft of the Articles of Federation of the United States of Earth." Twenty years later, however, Clarke’s United States of Earth remain elusive. And the global electronic common is still embryonic. The closest proxies, CNN, the satellite distributed news network, and the Internet, the trans- national matrix of computer networks, together reach less than 1 or 2% of the world’s population. Nor do we live in one electronic village. The telecommunities which are constantly forming and reforming on the network are as likely to share a common set of goals or values as those of any two households chosen randomly from the world’s telephone directories. To be fair, McLuhan’s legacy is easily misread. While McLuhan did indeed state that electronic media had contracted the world into "a single large village" he also argued that the "electric age, by involving all men deeply in one another" would lead to a new "tribalism" -- a tribalism which would make impossible Western "literate man’s dream of [a] solution to the problem of human differences." In 1990s terms, cyberspace is tribal space, universalist pleas notwithstanding. With the benefit of hindsight, we also know now that the promise of a global village has proven false for another reason both McLuhan and Clarke discussed -- ever cheaper international communications. The proliferation of electronic places in the world hence is largely a matter of economics. As the price of getting on the telecoms map has fallen, the demand for places has mushroomed. The luxury overseas call of the 1960s has become the routine global chat of the 1990s. A Generational Leap In January 1961, when President John F. Kennedy took office, a three minute phone call from the U.S. to Europe or Asia (assuming a connection could be made) typically cost $18 -- the equivalent of $90 today. The cost of calling in the other direction usually was much greater. Prices were high, in part, because the supply of international circuits was limited. High capacity communications satellites were a decade away and, as noted, the only trans-Atlantic telephone cable, completed but four years earlier, had less than 100 voice paths. (Radio telephone service began across the Atlantic in 1927 and until 1956 this very limited service was the only means by which Europe could talk to America.) The first trans- Pacific telephone cable did not begin service until 1964 and could handle just over 150 calls. However, when President Bill Clinton took office in January 1993, there was sufficient satellite and cable capacity to accommodate at least 800,000 simultaneous telephone calls across the Atlantic. Carrier tariffs for trans- Atlantic calls had fallen to $1.00-$1.50 a minute in many European states and half that in the U.S. Not surprisingly, these price changes and the expanding base of telephones across the world dramatically changed the electronic landscape. Again, the U.S. provides an example. In 1960, Americans made just 1.3 million overseas telephone calls, of which almost 80% were to Western Europe and the Caribbean. Only five countries -- the U.K., Germany, the Bahamas, Cuba and Japan -- accounted for 50,000 or more calls annually. By 1990, the U.S. made approximately 50,000 calls daily to a score of countries, including Korea, Brazil, Taiwan, Colombia, Australia, Italy and the Dominican Republic; the U.K. received over 250,000 calls each day. These statistics point to another fundamental change in our electronic terrain. Thirty years ago a few industrialized countries owned the vast majority of the world’s telephones; computer modems were the preserve of Big Science and Big Business and the fax machine was a poor substitute for telex. Yet, by the early 1990s, electronic mail and fax machines were consumer staples in many countries and the number of telephone lines had quadrupled to approximately 600 million. Further, in less -51 TeleGeography 1993 The Explosion of Place than a generation, developing countries from Indonesia to Iran, and from Brazil to China, had added more telephones than the world’s industrialized countries (excluding the U.S.) had installed during the telephone’s first 100 years! By comparison, the number of international airline routes, bounded as they are by the need for airports, connecting roads and the like, has yet to exceed a few thousand. And even that quintessential 20th century travel machine, the automobile, found itself hard put to compete with this explosion of new spaces on the network until, of course, it gained a phone of its own. Yet, this explosion of places on the global network tells only part of the story. Many of the same advances in micro-electronics and information processing which made international telephony affordable for "Our new electronic communications devices and the cybercompasses which direct them -- the telephone key pad, computer "mouse" and TV remote -- have created a vast new geographic overlay .... This virtual geography has a hold on our lives every bit as powerful as that of the places where we live or work or go to school. Indeed, in many cases, these places have become folded into this new electronic terrain." millions also dramatically expanded peoples’ personal communications space through other media. The Sony Walkman and the Video Cassette Recorder (VCR), almost unknown in 1970, now offer a novel electronic environment for tens of millions of people. So too, of course, do cable and satellite TV. Taken together, the worldwide take up of these new electronic communications devices and the cybercompasses which direct them -- the telephone key pad, computer "mouse" and TV remote -- have created a vast new geographic overlay. We call it by various names, depending on how we enter and exit or what we do when we are there. The network. Cyberspace. Television. But whatever the name, there is little doubt that this virtual geography has a hold on our lives every bit as powerful as that of the places where we live or work or go to school. Indeed, in many cases, these places have become folded into this new electronic terrain. With this background in mind, the next section of this essay looks at how the network’s simultaneous explosion (the piston’s upstroke) and compression of place (the downstroke) is affecting our views on three subjects: immigration, trade and community values. Immigration Immigration is as old as the nation state. But never before has it been possible for so many immigrants from so many different countries to call home. The upsurge of immigration in the 1980s has thus made the network’s contradictory dialectic an everyday fact of life for more and more nations. For example, a June 1993 U.N. population study reports that over 100 million people (2% of the world’s population) left their home country since 1980 in search of a better life. There are now more than 20 million legal immigrants in Western Europe and since 1990 the number has been growing by almost 2.5 million annually. Countries such as Italy, Spain and Greece, which historically had a net outflow of immigrants, now have a net influx. In the 1980’s, the U.S. gained approximately 9 million new immigrants. One third came from Asia, a quarter from Mexico and another quarter from Latin America and the Caribbean. While the new immigrants have had a hostile reception in some countries, they have been a boon to telephone carriers. Over the last few years, in both the U.S. and Europe, most of the fastest growing telephone routes have been to countries which have seen large scale emigration. This new market for social calling -- to Guatemala and the Philippines, to Turkey and Tunisia, to South Africa and India -- has boosted demand during a comparatively weak business market. With each call home, immigration has helped the world’s major carriers to build out their networks, to amortize the next generation of undersea cables and to recognize ever greater economies of scale. Thus, while calling home may be a very personal experience for the participants, from the standpoint of network economics, each immigrant’s call brings the world closer for everyone. The global reach of the network has also allowed immigrant communities to turn their very separateness into an asset by leveraging their multi-national contacts. Indeed, Joel Kotkin’s insightful new book, Tribes, argues that the most successful ethnic disaporas today provide a model for advancement in the 1990s. "As the conventional barriers of nation states and regions become less meaningful," writes Kotkin, "it is likely [that] dispersed peoples -- and their worldwide business and cultural networks -- will increasingly shape the economic destiny of mankind." Tribes focuses on the Japanese, the Chinese, the Indians and the Jews. But Kotkin sees other ethnic 52 TeleGeography 1993 The Explosion of Place ..... communities gaining ground as well, including Armenians, Koreans and Palestinians, to name but a few. Kotkin also contends that "[t]he power of global tribes derives from their successful coalescing of two principles that many people believe are contradictory: an "intrinsic tribal" sense of ethnic identity which is historically unique and, secondly, "the ability to adapt to a cosmopolitan global economy." To bridge these two worlds, Kotkin states, the most successful "global tribes" have placed a premium on education as well as religious or ethical training. Kotldn is only partially right. Education and religion often do help people feel at home abroad. But in our networked world, "tribal" and "cosmopolitan" values are no longer antagonistic. They are complementary. It is the network’s very ability to sustain people’s sense of separateness, to create a special place for their own way of life, while connecting them to the larger world that is its leitmotif. The network’s capacity to transform a community’s separateness into a global asset can also create problems. As the telephone makes next door around the world, immigrant families may be less likely to look across the street for social and economic support. Similarly, while telephone service may be invaluable in helping immigrants reach out to other local communities for jobs, social assistance, schooling and child care, it may also unite them with their own nearby electronic neighbors for shopping, entertainment and charity. One final point about calling home which has much wider implications: No passport is needed to traverse the global network. Its electronic highways and byways may be messaged by citizen and alien alike. The network thus unites families and factions; it attracts brides and bombers; and it allows doctors to follow their patients abroad as much as it does drug dealers their cargoes. It is this public openness, affording a kind of universal private asylum, in a world still ruled by border guards and immigration quotas, which exemplifies the contradictions at the network’s very core. Trade and the Economy The network’s ability to compress the world’s geography (the piston’s downstroke) is perhaps most "In our networked world, ’tribal’ and ’cosmopolitan’ values are no longer antagonistic. They are complementary. In fact, it is the network’s very ability to sustain people’s sense of separateness, to create a special place for their own way of life, while connecting them to the larger world that is its leitmotif." evident in the economic sphere. Here too, however, the contrary is true as well. Coca Cola, BMW, General Mills, Levi Strauss, Visa, Intel, Citicorp, Sony. The brand names which we-see around the world testify to the critical role which international telecommunications has played in unifying distant markets and in the rise of global corporations. The consumer benefits have been enjoyed by millions. There are also economic costs. Some companies and communities are able to leverage the network to open up new horizons and others find their vistas have been foreshortened by the new electronic space they share with people and factories an ocean away. On the one hand, instant global communications allows companies to sever their ties to nearby markets and to operate on a trans-national scale -- matching buyers and sellers, marshalling inventories, coordinating production runs and filling orders (just in time). These economies of scale and lower transaction costs have led to large economic savings. Yet, on the other hand, the network is a grim reaper. By enabling companies to exploit the differential resources and productivity of one locale vis-a-vis another, the network can impoverish whole communities even as it enriches others. "Reach out and touch someone," AT&T advertised in the 1980’s. But what happens when whole countries reach out and touch one another via faxes and bank wires? Just ask people in the American Midwest or any European manufacturing center. When Osaka (home of Nissan) or Ulasan (the Korea home of Hyundai Motors) uses these new information pipelines to reach out and touch Detroit or Lyon, the politics of nations are thrown into the balance. Japanese and Korean cars may be or may not be superior. But without the means which moderu telecommunication and broadcast facilities offer to finance and market these products quickly, their acceptance in foreign markets would probably take decades not years. And what is true of cars, is also true of a host of other products, from textiles to machine tools and foodstuffs to finance. John Naisbitt first pointed to the role of telecommunications in this restructuring of community and (work) place in his popular I982 book, Megatrends. Cheap global communications, wrote Naisbitt, "collapses 53 TeleGeography 1993 The Explosion of Place the information float--the time it takes information to travel from sender to receiver." The result? Take away the float and many domestic enterprises are left trying desperately to tread water. For absent a life raft of trade barriers, the tempo and cost of production in Indian or Korean workshops and factories immediately becomes the regulator of commerce a continent away. Lee Kuan Yew, architect of Singapore’s economic miracle, bluntly summed up this process as follows: "Globalization ... widen[s] income difference in each society. America’s top 10% will still enjoy the highest incomes in the world. But the wages of its less educated citizen will drop to those of workers in the developing countries with equal or higher educational standards. Telecommuting transfers jobs worldwide." Community Values The network’s piston-like power to compress peoples horizons even as it creates new spaces for others is not limited to the economic sphere. This dynamic is equally evident in the social arena as well. And it has brought conflict with it too. For much of the world’s "We may all live in electronic neighborhoods, but the residents of one digital street often know nothing of what the "Boyz in the Hood" are doing on the next one... When a newspaper in Texas drew attention to the fact that a local computer BBS had an archive of pornographic photos, ... the Usenet newsgroup ’alt.sex pictures’ instantly moved to Finland.... Of course, ten years from now, when many computers have wireless modems, and satellite telephones are a commonplace, the BBS computer itself could be moved around the world at will. Thus, for the guardians of community values, the world has suddenly gotten impossibly large." history, geography has provided something of a buffer, a "greenbelt," between the values of city and countryside and between one religious community and the next. Yet in much of the world the electronic network now threatens to plow under what remains of this greenbelt. Telegeography transports every villager’s home into the metropolis and settles the atheist next door to the true believer. To date, television, has suffered the brunt of the moral backlash loosed by this new electronic geography. Video cassettes are confiscated by border police; satellite channels are denied "landing rights;" TV shows with "too much" violence or sex are threatened with advertising boycotts; and broadcast tribunals are created to protect the national culture. The telecoms network, however, is fast becoming the next battleground. In its infancy, the telephone only brought foreign voices into our homes, and then but fleetingly. The marketplace for clothes, entertainment, banking etc. remained in the community. So too did the library and the newsstand. But as the network has begun to change from a purveyor of POTS (plain old telephone service) to PANS (pictures and new services), the clash between local and metropolitan values has heightened. By 1993, the network had made the life of the city, its markets and its fashions, its vices and its knowledge, a part of the common electronic landscape. This expansion of social space predictably has made many people very uneasy. But others have welcomed the opportunities which new network services are making available and are keen to test the limits of these electronic frontiers. The values of virtual communities are being pitted against the ethics of actual communities. People "on the line" are ranged against those "on the land." And, as often as not, there are also strong differences among the myriad communities which now call the net their home, just as there are among neighborhoods in the "real" world. A good overview of the these different communities is provided by the tens of thousands of free phone or "800" number services now available in North America and Europe. The directory for France’s teletex service, Mimtel, also is instructive. There are electronic neighborhoods for everyone; for psychotherapists, food fanciers, pet owners, language students, financiers and genealogists. The same is true of the computer world. According to Howard Rheingold’s new book, Virtual CommunitSr, in the U.S. alone there are now 60,000 electronic bulletin board services (BBSs) which allow people to chat with each other using electronic mail (E-mail). Each BBS supports a dozen to several hundred or even thousands of participants, notes Rheingold. "There are religious BBSs of every denomination, sex BBSs of every proclivity, political BBSs, from all parts of the spectrum, outlaw BBSs, law enforcement BBSs, BBSs for the disabled, for educators, for kids, for cults ... [the] list ... is dozens of pages long. [And] [t]he BBS culture has spread from the United States to Japan, Europe, Central and South America." Most people are unaware of this wildly-varied assortment of new cultures emerging on the network. We may all live in electronic neighborhoods, but the residents 54 TeleGeography 1993 The Explosion of Place of one digital street often know nothing of what the "Boyz in the Hood" are doing on the next one. When they do f’md out, the result can be outrage. One man’s computer game is another’s obscenity. One woman’s public data base is another’s national security threat. One community’s weekly electronic meeting is another’s blasphemy. Most of the activities on the network are private. Almost all countries have rules against wiretapping or eavesdropping on telephone or computer traffic. And that is one of the network’s great attractions; you can travel anonymously -- and leave your passport at home. For instance, many people who would not wish to be seen in an compact "adult" bookstore or a bordello integrated will happily spend a few minutes centralized or a few hours entertaining manageable themselves with pay-per-call sex prosperous services. Some nations have tried to control such network based "dial-a-porn" services by requiring the telephone companies to block pay-per-call numbers unless the worldly subscriber requests otherwise. But intimate the network is global. And dial-a- secure porn services, which are now a accessible multi-billion dollar business, can informed easily migrate off-shore. Blocking connected foreign dial-a-porn services is very difficult; it may mean blocking all telephone calls with the country housing the offending audio service. The technical difficulty of enforcing local values in a virtual world is even greater for computer-based services (i.e., the telephone network of tomorrow). For example, Howard Rheingold reports that when a newspaper in Texas drew attention to the fact that a local computer BBS had an archive of pornographic photos, as well, incidentally, as numerous mundane government documents, the "Usenet newsgroup ’alt. sex pictures’ instantly moved to Finland." Network traffic to Finland jumped significantly overnight, writes Rheingold, but there was little local authorities could do. Of course, ten years from now, when many computers have wireless modems, and satellite telephones are a commonplace, the BBS computer itself could be moved around the world at will. Thus, for the guardians of community values, the world has suddenly gotten impossibly large. The network has exploded the number of places which must be walled off. And. at the same time, it has collapsed the borders which formerly existed between one community and the next. Closing Thoughts How might the The Network’s Digital Dialectic Makes The World More .... dispersed balkanized decentralized chaotic divided And Makes One Feel More .... unique distant exposed alone confused displaced © TeleGeography, Inc. 1993 tele-communications industry and government respond to the new geographic era we have entered? I have two suggestions. First, the current debate in most countries regarding the future information infrastructure or "infostructure" deserves more candor. As one of America’s most beloved practical philosophers, Yogi Berra, said: "If you don’t know where your going, you might end up somewhere else." Too often, the job creating, new business, economic integration arguments advanced for extending and widening a nation’s electronic highways have been accepted at face value. The unwritten premise is that the nation which is able to transmit the most bits per square meter from one locale to another will be the economic victor in the 21st century. Perhaps. But so long as telecommunications networks remain open to all -- citizens and aliens, buyers and sellers -- any local expenditure will also yield substantial benefits to the global community. Or as economists put it, infostrncture expenditures have large externalities. Beyond that, the network dialectic described above strongly suggests that the infostructure of the 1990s will lead us in quite different directions, and simultaneously so. This may be unavoidable. But we need to probe more deeply any claim that tomorrow’s electronic highways are a one way ticket to prosperity and the global common. My second suggestion is to make sure that public policy and technical decisions about the network, from standards to telephone numbering plans to electronic mail gateways to new service offerings, enhance peoples own sense of place. In a networked world, it is all too easy to feel displaced -- literally and figuratively -- if one is not 55 TeleGeography 1993 The Explosion of Place connected. That is why access to the network and the terms on which access will be available to future network services (i.e., interactive data and video offerings) are not just telecommunications policy issues but critical social and economic policy matters for society as a whole. Similarly, the more time people spend in one electronic realm or another, the more they wish to make it their own. It may be virtual property, but it gives people as much joy and pain as real property and thus many people expect to have the same ability to personalize and protect the virtual places they now occupy. Telephone calling cards are one example. For years, except for the color, calling cards (like credit cards) were anonymous, almost identical plastic rectangles. Lots of people, however, wanted these network tokens to be particular and personal, marked with their own countries and signs. The Japanese responded first and other countries quickly followed. Tens of thousands of customized calling cards now circulate in the world and many more are on the way. Another example is automatic number identification (ANI), sometimes known as "caller ID." Although the telephone is a very intimate medium it can be rung randomly by friend and foe alike. Many people resent this invasion of privacy, but lack a secretary or voice mail system to protect them, especially at home. AIN and caller ID provides the "electronic doorman" they seek by disclosing the number of the calling party at the same time the telephone rings. Of course, caller ID also compromises the anonmity of the caller; but so too does the "peep hole" on people’s doors or the return address on an envelope. Conflicts like these between the privacy of the network traveler and the privacy of the host are likely to be increasingly common as people seek to make cyberspace a better place to live. In a world where communication is possible "any place, any time" the response of many people is, please, "not here, not now." Consumer interest in caller ID services also points to another category of network filters. With tens of thousands of databases and millions of terminals on line, information overload is a common complaint. Hence the demand for increasingly sophisticated audio and data screening services. The great popularity of Internet search and retrieval services, such as "gopher", and the personalized electronic newspaper services now coming to market, also respond to these needs. Consumer support for the deregulation of telephone equipment provides a final example. Most people want to own their own phones. And they also want to own their own telephone numbers as they change houses and jobs. In Beijing, for example, when local authorities auctioned off telephone numbers, many thousands of dollars were paid for the rights to use the most auspicious numbers. I am also reminded of a poem by the Ghanian writer, Atukwei Okai, which harks back to a much earlier-age: "Of course we are glad to be born to Universe. But we’d love to have our home address somewhere Specific directions about our house our home, Our little place in a monstrous world. Yes we’d like to hand our own address Up at the crossroads of this earth Lest the gods should one day come looking For us in the wrong place." Tribal space or cyberspace, the human desire for a sense of place runs deep. The more places there are in the world, the stronger this desire seems to be. Those governments and carriers which respond accordingly are unlikely to go far astray. 56 TELEMAPS 57 PetroPower Official Oil Reserves 1990 EUROPE 1.6% NON-ARAB AFRICA 2.0% U.S.S.R. AND CHINA 5.0% SAUDI ARABIA 26.6% AUSTRALIA AND FAR EAST ~R, UAE ASIA .7% 1.4% Source: International Petroleum Encyclopedia © Gregory C. Staple/llC 1992 TelePower Distribution of telephone main lines 1992 SOUTH AMERICA 3.8% EASTERN EUROPE AND U.S.S.R. 9.7% J1APAN 0.5% MIDDLE EAST (~ 1.6% AUSTRALIA AND NEW ZEALAND (OCEANIA) 1.8% Source: Siemens. © TeleGeography, Inc. 1993 58 PeoplePower World Population 1990 CANADA .5% EUROPE (including ex-USSR) 17.5% CENTRAl AMERICA 2.8% UNITED STATES 4,7% AMERICA 5.5% JAPAN 2,3% ASIA 54% Source: World Bank Development Report MIDDLE EAST 4.1% INDONESIA 3.4% AUSTRALIA AND NEW ZEALAND .5% © Gregory Staple/lIC 1992 GeoPower Percentage of Land Area EUROPE 6.6% Modified Peter’s Projection SOUTH AMERICA \~ 11,9% AFRICA 20,3% ASIA 29,9% AUSTRALIA5.1% 59 I Growth of |nternat|onal Calling (1990-1992) The Americas and Europe THE AMERICAS Country % Growth 1 Argentina 39.0’ 2 Brazd 4 2 3. Canada 6 6 4 Chde 9 0 5 Colombia 16,6 6. Mexico 27 4 7 Peru 14 2 8 Umted States 12 0 9. Uruguay 8,2 10 Venezuela 16 4 Growth Rate: 1990-1992 20% Growth 15 20% Growth 10- 15%Growth 10% Growth 1 Oro~vth rates based on outbound Mlnuteg of TelecommunicatlOl]g Tralll( (M~TT) on pubh( telephone network only 2. Astensk m(hcates ’91-’92 growth EUROPE (Inset) _C_o__ul_ !!r_y- % Growth 11 Austria 13.0 12. Belgium 11.6 13. Bulgaria 26.6~ 14. Czechoslovakia 39 0* 15. Denmark 18 3 16. Finland 12 1 17. France 7.3 18. Germany 13.3 19. Greece 18 4 20 Iceland 7 8 21. Ireland 6.5 22. Italy 18 8 23 Luxembourg 9.6 24. Netherlands 11 9 25 Norway 11 4 26. Portugal 16.4 27 Spare 14 7 28. Sweclen 5.9 29. Switzerland 6.9 30 UK. 61 31. Hungary 22.8 32 Poland >20.0 60 Growth of International Calling (1990-1992) ~Af~r!ca & Asia-Paciflc Countries I MIDDLE EAST/AFRICA Countly % (_]ro\vth I Isl~ol 1 3 9 2 Saclch Alal)ia 13 2" 3 Turkey 19 6 4 South Attica 9 4~ Growth Rate: 1990-1992 20% Growth 15 20% Growth l---I 10-15% Growth 10% Growth I Giowlh tat(,, ba,ed on OLItl)OLllld (MIT]) Oil pubh( tolephono notwolk only 2 i~s[Ollsk Indl(’alo$ ’91-~92 glo\,vtl~ ASIA/PACIFIC ~_..~1 % O~owlh Country % Giowth 7.1 I 2. Soutt~ Korea 27 5 388~ 13 Malaysia 22.1 44 3> 14 NewZealan(I 54 24 9 15. Philippines 16.1 33.0 I () Sillgal)oro 1 9.8" 18.1 17 -ralwan 3 1.9 2 ~,0 18. Thadand 19.8~ 61 I Balance of Telecommunications Traffic (1992 The Americas and Europe THE AMERICAS Worlcl SLuplus Or (Dehc~t) CouRtry in miNIons 1 Argentina 35.4 2 Brazil 160.7 3 Canada 420 I 4 Chde 30 6 5 Colombia 14.} 5 6 Mexico 45! S 7 Peru 94 0 8 Umted States (4875 O) 9 U~uguay 22 8 10. Venezuela 13 1 EUROPE (Inset) World SurlJh.ls ol (Belle it) CoLitltry i1] million~ 11 Aushla f21 2) 12 Cze(h & blovak 38 (t Repubh~ , I Der~malk C) 7 1-i Fran(e 9t 0 15. Greece 60 8 16 Hungary f33.}) 17 Iceland (0 4) 18 Iroland 86.4 19 Italy 67 6 20. Luxembouig (77.5) 21 Nethellands 4~4 9) 22. Norway (47.3) 23 Poland 15 9 24. Poltugal N/A 25 Russia (g5 I) 26 Spare 42 7 27 Sweden (96 0; 28 Switzerland ~59 S) 29. United IKingdon~ (60 0; 62 I Balance of Telecommunications Traffic (1992)1 Africa & Asia-Pacific Countries o MIDDLE EAST/AFRICA World Sclrplus or (Detroit) m m~l]ions 112.9 (172 5) 29.0 333 2 Surplus over outgoing tram( ~, and 11-1(()1]111~12, tratm i~ D 10% Balanced oi total two+way tlalilt (MI] I) Balance> ale based on pubh( lelOl)htme notwolk h allit only ASIA/PACIFIC World ~..~° SuII)Ius or (Deficit) Country in millions 5 Bangladesh 52.5 6. Hong Kong (127 3) (balanced) 7 India 95 0 8 Indonesia 47.4 9 Japan (323 7) 10. Malaygla 32.5 1 I. Pakistan 151 2 12. Phlhpl~ines 326.3 13. Tatwan 11S 7 14 Thailand 69.2 63 The United States Telecontinent AUSTRALIA © IIC 1992 This map reflects the volume of switched telecolnmunications traffic to and from the United States in 1990. The area of the Umted States overlapped by another country (eg. the U.K.) is roughly proportional to the percentage of United States traffic to that country (for the U.K.. 8%). The area of the correspondent country overlapping the United States reflects the percentage of that country’s inte,nat~onal traiTic sent to the United States (for the U.K., 21%, because 21% of U.K. traffic was sent to the Umted States). Note: Country maps are not to scale and overlapping is only approximate. 64 The German Telecontinent CANADA UNITED STATES LUXEMBOURG SWITZERLAND © IIC 1992 This map reflects the volume of switched telecomnmnlcations traffic to and from Germany in 1991. The area of Germany overlapped by another country (eg, Austria) is roughly proportional to the percentage of German traffic to that country (for Austria, 9%). The area of the correspondent country overlapping Germany reflects the percentage of that country’s international traffic sent to Germany (for Austria, 43%, because 43% of Austrian traffic was sent to Germany). Note: Country maps are not to scale and overlapping is only approximate. 65 The Japanese Telecontinent CHINA HONG KONG THAILAN INDONESh © IIC 1992 This map reflects the volume of switched telecommunications traffic to and from Japan in 1990. The area of Japan overlapped by another country (eg, the U.S.) is roughly proportional to the percentage of Japanese traffic to that country (for the U.S., 23%). The area of the correspondent country overlapping Japan reflects the percentage of that country’s international traffic sent to Japai1 (for the U.S., 4%, because 4% of U.S. traffic was sent to Japan). Note: Country maps are not to scale and overlapping is only approximate. 66 Traffic Database For TeleContinent Maps United States TeleContinent Calling Country % Outgoing MiTT To U,S, From, Calling Country % Outgoing MiTT From U,S. To Calling ,Country Canada 71,2 22~8 Mexico 88.8 10,0 United Kingdom 20.9 8,0 Germany 6.4 6,3 Japan 22,5 4.1 France 6.3 2.6 Italy 9.4 2.2 Dora, Republic 70,0 2,1 South Korea 27 ~6 2,1 Philippines 51.0 2,1 Colombia 59.0 1.7 Taiwan 26,5 1,5 Brazil 34,1 1,4 Israel 34,1 1.4 Jamaica 55,0 1.2 Australia 15.5 1.2 Hong Kong 9.6 1,1 E1 Salvador 60.0 1,1 Switzerland 4,3 1,0 Netherlands 5,9 0.9 t German TeleContinent Calling Country % Outgoing MiTT to Germany From Calling Country Austria 43 Switzerland 23 France 1 t United Kingdom 11 Turkey 40 Italy 16 United States 6 Netherlands 23 Yugoslavia 26 Spain I4 Belgium 13 Greece 22 Denmark 18 Sweden 10 Luxembourg 20 Portugal I6 Canada 2 Outgoing MiTT From Germany To CallinR Cotmtrv 9 8 8 8 8 7 7 7 4 3 3 2 2 2 1 1 1 Japanese TeleContinent % Outgoing MiTT to lapan Calling Country From Calling CountrJ~ % Outgoing MiTT From Japan To Calling Country United States 4, t 22.5 South Korea 33,6 12.1 Taiwan 20,4 7.2 Philippines 12,4 6,7 China 3,0 4,2 United Kingdom 1,6 4.2 Hong Kong 5,5 3.7 Thailand 20.2 3.1 Singapore 12,0 2.6 Germany 1.0 2.3 Australia 4,8 2,2 Brazil 3.0 2.2 France 0.5 1.8 Malaysia 16.3 1,7 Canada 0.4 1.3 Indonesia 11,5 1. t MiTT is Minutes of Telecommunications Traffic. Data areforpublic voice circuits only in 1990. 67 STATISTICAL TABLES ; ,,’ Table 1 CAPACITY AND COST, PER VOICE PATH, OF SELECTED TRANS-OCEANIC CABLES (1956-2000) Year Cable Cost ($ US) Capacity In-Service System per voice path in voice paths 1956 TAT-1 557,000 1965 TAT-4 365,000 1970 TAT-5 49,000 1983 TAT-7 23,000 1988 TAT-8 9,000 1989 PTAT 6,000 1993 TAT-10 4,000 1994 CANTAT-3 1,000 1996-97 TAT 12/13 1,000 89 138 1 440 8 400 37 800 85 000 75 600 302 000 600 000 1957 Hawaii 1 378,000 91 1964 TPC-1 406,000 167 1974 Hawaii 3 41,000 1690 1975 TPC-2 73,000 1690 1988 TPC-3 16,000 37,800 1991 North Pacific Cable 5,000 85,000 1992 TPC-4 5,500 75,600 1996 TPC-5 2,000 600,000 1997 FLAG 1,500 (Japan-Saudi _ Arabia-U.K) 600,000 Notes: Costs are capital and construction costs only stated in $US to the nearest $500 unadjusted for inflation. Current technology permits approximately 5 virtual voice paths to be derived from a digital channel operating at 64,000 bits per second (64 Kbit/s). Fiberoptic submarine cables are expected to have a useful life of at least 25 years. Table reports average cost per voice path for cables with multiple landing points. For example, the TAT-9 system interconnects the U.S. and Canada with the U.K., France and Spain. The average U.S.-U.K. cost per voice path is approximately $4000. Reserve capacity of cables generally is excluded. Source: FCC and Carriers © TeleGeography, Inc. 1993 69 Table 2 CAPACITY, IN VOICE PATHS, OF TRANS-OCEANIC CABLE SYSTEMS (1986-2000) Trans-Atlantic (North America to Europe) Year Cable Voice Paths 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997-2000 22000 22000 60000 145.000 145.000 221.000 296.600 372 200 674.000 1,274,000 1,274,000 1,274,000(minimumavailable) Trans-Pacific (North America to Japan via Hawaii or Guam) Cable Year Voice Paths 1986 2,000 1987 37,800 1988 37,800 1989 37,800 1990 37,800 1991 114,200 1992 190,520 1993 266,120 1994 266,120 1995 266,120 1996 866,120 1997-2000 1,466,120 (minimum available) Notes: Estimate of cable voice paths assume that 5 virtual voice paths can be derived from one 64kbit digital circuit; cable estimates do not include circuits held in reserve for cable/satellite restoration services. Estimates of trans-Atlantic capacity exclude proposed PTAT-2 cable in 1997-2000 timeframe. Estimates of trans-Pacific cable circuits exclude proposed, SE-ME-WE-3, CANPAC-1, and Trans-Siberian Link ~I"SL) cables, all scheduled for 1996-1997 timeframe. Regional capacity estimates do not necessarily imply that full capacity is available to satisfy demand on any given bilateral route. Source: FCC and Carriers © TeleGeography, Inc. 1993 70 Table 3 CAPACITY, IN VOICE PATHS, OF TRANS-OCEANIC SATELLITE SYSTEMS (1986-2000) Trans-Atlantic (North America to Europe) Year Satellite Voice Paths 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997-2000 78,000 78,000 78,000 93,000 283,000 283,000 496,600 620,800 620,800 710,800 710,800 737,500 (minimum available) Trans-Pacific (North America to East Asia) Satellite Year Voice Paths 1986 39,000 1987 39,000 1988 39,000 1989 39,000 1990 39,000 1991 27,000 1992 27,000 1993 83,300 1994 234,000 1995 234,000 1996 234,000 1997-2000 424,500 (minimum available) Notes: Estimates of satellite voice paths are based on lntelsat satellites only prior to 1993; satellite estimates exclude one Intelsat satellite in each region held in reserve. Estimates also assume one voice path per channel until 1989 deployment of lntelsat Vl series with 24,000 channels or 120, 000 voice paths using Digital Code Multiplication Equipment (D CME). The Intelsat Vll series, deployed in 1992, have a nominal capacity of 18,000 channels or 90,000 voice paths using DCME. For 1993-2000 time period, estimates assume full capacity of the following non-lntelsat systems is available: Trans-Atlantic PAS-1; PAS-2, Orion-1 and TDRS-41; Trans-Pacific PAS~I; Rimsat/Express (2 satellites) and TDRS-174. In the near term, some additional satellite capacity also is likely to be available from lntersputnik, Hispasat and, after 1997, from the Iridium and lnmarsat proposed mobile telephone satellite systems. Currently, non-Intelsat satellites are limited to 1250 64kbps circuits per satellite for public switched telephony. These limits are expected to be lifted or phased-out in 1997. Regional capacity estimates do not necessarily imply that full capacity is available to satisfy demand on any given bilateral route. Source: FCC and Carriers © TeleGeography, Inc. 1993 71 Table 4a THE TRAFFIC BASE OF THE TOP 20 INTERNATIONAL CARRIERS (1991-1992) 1992 1991 Percent Outgoing Outgoing Growth MiTT in MiTT in in MiTT Carrier Millions* Millions* 1991-92 Country 1 AT&T* 6984 6557 6.5 United States 2 DBP Telekom* 4087 3605 13.4 Germany 3 France Telecom 2449 2295 6.7 France 4 BT 2188 2105 3.9 United Kingdom 5 MCI* 2083 1600 30.2 United States 6 Swiss PTT 1551 1429 8.5 Switzerland 7 Stentor* 1520 1425 6.7 Canada 8 Hong Kong Telecom 1137 913 24.5 Hong Kong 9 Netherlands PTT* 1134 1018 11.4 Netherlands 10 Iritel* 1116 980 13.9 Italy 11 Sprint* 940 723 30.0 United States 12 Belgacom 911 823 10.7 Belgium 13 KDD 893 850 5.6 Japan 14 Telefonica 804 719 11.8 Spain 15 Teleglobe* 722 647 11.6 Canada 16 Austrian PTT 713 642 11.1 Austria 17 Telia AB* 691 672 3.5 Sweden 18 Telmex 684 500 36.8 Mexico 19 Mercury 661 493 34.1 United Kingdom 20 Telstra 659 610 8.0 Australia MiTT is Minutes of Telecommunications Traffic. Data are for public voice circuits only rounded to the nearest million MITT. For U.S. carriers (1991 and 1992) and the Netherlands PTT (1992 only) MiTT are based on billing point of call not originating point. Data for North American carriers include continental traffic. Stentor was formerly Telecom Canada; Stentor traffic is for U.S. and Mexico only of which approximately 70% was originated by Bell Canada. For Italy, Iritel (formerly ASST) handles intra-continental traffic only; Italcable carries overseas traffic (See Table 4b). For DBP Telekom, all data includes outgoing traffic from the former East Germany. Telia AB was formerly Televerket. Note: /3T, KDD, Hong Kong Telecom and Mercury data are for the Fiscal Year (April 1992 to March 1993). © TeleGeography, Inc. 1993 72- Table 4b THE TRAFFIC BASE OF THE SECOND 20 INTERNATIONAL CARR1F~RS (1991-1992) Rank 1992 1991 Percem Outgoing Outgoing Growth MiTT in MiTT in in MiTT Carrier Millions* Millions* 1991-92 Country 1 China PTT 635 440 44.3 China 2 Saudi Com. Ministry 465 411 13.1 Saudi Arabia 3 Danish PTT 425 395 7.6 Denmark 4 Singapore Telecom* 412 344 19.8 Singapore 5 DGT Taiwan 369 255 44.7 Taiwan 6 Italcable 358 239 49.8 Italy 7 Norwegian Telecom 349 309 12.9 Norway 8 Hellenic Telecom. Org 299 245 22.0 Greece 9 U.A.E. Comm. Ministry 299 262 14.1 U.A.E. 10 Telecom Eireann* 297 273 8.8 Ireland 11 Videsh Sanchar* 260 186 39.8 India 12 Korea Telecom 245 229 7.0 Rep. of Korea 13 Finnish PTT 235 215 9.3 Finland 14 Turkish PTT 227 198 14.6 Turkey 15 Telkom South Africa 222 203 9.4 South Africa 16 Telekom Malaysia 217 176 23.3 Malaysia 17 Telekom Poland 213 N/A N/A Poland 18 IDC Japan 197 154 27.9 Japan 19 ITJ Japan 193 156 23.7 Japan 20 Hungarian PTT 184 165 11.5 Hungary MiTT is Minutes of Telecommunications Traffic. Data are for public voice circuits only rounded to the nearest million MITT. Telecom Eireann data includes traffic to the U.K.; Videsh Sanchar data excludes traffic to Pakistan; Singapore Telecom data includes traffic to Malaysia (except local border traffic). Telecom Portugal only handles intm-confinental traffic; overseas traffic is carried by CPRM which had 41 million MiTT outbound in 1992. Additional Notes: 1. The next largest carriers ranked by 1992 outgoing MiTT in millions ( ) include: the Luxembourg PTT (181); Czech Telecom International (178); Intertelecom Russia (176), Telecom Portugal (171); Embratel Brazil (170); Bezek Israel (153); New Zealand Telecom (149), Communication Authority of Thailand (132), PLDT Philippines (124); and Telintar Argentina (124). 2. Data for Singapore Telecom, Telecom New Zealand, Videsh Sanchar, ITJ and IDC are for Fiscal Year (April 1992 March 1993). © TeleGeography, Inc. 1993 73 Table 4c THE TRAFFIC BASE OF THE TOP 20 INTERNATIONAL CARRIERS (1988-1992) 1992 1988 Cumulative Outgoing Outgoing Annual Growth MiTT in MiTT in in Outgoing Carrier Millions* Millions* Traffic 1988-92 Country 1 AT&T* 6984 4778 7.9 United States 2 DBP Telekom 4087 2479 10.6 Germany 3 France Telecom 2449 1570 9.3 France 4 BT 2188 1654 5.8 United Kingdom 5 MCI* 2083 262 51.4 United States 6 Swiss PTT 1551 1014 8.9 Switzerland 7 Stentor* 1520 1054 7.6 Canada 8 Hong Kong Telecom 1137 441 20.9 Hong Kong 9 Netherlands PTT* 1134 706 9.9 Netherlands 10 Iritel, 1116 734 11.0 Italy 11 Sprint* 940 131 48.3 United States 12 Belgacom 911 561 10.2 Belgium 13 KDD 900 529 11.2 Japan 14 Telefonica 804 330 19.5 Spain 15 Teleglobe* 722 358 15.1 Canada 16 Austrian PTT 713 401 12.2 Austria 17 Telia AB* 691 485 7.1 Sweden 18 Telmex* 684 211 26.5 Mexico 19 Mercury 661 75 54.5 United Kingdom 20 Telstra 659 415 9.7 Australia MiTT is Minutes of Telecommunications Traffic. Data are for public voice circuits only rounded to the nearest million MITT. For U.S. carriers and Netherlands PTT, 1992 data is based on billing point of call not point of origin. ¯ 1989 data/CAGR 1989-92. Data for North American carriers include continental traffic. Stentor was formerly Telecom Canada; Stentor traffic is for U.S. and Mexico only of which approximately 70% was originated by Bell Canada. For Italy, Iritel (formerly ASST) handles intra-continental traffic only; Italcable carries overseas traffic (See Table 4b). For DBP Telekom, 1992 data includes outgoing traffic from the former East Germany. Telia AB was formerly Televerket. Note: BT, KDD, Hong Kong Telecom and Mercury data are for the Fiscal Year (April 1992 to March 1993). © TeleGeography, Inc. 1993 74 Table 5 Market Share of Competing International Carriers Percent Outgoing MiTT (1986-1993) Country/Carrier 1986 1987 1988 1989 1990 1991 1992 UoS, AT&T 94.3 93.0 89.1 83.3 78.4 74.8 70.3 MCI Sprint 4.0 4.7 7.0 10.2 14.6 17.8 21.2 1.6 2.3 3.5 5,8 6.4 6.3 7.3 U.K. (F.Y.) BT Mercury Japan (F.Y.) KDD IDC ITJ New Zealand (F.Y.) TNZ ClearCom Kor~ Korea Telecom Dacom 99.8 98.5 95.5 91.0 86.0 80.7 76.0 Chile (Est, for 1993 to 30 June) Entel Chilesat VTRTelecom Philippines PLDT Eastern Telecom 0.2 1.5 4.5 9.0 14.0 19.5 24.0 Sweden (Est. for 1993 to 30 June) Televerket Tele-2 -,’,,, ......... ’ .................... ::.i :i::..i...~..: 93.3 88.0 73,7 ¯ ~-- 3.7 6.5 13.5 ¯ b.- .... ..... ’."~ 3.0 5.5 12.8 I .. -t 92.0 82.0 ~-%--.- .- .-,.4---.-.-..---1 .... .~ ....... :-",:":,~ ....... ’ ...... "~ ...... ’ ................ ~" 8.0 18.0 ¯ I’" ." i’: "- .:" t .’."| ........ ::: b -" I’. ... J .. t .... ", .t.::" :" .~ ! ! r- .... ’:, :~I,: ’ ..::: ;. ::" I I I :I :.- ’I I I t ..... I ...’.’.’:’". .... I I ! t ’ : :’:’:".:’ I’.’..’":’ ’r ! ’--.--.’~-...’F’--..--.--...--.:.--..~,’--~--’--.--~ .- .... I I : ......... .’ .’: l": ....L:’ l .: ’ :,",’t’:" i ...... .’ ’t .... ’r r’--- .-?~-r’ r’.---.~,~.’r" .... "7. "I ....... I’ 1 ’t’ I I I ! I .:.:--t ::~: I I t t -:- -- 1 .1- ..... ..... ..... @~---+-,-,:- .... 69.7 15.3 15.0 80.0 20.0 79,9 20.1 73.0 25.0 2.0 91.6 8.4 95.0 5.0 Notes: 1. MiTT is Minutes of Telecommunications Traffic. Data based on international outgoing traffic for the public switched network only. Market shares are for the full year, beginning in the first year of competition. Market shares for U.S. carriers exclude traffic to Canada and Mexico; for U.K. carriers traffic to Ireland is excluded. Competing international telephone carriers (bold type) have also been authorized in the following markets, but were not operational or had very limited international traffic at 1 July 1993: Australia (Telstra; Optus, AAP Telecommunications); Malaysia (Telekom Malaysia; Technology Resource Industries tTRI); Indonesia (PT Indosat; PT Satelindo); Israel (Bezeq; DARCOM Communications Ltd.); Finland (Telecom Finland, HelsinkiTelephone and others). In addition, since September 1992, TeleGlobe, Canada’s overseas carrier, has competed with Stentor, the Canadian consortium of provincial telephone companies, for service between Canada and Mexico. © TeleGeography, Inc. 1993 75 Table 6 INTERNATIONAL ROUTES WITH THE LARGEST VOLUME OF TELECOMS TRAFFIC (1992) Continental (out/in) Imercontinental (out/in) U.S Canada U.S.- Mexico (2226/1511) (1277/609) U.S. U.K.+ 034/502) Austria- Germany Switzerland Germany France Germany France U.K.÷ Germany U.K.÷ Germany-Italy Netherlands Germany Hong Kong÷ China (PRC) (303/329) (358/315) (3o3/818) (275/269) (287/325) (288/227) (264/250) (535/412) U.S. Germany+ (563/236) U.S.- Japan+ t36"//282) France Italy France Belgium France Spain France Switzerland Netherlands U.K.÷ Netherlands Belgium Switzerland Italy U.K.÷ Ireland÷ U.K.÷ Italy (217/209) (216/221) (157/139) (196/264) (147/134) (189/185) (224/149) (218/206) (135/118) U.S. France+ t240/lb~ U.S. S. Korea+ (208/96) U.S. Italy+ t207/1181 Germany Turkey+ (295/83) MiIT is Minutes of Teleconmmnications Traffic. Data are for two-way traffic for switched voice circuits only. On U.S. routes, to reduce double counting, data from U.S. carriers has been used for outbound and inbound traffic. See Methodology pages. Data for outgoing routes from these countries are for F.Y. 1992/93 © TeleGeography, Inc. 1993 76 AMERICAS TRAFFIC TABLES 77 Table 7 Argentina Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A ......................... 27.6 2. Uruguay .................... 21.1 3. Brazil ......................... 12.5 4. Chile .......................... 10.2 5. Spain ........................... 8.9 6. Italy ............................... 8.1 7. Paraguay .................... 5.8 8. France ......................... 2.9 9. Germany ..................... 2.7 10. Canada ...................... 2.3 11, U, K ............................ 2.2 12. Peru ............................ 2.1 13. Bolivia ........................ 2.1 14. Mexico ....................... 1.9 15. Switzerland ............... 1.5 Percent of Outgoing Traffic ~/~ 22.2O/o 17.0%o 10.0% 8.2% 7.1% 6.5% 4.7% 2.4% 2.2% []1.9% []1.8% []1.7% ~1.7% ~1.5% .2% Other .......................... 7.2 5.8% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 1991 N.A. 132.5 N.A. 89.4 N.A. 43.1 N.A. 221.9 1992 159.7 124.3 35.4 284.1 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 78 Table 8 Brazil Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A ......................... 61.3 2. Argentina .................. 17.7 3. Germany ...................... 9.1 4. Italy ............................... 8.6 5. Portugal ....................... 7.5 6. U.K ................................ 6.4 7. France .......................... 6.2 8. Uruguay ....................... 4.7 9. Japan ........................... 4.4 10. Paraguay ................... 4.0 11. Switzerland ............... 3.7 12. Chile ........................... 3.6 13. Spain .......................... 3.6 14. Canada ...................... 2.7 15. Mexico ........................ 2.3 Percent of Outgoing Traffic 10.4% 5.4% 5.1% 4.4% 3.8% 3.6% N 2.7% N 2.6% [] 2.4% 2.2% 2.1% 2.1% 1.6% 1.4% 36.1% Other .......................... 24.2 14.2% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 156.5 N.A. N.A. 1991 261.6 171.2 90.4 432.8 1992 330.6 169.9 160.7 500.5 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 79 Table 9 Canada Largest Telecommunication Routes (1992) Destination MITT 1. U.S.A ...................... 1511.4 2. U.K ............................ 117.2 3. Hong Kong ................ 43.0 4. France ........................ 41.3 5. Germany .................... 38.9 6. Italy ............................. 30.1 7. Australia .................... 19.8 8. Jamaica ..................... 18.7 9. Japan ......................... 17.9 10. India .......................... 17.5 11. Philippines ............... 18.6 12. Netherlands ............. 14.3 13. Mexico ...................... 13.0 14. Portugal .................... 12.9 15. Switzerland ............. 11.9 Percent of Outgoing Traffic []5.2% ~1.9% 1.8% ~1.7% 1.3% 0.9% ~0.8% 0.8% ~0.8% 0.7% 0.6% 0.6% 0.6% 0.5% Other ........................ 322.2 14.3% National Traffic Balance *MITT 1 990 Incoming 358.5 Outgoing 565.2 S~rplus(Deficit) (206.7) Total Volume 923.7 1991 398.2 647.3 (249.1) 1045.5 1992 441.2 722.3 (281.1) 1163.5 Note: Incoming and outgoing totals exclude all Canada-U.S. and Canada-Mexico traffic. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 80 Table 10 Chile Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A .......................... 15.3 2. Argentina ..................... 8.9 3. Brazil ............................ 3.1 4. Spain ........................... 2,6 Percent of Outgoing Traffic 16,2% 5.6% 4.7% 5. Peru ............................. 2.2 4.0% 6. Germany ..................... 1.9 3.5% 7. Canada ....................... 1.4 [] 2.5% 8. U.K ............................... 1.3 [] 2.4% 9, France ......................... 1.3 2.4% 10. Bolivia ........................ 1.2 [] 2.2% 11. Italy ............................. 1.2 [] 2.2% 12. Mexico ........................ 1.0 [] 1.8% 13. Venezuela ................. 0.9 1.6% 14. Colombia ................... 0.8 1.5% 15. Uruguay ..................... 0.7 1.3% 27.8% Other ......................... 11.2 20.4% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 61.8 49.0 12.8 110.8 1992 85.6 55.0 30.6 140.6 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 81 Table 11 Colombia Largest Telecommunication Routes (1992) Destination Ml’l’r Percent of Outgoing Traffic 1. U.S.A ......................... 50.0 52.9% 2. Venezuela .................. 9.2 9"7% 3. Spain ........................... 3.4 I~ 3.6% 4. Panama ...................... 3.3 [] 5. Ecuador ...................... 3.0 ~J 3.2% 6. Mexico ........................ 2.6 I~ 2.6% 7. Italy .............................. 2.3 I~ 2.4% 8. U.K .............................. 2.0 ~J 2.1% 9. Brazil ........................... 2.0 ~J 2.1% 10. France ....................... 1.7 1.6% 11.Germany ................... 1.7 1.8% 12. Argentina .................. 1.2 1.3% 13. Canada ..................... 1.0 I~ 1.1% 14. Chile .......................... 0.9 1.o% 15. Bolivia ........................ 0.2 I o.2% Other ........................ 10.0 10.6% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 69.5 N.A. N.A. 1991 N.A. N.A. N.A. N.A. 1992 238.0 94.5 143.5 332.5 "Mi’l-[ is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 82 Table 12 Mexico Largest Telecommunication Routes (1992) Destination MiTr 1. U.S.A ......................... 610.0 2. Canada ......................... 8.9 3. Spain ............................. 7.6 4. Germany ....................... 4.7 5. Colombia ...................... 4.4 6. France ........................... 4.2 7. U.K ................................. 3.6 8. Italy ................................ 3.5 9. Guatemala ................... 2.9 10. Argentina .................... 2.6 11. Brazil ........................... 2.5 12. Venezuela .................. 2.1 13. Japan .......................... 2.1 14. Costa Rica .................. 1.9 15. Peru ............................. 1.8 Percent of Outgoing Traffic ~1.3% 1.1% 0.7% 0.6% 0.6% 0.5% 0.5% 0.4% 0.4% 0.4% 0.3% 0.3% 0.3% 0.3% 89.2% Other .......................... 20.7 3.0% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 421.1 N.A. N.A. 1991 N.A. 500.2 N.A. N.A. 1992 1115.0 683.5 431.5 1798.5 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 83 Table 13 Peru Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A ......................... 15.3 2. Chile ............................ 2.1 3. Spain ...... ................... 2.0 4. Argentina .................... 1.3 5. Venezuela .................. 1.2 6. Brazil ........................... 1.2 7. Italy .............................. 1.1 8. Colombia .................... 0.9 9. Canada ....................... 0.8 10. Ecuador ..................... 0.7 11. Mexico ....................... 0.7 12. Germany ................... 0.7 13. France ....................... 0.6 14. Bolivia ....................... 0.6 15. U.K ............................. 0.6 Percent of Outgoing Traffic 47.5% W 6.7% ~6.3% []3.9% []3.9% 3.7% ]3.3% 2.8% 2.5% 2.3% 2.2% 2.2% 1.9% 1.8% 1.8% Other .......................... 2.3 7.2% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 24.6 N.A. N.A. 1991 N.A. N.A. N.A. N.A. 1992 126.1 32.1 94.0 158.2 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 84 Table 14a United States--Outgoing Traffic Largest Telecommunication Routes (1992) Destination MiTT 1. Canada ....................... 2226.4 2. Mexico ........................ 1276.8 3. U.K ............................... 733.7 4. Germany ....................... 563.0 5. Japan ........................... 366.8 6. France ......................... 239.8 7. Dominican Republic ...... 231.3 8. South Korea ................. 207.8 9. Italy ............................... 207.3 10. Philippines ................... 200.3 11. Colombia ...................... 172.0 12. Taiwan ......................... 162.9 13. Brazil ........................... 144.7 14. Israel ............................ 137.7 15. Jamaica ........................ 124.4 16. Hong Kong .................. 114.3 17. Australia ....................... 113.6 18. Spain ........................... 104.9 19. El Salvador .................... 99.9 20. India .............................. 97.9 21. Switzerland .................... 93.5 22. Netherlands ................... 92.7 23. Guatemala ...................... 88.0 24. Peru .............................. 80.1 25. Venezuela ..................... 75.9 Other ........................ 2217.9 Percent of Outgoing Traffic 12.6% 7.2% 5.5% 3.6% 2.4% 2.0% 2.0% 2.0% 2.0% [] 1.7% ~] 1.6% ~’~ 1.4% [] 1.4% [] 1.2% 1.1% 1.1% 1.0% 1.0% 1.0% 0.9% 0.9% 0.9% 0.8% 0.7% 21.8% National Traffic Balance *MITT 1990 1991 Incoming 2542.8 2829.8 Out£1oin£1 5275.9 5984.5 Surplus(Deficit) (2733.1) (3154.7) Total Volume 7818.7 8814.3 1992 3149.4 6670.4 (3521.0) 9819.8 Note: Incoming and outgoing traffic totals exclude Canada and Mexico traffic. 1991 and 1992 data are based upon the bill~ng point of the traffic. See Methodology pages. "Mi’FI" is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 85 Table 14b United States--Incoming Traffic Largest Telecommunication Routes (1992) Country of Origin MITr 1. Canada ....................... 1511.4 2. Mexico .......................... 608.5 3. U.K ............................... 501.5 4. Japan ........................... 282.1 5. Germany ....................... 235.7 6. 7. 8. 9. Percent of Incoming Traffic 28.7% 11.5% 9.5% 5.4% 4.5% France .......................... 156.6 3.0% Italy ............................... 118.0 [] 2.2% Australia ........................ 104.3 [] 2.o% South Korea .................... 96.1 [] 1.8% 10. Hong Kong ..................... 82.6 B 1.6% 11. Taiwan ............................ 82.4 [] 1.6% 12. Netherlands .................... 66.7 1.3% 13. Switzerland ..................... 66.1 1.3% 14. India ............................... 60.1 I 1.1% 15. Sweden ......................... 53.4 B 1.o% 16. Venezuela ..................... 53.0 1.0% 17. Spain ............................. 52.7 B 1.o% 18. Colombia ....................... 50.4 I 1.0% 19. Brazil ............................. 49.3 II o.9% 20. Dominican Republic ....... 48.4 o.9% 21. Israel ............................. 48.1 B o.9% 22. Jamaica ......................... 42.4 0.8% 23. Singapore ..................... 34.2 o.6% 24. Bahamas ....................... 33.7 0.6% 25. Belgium ........................ 32.9 o.6% Other ......................... 798.7 15.2% Note: Data are based on billing country rather than physical point of origin. "Mi"FF is Minutes of Telecommunications Traffic Data are in millions of minutes for pubhc voice circuits ©TeleGeography, Inc. 1993 86 Table 14c United States--Growth in Traffic for Top 25 Routes (’90-’92) Outgoing COUNTRY CAGR ’90-’92 1. Canada .................. 23.5% 2, Mexico .................... 27.9% 3. U.K ......................... 8.1% 4. Germany ................. 6.7% 5. Japan ...................... 7.1% 6. France ................... 9.1% 7. Dominican Republic.18.4% 8. Italy ......................... 10.7% 9, South Korea ........... 12.3% 10, Philippines ............. 10,2% 11. Colombia ................ 13.9% 12. Taiwan ................... 16.4% 13. Brazil ..................... 15.3% 14. Israel ...................... 13.4% 15. Jamaica .................. 13.4% 16. Hong Kong ............. 15.7% 17. Australia ................. 10.2% 18. Spain ..................... 23.3% 19. El Salvador ............. 9.0% 20. India ....................... 29.0% 21. Switzerland ............. 9.8% 22. Netherlands ............ 14.3% 23. Guatemala .............. 11.7% 24. Peru ....................... 11.5% 25. Venezuela ............. 24.0% Incoming Total Volume CAGR ’90-’92 CAGR ’90-’92 (’92 Volume x 106) 3.2% 13.9% (3738) 27.6% 27.8% (1885) 3.4% 6.1% (1235) 9.2% 7.4% (799) 11.6% 9.0% (649) 7.5% 8.4% (396) 11.4% 17.1% (280) 23.9% 15.0% (325) 34.3% 18.0% (304) 14.4% 10.6% (224) 10.5% 13.1% (222) 16.2% 16.4% (245) -6.3% 8.5% (194) 8.8% 12.2% (186) 18.7% 14.6% (167) 13.4% 14.7% (197) 9.3% 9.8% (218) 22.5% 23.0% (158) 2.4% 8.4% (108) 26.6% 28.1% (158) 5.8% 8.1% (160) 11.1% 12.9% (159) 24.2% 13.6% (107) 6.1% 10.7% (93) 18.3% 21.5% (129) Note: Routes are ranked by 1992 outgoing traffic volume. CAGR is Cumulative Annual Growth Rate. 1992 data are based on billing country rather than physical point of origin. ¯ MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 87 Table 15 Uruguay Largest Telecommunication Routes (1992) Destination MiTT 1. Argentina .................. 17.8 2. Brazil ............................ 3.4 3. U.S.A ............................ 2.1 4. Spain ........................... 1.0 5. Chile ............................. 0.7 6. Paraguay ..................... 0.5 7. Italy ............................... 0.4 8. Germany ..................... 0.3 9. France ......................... 0.3 10. U.K .............................. 0.2 11. Canada ...................... 0.2 12. Israel ........................... 0.2 13. Switzerland ............... 0.2 14. Venezuela ................. 0.2 15. Mexico ....................... 0.2 Percent of Outgoing Traffic 59,1% 11.4% 7.1% [] 3.5% 2.3% 1.6% 1.5% 1.1% 1.0% 0.8% 0.7% 0.7% 0.7% 0.7% 0.6% Other .......................... 2.2 7.3% National Traffic Balance *MITT 1990 Incoming N.A. Outgoing 25.8 Surplus(Deficit) N.A. Total Volume N.A. 1991 48.0 29.3 18.7 77.3 1992 53.0 30.2 22.8 83.2 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 88 Table 16 Venezuela Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A ......................... 53.7 2. Colombia .................. 12.7 3. Spain ........................... 7.3 4. Italy ............................... 5.4 5. Peru ............................. 2.8 6. Portugal ...................... 2.3 7. Canada ....................... 2.3 8. Germany ..................... 2.0 9. France ......................... 2.0 10. Mexico ....................... 1.9 11. Argentina .................. 1.8 12. Brazil .......................... 1.8 13. Dominican Rep ........ 1.7 14. U.K ............................. 1.7 15. Chile .......................... 1.4 Percent of Outgoing Traffic 11.0% 6.3% 4.6% 2.4o/o 2.0% 2.0% 1.7% 1.7% 1.6% 1.6% 1.5% 1.5% 1.4% 1.2% 46.5% Other ........................ 14.8 I 12.8% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 85.3 N.A. N.A. 1991 103.9 91.7 12.2 195.6 1992 128.6 115.5 13.1 244.1 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 89 EUROPEAN COMMUNITY TRAFFIC TABLES ,).) .," ,, 90 Table 17 Belgium Largest Telecommunication Routes (1992) Destination MiTT 1. France ....................... 220.5 2. Netherlands ............. 184.9 3. Germany ................... 119.3 4. U.K ............................... 82.5 5. Italy .............................. 47.0 6. U.S.A ........................... 31.1 7. Luxembourg ............... 28.1 8. Spain ........................... 26.6 9. Switzerland ................ 23.5 10. Sweden ..................... 11.5 11. Portugal ....................... 9.5 12. Greece ......................... 9.0 13. Denmark ...................... 8.1 14. Austria .......................... 6.9 15. Ireland .......................... 5.1 Percent of Outgoing Traffic 24.2% 20.3% 13.1% 9.1% 5.2% 3.4% 3.1% [~ 2.9% 2.6% 1.3% 1.0% 1.0% 0.9% 0.8% 0.6% Other ........................... 97.4 10.7% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 1991 754.1 748.5 731.3 822.7 22.8 (74.2) 1485.4 1571.2 Note: 1992 route data are based on Belgacom provisional data. 1992 N.A. 911.0 N.A. N.A. "Mi’l-I- is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 -91 Table 18 Denmark Largest Telecommunication Routes (1992) 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Destination MiTF Percent of Outgoing Traffic Germany .................. 77.7 Sweden .................... 76.1 U.K ............................. 46.6 Norway ...................... 44.0 U.S.A ......................... 21.0 France ....................... 19.3 18.3% 17.9% 11.0% 10.3% 4.9% 4.5% Netherlands .............. 16.3 3.8% Italy ............................. 10.7~ 2.5% Switzerland ................. 9.7~ 2.3% Spain ........................... 8.9 2.1% Finland ........................ 8.8 2.1% Belgium ....................... 8.5 2.0% Poland ......................... 6.8 1-6% Turkey .......................... 4.6~ 1.1% Austria ................ : ........ 3.9 0.9% Other .......................... 61.5 14.5% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 343.2 362.3 (19.1) 705.5 1991 397.2 394.5 2.7 791.7 1992 425.2 424.5 0.7 849.7 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 92 Table 19 France Largest Telecommunication Routes (1992) Destination MiT[ 1. Germany .................. 303.0 2. U.K ............................ 275.3 3. Italy ........................... 217.1 4. Belgium .................... 215.6 5. Switzerland .............. 195.5 6. Spain ........................ 156.9 7. U.S.A ......................... 153.1 8. Portugal .................... 114.1 9. Netherlands ............... 80.4 10. Morocco .................... 61.6 11. Algeria ....................... 54.6 12. Tunisia ....................... 32.5 13. Canada ..................... 31.7 14. Turkey ........................ 29.3 15. Sweden ..................... 22.0 Percent of Outgoing Traffic 12.4% 11.2% 8.9% 8.8% 8.0% 6.4% 6.3% 4.7% 3.3% 2.5% 2.2% 1.3% 1.3% 1.2% 0.9% Other ......................... 506.3 20.7% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 2183.0 2126.0 57.0 4309.0 1991 2355.0 2295.0 60.0 4650.0 1992 2540.0 2449.0 91.0 4989.0 "MiFI" is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 93 Table 20 Germany Largest Telecommunication Routes (1992) Destination MiTT 1. Austria .......................... 329.0 2. France .......................... 318.0 3. Switzerland ................. 315.0 4. Turkey .......................... 295.0 5. Italy ............................... 288.0 6. U.K ................................ 287.0 7. Netherlands ................ 250.0 8. U.S.A ............................ 235.0 9. Yugoslavia (former)...150.0 10. Spain .......................... 118.0 11. Belgium ...................... 115.0 12. Poland ........................ 104.0 13. Czechoslovakia .......... 96.0 14. Greece .......................... 77.0 Percent of Outgoing Traffic 7.8% 7,7% 7.2% 7.0% 7.o% 6.1% 5.7% 3.7% 2.9% 2.8% 2.5% 2.3% [] 1.9% National Traffic Balance *MITT Incoming Outgoing ,,,,Surplus(Deficit) Total Volume 1990 2600.0 3185.0 (585.0) 5785.0 1991 2900.0 3605.0 (705.0) 6505.0 Note: Annual incoming MiTT are rounded to the nearest 100 million 1992 3100.0 4087.0 (987.0) 7187.0 all other data are rounded to nearest million MiT-r. Data include traffic from the former East Germany (i.e. all data are reported for the unified Germany). In 1990, prior to unification (September 1990), East Germany originated about 40 million MiTl to foreign countries (excluding West Germany); in 1991 Eastern L~nder accounted for approximately 48 million Mil-f; and in 1992 approximately 60 million MITT. ¯ MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits © Copyright TeleGeography 1993 94 Table 21 Greece Largest Telecommunication Routes (1992) Destination MiTT 1. Germany ................... 61.5 2. U.K ............................. 39.9 3. Italy ............................ 26.4 4. U.S.A ......................... 25.3 5. France ...................... 17.4 6. Cyprus ...................... 12.5 7. Canada ...................... 8.8 8. Switzerland ............... 8.5 9. Belgium ...................... 8.5 10. Australia .................... 8.1 11. Netherlands ............. 8.0 12. Austria ....................... 4.9 13. Sweden .................... 4.8 14. Turkey ....................... 4.5 15. Egypt ......................... 3.3 Percent of Outgoing Traffic 20.6% 13.4% 8.8% 8.5% 5.8% 4.2% 2.9% 2.8% 2.8% 2.7% 2.7% [] 1.7% I 1.6% I 1.5% I 1.1% Other ........................ 56.4 18.9% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 213.3 N.A. N.A. 1991 320.1 245.0 75.1 565.1 1992 359.7 298.9 60.8 658.5 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 95 Table 22 Ireland Largest Telecommunication Routes (FY 1992/93) Destination MWT 1. U.K ............................ 206.0 2. U.S.A .......................... 27.7 3. Germany .................... 12.3 4. France ........................ 10.0 5. Netherlands ................ 5.7 6. Italy ............................... 4.1 7. Spain ........................... 3.8 8. Belgium ....................... 3.7 9. Canada ....................... 3.0 10. Australia .................... 2.6 11. Switzerland .............. 2.2 12. Denmark ................... 1.7 13. Sweden ..................... 1.7 14. Japan ......................... 0.8 15. Austria ........................ 0.7 Percent of Outgoing Traffic 9.3% ~j 4.1% 3.4o/o 1.9% 1.4% 1,3% 1.2% 1.0% 0.8% 0.8% 0.6% 0.6% 0.3% 0.2% Other ......................... 10.5 3.5% National Traffic Balance *MITT FY 1990/91 Incoming 1 22.2 Outgoing 75.1 Surplus(Deficit) 47.1 Total Volume N.A. 1 97.3 FY Note: FY ends on 31/3. Traffic balances for 1990 and 1991 exclude the U.K. 1991/92 140.4 83.4 57.0 223.8 FY 1992/93 383.0 296.6 86.4 679.6 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 96 Table 23 Italy Largest Telecommunication Routes (1992) Destination MiTT 1. Germany .................. 226.5 2. France ...................... 209.1 3. U.S.A ........................ 155.0 4. Switzerland ............. 148.7 5. U.K ............................ 117.9 6. Spain .......................... 56.0 7. Belgium ...................... 44.4 8. Austria ........................ 39.1 9. Yugoslavia (former).39.0 10. Netherlands ............. 31.5 11. Greece .................... 22.9 12. Morocco .................... 20.1 13. Canada .................... 17.3 14. Hong Kong .............. 16.6 15. Tunisia ..................... 16.0 Percent of Outgoing Traffic I~~ 15.4% 14.2% lO.2% 10.1% 8.0% 3.8% 3.0% 2.7% 2.6% 2.1% 1.6% 1.4% 1.2% 1.1% 1.1% Other ....................... 313.3 21.3% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 1043.1 N.A. N.A. 1991 N.A. 1219.5 N.A. N.A. 1992 1541.0 1473.4 67.6 3014.4 "MiTF is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 97 Table 24 Luxembourg Largest Telecommunication Routes (1992) Destination MiTT 1. Belgium ..................... 41.6 2. Germany ................... 36.0 3. France ....................... 35.2 4. Portugal .................... 11.8 5. U.K ............................. 10.9 6. Italy .............................. 9.1 7. Netherlands ............... 6.7 8. Switzerland ................ 5.7 9. U.S.A ........................... 4.1 10. Denmark .................... 3.1 11. Spain .......................... 2.9 12. Sweden ..................... 1.9 13. Austria ........................ 1.8 14. Greece ....................... 1.2 15. Japan ......................... 0.7 Percent of Outgoing Traffic 19.5% 6.5% 6.0% 5.0% 3.7% 3.2% 2.3% [] 1.7% ~] 1.6% L~ 1.0% L~ 1.0% 0.7% ~] 0.4% Other ........................... 8.2 4.5% National Traffic Balance *MITT 1990 1991 Incoming N.A. N.A. Outgoing 150.6 165.3 Surplus(Deficit) N.A. N.A. Total Volume N.A. N.A. 1992 107.5 181.0 (73.5) 288.5 "Mi]-r is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 98 Table 25 Netherlands Largest Telecommunication Routes (1992) Destination MiTT 1. Germany .................. 263.7 2. Belgium .................... 189.3 3. U.K ............................ 146.8 4. France ........................ 86.6 5. U.S.A .......................... 66.1 6. Italy ............................. 36.5 7. Switzerland ............... 32.0 8. Spain ......................... 30.8 9. Turkey ........................ 27.7 10. Sweden .................... 18.1 11. Denmark .................. 16.8 12. Austria ...................... 16.7 13. Canada .................... 11.9 14. Norway ..................... 10.0 15. Poland ........................ 9.7 Percent of Outgoing Traffic 16.7% 12.9% 5.8% 3.2% 2.8% @ 2.7% 2.4% [] 1.6% 1.5% [] 1.5% 1.0% 0.9% 0.9% 23.3% Other ....................... 181.1 ..................... I 16.o% National Traffic Balance *MITT Incoming Outgoing Su rplus(Deficit) Total Volume 1990 851.8 905.2 (53.4) 1757.0 1991 N.A. 1017.7 N.A. N.A. 1992 1039.0 1133.9 (94.9) 2172.9 Note: 1992 data are based upon the bdhng point of the traffic. See Methodology pages. *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 99 Table 26 Portugal Largest Telecommunication Routes (1992) Destination MiTT Percent of Outgoing Traffic 1. France ........................ 44.5 2. Spain .......................... 32.5 3. U.K .............................. 23.4 11.o% 4. Germany .................... 20.9 9.9% 5. Switzerland ............... 12.0 5.7% 6. U.S.A ............................ 9.2 ~i 4.3% 7. Italy ................................ 8.3 3.9% 8. Brazil ............................. 7.7 3.6% 9. Belgium ........................ 7.5 3.5% 10. Netherlands ............... 7.0 3.3% 11.Angola ......................... 4.5 [] 2.1% 12. Canada ....................... 4.2 [] 2.0% 13. Sweden ..................... 3.0 a 1.4% 14. Denmark .................... 2.0 o.9% 15. Greece ....................... 1.8 o.6% 15.3% 21.0% Other .......................... 23.5 11.1% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 325.9 156.5 169.4 482.4 1991 384.0 186.1 197.9 570.1 1992 N.A. 212.0 N.A. N.A. "Mi’l-I" is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 i00 Table 27 Spain Largest Telecommunication Routes (1992) Destination Mil-r 1, France ...................... 139.4 2. U.K ............................ 115.0 3. Germany .................. 112.4 4. U.S.A .......................... 75.1 5. Italy ............................. 57.7 6. Netherlands .............. 31.6 7. Portugal ..................... 30.6 8. Switzerland ............... 29.7 9. Belgium ...................... 27.2 10. Morocco .................... 16.6 11. Argentina .................. 14.1 12. Sweden .................... 13.1 13. Denmark ..................... 8.2 14. Venezuela .................. 7.7 15. Colombia .................... 7.5 Percent of Outgoing Traffic I 17.3% I 14.3% 14.0% 9.3% 7.2% 3.9% 3.8% 3.7% 3.4% 2.1% 1.8% [] 1.6% 1.0% 1.0% 0.9% Other ....................... 118.7 14.8% National Traffic Balance *MITT 1990 Incoming 653.0 Outgoing 611.0 Surplus(Deficit) 42.0 Total Volume 1264.0 1991 736.9 718.7 18.2 1455.6 1992 847.2 804.5 42.7 1651.7 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 i01 Table 28 United Kingdom Largest Telecommunication Routes (FY 1992/93) Destination MiTT 1. U.S.A .......................... 575.0 2. Germany .................... 325.0 3. France ........................ 269.0 4. Ireland ........................ 218.0 5. Italy ............................. 135.0 6. Netherlands .............. 134.0 7. Spain ......................... 115.0 8. Canada ....................... 95.0 9. Australia ...................... 90.0 10. Switzerland ............... 86.0 11. Belgium ..................... 79.0 12. Sweden ..................... 54.0 13. India ........................... 53.0 14. Denmark ................... 47.0 15. Japan ........................ 43.0 16. Greece ...................... 43.0 17. Norway ..................... 42.0 18. South Africa ............. 39.0 19. Portugal .................... 36.0 20. Hong Kong ............... 35.0 Other ....................... 336.0 Percent of Outgoing Traffic _ 20.2% 11.2% 9.3% 7.5% 4.7% 4.6% 4.0% 3.3% 3.1% 3.0% D 2.7% 1.9% [] 1.8% !~ 1.6% B 1.5% [] 1.5% l~ 1.5% [] 1.3% ~! 1.2% 1.2% 11.8% National Traffic Balance *MITT FY 1989/90 Incoming 2 330.0 Outgoing 2253.0 Surplus(Deficit) (7 7.0) Total Volume 45 83.0 Note" FY ends on 31/3. FY 1991/92 2692.0 2598.0 94.0 5286.0 "MiTI" is Minutes of Telecommunications Traffic FY 1992/93 2789.0 2849.0 (6o.o) 5638.0 Data are in millions of minutes for public voice circuits and have been rounded to the nearest million. ©TeleGeography, Inc. 1993 102 Z o -- z o o -- z o 103 Z 104 EUROPEAN FREE TRADE ASSOCIATION TRAFFIC TABLES 105 Table 30 Austria Largest Telecommunication Routes (1992) Destination MH’T 1. Germany ................... 303.0 2. Switzerland ................ 53.3 3. Italy .............................. 35.4 4. Serbia/Montenegro..28.7 5. Hungary ...................... 24.5 6. Czechoslovakia ......... 23.4 7. U.S.A ........................... 23.4 8. U.K ............................... 22.9 9. Turkey .......................... 22.3 10. France ........................ 19.2 11. Netherlands .............. 17.2 12. Croatia ....................... 17.0 13. Poland ....................... 16.6 14. Slovenia .................... 11.8 15. Belgium ........................ 7.5 Percent of Outgoing Traffic 42.5% 7.5% 5.0% [] 4.0% Li~ 3.4% L~ 3.3% [] 3.3% I~ 3.2% L~ 3.1% 2.7% L~I 2.4% L~ 2.4% 2.3% L~ 1.6% I~ 1.0% Other ........................... 87.4 12.2% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 513.3 558.6 (45.3) 1071.9 1991 574.0 641.9 (67.9) 1215.9 1992 692.3 713.4 (21.1) 1405.7 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 106 Table 31 Finland Largest Telecommunication Routes (1991) Destination MiTT 1. Sweden .................... 89.0 2. Germany ................... 19.0 3. U.S.A ......................... 14.0 4. U.K ............................. 13.0 5. U.S.S.R.(former) ...... 12.0 6. Denmark ..................... 7.0 7. Norway ........................ 7.0 8. France ......................... 6.0 9. Switzerland ................ 4.0 10. Netherlands .............. 4.0 11. Italy ............................. 4.0 12. Spain .......................... 4.0 Percent of Outgoing Traffic 8.8% 6.5% 6.0% 5.6% 3.3% 3.3% 2.8% 1.9% 1.9% 1.9% 1.9% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 213.0 187.0 26.0 400.0 1991 N.A. 215.0 N.A. N.A. 1992 N.A. 235.0 N.A. N.A. *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits rounded to the nearest million ©TeleGeography, Inc. 1993 107 Table 32 Iceland Largest Telecommunication Routes (1992) Destination MiTT 1, U,S,A .......................... 6,1 2. Denmark .................... 3.0 3. U.K .............................. 2.5 4. Sweden ..................... 2.5 5. Norway ....................... 1.9 6. Germany .................... 1.7 7. France ........................ 0.7 8. Netherlands .............. 0.6 9. Faroe Islands ............ 0.5 10. Canada ..................... 0.3 11. Spain ........................ 0.3 12. Italy ............................ 0.3 13. Belgium .................... 0.2 14. Finland ..................... 0.2 15. Switzerland ............. 0.2 Percent of Outgoing Traffic ~.~~’~’~ 1 1.5% ~’XXX~,~’~"~ 11.4% 8.5% 7.7% ~:~ 3.1% 2.8% ~] 2.3% 1.4% 1.3% 1.3% 1.1% 1.1% 1.0% Other ......................... 1.1 5.o% National Traffic Balance *MITT Incoming Outgoing Su rplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 N.A. N.A. N.A. N.A. 1992 21.7 22.1 (0.4) 43.8 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for pubhc voice circuits ©TeleGeography, Inc. 1993 108 Table 33 Norway Largest Telecommunication Routes (1991) Destination MiTT 1. Sweden ...................... 90.0 2, Denmark ..................... 45,0 3. U.K ............................... 40.0 4. U.S.A ........................... 24.0 5. Germany ..................... 23.0 6. France ......................... 10.0 Percent of Outgoing Traffic 29.2% 14.6% 13% 7.8% 7.5% [] 3.7% 7. Finland .......................... 8.0 I 2.6% 8. Spain ............................. 6.0 [] 1.9% 9. Switzerland .................. 5.0 1.6% 10. Italy ............................... 5.0 I 1.6% 11. Belgium ....................... 4.0 1.3% 12. Poland ......................... 3.0 13. Canada ....................... 3.0 14. Yugoslavia (former)..2.0 15. Austria ......................... 2.0 16. Iceland ........................ 2.0 17. Turkey ......................... 2.0 Other ......................... 34.5 1.0% 1.0% 0.6% 0.6% 0.6% 0.6% 11.2% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. 281.2 N.A. N.A. 1991 N.A. 308.5 N.A. N.A. 1992 301.0 349.0 (48.0) 650.0 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits rounded to the nearest million ©TeleGeography, Inc. 1993 109 Table 34 Sweden Largest Telecommunication Routes (1992) Destination MiTT Percent of Outgoing Traffic 1. Finland ...................... 98.01 14.4% 2. Norway ..................... 95.0 I I13-9°/o 3. Denmark ................... 84.0 ~-................ -~ 12.3% 4. Germany ................... 70.0f lO.3% 5. U.K ............................. 60.0 ~-..... T_--~=_-=--~-_-_J 8.8% 6. U.S.A ......................... 53.0[ 7.8°/o 7. France ....................... 26.0[ I 3.8% 8. Spain ......................... 17.0 2.5% 9. Italy ............................. 16.0 2.3% 10. Netherlands ............ 15.0 2.2% 11. Switzerland ............. 14.0 2.1% 12. Belgium .................... 12.0 1.8% 13. Austria ........................ 9.0 1.3% 14. Turkey ........................ 6.0 [] o.9% 15. Canada ...................... 6.0 [] o.9% 16. Portugal ..................... 4.0 0.6% 17. Czechoslovakia ....... 4.0 0.6% Other ....................... 93.0 f I13"6% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 1991 N.A. 571.0 616.0 672.0 N.A. (101.0) N.A. 1 243 Note: Data are estimated to nearest million Mi’FT. 1992 595.0 691.0 (96.0) 1286.0 "Mi’l-r is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 ii0 Table 35 Switzerland Largest Telecommunication Routes (1992) Destination MiTT 1. Germany .................. 357.7 2. France ...................... 264.3 3. Italy ........................... 224.4 4. U.K .............................. 82.4 5. Austria ........................ 69.1 6. U.S.A .......................... 65.3 7. Portugal ..................... 58.8 8. Spain ......................... 52.8 9. Yugoslavia ................ 45.5 10. Netherlands ............. 36.5 11. Turkey ....................... 29.4 12. Belgium .................... 27.4 13. Sweden .................... 16.1 14. Canada ..................... 11.7 15. Czechoslovakia ....... 11.0 Percent of Outgoing Traffic 23.1% 17.0% 14.5% 5.3% 4.5% 4.2% 3.8% 3.4% 2.9% 2.4% 1.9% 1.8% 1.0% 0.8% 0.7% Other ........................ 198.5 12.8% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 1015.7 1356.3 (340.6) 2372.0 1991 1124.0 1429.4 (305.4) 2553.4 1992 1191.5 1551.0 (359.5) 2742.5 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 iii EASTERN EUROPE TRAFFIC TABLES 112 Table 36 Czechoslovakia Largest Telecommunication Routes (1992) Destination MiTT 1. Germany .................... 43.8 2, Austria ........................ 19.6 3. U.S.A .......................... 10.6 4. Italy ............................... 8.7 5. Hungary ...................... 8.2 6. U. K .............................. 7.3 7. U.S.S.R. (former) ....... 7.1 8. Poland ......................... 6.7 9. Switzerland ................. 6.1 10. France ......................... 5.9 11. Yugoslavia (former)...5.2 12. Netherlands ................ 4.2 13. Belgium ....................... 2.5 14. Canada ........................ 2.4 15. Sweden ....................... 2.0 Percent of Outgoing Traffic 24.6% 11.0% 5.9% 4.9% 4.6% 4.1% 4.0% 3.8% 3.4% 3.3% 2.9% 2.3% [] 1.4% [~ 1.3% 1.1% Other ........................... 37.6 21.1% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 179.5 128.1 51.4 307.6 1992 236.0 178.0 58.0 414.0 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 113 Table 37 Hungary Largest Telecommunication Routes (1992) Destination MITT 1. Germany ................... 45.9 2. Austria ....................... 24.9 3. U.S.A .......................... 12.9 4. Italy ............................. 10.3 5. Romania ..................... 8.9 6. U.K ............................... 8.8 7. Czechoslovakia ........ 7.2 8. Switzerland ................ 6.5 9. France ......................... 6.3 Percent of Outgoing Traffic 25.0% 13.6% 7.0% 5.6% 4.8% 4.8% 3,9% 3.5% 3.4% 10. Russia ........................ 5.8 3.1% 11. Netherlands .............. 4.2 [] 2.3% 12. Sweden ..................... 4.2 [] 2.3% 13. Canada ..................... 3.4 [] 1.9% 14. Belgium ..................... 3.0 [] 1.6% 15. Australia .................... 2.4 i 1.3% Other ........................ 30.0 16.3 National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 109.4 121.9 (12.5) 231.3 1991 N,A. N,A, N.A. N.A. 1992 150.5 183.8 (33.3) 334.3 "Mi’l-I" is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 114 Table 38 Poland Largest Telecommunication Routes (1992) Destination Mi3"r 1. Germany ................... 63.0 2. U.S.A ......................... 25.3 3. France ....................... 12.6 4. U.K ............................. 12.1 5. Russia ....................... 11.5 6. Italy .............................. 9.4 7. Sweden ...................... 8.6 8. Austria ......................... 8.6 9. Netherlands ............... 6.4 10. Canada ..................... 5.9 11. Czechoslovakia ....... 5.4 12. Belgium ..................... 4.8 13. Denmark ................... 4.2 14. Switzerland .............. 3.4 15. Ukraine ..................... 3.0 Percent of Outgoing Traffic ~//’//////////,///~ 11.9% 5.9% 5.7% 5.4% 4.4% 4.1% 4.0% 3.0% 2.8% 2.5% [] 2.3% N 2.0% 1.6% ~J 1.4% Other ........................ 28.5 ~~~ 13.4°/o National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 N.A. N.A. N.A. N.A. 1992 366.6 212.7 153.9 579.3 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 115 Table 39 Russia Largest Telecommunication Routes (1992) Destination MiTT 1. Germany ..................... 24,5 2. U.S.A ........................... 20.6 3. Poland ......................... 11.4 4. Yugoslavia (former) .... 8.4 5. Bulgaria ........................ 6.9 6. Finland .......................... 6.8 7. U.K ................................. 6.2 8. Czechoslovakia ........... 6,0 9. Hungary ........................ 5.9 10. France .......................... 5.9 11. Belgium ....................... 3.5 12. Sweden ....................... 2.9 13. Spain ........................... 2.4 14. Canada ....................... 2.3 15. Switzerland ................ 2.3 Percent of Outgoing Traffic ~~ 14.0O/o 11.7% 6.5% 4.8% U 3.9% 3.9% 3.5% 3.4% ]3.4% [] 3.4% W 2.0% 1.7% 1.4% 1.3% 1.3% Other .......................... 59.6 33.9% National Traffic Balance *MITT 1990 Incoming N.A. Outgoing N.A. Su rpl us(Deficit) N.A. Total Volume N.A. 1991 N.A. N.A. N.A. N.A. 1992 230.7 175.6 55.1 406.3 Note: Outgoing total and ’other’ include traffic between Russia and the other states of the former U.S.S.R. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 I16 MIDDLE EAST & AFRICA TRAFFIC TABLES !17 Table 40 israel Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A ......................... 48.3 2. U.K ............................. 15.1 3. France ....................... 11.8 4. Germany ................... 10.8 5. U.S.S.R. (former) ....... 6.2 6. Italy ............................... 5.6 7. Canada ....................... 4.6 8. Switzerland ................ 4.5 9. Netherlands ............... 4.0 10. Belgium ..................... 3.8 11. South Africa .............. 2.6 12. Australia .................... 2.4 13. Romania .................... 2.1 14. Turkey ........................ 1.8 15. Argentina ................... 1.7 Percent of Outgoing Traffic 9.9% 7.7% 7.1% ~4.1% 3.7% 3.0% 2.9% N 2.6% N2.5% 1.7% 1.6% 1.3% 1.2% ~1.1% National Traffic Balance *MITT 1 990 Incoming 202.0 Outgoing 118.0. Surplus(Deficit) 84.0 Total Volume 320.0 1991 247.0 139.4 107.6 386.4 1992 266.0 153.1 112.9 419.1 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 1.18 Table 41 Saudi Arabia Largest Telecommunication Routes (1992) Destination MiTt 1. Egypt ........................ 123.8 2. Pakistan ..................... 49.0 3. U.S.A .......................... 30.3 4. India ............................ 23.8 5. U.K .............................. 21.4 6. U.A.E .......................... 20.7 7. Yemen ....................... 17.4 8. Jordan ....................... 16.5 9. Kuwait ....................... 15.3 10. Philippines .............. 13.1 11. Bahrain .................... 12.4 12. Turkey ...................... 11.7 13. Syria .......................... 9.7 14. Bangladesh .............. 9.3 15. France ....................... 9.2 Percent of Outgoing Traffic --1 10.5o/o 6.5% 5.1% 4.6% 4.4% 3.8% 3.5% 3.3% 2.8% ~’~ 2.7% 2.5% [] 2.1% [] 2.0% [] 2.0% 26.6% Other ........................ 81.2 17.5% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 266.7 410.6 (143.9) 677.3 1992 292.1 464.6 (172.5) 756.7 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 119 Table 42 South Africa Largest Telecommunication Routes (1992) Destination MiTI" Percent of Outgoing Traffic 1. U.K ............................. 37.1 2. Namibia .................... 27.5 3. Zimbabwe ................ 18.8 8.5 4. U.S.A ......................... 17.6 7.9 5. Botswana .................. 12.8 5.8 6. Germany .................... 10.8 4.9 7. Swaziland ................... 9.9 4.5 8. Mozambique ............... 9.4 4.2 9. Lesotho ........................ 7.1 3.2 10. Australia .................... 5.4 [] 2.4 11. Portugal ...................... 4.1 I~ 1.8 12. France ........................ 3.9 I~ 1.8 13. Israel ........................... 3.7 [] 1.7 14. Italy ............................. 3.6 ~1 1.6 15. Canada ...................... 3.6 I~ 1.6 12.4 16.7 Other ......................... 46.4 20.9 National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N~. N.A. N.A. N.A. 1991 N.A. 202.6 N.A. N.A. 1992 250.7 221.7 29.0 472.4 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 120 Table 43 Turkey Largest Telecommunication Routes (1992) Destination MiTT 1. Germany ................... 82.8 2. U.K ............................. 18.0 3. U.S.A ......................... 15.9 4. France ....................... 14.9 5. Netherlands .............. 10.8 6. Switzerland ................. 8.8 7. Italy ............................... 7.9 8. Austria .......................... 7.0 9. Bulgaria ....................... 5.0 10. Belgium ...................... 4.7 11. U.S.S.R (former) ....... 4.6 12. Romania ..................... 4.3 13. Greece ........................ 3.9 14. Saudi Arabia ............. 3.6 15. Iran .............................. 3.2 Percent of Outgoing Traffic 36.5% 8.0% 7.0% 6.6% 4.8% 3.9% 3.5% 3.1% 2.2% 2.1% 2.0% 1.9% 1.7% 1.6% 1.4% Other ......................... 31.4 13.8% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 441.2 158.5 282.7 559.7 1991 470.0 198.1 271.9 668.1 1992 560.0 226.8 333.2 786.8 "Mi’FI- is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 121 ASIA TRAFFIC TABLES ¯ ~22 Table 44 Australia Largest Telecommunication Routes (1992) Destination MiTT 1. U.K ............................. 112.0 2. U.S.A ......................... 104.0 3. New Zealand .............. 90.0 4. Japan ........................... 32.0 5. Hong Kong .................. 28.0 6. Singapore ................... 20.0 7. Italy ............................... 19.0 8. Germany ...................... 17.0 9. Canada ........................ 16.0 10. Philippines ................. 14.0 11. Greece ....................... 12.0 12. Indonesia .................... 11.0 13. Papua New Guinea..11.0 Percent of Outgoing Traffic 17.0% ~~ 15-8°/o 13.7% 4.9% 4.2% 3.0% [~ 2.9% [] 2.6% W 2.4% [] 2.1% 1.8% 1.7% 1.7% r~~ :7~ 31.9% Other .......................... 210.0 National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 439.0 574.7 (135.7) 1013.7 1991 N.A. 610.0 N.A. N.A. 1992 N.A. 659.0 N.A. N.A. Note: Data are estimated to the nearest million MITT. The next largest outgoing traffic streams are to Malaysia; China; Yugoslavia; Taiwan; Fiji; France; Thailand; Netherlands; South Africa; South Korea; Ireland; Switzerland; India; Sweden; Turkey; Sri Lanka; and Israel. "Mi-I-I- is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 123 Table 45 Bangladesh Largest Telecommunication Routes (1992) Destination MiTI" 1. U.K ............................... 2.8 2. U.S.A ........................... 1.7 3. Hong Kong ................. 1.7 4. India ............................ 1.4 5. Japan .......................... 1.3 6. Singapore .................. 1.2 7. Saudi Arabia ............. 0.7 8. Pakistan ...................... 0.6 9. Australia ...................... 0.4 10. Germany .................... 0.4 11. Thailand ..................... 0.4 12. U.A.E .......................... 0.4 13. Italy ............................. 0.3 14. South Korea ............. 0.2 15. Canada ...................... 0.1 PercentofOutgoingTraffic 9.2 8.6 4.8% 4.5% 2.9% 2.9% 2.7% 2.5% 2.4% 1.5% 0.8% 19.5% Other .......................... 7.0 4.9% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 NA. NA. NA. N.A. 1991 N.A. 10.3 N.A. N.A. 1992 66.8 14.3 52.5 81.1 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 124 Table 46 China Largest Telecommunication Routes (1992) Destination MiTT 1. Hong Kong .............. 412.0 2. Japan .......................... 56.9 3. Taiwan ........................ 47.1 4. U.S.A ........................... 30.3 5. Macao ......................... 24.8 6. South Korea ................ 9.3 7. Singapore .................... 7.3 8. Australia ....................... 6.2 9. Germany ...................... 5.2 10. Canada ....................... 4.9 11. Thailand ...................... 3.5 12. France ......................... 3.5 13. U.K ............................... 3.2 14. Italy .............................. 2.6 15. Malaysia ..................... 2.1 Percent of Outgoing Traffic 1 .9% ~9.0% ~7.4% N 4.8% 3.9% 1.5% 1.2% 1.0% 0.8% 0.8% 0.6% 0.5% 0.5% 0.4% 0.3% Other ......................... 16.2 2.5% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 N.A. 440.0 N.A. N.A. 1992 N.A. 635.1 N.A. N.A. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 125 Table 47 Hong Kong Largest Telecommunication Routes (FY 1992/93) Destination MITT Percent of Outgoing Traffic 1. China ........................ 535.0 47.0% 2. U.S.A .......................... 80.0 7.0% 3. Taiwan ....................... 68.0 6.0% 4. Canada ...................... 57.0 ~i 5.0% 5. Japan ......................... 45.0 [] 4.o% 6. U.K .............................. 45.0 [] 4.0% 7. Macao ........................ 34.0 [] 3.0% 8. Singapore ................. 34.0 [] 3.0% 9. Australia .................... 34.0 3.o% 10. Philippines .............. 34.0 IB 3.0% Other ....................... 171.0 15.o% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume FY 1990/91 645.5 729.4 (83.9) 1374.9 FY 1991/92 783.0 913.2 (130.2) 1696.2 FY 1992/93 1009.4 1136.6 (127.2) 2146.0 Note: Data are estimated to nearest mdhon MITT, based on market share data rounded to nearest percent. Fiscal year ends on 31/3. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 126 Table 48 India Largest Telecommunication Routes (FY 1992/93) Destination MiTT 1. U.S.A .......................... 62.1 2. Saudi Arabia ............ 28.4 3. U.A.E .......................... 28.2 4. U.K ............................. 24.8 5. Singapore .................. 8.9 6. Canada ....................... 6.4 7. Germany ..................... 5.5 8. Hong Kong ................. 5.2 9. Japan .......................... 4.4 10. Australia .................... 4.2 11. Oman ......................... 3.2 12. France ........................ 2.9 13. Qatar ........................... 2.8 14. Malaysia .................... 2.4 15. Italy ............................. 2.2 Percent of Outgoing Traffic 24.0% 10.9% 10.9% 9.6% 3.4% 2.5% N 2.1% 2.0% 1.7% 1.6% 1.2% 1.1% 1.1% 0.9% 0.9% Other ......................... 68.2 26.3% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume FY 1990/91 N.A. N.A. N.A. N.A. FY 1991/92 FY N.A. 185.8 N.A. N.A. Note: Outgoing totals and route data do not include India-Pakistan traffic. Fiscal year ends on 31/3. 1992/93 354.6 259.6 95.0 614.2 *MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 127 Table 49 Indonesia Largest Telecommunication Routes (1992) Destination M[Tr 1. Singapore ................ 28.8 2. U.S.A ......................... 12.5 3. 4. 5. 6. 7. Taiwan .......................... 5.9 8. Malaysia ...................... 5.2 9. U.K ................................ 4.1 10. Germany ..................... 3.4 11. Netherlands ............... 3.2 12. France ......................... 2.5 13. Philippines ................. 1.9 14. Canada ....................... 1.6 15. Thailand ...................... 1.5 Percent of Outgoing Traffic Japan ......................... 12.3 10.4% Australia ....................... 8.5 ~’~,~.~ 7.2% Hong Kong .................. 8.3 7.0°/0 South Korea ................ 6.0 5.1% 5.0% 4.4% 3.4% 2.9% ~--~ 2.7% 2.1% 1.6% 1.3% 1.2% 24.4% Other .......................... 12.5 10.6% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 NA. NA. NA. NA. 1991 135.7 100.0 35.7 235.7 1992 165.5 118.1 47.4 283.6 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 128 Table 50 Japan Largest Telecommunication Routes (FY 1992/93) Destination MiTT 1. U.S.A ........................ 258.8 2. South Korea ............ 133.1 3. China .......................... 93.8 4. Philippines ................ 89.3 5. Taiwan ........................ 79.0 6. Thailand ..................... 55.5 7. Brazil ........................... 49.3 8. Hong Kong ................. 43.8 9. U.K ............................... 42.7 10. Iran ............................. 37.2 11. Malaysia .................... 31.0 12. Sigapore ................... 30.5 13. Australia .................... 25.8 14. Germany ................... 25.7 15. France ....................... 18.7 16. Canada ..................... 16.6 17. Indonesia .................. 16.3 18. Peru ........................... 16.1 19. Pakistan .................... 15.0 20. Italy .............................. 8.9 Other ....................... 196.4 Percent of Outgoing Traffic 20.2% 10.4% 7.3% 7.0% 6.2% 4.3% 3.8% 3.4% 3.3% 2.9% [] 2.4% [] 2.4% [] 2.0% [] 2.0% ~1 1.5% [] 1.3% [] 1.3% 1.3% 1.2% 0.7% 15.3% *Mi’l-r Incoming Outgoing Surplus(Deficit) Total Volume Note: FY ends on 31F3, FY 1990/91 747.0 937.0 (190.0) 1684.0 FY 1991/92 836.8 1160.5 (323.7) 1997.3 FY 1992/93 891.5 1283.3 (391.8) 2174.8 "Mi’l-r is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits and have been rounded to the nearest million. ©TeleGeography, Inc. 1993 129- Table 51 South Korea Largest Telecommunication Routes (1992) Destination MiTT 1. U.S.A ......................... 93.5 2. Japan ......................... 80.1 3. Hong Kong ................ 14.6 4. Philippines ................. 12.6 5. China ......................... 10.8 6. Germany ...................... 7.5 7. Taiwan ......................... 7.3 8. U.K ................................ 6.6 9. Australia ...................... 5.7 10. Singapore .................. 5.7 11. Indonesia .................... 5.5 12. Canada ....................... 5.1 13. France ......................... 4.3 Percent of Outgoing Traffic -_----:--~ 30.6% j 25.2O/o 4.8% 4.1% 3.5% ~-~ 2.5% 2.4% ~’~ 2.2% 1.9% 1.9% 1.8% 1.7% 1.4% Other .......................... 46.6 5.2% National Traffic Balance *MITT 1 990 Incoming 350.2 Outgoing 188.1 Surplus(Deficit) 162.1 Total Volume 538.3 Note: 1990 and 1991 figures are for Korea Telecom only. 1991 425.1 229.2 195.9 654.3 1992 N.A. 305.9 N.A. N.A. "Mi1-r is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 130 Table 52 IVlalaysia Largest Telecommunication Routes (1992) Destination MiTT 1. Singapore ................ 97.3 2, Japan ........................ 16.3 3. U.S.A ........................ 12,7 4. Australia ................... 12.0 5. U.K ............................ 11.2 6. Hong Kong .............. 10.4 7. Taiwan ........................ 9.2 8. Thailand ..................... 7.4 9. Indonesia ................... 6.0 10. India ............................ 4.5 11. Canada ...................... 2.2 Percent of Outgoing Traffic 44.9% I 7.5% I 5.9% I 5.5% I 5.2% I 4.8% I 4.2% I 3.4% 2.8% 2.1% 1.0% Other ........................... 9.8 4,5% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 163.4 140.9 22.5 304.3 1991 N.A. 176.2 N.A. N.A. 1992 249.0 216.5 32.5 465.5 Note: Above totals do not include local Malaysia-Singapore border traffic. Incoming totals for 1990 and 1992 are estimated to the nearest million MITT. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 131 Table 53 New Zealand Largest Telecommunication Routes (FY 1992/93) Destination MiTr 1. Australia .................... 72.0 2. U.S.A ......................... 16.0 3. U.K ............................. 16.0 4. Fiji ................................ 4.0 5. Hong Kong ................. 4.0 6. Japan .......................... 4.0 7. Canada ..................... 3.0 8. Singapore .................. 3.0 9. Taiwan ........................ 2.0 10. Germany .................... 2.0 11. West Samoa ............ 1.0 12. Malaysia .................... 1.0 13. Netherlands .............. 1.0 14. France ........................ 1.0 15. Switzerland ............... 1.0 Percent of Outgoing Traffic 48.3% 10.7% 10.7% 2.7% 2,7o/o 2.7% 2.0% 2.0% 1.3% 1.3% 0.7% 0.7% 0.7% 0.7% 0.7% Other ......................... 18.0 12.1% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume FY 1990/91 N.A. N.A. N.A. N.A. FY 1991/92 N.A. N.A. N.A. N.A. FY 1992/93 N.A. 149.0 N.A. N.A. Note: Total outgoing traffic has been estimated to the nearest million MITT. Fiscal year ends on 31/3. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 132 Table 54 Philippines Largest Telecommunication Routes (1992) Destination Mi~r 1. U.S.A ......................... 53.5 2. Japan ........................ 21.1 3. Hong Kong ............... 11.8 4. Canada ....................... 7.3 5. Australia ...................... 6.0 Percent of Outgoing Traffic 15.5% 8.7% 5.4% 4.4% 6. Singapore ................... 4.4[] 3.2% 7. Taiwan ......................... 3.9 [] 2.9% 8. South Korea ............... 3.3 [] 2.4% 9. Saudi Arabia .............. 3.0 I 2.2% 10. U.K .............................. 2.5~ 1"8% 11. Italy ............................. 2.3 1.7% 12. Guam .......................... 2.1 1.5% 13. Malaysia .................... 1.8 ~1 1.3% 14. Germany .................... 1.3 1.°% 39.4% Other ......................... 11.6~ 8.5% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 255.4 100.7 154.7 356.1 1991 385.4 128.5 256.9 513.9 1992 462.1 135.8 326.3 597.9 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 133 Table 55 Singapore Largest Telecommunication Routes (FY 1992/93) Destination MiTT 1. Malaysia .................. 110.0 2. Indonesia ................... 40.0 3. U.S.A .......................... 35.0 4. Hong Kong ................ 34.0 5. Japan ......................... 34.0 6. U.K .............................. 22.0 7. Thailand ..................... 19.0 8. Australia ..................... 17.0 9. India ............................ 14.0 10. Philippines ............... 12.0 11. China ......................... 10.0 12. Germany ..................... 9.0 13. France ......................... 8.0 Percent of Outgoing Traffic ~!~i~~~1 26.7o/o 9.7% ~~’:’:’, 8.5% 8.3% 8.3% 5.3% 4.6% 4.1% 3.4% 2.9% [] 2.4% 2.2% 1.9% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume FY 1990/91 N.A. N.A. N.A. N.A. FY 1991/92 N.A. 344.0 N.A. N.A. FY 1992/93 NA. 412.0 NA. N.A. Note: Data are estimated to the nearest million MITT. Fiscal year ends on 31/3. Above totals do not include local Malaysia-Singapore border traffic. °Mi’FI" is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits @TeleGeography, Inc. 1993 134 Table 56 Sri Lanka Largest Telecommunication Routes (1991) Destination Mi’l-r Percent of Outgoing Traffic 1. India ............................ 1.6 I I 10.3%o 2. U.K .............................. 1.6 I I 1°.3°/0 3, Singapore ................. 1.5 9.6% 4. Japan ......................... 1.0 6-4% 5. U.S.A .......................... 1.0 6"4% 6. Hong Kong ................ 0.9 5.8% 7. Germany .................... 0.8 5-1% 8. U.A.E .......................... 0.6 3.8% 9. Australia .................... 0.5 [] 3.2% 10. Canada .................... 0.4 [] 2.6% Other ......................... 5.7I 36.5% National Traffic Balance *MITT 1990 Incoming N.A. Outgoing 9.0 Surplus(Deficit) N.A. Total Volume N.A. 1991 N.A. 15.0 N.A. N.A. 1992 N.A. N.A. N.A. N.A. "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 135 Table 57 Taiwan (Republic of China) Largest Telecommunication Routes (FY 1992/92) Destination Mil-r Percent of Outgoing Traffic 1. U.S.A ......................... 82.2 22.3% 2. Hong Kong ............... 64.3 3. Japan ......................... 55.1 4. China (P.R.C.) .......... 52.9 5. Singapore ................. 13.4 6. Thailand ...................... 9.0 7. Canada ....................... 8.8 17.5% 14.9% 14.3% 3.6% 2.4% [] 2.4% 8. Philippines .................. 8.5 [] 2.3% 9. Malaysia ..................... 8.3 2-2% 10. Australia ..................... 6.8 11. Germany .................... 6.5 12. Indonesia ................... 6.0 I 1.6% 13. South Korea .............. 6.0 [] 1.6% 14. U.K ............................... 5.5 [] 1.5% 15. France ......................... 3.9 1.o% Other .......................... 31.4 8.5% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume FY 1989/90 301.0 212.0 89.0 513.0 FY 1990/91 415 254.7 160.3 669.7 Note: Fiscal year 1989/90 ends on 30/6/1990. Later fiscal years end on 31/3. FY "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 1992/93 484.4 368.7 115.7 853.1 136 Table 58 Thailand Largest Telecommunication Routes (1992) Destination MiTT 1. Japan ........................ 23.1 2. U.S.A ......................... 20.5 3. Singapore ................ 13.4 4. Hong Kong ............... 11.7 5. Taiwan ........................ 9.2 6. U.K ............................... 6.6 7. Australia ...................... 5.2 8. Germany ..................... 4.6 9. China .......................... 3.9 10. France ....................... 3.5 11. Korea ........................ 3.0 12. Italy ............................ 2.4 13. India .......................... 2.3 14. Canada .................... 1.8 15. Switzerland ............. 1.7 Percent of Outgoing Traffic 17.4% 15.5% 10.1% 8.8% 6.9% 5.0% 3.9% 3.5% 2.9% 2.6% 2.3% ~--~ 1.8% 1.7% N 1.4% N 1.3% Other ....................... 19.7 14.9% National Traffic Balance *MITT Incoming Outgoing Surplus(Deficit) Total Volume 1990 N.A. N.A. N.A. N.A. 1991 N.A. 110.5 N.A. N.A. 1992 201.6 132.4 69.2 334.0 "MITT is Minutes of Telecommunications Traffic Data are in millions of minutes for public voice circuits ©TeleGeography, Inc. 1993 137 The traffic statistics in TeleGeography 1993 were compiled primarily from an independent survey of telecommunications service providers. In a few cases (Australia, Singapore, Sri Lanka, Sweden), traffic data have been estimated based upon annual reports, government publications and industry interviews. The following publications were also consulted in preparing this report: Asia-Pacific Telecommunication Indicators (ITU, Geneva, 1993); European Telecommunication Indicators (ITU, Geneva, 1992); Arab States Telecommunication Indicators (ITU, Geneva 1992), Internationale Fernmeldestatistik (Siemens, Munich, 1993); The World’s Telephones A Statistical Compilation as of January 1991-1992, (AT&T, Indianapolis, IN, 1993); and The World’s Telephones A Statistical Completion as of January 1990 (AT&T, Indianapolis, IN, 1992). A common accounting unit known as MiTT -- Minutes of Telecommunications Traffic -- is used throughout the report for traffic on public switched voice circuits. MiTT includes operator assisted and credit card calls; it may also include voice as well as non-voice (facsimile, data) traffic. For a discussion of the origins of MiTT and its various applications (eg., economic forecasting, competition policy, geography), see G. Staple and M. Mullins "Telecom Traffic Statistics MiTT Matter", Telecommunications Policy, Vol. 14, No. 2, June 1989, pp. 105-128. See also Communications Outlook 1993 (OECD, Paris, 1993) for a comparison of MiTT with other telecommunication indicators. The outbound MiTT volumes reported for countries in TeleGeography 1993 generally reflect outbound traffic metered by national carriers regardless of whether the traffic was billed domestically or to a foreign location (e.g., to a foreign calling card or credit card.) However, the outbound MiTT data for the U.S. (in 1991 and 1992) and the Netherlands (in 1992) are based on the billing point of the call. Thus, for example, a "Country Direct" call from Thailand to the U.S., billed to a U.S. calling card or credit card, would be counted as outbound (not inbound) U.S. MITT; the same call also would be included as outbound MiTT by Thailand. Consequently, in some cases, the incoming traffic data reported in the U.S. MiTT table is not directly comparable with the outgoing MiTT to the U.S. reported in the tables for other countries. 138- There may also be other reasons why the inbound traffic on a given route differs from the outbound traffic stated for the originating country (e.g., calendar vs. fiscal year data). In any event, the route-by-route traffic data reported in TeleGeography for each country generally is based upon the survey data supplied to TeleGeography by the originating country not the terminating country. Calendar year data have been used wherever possible; fiscal year data are used elsewhere (e.g., the U.K., Ireland, Japan, Taiwan). Carrier traffic statistics do not include traffic from foreign subsidiaries or investment interests, unless stated. Some differences exist between the historical statistics (i.e., 1988 to 1991) reported in TeleGeography 1993 and data stated in prior reports. The variations reflect corrections and/or revised data subsequently provided to TeleGeography. 139 TeleGeography 1993 140 TELEGEOGRAPHY 1993 BLUE PAGES TELEGEOGRAPHY 1993 BLUE PAGES Table l(b) NATIONAL TELECOMMUNICATION INDICATORS (1992) (Data at 111/93 except as indicated) Telephone Lines Mobile Telephone Fax Machines Intemet Host Country (Milliom) Subscribers (Thousands) (Thousands) Computers Australia 8.3 440.0 450.0 (1991) 346 Brazil 10.7 30.7 150.0 68 Czech and Slovak Republics 2.6 5.0 29.4 106 Hong Kong 2.8 202.0 (1991) 172.2 9 ~dones~ 1.3 (1991) 24.6 (1991) 25.0 0 Luxembourg 0.2 1.1 5.0 4 Pakistan 1.1 (1991) 8.5 (1991) 4.1 (1991) 0 Po~gal 3.2 37.3 27.8 35 Sri ~a .1 (1991) 2.0 (1991) 5.9 (1991) 0 United Arab Emka~s 0.5 48.9 23.9 0 Venezuela 1.8 7.4 30.0 (19~) 6 Sources: Columns 2, 3 and 4, ITU; Column 5, Interact Society Figures in Columns 3 and 4 are approximate only. © TeleGeography, Inc. 1993 Table l(a) NATIONAL TELECOMMUNICATION INDICATORS (1992) (Data at 1/1/93 except as indicated) Coumry Population (milliom) Per Capita GNP Area Telephone Lines ($ US) (000 sq. km) per 100 people Australia 17.5 17,050 7,687 47.43 Bangladesh 119.3 220 144 0.25 Brazil 156.3 2,940 8,512 6.85 China (PRC) 1155.8 (1991) 370 9,561 .99 Czech and Slovak Republics 15.7 2,470 128 16.56 Germany 80.6 23,650 357 43.92 Iceland 0.3 24,960 103 33.33 Indonesia 191.2 610 1,905 0.68 Japan 123.8 26,930 378 46.43 Luxembourg 0.4 25,963 1 50 New Zealand 3.4 12,350 269 44.12 Philippines 64.3 730 300 1.09 Portugal 9.9 5,930 92 32.32 Saudi Arabia 15.9 7,820 2,150 10.06 South Africa 39.8 2,560 1,221 8.79 Sri Lanka 17.4 500 66 .57 Switzerland 6.8 33,610 41 61.76 Thailand 57.8 1,570 513 3.11 United Arab Emirates 1.7 19,870 (1990) 84 29.41 United States 255.0 22,240 9,373 56.51 Venezuela 20.7 2,730 912 8.70 Sources: Columns 2 and 5, ITU; Columns 3 and 4, World Bank Development Report © TeleGeography, Inc. 1993 Country Table l(c) NATIONAL TELECOMMUNICATION INDICATORS (1992) Outgoing Outgoing Balance of MiTT MiTT Outgoing MiTT Per (deficit)/surplus (Millions) MiTT Per Capita Access Line (millions) Australia 659.0 37.7 79.4 N/A Bangladesh 14.3 0.1 47.7 52.5 Brazil 169.9 1.1 15.9 160.7 Canada 2246.5 82.0 138.7 429.8 China (PRC) 635.1 0.5 55.2 N/A Czech and Slovak Republics 178.0 11.3 68.5 58.0 Finland 235.0 47.0 87.0 N/A Germany 4087.0 50.7 115.5 (987.0) Hong Kong 1136.6 196.0 405.9 (127.2) Iceland 22.1 73.7 221.0 (0.4) Indonesia 118.1 0.6 90.8 47.4 Israel 153.1 29.4 90.1 112.9 :: :::::: :...: :: ::::::::::::::::::::::::::::::: ::::: :::::: ::::::::::::::::::::::::::::::: :::: ::::::: :::::::::: :::::::::::: :::::::::::::::::::::::::::::::::::::::: :: :::::::: :::::::::::::::::::::::::::::::::::::::::::: ::::::: :::::::::::::::: :::::: ::::: :: Japan 1283.0 10.5 22.5 (391.8) Luxembourg 181.0 452.5 905.0 (73.5) Mexico 683.5 7.6 100.5 431.5 ~i~i===============================================================================================ii~:::::::::::::::::::::::::::::::::::::::::::::::ii~ii~ii~i~i~iii~i~iiii~!~i~i~i~I..~:::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: New Zealand 149.0 43.8 99.3 NIA .... ......................................................................... ...................................... ............................................... ........................................ :: :: :: :: :: :: ~::~::::::::::::::::::::~:::::::::~:~::::::~::~:::~:~::::::~:~:~:::::~::~:::::~::::~:::~:::::::::::::~:::~:::~:::::::::::~:::::::~:::::::~:::~::::::::::~::::::::::~:::::::::::::~::~:::~::~:~:~:~ ~::~ ~:~:~:~:~:::~:::::::::::::::::::::::~:::::::::::::~::~::::~::~:::~::::::::~::::::::::~:~:~:~::::::~:~:~:~:~: i~ :: :: :: :: :: :: :: :: :: :: :: :: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::: Philippines 135.8 2.1 194.0 326.3 :: ::::::~.l.~ii~! ::::::::::::::::::::::::::::::::::::::::: ::i:::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: Portugal 212.0 21.4 66.3 NIA Saudi Arabia 464.6 29.2 290.4 (172.5) South Africa 221.7 5.6 63.3 29.0 Sri Lanka 15.5 0.9 155.0 NIA Thailand 132.4 2.3 73.6 69.2 United Arab Emirates 299.0 175.9 598.0 N/A "" ÷ ...... ~+:’ ::: ...... iiiiiiiiiiiiiii’" "i "iiiiiiii!!il i~ ....... ’’" ..................... :’:’:’:’:’: ....................................... United States 10173.6 39.9 70.6 (4904.3) Venezuela 115.5 5.6 64.2 13.1 Source: TeleGeography 1993 © TeleGeography, Inc. 1993 Code 1 1 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 1-809 20 212 213 216 218 22o 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 240 241 242 243 244 245 247 248 249 250 251 252 INTERNATIONAL TELEPHONE DIALING CODES Canada United States Anguilla Antigua & Barbuda Bahamas Barbados Bermuda British Virgin hlands Carriacou Cayman Islands Dominca Dominican Republic Grenada Jamaica Montaerrat Nevis Islands Puerto Rico St. Kitts St. Lucia St. Vincent & the Grenadines Trinidad & Tobago Turks & Caicos Islands Egypt Morocco Algeria Tunisia Libya Gambia Senegal Mauritania Mall Guinea Ivory Coast Burkina Faso Niger Togo Benin Mauritius Liberia Sierra Leone Ghana Nigeria Chad Central African Republic Cameroon Cape Verde Islands Equatorial Guinea Gabon Congo Zaire Angola Guinea-Bissau Ascension Island Seychelles Sudan Rwanda Ethiopia Somalia Capital (city code) Ouawa (613) Washington, D.C. (202) Nassau Bridgetown Santo Domingo Kingston San Juan Port of Spain Cairo (2) Rabat (7) Algiers (2) Tunis (1) Tripoli (21) Banjui Dakar Nouakchott Bamako Conakry (4) Yamoussoukro Otlagadougou Niamey Porto Novo Port Louis Monrovia Freetown (22) Accra (21) Abuja N’Djamena Bangui (61) Yaoonde Praia Bata Libreville Brazzaville (81, 82, 83) Kinshasa (12) Luanda (2) Bissau Khartoum (11) KJgali Addis Ababa (1) Mogadishu (1) Code 253 254 255 256 257 258 260 261 262 263 264 265 266 267 268 27 297 298 299 3o 31 32 33 33 33 34 350 351 352 353 354 355 356 357 358 359 36 370 371 372 373 385 386 39 41 41 42 42 43 45 46 47 48 49 501 502 Country DJibouti Kenya Tanzania Uganda Bumndi Mozambique Zambia Madagascar Reunion Island Zimbabwe Namibia Malawi Lesotho Botswana Swaziland South Africa Aruba Faeroe Islands Greenland Netherlands Belgium Andorra France Monaco Spain Gibraltar Portugal, Azores & Madeira hla Luxembourg Ireland Iceland Albania Cyprus Finland Bulgaria Hungary Lithuania Latvia Estonia Moldova Croatia Slovcnia Italy Romania Licchtenstein Switzerland Czech Republic Slovakia Austria Great Britain Denmark Sweden Norway Poland Germany Belize Guatemala Capital (city code) Djibouti Nairobi (2) Dar Es Salaam (51) Kampala (41) Bujumbura Maputo Lusaka (1) Antananarivo (2) Harare (4) Windhoek (61) Liiongwe Maseru Gaborone (31) Mbabane Pretoria (12) all Iocatiom (8) Athens (1) Amsterdam (20) Brussels (2) all locations (628) Paris (1) all locations (93) Madrid (1) Lisbon (1) Dublin (1) Reykjavik (1) Tirana (42) Valletta Nicosia (2) Helsinki (0) Sofia (2) Budapest (I) Vilinins (2) Riga (2) Tallinn (2) Chisinau Zagreb (41) Ljubljana (61) Rome (6) Bucharest (0) all locations (75) Bern (31) Prague (2) Bratislava (7) Vienna (1) London (71.81) Copenhagen Stockholm (8) Oslo (2) Warsaw (22) Bonn (228) Punta Gurda (7) Guatemala City (2) © TeleGeography, Inc. 1993 Code 503 5O4 5O5 506 507 5o8 509 51 52 53 54 55 56 57 58 590 591 592 593 594 595 596 597 598 599 61 62 63 65 66 670 671 673 674 675 676 679 684 685 686 687 689 689 689 INTERNATIONAL TELEPHONE DIALING CODES Country El Salvador Honduras Nicaragua Costa Rica Panama St. Pierre & Miquelon Haiti Peru Mexico Cuba Argentina Brazil Chile Colombia Venezuela Guadeloupe Bolivia Guyana Ecuador French Guiana Paraguay French Antilles Surinam Uruguay Netherlands Antilles Malaysia Australia Indonesia Philippines New Zealand & Chatham Islands Singapore Thailand Saipan, Rota & Tinian Guam Brunei Nauru Papua New Guinea Tonga Islands Fiji American Samoa Western Samoa Kiribati New Caledonia French Polynesia Moorea Tahiti Capital (city code) San Salvador Tegucigalpa Managua (2) San Jose Panama Port au Prince (1) Lima (14) Mexico City (5) Havana (7) Buenos Aires (1) Brazilia (61) Santiago (2) Bogota (1) Caracas (2) La Paz (2) Georgetown (2) Quito (2) Cayenne Asuncion (21) Paramibo Montevideo (2) Kuala Lumpur (3) Canberra (62) Jakar~ (21) Manila (2) Wellington (4) Bangkok(:7.) Bandar Seri Begawan Port Moresby Code 691 692 7 7 7 7 7 7 7 7 7 7 7 81 82 850 852 853 855 856 86 88O 886 91 93 95 96O 961 962 963 964 965 966 967 968 971 972 973 974 975 976 977 98 Country Micronesia Marshall Island Belarus Ukraine Kyrgyzstan Russia Turkmenismn Uzbekistan Tajikistan Georgia Armenia Azerbaijan (future code: 994) Japan & Okinawa Republic of Korea Vietnam North Korea Hong Kong Macao Carabodia Laos China, People’s Republic of Bangladesh Taiwan Turkey India Pakistan Afghanistan Sri Lanka Myanmar (Burma) Maldives Lebanon Jordan Syria Kuwait Saudi Arabia Yemen United Arab Emirates Israel Bahrain Bhutan Mongolia Nepal Capital (city code) Minsk (0172) Kiev (044) Alma Ata (3272) Bishkek 0312) Moscow (095) Ashkhabad (363) Tashkent (3712) Dushanbe (3772) Tbilisi (8832) Yerevan (8852) Balm (8922) Tokyo (3) Seoul (2) Hanoi (4) Pyongyang (5) Phnom-Penh Vientiane Seijing (1) Dhaka (2) Taipei (2) Ankara (4) New Delhi (11) Islamabad (51) Kabul Colombo (1) Rangoon Beirut (1) Amman (6) Damascus (11) Baghdad (1) Riyadh (I) Sarm (2) Muscat Dubal (4) Jerusalem (2) AI Manama Doha Thimphu Ulan Bator Kathmandu Tehran (21) Note: No city code is required for capital cities unless indicated. © TeleGeography, Inc. 1993 Code 201 202 203 204 205 206 207 2o8 209 210 212 213 214 215 216 217 218 219 3ol 303 3o5 308 31o 312 313 314 315 316 317 318 319 4Ol 405 408 41o 412 413 414 415 416 417 418 419 501 502 503 505 5o6 508 510 512 513 514 515 516 517 518 State New Jersey, Northern District of Columbia Con~ec~cot Mmitoba (Canada) W~¢hln~ West (Sea.e) Maine Idaho Texas, Somh Cantral New York City Texas, Ce~ral Pennsylvania, Eastern Ohio, Northeastern Minneu~, Nortt~-m Indi~n~, Northern Maryland, Northern Delaware Colorado, Northern West Virginia Sas~tchewan wym~ N~, We~-m Illinois, Northeast Michigan, Eas~rn Missouri, Eastern New York, Northern Lou~iana. Western iowa, Eastem Nebraska, Eastern Alberta (Canada) Oklahoma, Somhwe~’rn Florida, Ea~em Te~, S~uthea~n Maryland, Southern Pennsylvania, Smthwe~em Ma.~qc.lmsem, Western Wisc~min, Eastem California, Northern Ontario, So~r.astern (Canada) Mmousi, Southern Quebec, Nor~em (Canada) Ohio, Nor~we~em Arkan~. Kentucky, Western Oregon New Mexico New Br~swick (Canada) Mume~a, Sonthem Washin~on, East (Spohne) California, Not,~hem Texas, South Cemral Ohio, Sonthwestem Quet~ec, Southern (Canada) Iowa, Cenlral New York, Northeastern Michigan, Central New York, Northeastern NORTH AMERICAN AREA CODES c~ Newark Wa~hln~ all all all Sea~e all all Fre~o San Ammio Ma~nhsnsn Los Angeles Dall~ Cleveland Gary Rockville Miami North Plaue Peoria Syra~me Wichita A~ Oklahoma City We~ Palm Beach Galvemm Baltm~xe Springfteld Milwaukee San Franci.~o Toledo Louisville New Orleans Roches~r Worcester Spokane Oakland Austin Cin~mmti Monlreal Des Moines Hempstead Lansing Albany Code 519 6Ol 6o3 612 613 614 615 616 617 618 619 7o2 7o3 7o5 7o7 7o8 7o9 712 713 714 715 ?16 719 8o3 ~o7 8o8 812 814 815 816 819 9Ol 9o3 9o8 91o 912 913 914 916 917 918 919 State O~ar~, Smthern (Canada) Mmmippi Aroma New Hampshire British Columbia (C~n~b) Nozlhwest Temmry South Dakota Kentucky, Ea~rn New York, Sonthem Wisconsin, Son,hem New Jersey, Sonthern Mim~eu~a, Ccotr~ Omario, Sonthe~t~m (Canada) Ohio, Sonltz~t~m Temessee, Easmm Mich~m, Western Caliomm, Sou~em & Central North Dak~a Nevada Virginia, Norll~m North Carolina, Western Ommo, No~wats~m (Canada) Georgia, Nor~em Cafifornia, NoPd~’rn Newfona~and (Omada) Iowa, Western Texas, $ontheastem California, Sontlm~ Wisc~min. Northern New York, Wesmm Penmylvania, Cenl~l New York City Utah Vermol~ South Carolina Virg~a, Sonthem Texas, Northern Ommo, Northwesmm (Canada) Hawaii Puerto Rico Pera-.sylvania, Central Illinois, Northern Missouri, Northern Texas, Central California, Somhem Quebec, Southern (Canada) Tennessee, Western Nova Scaxia (Caaada) Prince Edward Island (Canada) Texas, Eastern Florida, Northern Michigan, Northern New J~rsey, Central Caltorrua, Central North Carolina, C~ntral Georgia, Sou~em Kansas, Northern New York, Sontheastem Texas, Western California, Nord~em New York City Oklahoma, Northeast North Carolina, Eas~:rn © TeleGeography, Inc. 1993 TeleGeography 1993 Blue Pages 307 TeleGeography, Inc., Suite ZOO0 1150 Connecticut Avenue, N. W., Washington D.C. 200 J6 Tel. 202-467-0017; lax 202-467-5915 Hawaii 8O8 ~New Orleans $O4 Dallas/Ft. Worth 214 ~Miami 3O5 ~lewfoundland Boston 617 $08 City 212 & S.I. 718 302 201 ~410 D.C. 202 Telephone Numbering Zones of the World Nassa~J 502 city (2) Managua (2) 5O9 1-809 Port Au Prince Caribbean (1) [Except as stated] 5~0 Guadelupe ’~,/7 596 [Martinique, St. Barth, St. Martin] 297 Arub~a(8) etown 597 593 Quito (2) 56 Santiagc (2) 598 Montevideo (2) Reykjavik (1) Dublin London (71,81 Rabat 220 24S 22~ (4) Freetown (22) (21 ) 228 (81,82,83) (2) (61) (6) 7-3272 hts Reserved. KEY FOR EUROPE INSET NuKn~e~er Co..try Tel~oPdhe°"e C~i~;al cCoi~l/e Czech Republic 42 Prague 2 Slovakia 42 Bratislava 7 Moldova ~173 Kishini~v 2 Slovenia 386 Ljubljana 61 Croatia 385 Zagreb 41 Bosnia/Herzegovina 387 Sarajevo 71 Serbia & Montenegro 381 Belgrade 11 Macedonia 389 Chisinau Malta 356 Valletta Cyprus 351 Nicosia 2 7-3312 Ulan~Sator 81 (3) 252 Mogadishu (1) Nairobi )ar Es Salaam (Sl) New Delhi (11) Bangkok (2) Phnom-Penh 85S (~) 94 Kuala Lure (3) 62 Hanoi (2) ° ’23o (31) Note: No city code is required for listed cities unless stated. Pc Moresby International San Santiago de St.John’s ~aulo ATLANTIC OCEAN BAHAMAS ATLANTIC OCEAN ABCTIC SEYCH£LLES INDIAN MAIJRITIUS R[UNION OCEAN PACIFIC OCEAN NEW ZEALAND ./ EUROPE 1. NETHERLANDS 2. BELGIUM 3. LUXEMBOURG 4. CZECH REPUBLIC 5. SLOVAKIA 6. SWITZERLAND 7. LIECHTENSTEIN 8. AUSTRIA 9. HUNGARY 10. SLOVENIA 11. CROATIA 12. BOSNIA 13. YUGOSLAVIA 14. ALBANIA 15. MACEDONIA 16. MOLDOVA ASIA 17. GEORGIA 18. ARMENIA 19. AZERBAIJAN 20. TURKMENISTAN 21. UZBEKISTAN 22. TAJIKISTAN 23. KYRGYZSTAN AFRICA 24. BURKINA FASO 25. TOGO 26. EQUATORIAL GUINEA TI IIS REPORT IS PUBLISHED ANNUALLY BY TELEGEOGRAPHY, INC., WASHINGTON, D.C. IN (’OOPE!~ATION WITH THE INTERNATIONAL INSTITUTE OF COMMUNICATIONS (IIC), LONDON