TeleGeography 1994 Global Telecommunications Traffic Statistics & Commentary Gregory C. Staple, Editor This report is based upon sources believed to be reliable, but the pub- lisher does not warrant the accuracy or completeness of any informa- tion for any purpose and is not responsible for any errors or omissions. This report is for the confidential use of subscribers. 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, pho- tocopying, recording or otherwise, without written consent from the pub- fisher. All rights reserved. Copyright © TeleGeography, Inc. 1994 First Printing, October 1994 Second Printing, December 1994 z’~ This is copy x,~ Do not reproduce. ISBN 1-886142-01-7 Printed in the United States of America (~LOBAL TELECOMMUNICATIONS HAS NOT LED TO THE END OF GEOGRAPHY SO MUCH AS THE EXPLOSION OF PLACE. The Front Cover: Digital Planets Back Page (p. 170): E-Mail Comet Cover Illustration: Annemarie Feld, Gregory C. Staple, Jessica A. Perlman, Zachary Schrag © TeleGeography, Inc. 1994 All rights reserved. telegeography\t~l’a-j~-~g’ra-f~ n (1990) abbrv, of telecommunications geography [fr. Gk tele, far off, at a dis- tance and L. communicatus, pp. of communicare to impart fr. Gk geo (earth) 9raphein, (to write)] 1. a new branch of geography that maps the global pattern of telephone traffic and other electronic communication flows; 2. places created by or perceived solely via telecommunications (e.g., a com- puter network address); 3. the telecommunications artifacts (radio antennae, terminals, signs) on a site; 4. the balance of telecommunications power in one country or region vis-a-vis another (cf. geopolitics, archaic). © TeleGeography, Inc. 1994. TeleGeography 1994 Contents Preface ............................................................. vii Acknowledgments ...................................................... viii Introduction .......................................................... ix A~icles Elliot M. Gold The Market for Picture Phones: Prospects for Sales and Traffic .... 3 Gregory C. Staple International Communications: The Challenge of Convergence .... 1 l Graham Finnie Competition in Europe .............................. 38 Stuart Corner New Carriers in Asia ................................ 39 Zachary M. Schrag Navigating Cyberspace: Maps and Agents ................. 64 Maps Cartograms: Teledensity vs. Population ................... 55 Tariff maps (USA, U.K., Germany, Japan) .................. 56 Intercontinental Telecommunications Traffic Flows ............ 58 European Telecommunications Traffic Flows ................ 59 East Asian Telecommunications Traffic Flows ................ 60 South American Telecommunications Traffic Flows ............ 61 Traffic Balances .................................. 62 Major Submarine Cables ............................. 66 Global Phone Sex Traffic ............................. 66 Tables Intercontinental Transmission Facilities {Cables and Satellites) .... 69 The Top 40 Carriers ................................ 71 Competing International Carriers ....................... 72 Stock Market Trends and Capitalization ................... 74 Internet Facilities and Traffic ........................... 76 The Top 50 International Routes ........................ 80 Outgoing International Traffic, by country Americas ............................ 81 Europe ............................. 97 Asia, Middle East and Africa ............. 25 Methodology and sources ........................... 49 Blue pages International Telephone Dialing Codes ................... 153 North American Area Codes .......................... 156 World Dialing Codes Map ........................... 158 World Time Zones ................................ 60 Political Geography ................................ 62 National Telecommunications Indicators ................. 166 The E-mail Comet ................................. 170 Back Page 111 TeleGeography 1994 Traffic Tables Americas Argentina ...................... 83 Brazil ........................ 84 Canada ....................... 85 Chile ......................... 86 Colombia ...................... 87 Cuba ......................... 88 Ecuador ...................... 89 Mexico ....................... 90 Peru ......................... 9 United States, outgoing ............ 92 United States, incoming ............ United States, other correspondents .94 Uruguay ...................... 95 Venezuela ..................... 96 Europe Austria ....................... 98 Belgium ....................... 99 Bulgaria ...................... 1 O0 Croatia ...................... 101 Czech and Slovak Republics ........ 102 Denmark ..................... 103 Finland ...................... 104 France ....................... 105 Germany ..................... 106 Greece ...................... 107 Hungary ..................... 108 Iceland ...................... 109 Ireland ...................... 110 Italy ........................ 111 Luxembourg ................... 112 Netherlands ................... 113 Norway ...................... 114 Poland ...................... 115 Portugal ..................... 116 Russia ....................... 117 Slovenia ..................... 118 Spain ....................... 119 Sweden ...................... 120 Switzerland ................... 121 United Kingdom ................ 122 Yugoslavia .................... 123 Traffic within the Commonwealth of Independent States .............. 124 Asia, Middle East, and Africa Australia ..................... 126 Bangladesh ................... 127 China ....................... 128 Cyprus ...................... 129 Hong Kong .................... 130 India ........................ 131 Indonesia ..................... 132 Israel ....................... 133 Japan ....................... 134 Korea, Republic of .............. 135 Macau ...................... 136 Malaysia ..................... 137 New Zealand .................. 138 Pakistan ..................... 139 Philippines .................... 140 Saudi Arabia .................. 141 Singapore .................... 142 South Africa ................... 143 Sri Lanka ..................... 144 Taiwan ....................... 145 Thailand ..................... 146 Turkey ....................... 147 United Arab Emirates ............ 148 TeleGeography 1994 Other Tables and Figuresl Global MITT: Growth and Origin ............................................. x Global MITT.- Growth by Country and Region .................................... xi Worldwide Picture Phone Equipment Sales (Units) ................................. 4 The Need for Bandwidth ................................................... 5 Videophone Traffic Volumes and Revenues, 1991-99 ............................... 6 International Telecom Services: A Paradigm Shift ................................. t 2 International Simple Resale (ISR)--1994 ...................................... 13 The Evolution of Network Intelligence ......................................... 16 Estimated Range for Standard Period International Telephone Service Accounting Rates in SDRs 17 Settlement Rates vs. Estimated National "Access Charges" Per Minute for Qutbound U.S. Switched Telephone Traffic ......................... 18 AT&T WorldPartner Association ............................................. 19 BT-MCI Joint Venture .................................................... 20 Unisource Alliance ...................................................... 21 Sprint-France Telecom-DBP Telekom Partnership .................................. 22 AT&T’s Software Dial Tone ................................................. 23 Who Will Carry North America’s International Traffic in 20017 ........................ 26 Trends in the Global Information Economy: Traffic, Revenues, Main Lines ................. 34 Peak Rate International Tariffs Between OECD Countries ............................ 35 World Bank Report on Telecom Sector Reform .................................. 41 eWorld Top Screen ...................................................... 45 Magic Cap Downtown ................................................... 48 The CompuServe Navigator ................................................ 50 The ImagiNation Network ................................................. 51 TeleGeography 1994 TeleGeography, Inc./lie This report was prepared by TeteGeography, Inc., Washington, D.C., an independent research and publishing company, in cooperation with the International Institute of Communications (llC), a London- based non-profit educational and conference organization with mem- bers in over 70 countries. The TeleGeography 1994 production team included: Gregory C. Staple ..................... Editor Zachary M. Schrag ............... Deputy Editor Barbara Frank ........... Administrative Assistant Annemarie Feld ............... Design Consultant The following reports also are available from TeleGeography, Inc.: TeleGeography 1993 (ISBN 886 t 42-00-9) Direction of Traffic: International Telephone Traffic, 1983-1992 (ISBN 92-61-04967-2). For more information, contact: TeleGeography, Inc. Suite 1000 1150 Connecticut Avenue, NW Washington, DC 20036 USA Tel.: +1 (202)467-0017 Fax: +1 (202) 467-5915 e-maih 4579195@mcimail.com For membership information and details on IIC publications, contact: The International Institute of Communications Tawstock House South Tavistock Square London WC1H 9LF, England, UK Tel.: +44 71 388 0671 Fax: +44 71 380 0623 Telex: 24578 IICLDN G TeleGeography 1994 Prefacel W ’elcome to the age of information superhighways, where there are networks of networks, and everything is becoming multimedia. Each year the pace of new telecommunications applications quickens, which makes tracking this industry so exciting. TeleGeography 1994 is an invaluable collection of facts and figures, graphs and charts, maps and tables that combine to give a clear picture of the worldwide telecommunications industry. Much of TeleGeography’s value is in the emerging trends it tracks. Today, the trend to follow most closely is the powerful convergence of telecommunications, computers, information services, and entertain- merit. Boundaries between these businesses are blurring and collaps- ing. Just as we once thought global telecommunications would make national boundaries meaningless, so it becomes increasingly difficult to tell where telecommunications leaves off, or where information ser- vices and entertainment begin. As the technology evolves, these industries are reaching beyond the narrow confines of their core busF nesses and creating entirely new businesses. TeleGeography 1994 is the road map that will carry us through these changes. By studying where the industry stands, we can gauge the direction in which it will evolve. Researchers, government agencies, regulators, and industry watchers have all come to rely on the data presented here. MCI is proud once again to underwrite this powerful information toot. Gerald H. Taylor President and Chief Operating Officer MCI Communications Corporation vii TeleGeography 1994 ( [Acknowledgments e are indebted to the numerous carriers, government departments, regulators and international organizations from around the world who responded to our requests for information. Special thanks are due as well to the dedi- cated people at these organizations and at the International Telecommunication 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 XlII, for underwriting our research on European telecommunications and for permitting us to publish certain Commission 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 publication grants from MCI Communications Corporation and Stentor Resource Centre, Inc. As in the past, however, these grants were made without any precondition; TeleGeography, Inc. is solely responsible for the report’s editorial con- tents. Gregory C. Staple Editor Vtll © TeleGeography, Inc. 1994 Introduction Introduction 1994 is a guide to people’s changing sense of place and identity in the world, and to some agents of those changes, international telecommuni- cation carriers. By monitoring traffic--the volume of telephone calls, faxes and e-mail between one coun- try and another--this report marks the shifting rela- tionships among the world’s peoples and the communica- tions businesses which serve them. In 1993 telephone subscribers (and their machines) spent approximately 47 billion minutes talking to other coun- tries. TeleGeography chronicles who was calling whom; how much France called Britain; Slovenia called Germany; India called Canada; the United States called Mexico. In all, there are tele- phone traffic statistics for over 60 coun- tries and 800 different routes. TeleGeography also contains national statistics on the growth of the lnternet, the world’s largest computer network. A companion volume of statistics and commentary, Direction of Traffic, pub- lished jointly with the International Telecommunication Union (tTU), con- tains route-by-route telephone traffic data for the prior ten years: 1983 to t992. its unique and temporary topography. They make every new telecommunications circuit not just a commercial mile- stone or engineering landmark but an organic part of our changing social and economic terrain. TRAFFIC FLOWS ARE MUCH MORE THAN TRADE STATISTICS. THEY CAN ALSO REVEAL THE MOVEMENTS OF PEOPLE~ AND SHOW WHERE THE NATURAL CHANNELS OF HUMAN COMMU- NICATIONS HAVE BEEN DAMMED AT THE BORDERS AND WHERE THEY FLOW FREELY, The rich array of data in Direction of Traffic and TeleGeography allows one to piece together the "who" and "where" behind buzzwords such as "globalization," the "information society" and "interdependence." The billions of minutes of international telecommunication traffic are themselves opaque; we do not know the content of the mes- sages, nor should we. But the new pattern of connections which they create--the telegeography--contains its own messages. Telecommunications geography is human geography. The flow of traffic between one country and another is much more than a trade statistic; it can also reveal movements of people--migrants, guest workers, tourists and refugees. Likewise, traffic statistics disclose where the natural channels of human commu- nications have been dammed at the borders and where they flow freely; where linguistic connections are strong and where they have thinned; which ethnic communities cohere and which have scattered across the electronic sea; which people of the world are rich in information and which have been left Telecommunications, like geography, is all about connec- tions. Just as islands, mountain passes, straits, and plateaus gain importance from their connection to, or isola- tion from, other places, so do telephones. But the geogra- phy of communications is far more fluid. The connections available between one terminal or network and another reflect the choices we make as a society about technology, the construction of telecommunication facilities, who may own them and how much it costs to use them. As impor- tantly, network connections largely rest on the uncertain ground of human conversation; the millions of words and texts, smiles and sobs, daily traversing the network create out. TeleGeography 1994 concentrates on the commercial aspects of this new geography. (Readers more interested in the social, linguistic and cultural dimen- s ions of the subject should consult Direction of Traffic as well and the sources cited therein.} The telecommu- ~[] nications data compiled here chart our modern trade routes--our global information highways and the changing fortunes of the carriers building them; their revenue base; market share and future prospects. As the industry’s traditional structures erode, TeleGeography offers a map to the new world. Slower growth and greater competition The last decade has been a period of unparalleled interna- tional traffic growth--15% annually from 1983 to 1992- spurred by international direct dialing IIDD), over 200 mil- lion new local subscriber lines, high capacity undersea cables and satellites, and new services, particularly facsimi- le. Traffic per line rose from an average of 30 minutes per year in 1983 to 71 minutes in 1992. However, since 1991, growth rates have fallen in almost all of the world’s regions. Global traffic growth declined from approximately 12% to 15% in 1992 to 11% in 1995 (47.4 billion minutes in 1993 vs. 42.8 billion minutes in 1992. See Figure 1.). Despite the general slowing of growth of international traffic since 199t, not every country and carrier fared the same. The differences point to significant shifts in power within the telecommunications industry. TeleGeography 1994 [] In 1993, Deutsche Telekom enjoyed traffic growth of 16.7%, despite a lingering recession (German GDP fell by 1.2% in the same year). Part of Telekom’s success is attrib- utable to the surge in telecommunications that has accom- panied the economic integration of Central and Eastern Europe. Notably, in !993 traffic from Poland, Hungary and the Czech Republic all grew over 15%--a plus for recent pri- vatizations (more on this below). Excluding Germany, inter- national traffic growth for members of the European Union (EU) was approximately 9%. In the U.K., for instance, out- bound traffic for the country’s two facilities-based carriers showed less than 5% growth on many routes, suggesting that alternative service providers--such as international pri- vate line (IPL) resellers, foreign based "call back" compa- nies, and Home Country Direct calling card services--are gaining a greater share of the pie. See the article beginning on p. 38 for further details on the new competition in Europe. [] Unlike Germany, Japan saw international traffic growth dampened by recession; outbound traffic rose only 9.9% in 1993 (it had averaged over 20% from 1985 to 1992) and grew 5.7% for the largest carrier, KDD. Yet, throughout the rest of Asia, growth rates of 15% to 20% per year remained the norm, as strong economies, emigration (to the West) and millions of new telephone lines translated into a contin- uing traffic boom. Outbound traffic from China grew over 40%, boosting the Chinese operator into the ranks of the top 15 carriers and adding new pressure for the government to license additional international gateways. See also Figure 2 and the article on Asia beginning on p. 39. [] Growth in public network traffic to and from the United States weakened in 1993; total traffic increased barely 10% over 1992. Furthermore, the volume of foreign billed (mainly incoming) traffic for U.S. carriers grew but 4.6% from 1992, suggesting that the 12.7% growth in U.S. billed (mostly outbound) traffic was mainly attributable to a change in the billing point of the call (e.g., from Home Country Direct calls). U.S.-originated outbound traffic probably grew at a 6-7% rate in 1993. The major benefi- ciaries of this more slowly growing U.S. market were com- peting facilities-based carriers; MCI, Sprint and IDB--cap- tured fully 75% of the additional 1.3 billion minutes of U.S. billed traffic added in 1993. MCI grew by 36% (adding three times the traffic of AT&T) to become the world’s third largest carrier, and Sprint grew by 23%. AT&T’s U.S.-billed traffic grew by only 4% and, for the first year since compe- tition began in the U.S., the volume of AT&T’s foreign billed traffic decreased from the prior year. [] Canada’s overseas carrier, Teleglobe, fared better than its U.S. counterparts, with total traffic growing 12.6%; for- eign-billed traffic increased by 14.1% from 1992. Stentor, the consortium of provincial telephone carriers which for- merly had a monopoly over traffic to the U.S., saw an absolute decline in traffic. In late 1992, regulators autho- rized the resale of IPLs for switched services on the U.S.- Canada route. Also in 1993, Canada’s new facilities based long distance carrier, Unitel, began operations. In 1994, Unitel and facilities-based resellers are expected to gain t 2% of the cross border traffic with other resellers capturing another 8%. Figure 1. Global MITT: Growth and Origin 1990 1991 1992 1993 1994 1995 (est.) (est.) Global MiTT (in billions), 1990-95 Note: Data include outbound minutes of telecommunication traffic (MITT) on public telephone network only. Global MiTT by origin, 1993 European Union (35 3%) Asia (121%) Other Europe (11 3%)_ M~ddle East!Africa (6 6%) -- Central & South America.(21%) -- Australia & N Z (2.0%) -- Other (1.2%) North America (29.3%) ©TeleGeography, Inc. 1994 IntroducUon Figure 2. Global MITT: Growth by Country and Region ©TeleGeography, Inc. 1994 Count~ ..... %.Gr0~ ~wt. h 1992-93 Belgium 7.5 France 5.2 6ermany 14.5 Hungary 16.0 Poland 28.2 Russia 14.5 U.K. 9.9 Brazil 7.4 Colombia 8.4 Venezuela 16.1 USA (out in) 10.0 China 40.9 India 9.4 Japan 9.9 Korea 15.7 Malaysia 20.0 Europe 12% Americas Asia Africa 10O/o ¯ 1991-92 World 1~O/o 1984-92 0% 5% 10% 15% 20% 25% 16% 23% Note: Growth rates based on outbound minutes of telecommunication traffic (MITT) on public telephone network only. d As competition grows from IPL resellers, foreign calling cards and other alternative service providers, it is becoming more difficult to measure accurately the size of international traf- fic flows and the market share of the main competitors. Further, carriers on each end of a given route may use dif- ferent reporting standards for the same traffic stream. Accordingly, in using this report readers should be sure to consult the footnotes to each table and the discussion in the Methodology section (p. 149). Capitalizing on growth When almost all the world’s international carriers were state-owned, yearly traffic growth generally affected only the operator’s budget or that of the central government. Corporatization and privatization have changed that. In more than 20 countries the shares of international carriers now may be bought and sold on local stock exchanges. Changes in a carriers’ international traffic and revenue now echo--affecting both the carrier’s share price and, given the shares’ weight on the local stock exchange, the value of other publicly listed enterprises as well. In many middle- income countries, telephone carriers have the highest capi- talization of any traded company and account for a sub- stantial share--often 10% or more--of the total value of listed issues. The new stock table which appears on p. 74 is designed to help readers evaluate the capital values placed on the inter- national traffic data documented elsewhere in TeleGeography. At September 1, 1994, the table shows that publicly quoted carriers providing international service had a tota! stock market valuation of approximately $400 billion. Further, the stock market price for most carriers was 15 to 20 times the company’s net earnings or profits. Assuming that each minute of international traffic generates average revenues of approximately $1 (see p. 34 below), and the average margin is at least 40%, then the 4.5 billion minutes of additional international traffic in 1993 had the potential to add approximately $30 to $35 billion to glob- al equities. The actual value added to world stock markets was somewhat less, however, because many large carriers (e.g., Deutsche Telekom, France Telecom) are as yet unlist- ed. Put another way, in t993 there were approximately 47 bil- lion minutes of international traffic, so that each 1% of addi- tional traffic was approximately equal to $470 million in revenues, on average, or $190 million plus in profits. If this profit were all capitalized at current rates, then each per- centage point of growth could add approximately $3-5 bil- lion in value to global telecommunication equities. Picture phones Competition for these billion dollar stakes is, not surprising- ly, driving carriers to find new ways to enlarge the interna- tional market. In the early 1990s, fax traffic and residential users helped to increase demand during an otherwise weak business market. What will take up the slack in the next decade? Will picture phones be the "killer application" of 20017 Elliot Gold, the editor of TeleSpan, a leading newsletter on video conferencing, provides some answers in the first essay xl TeleGeography 1994 of the Articles section. (See p. 3} Gold states that one of the promises of video phones is their "magnificent multipli- er." For example, a recent 18-month European videophone trial found that average videophone calls lasted up to ten minutes as compared to just over three minutes for voice calls. And video calls typically require twice as much band width, suggesting that revenue for video calls could be six times that of voice calls. The European trial and others in Japan and the U.S. lead Gold to project worldwide toll rev- enues from video phones of over $600 million by 1998 and more than $1 billion by 2000--a substantial sum, but still a very small fraction of future voice revenues. What will make videophone services a mass market? One answer is the growth of several niche markets, such as desk- top video conferencing (between office PCs) and network- based video games. However, before international video calls will become commonplace, manufacturers will need to agree on a standard; many U.S., Japanese and European video phones are still incompatible. Gold also rightly focus- es on the need for local carriers to install a switching plat- form that will permit users to have access to the same cus- tom calling features that are available for telephony--espe- cially voice mail and call conferencing. Absent such interac- tive features, the revenue multiplication from video phones may be slow to materialize. The challenge of convergence The next article carries forward the discussion from TeleGeography 1992 on how international carriers are cop- ing with the challenge of providing such multimedia services. The article starts by reviewing the paradigm shift already underway in the provision of international voice services. The old paradigm was based on monopolistic, facilities- intensive or Heavy Carriers which provided service in only one country. In contrast, the new paradigm is based on competitive, privately owned, Light Carriers which offer end- to-end international service by reselling, repackaging or reprogramming the facilities and services of established car- riers. But even as international carriers begin to accept this new paradigm for telephony, the agenda for change is shifting. Innovative technology and market liberalization are now eroding the boundaries between voice, text and video ser- vices as surely as they did in the early ]990s between one national market and another. Tomorrow’s international telecommunications network will be a conduit not only or even primarily for telephony but for multiple services: video telephony, computer software applications, color graphics and music. In today’s jargon, the network wilt become a global information highway. The role of international carriers and their new global alliances (Concert, AT&T WorldPartners, Unisource] in this network transformation deserves more attention: [] The current rate structure for international telephone services generates a $25 to $30 billion investment pool annually. How carriers use this surplus--the level and nature of the investment, and how much is passed on to telecom users and stockholders--will have a large impact on tomorrow’s service offerings. [] There is a risk that regulators’ preference for managed competition and traditional correspondent relations will encourage many international carriers to create a two-tier market structure for global services. Corporate users may benefit from new global alliances, whose multinational ser- vice platforms will pass on savings generated by end-to-end circuits, flexible routing and innovative software. Meanwhile, most international callers may be stuck with the higher rates and fewer options associated with the Heavy Carrier routing and interconnection arrangements of the past. [] The price and variety of multimedia will depend on the network architecture regulators encourage. Light Carriers have shown that there is more than one way to connect con- sumers to a network. So too has the Internet. In the tele- phone world, open access depends upon a user’s ability to plug in a telephone (or fax machine] and obtain a dial-tone to call any other terminal on the network. In a multimedia world, the test may well be whether a party can connect a competing computer file server to any other data, audio or video file server on the network. A key question is whether the largely one-way Video-On-Demand (VOD] plans of some public operators can be harmonized with this two-way model. On-line geography Much as multimedia is now (wrongly) said to portend the death of books, telecommunication has tong been hailed as the end of geography. But the opposite has occurred; by making almost any point on Earth (and beyondisee p. 170) accessible for a few dollars via a telephone or a point and click menu on a personal computer, the global network has led to the explosion of place. In a sense, travel or geog- raphy has become the common content of computer net- works. Hence as the Canadian scholar Marshall McLuhan (191 t-1980) first noted, rather than obliterating older forms, a new medium takes as its content the medium it replaces. Music and drama were the staples of early radio. Television gave new life to movies. And so the Internet, by abolishing the distance between the solitary desktops and computer screens of its twenty million users has given us a new geography on-line, a geography of rooms, malls, high- ways and frontiers. Xtl Introduction The challenge of navigating this new electronic terrain is attracting the attention of the telecoms and computer com- munities alike. Traffic on computer networks has grown dra- matically since the early 1990s, (see pp. 78-79) spurred in large part by the development of more user-friendly naviga- tional tools or screen interfaces. As Bruce R. Schatz and Joseph B. Hardin recently wrote in Science (t2 August 1994), computer networks have thereby been transformed from networks of machines into a distributed "information space." That means users can now "concentrate on the log- ical structure of the interconnection of information and data items rather than the underlying physical structure of com- puter and communication systems." Network geography has become information geography. But, whether geography can help users to orient themselves in this new information space is the focus of our third major article by Zachary Schrag, the Deputy Editor of TeleGeographg. According to Schrag, our current text-based interface with computer networks is being challenged by two movements. One group of programmers and designers is using graphical user interfaces (GUIs) based on geography (or geoGUIs) to make network information more visible. At the same time, another group is working on "intelligent agents" designed to make the computer networks invisible. Like computer macros and scripts, agents let computer users automate repetitive tasks such as logging onto a com- puter system or downloading information from a database. Unlike macros, agents also can go beyond a local computer to the network itself, retrieving information and sending messages and perhaps even conducting business. Schrag concludes that maps and agents each have a future in cyberspace; the challenge will be to match function and form, to use a geoGUl when it helps users find what they want and to abolish it when it interferes. For Schrag the future of cyberspace may not be unlike the history of radio. Conceived as a means of two way commu- nication (radio was first used to send distress signals from ships) radio later achieved great commercial success as a broadcast medium to transmit music, news and other pro- grams to millions of receivers. But it has never lost its orig- inal function. And ships, airplanes, trucks, taxi cabs and "ham" radio enthusiasts still depend upon two way trans- mission. Millions of new cellular telephone users do too. Thus, "if the radio, a single technology, could blossom as two radically different social phenomena, so could cyber- space," states Schrag. TeleGeography 1995 Next year’s edition will carry forward our main statistical enterprise: chronicling the changing pattern of internation- al telephone and data traffic. We will also continue to mon- itor multimedia traffic as the market begins to reach a criti- cal mass. In addition, we plan to expand the map section of the report to cover some of the vast new information spaces being opened up by the world’s burgeoning on-line communities. Mapping these new worlds--which have no north or south, up or down--is one of the great challenges of post-modern cartography. We invite suggestions and comments from our readers on this new endeavor as well as the contents of this year’s report. Gregory C. Staple Zachary M. Schrag September 1994 Washington, D.C. Xlll TeleGeography 1994 Articles TeleGeography 1994 © TeleGeography, Inc. 1994 THE MARKET FOR PICTURE PHONES Prospects for Sales and Traffic By Elliot M. Gold Picture Phones latson, stay put, I see you!" If Alexander Graham Bell had invented the video- phone instead of the voice phone, that would have been his first utterance. When he spilled the battery acid on his hand, he could have shown the burn to Watson instead of saying "... come here, I need you." And had Watson been a medical doctor, Bell would have gotten a free examination over the phone, followed by a prescrip- tion sent by fax, no doubt. Even though it didn’t happen that way, Alexander Graham Bell was on the same wavelength as later inventors of video- phones, Following his patent for the voice phone in 1876, Bell went on to invent and patent the Photophone, a voice phone that transmitted sound over light waves. With Bell’s Photophone, a person’s voice vibrated sunlight reflected on mirrors, literally flashing the message down the line, where it was converted back into audible sound waves for the lis- tener to hear. From Telephots to Videophone 2500 In the 88 years between Bell’s introduction of the voice phone and AT&T’s introduction of the Picturephone at the 1964 World’s Fair, many explored the notion of video- phones. Kit Galloway, possibly the world’s leading collector of videophone artwork and paraphernalia, has found refer- ences to videophones beginning one year after Bell’s voice phone patent was awarded. In one article located by Galloway, and appearing in the May 1918 Electrical Experimenter, author Hugo Gernsback lists some t 2 designs and patents worldwide for Telephots (from the Greek tele-- far--and photos--light). The first Telephot appears to have been invented by a Frenchman named d’Ardres in 1877, one year after Bell’s voice phone. Galloway’s collection contains references to the first video phonebooths as well, called Telephonoscopiques. In a refer- ence dated 1880, three years after d’Ardres’ vision of the Telephot, an inventor predicts that "Telephonoscopique kiosks are to be found on every street corner, in the better neigh- borhoods, throughout France in the mid-twentieth century. Not only can you speak with your distant friends and loved ones in all the corners of the Empire, you can also see them at the same time." Galloway’s collection includes artwork from the early 1900s showing the boy detective, Tom Swift, using a videophone in the story "Tom Swift and his Photo Telephone." The first trials of "audiovisual" communications were appar- ently those run between 1935 and 1938 by the Deutsche Bundespost Telekom. Participants in four German cities used the phones in public studios. Since 1964, a prolifera- tion of videophones have reached the market, not all of which would have satisfied the needs of the boy detective Tom Swift. In those 30 years, few, in fact, have attained the quality of AT&T’s initial Picturephone. Picturephone delivered smooth, full motion, black and white video (at a time when network television was still black and white) 30 years before jerky color was attained by others. But AT&T had to file a special construction tariff to offer 1 MHz circuits for Picturephone. The circuits proved too cost- ly for telephone users of the late 1960s, and AT&T quietly removed Picturephone from the market in the early ]970s. AT&T continued, however, to offer ful! motion black and white and color video over wideband circuits, principally to broadcasters. Dow Chemical, an early corporate user of two-way videoconferencing, paid AT&T $2,000 per hour in 1974 for a microwave video link between Dow’s headquar- ters in Midland, MI, and its chemical manufacturing facility in Freeport, TX. Over the next decade, prices actually went up, and not down. In 1983, AT&T opened its digital Picturephone Meeting Service (PMS), with occasional use rates tariffed at S2,300 per hour for T2 digital hookups from New York to San Francisco. Though this attempt was a failure, over the past decade PTTs worldwide have prof- itably operated several dozen rooms with videoconferencing facilities for public use. The major provider of public videoconferencing rooms in the United States today is Sprint, which has offered its Sprint Meeting Channel for over a decade. Using a combination of Sprint-owned public rooms, customer-owned shared rooms, and PTT public rooms, Sprint offers a network of scores of rooms worldwide. Elliot M. Gold is publisher of TeleSpan, a bulletin on teleconferencing. He can be reached by facsimile at 1 (818) 797-2035 or by e-marl at 2351706@mcimail. com. This past year both Sprint and AT&T expanded their efforts in public room networks. Sprint has launched a joint marketing venture with Kinko’s, an American chain of stores, many located near universi- ties, which offer photocopying and 3 TeleGeography 1994 computer services. Sprint is installing desktop videoconfer- encing equipment to offer lower cost public videoconferenc- ing services, as an extension of the existing Meeting Channel. AT&T, which closed down its public rooms in Hilton hotels in 1984, is back in the business, offering digital videoconferencing rooms and analog VideoPhone 2500s to guests of Westin Hotels. MCI’s small network of public rooms have failed to return a profit. But an entrepreneurial venture called Affinity Communications has been reasonably successful renting out private PictureTel rooms around the U.S. and the world since 1992. Affinity has kept its capital costs low by simply offer- ing a reservation service for companies with rooms to lease and for companies seeking to use them. The inconvenience of public facilities and the expense of installing special lines have inspired inventors to try to cut transmission costs by delivering video over ordinary phone lines. Small companies such as Colorado Video and Robot Research, and much larger ones like NEC and Stromberg Carlson tried slo scan television technology, later marketed as Slow Scan and Freeze Frame video. By limiting video to snapshots and by dividing these snapshots into sets of 128 to 520 television lines, the manufacturers were able to transmit the video a line at a time over the relatively eco- nomical Plain Old Telephone System. But early slo scan had serious limitations. A picture took longer to arrive than a Polaroid photo took to develop. Color slo scan transmission could take two or more minutes. And systems cost as much as $11,000 to $17,000 per end. While annual sales did reach thousands during the early 1980s, slo scan television attracted only very specialized applications such as tele-medicine in remote areas. In 1984 Widcom demonstrated the first videophone that delivered full color motion video over a single switched 56 kbit/s digital phone line. The $56,000 system was pur- chased by Westinghouse, Bell & Howell, and even the White House. Widcom, though, became a footnote in history when the company was forced into bankruptcy by its overly zealous competitor, Compression Labs. Figure1. Worldwide Picture Phone Equipment Sales (Units) GPT 9% Hitachi 5% VTEL 7% BT 19% CLI 19% PiatureTel 41% The Big Six Sales of group videoconfer- encing systems by the world’s six largest manufacturers for the first half of 1994. PictureTel 2,885 CLI 1,279 VTEL 451 Hitachi 315 GPT 596 BT 1,343 TOTAL 6,869 PC-based ISDN/POTS Analog Video Phones~ Group Conferencing Systems 1990 1991 1992 1993 1994 1995 1996 1997 Sales of systemsin facilities orrooms. © TeleGeography, Inc. 1994 500,000 450,000 400,000 350,000 300,000 250,000 200,000 150,000 100,000 50,000 0 1998 1999 i Picture Phones F 2. The need for Bandwidth Bandwidth 90 Mbit/s = 4700 telephone lines Resolution Full motion color television 2.0 Mbit/s = 104 telephone lines High quality videoconference 1.5 Mbit/s = 78 telephone lines Good quality image for business and industry meetings 384 kbit/s = 20 telephone lines Interviews/Education/Training 64 kbit/s = 1 ISDN bearer channel INDEO (PC-based), ISDN circuit 19.2 kbit/s = 1 4 kHz telephone line Videophones returned to the telephone network in 1986 when Mitsubishi introduced the Luma 1000, a snapshot video system with an extraordinary set of features, includ- ing an on-line phone directory. The Luma listed for $1,450 and delivered a postage-stamp-sized black and white image in 10 seconds. Although there were some niche sales, such as to law enforcement agencies, Mitsubishi abandoned the Luma the next year in favor of the less expensive VisiTel or Luma 500 which, by eliminating many of the Luma 1000’s features, managed a list price of $599 in 1987. Mitsubishi manufactured 40,000 VisiTels, selling only a few thousand. After writing them off at the end of the 1980s, Mitsubishi turned its inventory over to the Home Shopping Network, which closed them out in a matter of days, at prices as low as $87 a pair. AT&T returned to the market in January, 1992 with the VideoPhone 2500, which delivers jerky motion color video over an ordinary telephone line, and lists for $1,399 per end. AT&T has already sold 40,000 of the 60,000 VideoPhone 2500s manufactured. Today, buyers can purchase stand-alone or PC-based video- phones from more than four dozen telephone and comput- er manufacturers, including AT&T, BT, and Intel, with prices ranging from under $1,000 to $6,000 for PC add-on board sets, and from $1,399 to $6,000 for stand-alone systems. How videophones work Voice and full motion video begin their videophone journey as analog signals. While voice begins as a 3.0 kHz analog signal, video begins as a 4.6 MHz signal, some 1,500 times wider than voice signals. In order to preserve the quality of the signal and to abide by international telephony stan- dards, voice is converted to 64 kbit/s, or one ISDN Freeze frame (slo scan) telephone (Integrated Services Digital Network) basic rate circuit, while video is converted to 90 Mbit/s, the equivalent of two 45 Mbit/s or DS-3 channels, retaining the ratio of 1:1,500 between video and voice. This, principally, is the reason that until recently videophone transmission cost so much more than a voice call. During the 1960s and 1970s, mathematicians and engi- neers began in earnest to work on the disparity between the bandwidth or data rate requirements for video and voice. Due to the nature of television, they knew that the informa- tion in a picture was highly redundant. They knew, for example, that the color and texture of a wall or of a person’s skin would not vary greatly during a video call. And, because videophones were expected to view people seated in chairs, and not on top of race horses, their relative posi- tion wouldn’t change significantly from one frame to anoth- er. Therefore, they considered dividing frames of video into "blocks" of redundant information. After dividing the visual information into blocks, researchers theorized and then proved that the descriptive information in these blocks could be compressed greatly using mathe- matical algorithms, including some called "transforms." As the name "transform" suggests, the redundant picture information in the blocks is "transformed" into coefficients describing the content of and the relative location of the video blocks as they move in a motion sequence. By com- pressing the video information and converting it into coeffi- cients, the picture--or, more correctly, the description of the picture--occupies less storage and requires narrower trans- mission channels than it did when it was in its initial analog or digital format. TeleGeography 1994 Research was conducted broadly at telephone companies, mainly at AT&T (by Netravelli) and at NTT, and at govern- ment aerospace and defense agencies such as NASA and DARPA and related subcontractors, principally Philco-Ford (later Ford Aerospace). Researchers such as Widergren, Pratt, Fratick, and Tesher quickly became entrepreneurs, founding Compression Labs in ]979 and Widergren Communications in 1983 to build the first video compres- sion/decompression devices, called codecs. companies for ISDN, just as group videoconferencing boost- ed use of 56 kbit/s circuits. AT&T struggled with the launch of its 56 kbit/s circuit switched digital capacity until 1989, when usage revenues came in 170% ahead of forecast. The company found that PictureTel’s group videoconferencing, which AT&T sold through a joint marketing agreement for $.30,000, and AT&T’s own price cuts, making switched 56 kbit/s circuits as inexpensive as voice toll, were the winning combination. AT&T and NTT/NEC demonstrated color motion video over a pair of T1 (1.54 Mbit/s) channels in the late ]970s. In 1981, Compression Labs reduced the requirement to a single T1, with Widcom’s demonstration of 56 kbit/s transmission following in 1983. Today, through very specialized mathe- matics and more powerful chips, the transmission requirement for quality color motion video has been reduced to an ISDN basic rate interface of 128 kbit/s. Lower quality color and motion can even be attained over an ordinary voice channel, by compressing the video and audio to a single 9.6-19.2 kbit/s data stream. C ENS OF MILLIONS OF PERSONAL COMPUTERS ALREADY IN USE COULD BECOME TENS OF MILLIONS OF VIDEO PHONES~ IF USERS ARE WILLING TO PAY FOR ISDN. Such compression rates have yielded a new possibility for video communication. Instead of buying special video phones, people can now use the personal com- puters and modems already sitting on their desks, though each party will need to pay for an ISDN line. lntel’s INDEO System, which sells from $999 to $2,700 per computer, turns a $2,000 PC into a device to code video and voice, as well as for sharing documents with distant colleagues. The tens of millions of personal computers already in use could potentially become tens of millions of video phones, if users are willing to pay for ISDN. Indeed, desktop video might be the "killer application" long sought by the phone Video phones could drive network traffic Videophones provide carriers with a magnificent multiplier, one that multiplies the revenues they can receive from phone calls. Simply put, video calls are longer and often require twice as much bandwidth as voice calls, therefore generating six to twelve times as much toll as ordinary voice calls. This fact came to light after an 18- month trial of ISDN videophones throughout Europe between 1992 and 1 993 under a program called the European Videophone Experiment, EVE2. In the trial, PTTs in the United Kingdom, France, Germany, Italy, the Netherlands, and Norway installed videophones that communicated over an ISDN line, relying on standards set by the International Telecommunication Union (ITU) for group videoconferencing, called H.520. The EVE2 experience in the United Kingdom and in Norway was that the average videophone call lasted up to 10 min- utes, versus voice calls which typically last just over three minutes. These "holding times," as they are termed in tele- phony, were somewhat shorter in Italy and in Germany, but varied mostly by user and application. While users from some electronics firms, universities, and service companies reported average holding times of between eight and ten hone Traffic Volumes and Revenues, 1991-99 Analog Videophones1 PC Based ISDN/POTS1 Average Minutes of Use/Week2 Minutes of Toll/Yr 48 Weeks2 Hours Toll/Yr 2 Revenues2 1991 1992 1993 1994 1995 100 12,100 58,100 78,100 108,100 200 4,300 11,500 29,500 89,500 50 70 1996 1997 1998 1999 148,100 208,100 278,100 353,100 209,500 389,500 649,500 1,024,500 110 150 200 240 1,888,128 4,302,720 8,904,960 15,869,952 31,469 71,712 148,416 264,499 $132,169 $301,190 $623,347 $1,110,897 258,240 663,936 4,304 11,066 $18,077 $46,475 1Cumulative 2Thousands © TeleGeography, Inc. 1994 Picture Phones minutes, users from other industries reported holding times of only four minutes, or just over the length of their ordinary voice calls. Managers from BT, which participated in the trial, had average holding times of over 12 minutes, obvi- ously affected by their own enthusiasm for the medium of the trial. If we assume that holding times for videophone calls will gravitate to three times that of an ordinary voice call, and will use ISDN, which is generally tariffed as two voice cir- cuits, we could expect revenues to be on the order of six times that of voice-only calls. Others expect revenues to far exceed this, due to an anticipated need for video phone calls to bring together indi- viduals in group conference calls. In Japan, NTT is a proponent of this theo- ry. This past spring, NTT began to offer an ISDN videophone called the Pic Send-R (manufactured for NTT by Hitachi), which can be used to link up to five homes in a videoconference. The consumer videophone can display the faces of up to four callers, each in a quadrant of the screen. An ordinary Pic Send-R call within Tokyo cost 10 yen ($0.10) for three minutes, or virtually the same as an ordinary voice call. This is because a point-to- point videophone call requires only a single 64 kbit/s chan- nel on NTT’s ISDN. When a caller adds a second site, how- ever, the caller must convert the call to use two 64 kbit/s channels multiplying the number of yen spent every three minutes to 20 for the first link, 20 for the second link, or four times the cost of a simple voice call. With a five-way call, the bill quickly rises to t 00 yen per three-minute seg- ment, or a ten-fold increase over the toll generated by an ordinary consumer voice call. My estimate is that worldwide toll revenues from video- phones, both the analog variety and the PC-based ISDN variety, will exceed $1 billion a year by 999 (see Figure 2). Why haven’t videophones sold yet? VIDEOPHONE USERS WANT TO SEE MORE THAN JUST SOME- ONE’S FACE DURING A VIDEO- PHONE CALL. THEY WANT TO SHARE DATA AND GRAPHICS~ ESPECIALLY IF THE CALLS ARE MADE FROM ONE PLACE OF BUSINESS TO ANOTHER. My data indicate that, since the middle of 1992, approxi- mately 80,000 videophones had been shipped worldwide. (Again, see Figure 1). Based on sales during the first six months of 1994, videophones willYepresent approximately one-third of the 60,000 visual communications devices sold worldwide in 1994. Just over 20,000 group videoconfer- encing devices will be sold for the year, up from the 12,000 to 16,000 sold in 1993. Between 16,000 and 20,000 desktop or PC-based videoconferencing systems will sell, along with no more than 20,000 videophones. This will grow the installed base of videophones, worldwide, to somewhere between 90,000 and 100,000, a mere pixel in the sales picture predicted by others. There are several impediments to the sales of videophones. First, there is no one to call, yet. While the corded, cord- less, and cellular phones sold each year can be used immediately to make voice connections to hundreds of millions of locations, videophone callers have no one to call and see. Second, quality videophones are still much too expensive. Quality voice phones can be purchased for $20 to $30, but quality videophones can cost $3,000 to $6,000 per terminal (includ- ing the PC). Even poor quality video- phones, like the AT&T VideoPhone 2500, cost $1,399 per terminal. Regular users of visual communications, in partic- ular group videoconferencing systems, report that it is much easier to justify a $45,000 videoconferencing system to be used by many people, than a $3,000 to $6,000 device that sits on a desk for one person’s use. Finally, the use of videophones in Europe during the EVE2 trials indicates that videophone users want to see more than just someone’s face during a videophone call. They want to share data and graphics, especially if the calls are made from one place of business to another. This is why the majority of videophones brought to the market in 1994 car- ried graphics capability for what is termed "group collabo- ration." Since Bell’s voice phone patent was granted in 1876, almost 120 years ago, hundreds of millions of voice phones have been sold. As a measure of the voice phone’s success, the Electronic Industries Association estimates that 25.8 million corded phones, 19.6 million cordless phones, and 4.5 million cellular phones were sold to consumers in the United States during 1993 alone. Why is it then that video- phones, which are much flashier than voice phones, are see- ing annual sales of only 20,000 to 30,000 worldwide? Where videophones do sell The early adoption of new communications systems always clusters around a few niche markets. These markets typical- ly include the military, government security agencies, law enforcement, high technology companies, and, last but not least, pornographers. Government and defense agencies with a need for secrecy have been interested in videophones since the mid-1970s. After a year-long study, NATO contracted with the Jet TeleGeography 1994 Propulsion Laboratory (JPL) in 1977 to install slo scan videophones, manufactured by Colorado Video, in Belgium for war games. In 1979 the Defense Advanced Research Projects Agency (DARPA) funded CLI to develop a 9.6 kbit/s motion videophone called the "Sketch Coder" which deliv- ered outlines of participants and encrypted voice over secure voice circuits run under the streets of Washington, DC. The Sketch Coder was replaced at DARPA in 1983 by Widcom’s 56 kbit/s videophone, which was developed par- tially through a DARPA grant to Bell & Howell, which held the license for a short period of time. As more and more technology was applied to videophones, highly secret or "black agencies" of the federal govern- ment purchased hundreds of Widcom and PictureTel videoconferencing sys- tems for use as videophones. They were sold through Datapoint as the Multimedia Information Network Exchange, MINX. Secure federal agencies continue to be one of the largest users of videophones, with some PC-based videophones built by In-Soft connecting military vehicles, while several thousand encrypted Comtech Labs videophones have been merit by AT&T and Motorola. principally made up of still videophones, is now approach- ing t 0,000. High technology companies have always been among the first to use new communications, such as videophones. Unisys, the Jet Propulsion Laboratory, and Martin-Marietta are among more than a dozen high technology centers that have tested the use of videophones. Sales to these institu- tions amount to less than 100 systems, however. F JOR HURDLE PREVENT- ING THE LARGE-SCALE WORLD- WIDE DEPLOYMENT OF VIDEO- PHONES IS A LACK OF CONSEN- SUS ON STANDARDS. sold to the govern- Law enforcement agencies have used analog full motion videophones since ]973, when the City of Philadelphia strung coaxial cable under the city streets to connect city jails with the downtown courthouse for arraignment hear- ings. The city was forced to do so when the state legislature passed a law requiring that arraignment hearings be held within 36 hours of the arrest of a felony suspect. Similar sys- tems are now in operation throughout California, Texas, New Mexico, and the Boroughs of Brooklyn and Manhattan in New York City. There is also a large installation of still videophones today in the law enforcement industry for "house arrest" and "home detention" programs. Since the late 1980s, parole agencies have experimented with videophones as an alternative to electronic anklets and bracelets, in order to tether parolees to their homes or halfway houses. By using random phone calls, parole agents now talk to and see for themselves whether parolees are where they are supposed to be. Unfortunately, abuses abound. One agency, the Justice Network, Inc., in Memphis, Tenn., places 24-65 calls to parolees within each 24-hour period, sending parolees who complain (due to a lack of sleep, no doubt) back to prison. Over 4,000 still videophones were in use for home deten- tion in mid-1992, with the number now estimated to be closer to 6,000. The total law enforcement installed base, It is always a surprise to the marketing managers of new communications products when their first buyers come from the pornography industry. This process has been going on for millenia, most recently in the use or abuse of 900 or "dial-a-porn" services. (See also the map on page 66.) Videophones are no exception, and in the mid-1980s, manufacturers of still videophones found that many of their buyers wanted them for "party lines," or video bulletin boards. These services allowed partner seekers to post still videophone images, in any pose, on electronic bulletin boards. Hugh M. Hefner, founder and chairman of Playboy Enterprises, had his Playboy Mansion wired for Picturephone in the late 1960s, allowing Her and his editors to review pictures being considered for upcoming issues of Playboy magazine. The rich and famous are another reliably visible market for new communications technologies. AT&T reports that it has sold a good number of VideoPhone 2500s to basketball players in the National Basketball Association (NBA), for example. Market catalysts Analysts forecast that business will spend $1.5 billion on the purchase and use of videophones and videoconferencing in 1994 (ITCA). Yet, consumer markets may actually be more predictive of how videophone revenues could grow over the next 10 years. Teenagers, for example, are already large consumers of con- ferencing and interactive graphics. In North America alone, the annual phone bill for three-way conference calling for teenagers will be $264 million in 1994. Three-way toll is estimated to add another $264 million, providing phone companies total revenues of $528 million from these young consumers. Interviews with public telephone carriers in Germany, the U.K., and Japan indicate, though, that teenage three-way calling might be a peculiarly American habit. A more international phenomenon is the ease with which teens navigate through extremely complex graphical interfaces--also known as video games--resulting in $6.3 billion in 1993 American sales for Nintendo and Sega. This Picture Phones affection for video games could end up generating video- phone revenues, with the coming of new, communicating- interactive video games with the look and feel of video- phones. Already, ImagiNation, the joint venture between AT&T and Sierra OnLine, reported 40,000 households paying month- ly subscriptions in excess of $400,000 by January, 1994. ImagiNation allows players to play video games with distant subscribers by linking them through a data and phone net- work. By applying a modem technology called VoiceSpan, developed by AT&T’s Paradyne, ImagiNation subscribers can even talk to their fellow combatants while trying to mutilate them on a video screen shared over a phone line. Doom, the extremely popular video game, is increasingly being played as a communicating-interactive video game over pairs of ordinary voice telephone lines, one for 14.4 kbit/s sequential data and one for sequences of teenage squeals. Like video games, cable television could act as a spearhead for video telephony. In the United States, the Federal Communications Commission has authorized local telephone companies to provide "video dial-tones" to residences, Though the experiments now underway are only one-way, with consumers receiving television programs and other video from their phone company, the project will result in much greater deployment of fiber optic lines, if only to the curbside, potentially allowing two-way, switched video com- munication. However, the ability (and desire) of local tele- phone companies to develop a video dial-tone service plat- form which is capable of providing two-way videophone ser- vice is a very big, unresolved issue. The need for a standard Another major hurdle preventing the large-scale worldwide deployment of videophones is a lack of consensus on stan- dards. While the world’s group videoconferencing systems manufacturers have adopted the ITU’s H.320 standard, manufacturers in Europe, the United States, and Asia have adopted three different standards for personal video and videophone systems. In Europe, the consortium that conducted EVE2 has adopt- ed the H.320 standard for ISDN videophones. BT, Deutsche Bundespost Telekom, France Telecom, Norway’s Televerket, PTT Telecom Netherlands, and the Societa Italiana per EEsercizio delle Telecomunicazioni p.a. (SIP) are now selling H.320-compliant ISDN videophones. In Asia, AT&T has convinced several Japanese manufactur- ers to adopt its Global VideoPhone Standard, GVS. This standard describes the mathematics that AT&T’s VideoPhone 2500 uses to communicate voice and video over ordinary telephone lines. AT&T has already licensed Panasonic and Hitachi to manufacture videophones using GVS, while others, such as Sanyo, JVC, and Sharp, have signed agreements to resell the AT&T-manufactured VideoPhone 2500 in Japan. Meanwhile, GEC Marconi, which has shipped nearly as many analog videophones worldwide as AT&T, is pushing its Marconi Video Telephone Standard, M-VTS, through carriers in the Pacific Rim, outside of Japan, and in Europe to sell against the H.320 ISDN videophones. In the United States, Intel is promoting its own ISDN video- phone standard, called INDEO (INtel viDEO). So far Intel has convinced nearly two dozen PC manufacturers, software publishers, and telephone companies, including AT&T, to produce products and services around INDEO. While both H.320 and INDEO use ISDN circuits, they are incompatible with one another. Likewise, both GVS and M-VTS use ana- log circuits but are incompatible. AT&T has offered some hope of interconnectivity among these different standards. Its WorldWorx network offering, which is scheduled to be released in 1995, will provide con- version services between ISDN videophones using H.320 and INDEO by next year. Hewlett-Packard, Novell, PictureTel, and VideoServer are also members of the WorldWorx group. Still, the lack of consensus is a major roadblock to the growth of the videophone market. Lastly, widespread videophone service will not be possible without significantly upgrading network switching capacities. Consumers will likely demand that videophones offer at least as many special features as today’s voice phones. Anticipating this, carriers, most notably AT&T, have begun to develop software for network switches that will allow them to adapt the Custom Calling features offered today for voice calls to video, offering videophone call forwarding, three- way videophone calls, and videophone mail. AT&T expects to add these services, not currently available on networks, to its ISDN by late 1995, but Regional Bell Operating Companies are unlikely to offer them until 1996 or beyond. Toward 2001 Videophones have had a slow start. But the resolution of compatibility issues, likely by 1996, should allow a sizable market to emerge. Consumer videophones, as part of video games, could present a very large market for local exchange carriers, if they lower access charges for ISDN for intra-LATA calls. For business, still expected to be the largest market, as many as 375,000 ISDN videophones could be sold annu- ally by 1999. And once consumers and business people have others to call, revenues from sales and toll will be counted in billions of dollars--long overdue for the visionaries of 1877. TeleGeography 1994 For Further Reading... Elliot Gold’s Telespan, published 40 times a year, tracks the latest developments in picture phones and videoconferencing. For background reading, see the following books and articles: Rochelle Broder, "H.320 Sharpens Videoconferencing Picture," Network World, August 8, 1994, p.55. E. Dickerson and R. Bowers, The Video Telephone, Impact of a New Era in Telecommunications (Praeger Publishers, 1974). J. Pierce and A. Noll, Signals, The Science of Telecommunications (Scientific American Library) 1990. R. Pye and E. Williams, "Teleconferencing: Is Video Valuable or is Audio Adequate?" Telecommunications Policy, June 1977. Andy Reinhardt, "Desktop Conferencing Takes Off," Byte, June 1994, pp.24-25. -- "Video Conquers the Desktop," Byte, September 1993, pp.64-80. © TeleGeography, Inc. 1994 The Challenge of Convergence INTERNATIONAL TELECOMMUNICATIONS The Challenge of Convergence by Gregory C. Staple ~|~ s network technologies redraw the boundaries between different national markets and communi- cations services, international carriers face a dual I challenge: how to compete globally in a multimedia 1market. Outdated regulations and uncertain corn- /petition rules have compounded the task. This essay explores these challenges. It begins with the industry paradigm shift already underway. See Figure 1. Part 1. Introduction Paradigm Shift The old paradigm governing international telecommunica- tions was characterized by state-owned national monopo- lies, typically Post Telegraph & Telephone (PTT) operators. Historically, these carriers did not operate directly in other countries but provided international service by connecting their cable and satellite "half circuits" with the matching facilities of a foreign carrier. As facilities based or "Heavy Carriers," the PTTs compensated one another by a 50/50 division of a wholesale facilities charge or accounting rate per minute of traffic. Each national carrier recouped this charge, plus a mark-up, which might be 150% or more, through its own international tariff. At the beginning of the 1990s, however, pressure from new operators, rapidly falling transmission and switching costs and the rising demands of multinational users spurred an industry paradigm shift. The new paradigm is characterized by privately owned competitive carriers which offer direct end-to-end international service by reselling, repackaging or reprogramming the facilities and services of established car- riers. This alternative regime has been termed the "Light Carrier" model. These two international service models were explored at length in TeleGeography 1992.1 The Light Carrier challenge to the status quo is largely based on economics. Fiber optic cables and high capacity digital switches increased trans-oceanic circuit capacity at least 50 fold during the 1980s and radically cut connection costs. Yet, in most countries, only a very small part of these sav- ings were passed on to users. Light Carriers exploit this fact by acquiring (by lease or bulk tariff) the facilities necessary to provide an international dial-tone and then reselling them at a discount to the standard tariffs of the Heavy Carriers. For example, numerous companies resell the switched tele- phone service of U.S. international carriers; many resellers also solicit foreign customers through "call back" arrange- ments which offer foreigners a lower cost (U.S.) dial-tone for international calls upon dialing a pre-designated U.S. num- ber. Other European Light Carriers offer multinational companies services for refiling their outbound traffic via less expensive national carriers. Refiling, sometimes known as hubbing, generally refers to the practice whereby international traffic originating in Country A is collected by a company and transmitted (usually by private line) to Country B, where it is handed off to an international carrier for onward routing to take advantage of Country B’s lower international service tariffs.2 Analogous practices exist in the postal field. Foreign remailing--for example, bulk posting in the Netherlands of North America-bound business mail from the U.K. or Italy--is now common. It has not taken long for Heavy Carriers to respond to these challenges. Since 1993, eleven of the fifteen largest inter- national carriers, accounting for 55% of the world’s interna- tional traffic, have entered alliances to provide a common brand of "seamless global services." These ventures include Concert, the BT/MCI company; the Sprint-France Telecom- DBP Telecom Alliance; AT&T World Partners and Unisource-- the Swiss, Dutch, Swedish, Spanish consortium. See Part 111 below and Figures 5 to 8. Heavy Carriers also have begun more aggressively to adopt Light Carrier strategies to establish a global presence. For instance, international simple resale (ISR)--Ieasing dedicat- ed private line capacity to provide public switched telepho- ny-was once viewed as a low cost entry vehicle for new carriers. But ISR licenses are now held by Sweden’s Telia AB, Australia’s Telstra and Singapore Telecom. Other Heavy Carriers--AT&T, C&W and BT--have ISR applications pend- ing. See Figure 2. Enter Multimedia Gregory C. Staple is a Partner in the com- munications law firm of Koteen & Naftalin, Washington, D.C. This article does not necessarily represent the position of Koteen & Naftalin or its clients. But even as international carriers begin to accept a new ser- vice paradigm for telephony, the agenda for change has shifted. Digital signal processing and transmission technologies are eroding the boundaries between voice, text and video services. Tomorrow’s international telecom- munications network will be a con- duit not only or even primarily for [] 11 TeleGeography 1994 telephony but for multiple services--video telephony, com- puter software applications, color graphics and music. In the current vernacular, the network will become a global information infrastructure (GI!) or infobahn. 3 Thus the chal- lenge for the world’s major telephone operators is how prof- itably to build their own section of this superhighway.4 The network’s multimedia future, like the pressure for alization, has a differing priority from country to country. Outside of the wealthiest nations, a multimedia, broadband digital network has a limited attraction. For most carriers, whose penetration is typically less than 5 lines per 100 peo- ple, the priority is still providing basic telecommunication services.5 Moreover, even in richer nations, the near term multimedia challenge for many operators centers on the market for new enterprise-based services--voice mail, e- mail, data processing and work group software. Yet, by the early 21st century, it appears that more and more carriers will restructure their network service arrangements to satis- fy the needs for a robust multinational platform which can de~iver not only voice but a variety of data and video ser- vices as well. The public’s stake in this network transformation and the role of international carriers deserves more attention: [] The current telephone rate structure for international telephony insures that the world’s 600 million plus tele- phone subscribers provide an almost unparalleled global savings and investment pool for the industry. International telecom services will generate profits of $25-30 billion annually for the rest of the decade (1994 carrier revenues will total $55-60 billion. See Appendix 1.) How much of this surplus is invested in the network, what it is invested in, and how much is passed on to telecom users and stock- holders will decisively affect tomorrow’s network service offerings. [] There is a distinct risk that regulators’ preference for managed competition, intentional or not, will lead many international carriers to limit rather than expand competi- tion, and indeed to create a two-tier market structure for global services. The top tier of corporate users, their sup- pliers and customers will be served by new global alliances whose multinational service platform will pass on the cost savings from whole circuit provision, hubbing arrangements and innovative software. In contrast, regulators’ support for traditional international routing and settlement rules-- ostensibly designed to ensure a level playing field for local and foreign carriers alike--may relegate most international callers to a second tier of services with fewer options and higher rates. [] The service model for the Gll--for global multimedia networksias with the paradigm for international telephony before it, will affect the price and scope of services for years to come. The Heavy Carrier regime which built today’s international network, despite its successes, severely limited network access by tens of millions of people through monopolistic pricing policies and restrictive interconnection. Light Carriers have shown there are other ways to bring international networks services to consumers. So too has the lnternet, which has already demonstrated that one need not create a software company, be a record producer or build a world wide television network to become a global multimedia player. One need only be able to connect a computer file server with the audio or video or data bits one wishes to sell to other network users. Tomorrow’s global multimedia networks should build on these foundations to F lnternational Telecom Services: A Paradigm Shift The Old World HEAVY FACILITIES BASED (owned) HARDWARE INTENSIVE MONOPOLISTIC NATIONAL MARKETS HALF-CIRCUITS ACCOUNTING RATES PARALLEL TERMS FOR COMPETITORS VOICE SERVICES The New World LIGHT NON-FACILITIES BASED (leased) SOFTWARE DRIVEN COMPETITIVE MULTINATIONAL MARKETS END-TO END (WHOLE) CIRCUITS TRAFFIC CONTRACTS/ACCESS CHARGES PREFERENTIAL TERMS FOR "PARTNERS" MULTIMEDIA SERVICES © TeleGeography, Inc. 1994 The Cflaflenge of Convergence Figure 2. International Simple Resale (ISR)--1994 To Austraha, NZ Denmark W Europe to UK Legend Two-way ISR (October 1994) I I I Two-way ISR apphcatlon pend=ng I nbound flows for hubbmg Shaded countrles permit one-ended PSTN interconnectmn Note: ISR also possible in 1995-96 with Chile, New Zealand, and the Netherlands. © TeleGeography, Inc, 1994 To Canada provide reasonably priced, non-discriminatory access to users around the world. The remainder of this article is organized as follows: The next section, Part I1, updates the breakdown of the one car- rier, one country paradigm for international service and briefly discusses the key enabling technologies, such as ATM (Asynchronous Transfer Mode) switches, which are bringing together the market for voice, data and video services. Part II1 looks at the ways in which Heavy Carriers and Light Carriers are responding to the convergence between differ- ent national markets and previously separate communica- tions services. It also profiles the new global alliances noted above. Part IV examines the regulatory response to con- vergence, focusing on the United States. Because the U.S. originates and terminates approximately 35% of the world’s international traffic, U.S. regulation affects almost every other country. Part V offers some concluding thoughts about the regulatory imperative presented by market con- vergence. Part II. Global markets, digital services From National to Global Markets Since the mid-19th century international telecommunica- tions has been typified by joint ventures among national, often sovereign, monopolies or oligopolies. Through service was provided by interconnecting half circuits in internation- al cable and, later, satellite facilities. This half-circuit regime, coupled with restrictive foreign investment rules, ensured that the market for international services was seg- mented almost exclusively along national lines. A carrier from Country A could not pick up traffic in Country B with- out the cooperation of Country B’s national carrier(s) and vice versa. As noted earlier, since the 1980s, technology and regulato- ry change have enlarged the geographic scope of carriers’ markets, turning many national partners into global com- petitors. There are several causes. One is the demand of major telecom users to have greater control over the trans- national communications networks used to support their global operations. These demands gradually have led gov- ernments to relax national restrictions on the lease, public network connection and resale of international private lines, thus allowing private companies to create de facto whole circuit networks.6 For example, it is now lawful for a European multinational to acquire private lines among its Rome, Frankfurt, London and New York offices; to share part of the London-Frankfurt line with an affiliate; and to "refile" both companies’ outbound international traffic in the U.S. via the New York office’s Private Branch Exchange (PBX) which is directly connected to the public switched net- work. Competition between national carriers has also increased as multinationals become more adept at locating telecom- intensive operations in countries with the most favorable regulations and tariffs. In so doing, users have introduced some degree of price competition among national carriers even though these carriers do not serve other countries directly. Carriers have also been led to make their offerings TeleGeography 1994 more responsive to the geographic scope of users’ corporate Telefonos de Chile (CTC), a local carrier; and a minority activities rather than that of the carrier’s franchise, interest in CANTV. Second, as major carriers try to follow their prime business and residential customers when they travel, they further erode the barriers between national markets. International direct dialing (IDD) and "intelligent" software-based switch- es have enabled carriers to develop innovative billing and call set up arrangements to serve customers outside the operator’s home market. The best known services--Home Country Direct and Home Country Beyond--permit calling card and credit card holders to use their national carrier to place international calls from almost anywhere in the world. Access is via local "free phone" numbers and interna- tional "800" (toll free) services. These services still preserve the underlying half-circuit, accounting rate regime. (The carrier billing a Home Country Direct call pays the carrier where the call originates a per minute settlement charge at least equal to the settlement it would retain from an outgoing call.) But they may expose a national carrier to direct tariff competition from numer- ous foreign carriers. As one U.S. carrier advertises, "How do you call a foreign country when you’re already in one? Pick a card, any card." Even where direct foreign investment is not permitted, a for- eign carrier may be permitted to lease a matching interna- tional half-circuit and to resell the capacity for end-to-end switched services by interconnecting the whole circuits with the switched network at both ends. Again, see Figure 2. Taken together, these processes are gradually transforming the international service business. Markets once defined by national boundaries have become more elastic; user needs, billing and call set up practices, and foreign investment rules--not political geography or half- circuits--set the limits. THE DISAPPEARANCE OF DIS- TINCT NATIONAL MARKETS IS MIRRORED BY TIlE EROSION OF DISCRETE TELECOM SERVICE OFFERINGS. Finally, national markets are converging because at least twenty countries now permit foreign operators to acquire a stake in local international carriers and thereby own the half circuit at each end of the transmission link. This type of facilities-based geographic convergence is still the exception rather than the rule, but it is already affecting business strategy (and the regulatory response) on several routes. For example, the U.K.-based company, Cable & Wireless (C&W), owns 80% of Britain’s Mercury (BCE Inc., the major shareholder in Canada’s overseas carrier, Teleglobe, owns the other 20%); C&W also holds 17% of IDC (Japan), 40% of Tele-2 (Sweden), 24.5% of Optus (Australia), 58% of Hong Kong Telephone and a majority share of 17 different Caribbean carriers. Likewise, AT&T owns 20% of the Canadian long distance carrier Unitel (which also offers ser- vice to the U.S.); and a minority share in CANTV (Venezuela). France Telecom owns a minority share of the consortium controlling Telmex and 25% of Telecom Argentina, which serves the country’s northern states; it also has proposed to buy a t0% stake in Sprint (U.S.). Telefonica of Spain owns a majority interest in the U.S. Puerto Rican carrier; a minority interest in Telefonica de Argentina, which serves the country’s southern half; a 35% interest in Entel Peru; a controlling interest in Compania de From Telephony to Digital Services The disappearance of distinct national markets is mirrored by the erosion of discrete telecom service offerings. Again technology is the major driver. During most of the 20th century, inter- national communications typically con- sisted of two or three distinct publi, c ser- vices; telephony, telegraphy and telex (an automated tele-typewriter service). These services were based on different analog technologies, required dif.ferent terminals; often used different international transr, nission facilities; and in several countries, notably the U.S., they were furnished by different record (telegram, telex) and voice (telephone) carriers. During the 1970s new technologies began to challenge the basis for these distinct service categories and hence the via- bility of the underlying carriers. Of principal importance was the widespread business adoption of computers and the associated demand for communication links based on the digital standards used by computers to process information. Initially computer-to-computer (data) communications was carried on dedicated leased lines. The direct effect on the market for international telephone and telex service was limited. But, as advances in semiconductor and digital sig- nal processing continued, computer technologies began to transform the telecommunication industry itself from an analog world of facilities-specific networks and services to a world of digital transmission and switching able to transmit voice, text, graphic and video information on the same physical network. Ever cheaper and more powerful semi- conductor chips also led to the development of new termi- nals for electronically creating, storing and transmitting data over the public telephone network (e.g., via high speed fac- simile machines and computer modems.) By the early 1980s telecom carriers realized that they would need multi-service networks, based on digital technologies, in order to prosper in the long term. Otherwise the carrier’s The Challenge of Convergence network would not be able to meet the needs of most busi- hess users and many new services, with potentially large revenues, would be impractical. Moreover, if the telephone company networks failed to keep pace with technology, then they risked being bypassed by the computer industry or even cable television providers which would create their own transmission networks for key clients. This threat was underscored in ]985 by IBM’s acquisition of a minority stake (later sold) in MCI. To their credit, in the 1980s leading telecom engineers sought to keep the network one step ahead of the demand for basic digital connectivity by promot- ing a bold new vision known as ISDN-- the Integrated Digital Services Network (ISDN). ISDN entails a common set of digital transmission and switching stan- dards which use 64 kilobit per second (kbit/s] circuit building blocks plus a sig- nalling channel to produce a digital pipeline capable of carrying a wide range of services. And because ISDN standards are agreed internationally, in theory, ISDN permits new multimedia digital terminals to interact seamlessly around the world.7 This ISDN vision proved to be both too ambitious and not ambitious enough. On the one hand, international opera- tors already committed to moving from an analog to a dig- ital network, were reluctant to adopt ISDN as the basis for their new network when telephony--which still provided over 90% of their revenues--could be accommodated on non-ISDN digital channels. ISDN deployment was thus uneven and generally viewed as a "special" service with pre- mium tariffs. Similarly, local and regional interests led actu- al ISDN standards to vary somewhat from country-to-coun- try. enables a network to carry efficiently very large streams of text, audio, graphics and video information over the same electronic channel. ERGENCE IMPLIES NOT ONLY THE ENTRY OF TELECOM COMPANIES INTO DATA NET~ WORKING AND BROADCAST MARKETS, BUT ALSO THE ENTRY OF THESE INDUSTRIES INTO TELECOMMUNICATIONS. J On the other hand, even as operators began to implement ISDN, the transmission rates for basic rate ISDN services (two 64 kbit/s B circuits one 16 kbit/s D signalling chan- nel) were increasingly viewed as too limited and inflexible to accommodate the new demands of very high speed com- puter workstations and real time video networks. Thus, by the early 990’s, a digital transmission and switching stan- dard known at ATM--Asynchronous Transmission Mode-- gained widespread support as the key enabling technology for broadband ISDN or B-ISDN (networks carrying 2 Mbit/s or more}.8 ATM is a flexible, software-defined switching protocol championed initially by the computer and network- ing industries. By breaking all transmissions into small uni- form digital information packets or "ceils" of 53 bytes bytes of user data and a 5 byte "header" or address) ATM Individual ATM cells are routed across a network according to the information in the header. As each cell arrives, an ATM switch reads the header and then routes the cell to a designated output port which is part of the ATM network’s virtual path between the caller and receiver. This process occurs so rapidly--ATM switches may handle 5 Gbit/s or more, equivalent to over 75,000 uncompressed voice chan- nels-that a user’s data flow arrives in synch, although the data in neighboring ATM cells may have no relation to one another. ATM thus transforms the old analog, service spe- cific (circuit switched) telephone net- works of the world into digital, service independent (software controlled) virtu- al conduits for innumerable products. The technologies underlying broadband digital networks and their ability to break down barriers between different service markets have profound implica- tions for today’s telecoms operators: [] As the network becomes an end- to-end digital highway, it will offer a new or alternative "route-to-market" for businesses that create or distribute information and entertainment: video rental shops, com- puter software manufacturers, record and CD stores, video game makers, broadcasters, film studios and book publish- ers. Convergence thus implies not only the entry of telecom companies into data networking (computing) and broad- cast, video on demand (VOD), markets, but also the entry of these industries into telecommunications. This can already be seen in several countries as cable TV companies begin to offer telephone services, and computer companies begin to offer telephone switching and network-based soft- ware applications. [] Switching may become both more centralized and decentralized as advances in semi-conductor and software technologies increase the capacity of digital switches while sharply reducing their costs. Thus, on the one hand, new scale economies to switching may enhance the position of established operators. On the other hand, these technical trends suggest that the barriers to entry for new carriers will continue to fall and that more and more switching functions may be resident in users’ terminals and local networks, making trunk transmission networks and switches compara- tively "dumb." See Box 1. For example, in 1993 Synoptics Communications, Inc., a Santa Clara, California company, announced an ATM switch TeleGeography 1994 with the ability to carry, receive and route 5 Gigabytes of information a second--about half the telephone switching capacity of all the central offices in New York City. Also in 1993 Microsoft, the world’s largest software company, announced new application program interfaces (APIs) between its Windows PC operating systems and the tele- phone network. APIs permit software developers to make corporate PBXs into local network switches by adding net- work management software.9 The entry of Microsoft and semiconductor companies, such as Motorola and IBM, into the telecoms market means that it will become easier and easier for any company independently to acquire telecoms switches simply by purchasing the right silicon chips and software modules. [] The increasingly decentralized and plural telecoms architectures permitted by digital technologies also suggests that regulators will require existing operators to make their networks more and more "open" (i.e., accessible) to com- peting transmission and switching providers. In addition, the more the telephone network begins to resemble a dis- tributed computer network, the more computer networks, such as the Internet--which now permits vendor and orga- nization independent two-way access to over three million networked host computers or file servers--become the pre- ferred model for the future. In the telephone world, open access depends upon a user’s ability to plug in a telephone and obtain a dial-tone to call any other terminal on the net- work. For a multimedia network, the test may well be whether someone can plug in an independent file server which can connect with any other data, audio or video file server on the network. [] Widespread adoption of ATM technologies may also lead to pressure for carriers to move beyond current "open network" architectures by unbundling the network into var- ious service-independent transmission and switching mod- ules. Stephen J. Downs and others contend, for example, that an ATM network is best understood as a neutral com- munication platform which enables the provision of multiple services rather than a network which provides multiple inte- grated services,l° In this view, the provision of telephone service or video on demand or high speed data transmission services is logically separate from the provision of ATM transmission, switching and control functions--functions which might be provided by a whole new set of competitors (i.e., non-telecom companies). [] Because the transition to ATM networks will be costly, involving installation of very complex new software based switches, there wilt be a transition period in which narrow- band ISDN facilities and services enjoy favor. Advances in digital compression technologies, allowing full motion color video services to be distributed over a basic rate ISDN cir- cuit may also give narrowband ISDN a boost. Thus, ISDN appears to be coming into its own even as new ATM tech- nologies attract headlines. By the end of 1994, for exam- ple, approximately 625,000 ISDN lines (at 64 kbit/s) will be working in Europe and a like number in the U.S. A primary service driver for these ISDN lines is desktop (PC-to-PC) Bll~lox1. The Evolution of Network Intelligence Toward "Smarter" Terminals and "Dumber" Networks George Glider, the prol/hc U S essay/st and writer on reformation technologies--author of Microcosm (Slmon & Schuster, 1989) and Life After Television {Whittle Direct Books, 1990)--is a leading advocate of unbundhng network "intelligence" (switching and s~gnal processing) from network transmission capacity. In his new book, Telecosm (Simon & Shuster, 1994), Gilder sees th~s trend as an inevitable port of a larger evolution. "In the TV world," Glider states, "you had a stripped- down receiver that could be cheaply manufactured by the mllflons and that relegated almost all the inte!flgent functions to the [broadcasting] station [T]elephone campaniles] had al! the intelligence in the central switches and the telephone was a dumb terminal. Even in the computer world, until the last decade you essentially had smart centralized mainframes and lots of dumb terminals attached Now all those models are collapsing Moore’s law [posJted by Intel’s Chairman Gordon Moore], assuming the doubhng of [semiconductor processing] capabifltles every 18 months, [means] you [soon] have super computer terminals. When you have super computer terminals capable of sending out huge floods of bits, what you really want ~s just dumb bandwidth to accommodate it [or] what’s coiled ’dark hber’ The general evolution of networks will be toward hawng more intelflgence relegated to servers on the edge of the network (which can be used for very speclhc appflcat/ons) and more of the m~ddle of the network will be raw bandwidth Excerpted from "George G/lder, an Interview with Eric Nee," Part II, Upside, June 1994, pp 40-41. The Challenge of Convergence ~3. Estimated Range for Standard Period International Telephone Service Accounting Rates in SDRs by Carrier and Geographic Zone (1993) Carrier Estimated Ac c o u nti n q_R.p~es=(To!From) North America South America Japan India/Pakistan Western Europe "Tier I" European carrier 0.4-0.7 (e.g. DBP Telekom; France Telecom) 1.4-1.8 1.4-1.5 1.5-1.8 0.2-0.35 "Tier II" European carrier 0.5-1.1 1.6-1.8 1.6-1.8 1.7-1.9 0.3-0.45 (e.g. Italcable, OTE) Middle Easter carrier (e.g. Saudi Arabia PTT) 1.4-1.6 1.8-2.2 1.6-2.2 1.4-1.7 0.4-1.0 Southern European/ Mediterranean carrier (e.g. Turkey P]-I-) !.0-1.4 1.6-2.2 1.6-2.0 1.6-1.8 0.4-0.6 Notes: All rates stated m SDRs (In 1993, on average, 1 SDR = U.S. $1.37). On direct intercontinental routes, the settlement rate generally is equal to one-half the accounting rate, i.e., on a direct route where the accounting rate is 2 SDRs, the originating and terminating carrier typically receive 1 SDR each. Settlement rates for intra-European traffic, however, generally are asymmetric with the originating carrier receiving up to 60% of the accounting rate, depending on the share paid to trans=t carriers. Estimates are based on industry and government sources deemed to be reliable, but actual rates on any given route may vary. © TeleGeography, Inc. 1994 J video conferencing, aided by the recent commitment of computer and networking vendors. Part !!1. Responding to convergence It is an extraordinary challenge--geographical, technological and organizational-- for operators to respond simultane- ously to the erosion of national market boundaries and the convergence of voice, data and video services. Only a handful of Heavy Carriers appear to have an integrated strategy and, even then, the initiatives are still tentative. Further, as discussed below, no carrier wishes to move too far too fast into the new world without regulatory guidance. Price Reform Because national tariff disparities have played a large role in the breakdown of the one-country, one carrier market struc- ture for international service, price reform and rate rebal- ancing (to reduce the share of network costs borne by long distance and international services) has formed an impor- tant part of many carriers’ counter-strategies. On average, nominal international call prices have declined approxi- mately 4% annually (in U.S. $) since 1983, suggesting that real prices (after inflation) are probably falling at almost 10% annually. See Appendix 1. Further, a comparison of two-way call prices among the 24 industrialized nations of the Paris based Organization for Economic Cooperation and Development (OECD) showed that in January 1994 only 17 out of 552 routes had outbound tariffs which were double the rate of the inbound tariff (i.e., a call in the other direc- tion). See Appendix 2. Moreover, whereas in 1991 a typical peak-rate (daytime) tariff from the U.S. to other OECD states was always less than the tariff in the opposite direction, by 1994 the U.S. peak-rate tariff was greater on a majority of routes. Again, see Appendix 2. (However, these rate comparisons do not reflect substantial off-peak discount programs and con- sumer calling plans available in the U.S. which often are not available in other countries.) The price reforms initiated by some carriers have been matched by a slow, but continuing, downward adjustment of accounting rates. In t992, the ITU-T Study Group III, for- merly under the International Telegraph and Telephone Consultative Committee (CCITT), adopted Recommendation D. 140 which provides that "accounting rates for interna- tional telephone services should be cost-oriented" reflecting "relevant cost trends" and that these principles "should apply ... to all [routes] on a non-discriminatory basis.’’tl ("Cost-oriented," of course, does not necessarily mean "cost-based," but few countries were willing to support a more strongly worded recommendation.) D. 140 also com- mits countries to implement "cost-oriented" accounting TeleGeography 1994 rates expeditiously, with phased reduction expected over not more than five years, except possibly for developing nations. A profile of current international telephone accounting rates appears in Figure 3. In 1992, the ITU-T tariff group for Europe and the Mediterranean (TEUREM) also adopted a revised account- ing rate Recommendation D.300R, to reflect the cost reduc- tions arising from progressive digitalization of transmission and switching facilities.12 The new recommendation is expected to reduce accounting rates for TEUREM members by up to 20% between 1992 and 1995; this follows a 10% reduction in 1991-1992 under the prior version of D.300R. These actions by TEUREM have made intra-European accounting rates markedly more cost-oriented than inter- continental routes, creating wide disparities in the "landing fees" charged by European countries for regional traffic as compared to North American or Asian traffic. See Figure 4. These disparities, of course, provide a substantial incentive for extra-regional carriers to seek a European presence as well as for various forms of accounting rate bypass.13 Global Alliances Since 1992, market pressure has led the world’s largest car- riers to create several global service platforms. These new joint ventures or partnerships seek to address the threat of private line bypass as well as the demands of multinational I companies for a set of commonly managed services which are standard ("seamless") across national boundaries and networks. The form of these alliances, the structure of the global platforms, and the services on offer generally remain uncertain, in part because the regulatory ground rules for these new trans-national platforms have not been resolved. At this writing, U.S. and European regulators have provi- sionally decided that, on balance, the alliances are in the public interest, provided certain competitive safeguards are adopted, e.g., to prevent preferential interconnection arrangements between global partners where one partner controls bottleneck facilities or holds a dominant position in its home market. As discussed below, however, the ratio- nale for these new competition rules and the "no prefer- ences" and "no special concessions" policy adopted by U.S. regulators are controversial. Special concessions between interconnecting service providers are an accepted practice in the international airline, banking and shipping industries where entry may also be restricted. Indeed, the concessions which airlines pass through to consumers (e.g., frequent flyer mileage credit, airport gate sharing, preferential bag- gage handling) are highly valued by users and enable one global service provider to distinguish its offerings from its competitors. Figure 4. Settlement Rates vs. Estimated National "Access Charges" per Minute For Outbound U.S. Switched Voice Traffic (June 1994) International Private Line Termination Country Settlement Rate (Foreign End) Fst’d PSN Breakout = Total U.K. (BT) $0.32 $0.05 $0.07 = $0.12 Italy $0.32 $0.07 $0.10 = $0.t7 Germany $0.32 $0.06 $0.10 = $0.16 France $0.32 $0.06 $0.10 = $0.16 Spain $0.94 $0.07 $0.10 = $0.17 Chile (Entel) $0.80 $0.04 $0.07 = $0.11 Argentina $0.83 $0.09 $0.07 = $0.16 Japan $0.84 $0.08 $0.21 = $0.29 Korea $0.72 $0.08 $0.13 = $0.21 Australia $0.40 $0.05 $0.15 = $0.20 Notes: Settlement rate =s one-half peak period accounhng rate; 1 SDR = $1.45 (June 1994). Private line cost based on tariff for 1.5 Mbit (T-l) cable circuit except for Argentina where 256 kbit leased satellite c~rcuit used. Each 64 kb~t c~rcuit is used to derive 4 voice channels carrying 6000 minutes per month per channel (7200 minutes for Chile and Argentina). PSN breakout costs esti- mated from local tariffs, TEUREM settlement rates and industry reports. Sources: FCC, AT&T @ TeleGeography, Inc. 1994 The Challenge of Convergence AT&T WorldPartner "Association" AT&T 14o%1 KDD (24%) Singapore Telecom (16%) Unisource ¯ PTT Netherlands ¯ Telia AB ° Swiss Telecom ° Telefonica & Associate Members WorldPartner Company (US) ¯ Manages 61obal Operating Centers: Tokyo; Singapore; South Plainfield, NJ; Atlanta; Washington, DC ° Owns intellectual property for World Source services ¯ Provides "back-office" global billing and account management services ¯ Certifies distributors for World Source services in designated markets **WorldSource Services: Private Lines Virtual Networks I I Fees Distributors J AT&T Unitel Unisource Telstra Singapore NZ HK KDD (Canada) (Australia) Telecom Telecom Telecom Customers Busi~ tinational Companies and **A WorldPartner may join another global alliance. But WorldPartner Co. will not authorize competing distribution of WorldSource services, in territory where Partner has been certified. © TeleGeography, Inc. 1994 First, the Concert, Sprint and, to a lesser degree, Unisource alliances involve the creation of a new, commonly owned company for acquiring half circuit transmission capacity and services which, in turn, will be made available to selected national distributors for providing customized whole circuit (end-to-end) service. This whole circuit model, of course, marks a break from the traditional half-circuit world of cor- respondent carriers today. But, pending relaxation of the rules on resale of international private lines, none of the alliance will offer end-to-end voice services apart from par- Platforms and Services There are four main global platforms under development: Concert, the BT/MCI joint venture; AT&T World Partners; Unisource, also a member of the AT&T Partnership; and the Sprint-France Telecom-DBP Telecom alliance. As shown in Figures 5 to 8, the operating structures of these four ven- tures share several common elements but there are impor- tant differences as well.14 TeleGeography 1994 Joint Venture MC! 24.9% BT 75.1% Intellectual Property Agreement ~~ Technology Concert License Fees| Common Platform (5% of Concert revenue) for Global Products I I BTSyncordia sold to Joint5VentUreyears,$1AgreementBillion I I Concert for $62 million ~,~" Services Agreement Transmission Facilities Fees Global Products** Value Added Services or "telecommunication services management" to a customer in two or more countries (excluding IDD and simple resale of voice) Product Charges & License Fees (3.7% of each distributor’s forecast revenues) DistributorsI MCI BT Customers Multinationa! Companies and **Agreements between the parties restrict BT’s provision of Concert’s Global Products or similar services in MCI’s territory (the Americas); MCI’s provision of such products and services is restricted in BT’s territory (the rest of the world). ticipating carriers’ existing Virtual Private Network (VPN) products, which reflect existing half-circuit, accounting rate arrangements. In the near term, therefore, enhanced ser- vices-data transmission, video conferencing--are likely to be the primary, stand-alone offerings of the new alliance companies. The exclusivity and territorial allocation arrangements adopted by the global alliances vary. BT and MCI have each agreed not to market "core business" services provided by © TeleGeography, Inc. 1994 Concert or otherwise in each other’s territory, (MCI’s terri- tory is the Americas and the Caribbean with BT serving the rest of the world). Breach of this agreement would cause a party to lose significant shareholder rights. (The exclusivity agreement, however, does not apply to Country Direct ser- vices.) The AT&T WorldPartners and Unisource alliances apparently are non-exclusive. For example, a Unisource company may provide competing international services directly in the U.S. or other markets where Unisource and AT&T WorldPartners operate. The Challenge of Convergence Figure 7. Unisource Alliance Telia AB Sweden (25%) Swiss Telecom (25%) PTT Telecom Netherlands (25%) Telefonica Spa~n (25%) KDD Cooperation Agreement Unisource nv SITA Soci~t~ Internationate de T~l~communicat~ons Aeronautique Joint Services Agreement makes SITA leading distributor of Unisource services outside Europe I I I I Unisource Card Unisource Business Unisource Satellite Unisource Mobile Services Networks nv Networks nv Services Unisource Services** International VPNs ¯ Managed bandwidth ¯ Packet switched services ¯ EDI services ° x400 message handling I I Unisource KDD Companies and other local affiliates I Distributors I I I I I I I I TeliaAB Swiss Telefonica SITA PTTTelecom Helsinki Tele~com ; Netherlands Telephone Company ~u Customers Busine~ ltinational Companies and **Distributors of Unisource Services are not precluded from handling other global services, and Unisource members will also distribute AT&T WorldSource services in Europe. 1994 ©TeleGeography, Inc. The limited scope of these ventures’ initial services reflects the fact that the new global platforms do not fit into the cur- rent financial and regulatory mold for public operators. Branded services imply exclusivity. Partnerships imply pref- erences; partners receive better terms and conditions than non-partners. In contrast, the common carrier world of international telephony generally requires carriers to pro- vide a standard service to all interconnecting carriers and, in many countries, to resell basic services to any third party upon request. In the short run, therefore, these new global platforms are likely to lead to a two-tier international market--one where basic switched telephony largely remains in the half-circuit accounting rate world, while "value added" (data, video) and private network service, including voice services which are part of these private nets, are shifted to a new platform. Because international telephony is a large profit center for most carriers, there is little incentive to shift switched voice services to a new joint venture company so long as an oper- ator’s major correspondents don’t seek to bypass it via ISR TeleGeography 1994 print-France Telecom-DBP Telekom Partnership Sprint France Telecom (FT) DBP Telekom (DT) I Global Partnership Board Votes: 33.3% Sprint, 33.3% FT, 33.3% DT Network Operating Group Third Party Votes: 50% Sprint; 50% F and lIT Facilities Facilities Facilities Distributor European Operating Group 33% Sprint, 33.5% FT, 33.5% DT Europe, except France and Germany* Global Services** Global Virtual Private Netwoks (GVPNs); international private lines, frame relay, x.25 data services; credit and debit cards Distributor Worldwide Operating Group 50% Sprint; 50% FT and DT All countries outside US and Europe* Customers Busi~ ionat companies and **Partnership has said that European and Worldwide Operating Groups will be the exclusive vehicle for providing each company’s Global Services in designated markets. © TeleGeography, Inc. 1994 or otherwise. In the longer run, however, the carriers’ new global platforms will almost certainly provide a vehicle for a mix of voice, data and video services. AT&T’s recent domestic partnerships with major software and networking companies provides some indication of how such a multina- tional, multimedia platform may develop. See Box 2. Light Carrier Activities The main facilities-based responses to market convergence described above also have several Light Carrier counter- d parts, including ISR, calling cards, country direct calling pro- grams and call-back. Since 1992, a growing range of carri- ers have begun to make use of ISR. Established operators (Telia, BT, C&W, Singapore Telecom) as well as new carriers (ACC, MFS and Pacific Gateway in the U.S.; fONOROLA in Canada ; and AAP in Australia) have all announced ISR ser- vices. Use of ISR to hub or refile regional traffic via the U.K. and Australia, where one-ended interconnection with the public network is permitted, has proven to be particularly popular. 15 The Challenge of Convergence Box 2. AT&T’s Software Dial-Tone AT&T’s agreements with the Lotus Development Carp, Novel Inc. and Xerox suggest that computer software will be among the key dial-up services offered by AT&T’s future global network Even if a software or wdeo dial tone is not distributed by the company’s global partners, as with today’s/nterna- t/onal call back serwces, entrepreneurs are likely to hnd a way to resell AT&T’s new services around the world. AT&T Network Notes Part of AT&T’s strategy is its March 1994 agree- ment wsth Lotus to make Lotus/Votes ovo/labh on AT&T’s public network. Lotus Notes is a suite of programs for/mprowng the efficiency of work groups and orgamzabons Today the Notes "groupware" programs typically reside on an orgamzation’s central computer hh server, but they were based on the pubhc network, users could access, track, share and orgamze reforma- t/on together even/f they were not part of a porate network or only occasionally part of a net- work team The Notes "work space" consists of several "w/n- daws" or icons on a computer screen (right). Each icon represents a Notes software application or database such as a report to be edited, an e-marl hie, a document archive, customer service records or a price directory An adm/mstrator dec/des which/ndw/duals and work groups may edit or add to a parbcular database, which may include text, spreadsheets, graphics, sound, or wdeo Notes is now used by approximately I milhon people /n 3500 companies But Lotus beheves that Notes has a much wider appeal as a tool for the new "fiat- ter, leaner corporabon" for which horizontal commu- mcabon between changing teams of people has become more common than top-down hnks, Lotus also contends that Notes suits an economy in which more and more people work together despite being m different cities or time zones. For examph, Notes could provide a common customer~support database for geographically dispersed markebng and sales staff Marl-order compames could create a common set of order forms and updated product hterature access/bh by anyone with a PC or a touch tone phone. Following the AT&T announcement, Lotus began to ship "Phone Notes" to enable software developers to incorporate voice messages Into a Notes document. [4hth Phone Notes, one can have a text hh translated by computer into a voice message and then leave a voice response. AT&T’s software dial tone strategy is likely to have at least two benefits for the company First, it should stimulate long d/stance traffic. Most computer users spend far more t/me/n front of their terminals or con- nected to a company database than they do on the telephone Second, by making Notes a gateway for telephone users, the company may be able to convert more basic telephone traffic into computer traffic-- value added and unregulated As James Cosgrove, AT& T’s Vice President of Data Commumcat/on Serwces stated, AT&T’s goal is to "move from provi- sion circuits to hosting apphcat/ons’" on its network, AT&T Netware Connect AT&T’s software dial tone strategy also has led to an agreement with Novel to enable Novel’s Netware, commumcatlons software, which now hnks over 30 mill!on users, to operate over the AT&T network. Instead of hosing dedicated phone lines, o company seeking to establish a new dots hnk to a supplier or a remote site could simply dial mrs AT&T’s pubhc network and communicate with any site that sub- scribes to the new AT&T Netware ComTect Service In add/t/on, AT& T has announced an agreement with the Xerox Corporat/on to deploy new Xerox software which will allow printers and computers to transmit images over telephone lines. AT& T’s new software and video network applications are supported by a partnership of companies com- mitted to providing a choice of compat/bh hardware and software for network access under the AT&T "QVorld Worx" brand In addition to Novel and Lotus, partners include IBM, PictureTel Corp. and Sun M/crosystems © TehGeography, Inc. I994 --G S TeleGeography 1994 The demand for pre-paid telephone debit cards--in effect another form of service resale--has also grown at over 20% each year since 1990. Over 1 billion pre-paid cards are now bought annually throughout the world with an estimat- ed 10%-20% of the cards’ value used for international calls. During the last two years, several U.S. based companies, including USWest and Ameritech, have begun to play a sig- nificant role by marketing so called remote memory cards. These cards may be used to place long distance or interna- tional calls from any telephone (a central computer keeps track of the card’s value) and thus have greater appeal in the U.S. where, unlike Europe and Japan, most pay telephones are still coin operated. 6 The proliferation of different national telephone debit and credit cards by car- riers and non-carriers alike (the market is largely unregulated) has led many users to look for a single card. One option is the new Unisource Calling Card recently announced by the Card Services division of the Unisource alliance. Unisource claims that its card is truly multi-national and will be accepted by all Unisource parent com- panies as well as cooperating carriers in other countries. The Dutch PTT, a Unisource member, has also introduced the first bi-national (Dutch-German) debit card which is accepted by card phones in both coun- tries. To BE A GLOBAL MULTIMEDIA PLAYER IN THE AGE OF THE INTERNE~ ONE NEED ONLY PUT THE SOFTWARE OR VIDEO FOOTAGE WHICH PEOPLE WANT ON A PUBLICLY ACCESSIBLE NET- WORK FILE SERVER AND CHARGE A VIABLE ACCESS FEE. In a similar vein, AT&T has announced a second generation of integrated Home Country Direct services to be offered by a new WorldPlus, Inc. subsidiary. WorldPlus plans to license a global Country-Direct and traveler’s assistance platform to national operators. Wherever WorldPlus services are available, subscribers can dial a toll-free access number, key in personal account and identification data, and then use voice commands (in their chosen language) to chose a ser- vice. Options include in-country or third country calls, con- ference calling and global messaging service, including a personal voice mail box. WorldPlus calls generally will be priced at a premium to IDD: $1.99 per minute within Europe and S3.00 to $4.00 on most routes to Asia and South America. However, for many travelers, AT&T is plain- ly hoping that easy access, standardized billing and pay- ment options outweigh the additional charges. A new MCI WorldPhone card offers similar service to those of AT&T WorldPlus. Begond Call Back The spread of ISR and various card services has also boost- ed the international call back business by increasing access options and reducing return traffic costs. Many call back services rely upon uncompleted call signalling between cus- tomers in high tariff countries and operators in a low tariff one (e.g., a foreign call back subscriber signals a U.S. com- pany that she wants an open line to the U.S. by dialing a designated U.S. number and hanging up after several rings). As discussed in Part IV below, in April 1994, the FCC gave call-back services a "green light" by rejecting AT&T objec- tions to FCC resale applications filed by two such compa- nies.17 Although most call back companies are still U.S. based, several businesses now operate in Europe and Asia, making access to the lowest cost inter- national dial-tone a more and more competitive business. That may prompt some Light Carriers to take on the muF timedia challenge too. I of the For if network technologies and smart terminals permit call-back companies and other resellers to successfully pro- vide international telephone service without owning any undersea cables or satellites, it is likely that a similar vehi- cle can be developed to provide a glob- al "software" or "video" dial-tone. One natural home for such new services is the Internet, the worldwide computer network, rather than the public tele- phone system. However, International Discount Telecommunication (IDT), one original call-back companies, now based in Hackensack, New Jersey, has developed a promising hybrid. To generate more interest in its main international calling business, the company has begun providing free global access (via call back, if desired) to a U.S. based Internet gateway. And, to spread its "pax communications" mes- sage, IDT has begun working with international human rights groups to make available "banned" political and literary texts on a U.S.-based lnternet host computer. The docu- ments can then be dialed up from anywhere in the world.18 Although IDT’s Internet service is anything but commercial, the company has recognized that convergence will require new carriers to be both "Light" and "Soft"--that is, to mar- ket something more than low cost bandwidth (although there will always be a market for that). IDT’s experiment also suggests another lesson: To be a global multimedia player in the age of the Internet, one need not build a glob- al network, a film studio, or a software company. Instead, just put the software or video footage which people want on a publicly accessible network file server and charge a viable access fee, preferably on a per minute basis, much as the telephone companies do today for long distance talk.19 The Challenge of Convergence The lnternet, like today’s public telephone network, also provides a preferable multimedia model for tomorrow because it is two-way. Any telephone terminal can be called by any other terminal; any Internet file server can reach any other. Which, as discussed in the next section, is why some of the most difficult regulatory battles ahead are likely to center on access to the new global multimedia platform. Will the multimedia platforms developed by new global alliances be considered proprietary (private carriers) or must the basic components (international transmission facil- ities, local transport, switching) be unbundled and made available for resale on a common carri- er basis? Will local telephone compa- nies seeking to provide one-way Video on Demand (VOD) services also be required to offer a comparable video platform for two way video conferencing throughout their franchised area? Or, from another perspective, will today’s oligopoly markets for telephony and "free" markets for computer and video services both be replaced by managed competition or will one or the other "pure" model prevail? phone and broadcast interests). Beyond that, outside the larger industrialized countries, the forces of convergence are still weak; as mentioned earlier, in a majority of countries the telephone network (let alone computer networking and cable TV) still connects less than 5 people in 100. Yet, wherever located, as carriers evolve from national telephone utilities to pan-national electronic service businesses, the process will have a global impact; that is one of the m~an- ings of convergence, tt is not too early to ask therefore how well regulators are coping. Our point of departure is the U.S. for several reasons.2° TECHNOLOGY AND ECONOMICS ARE RAPIDLY UNDERCUTTING THE PUBLIC INTEREST RATIO- NALE FOR GEOGRAPHY AND SERVICE-SPECIFIC REGULATION. These questions bring us to the last section of this essay. Part IV. Regulation As national carriers seek to compete in each other’s home markets and digital technologies make telecommunication part of a much larger market for electronic services, regu- lators face choices as difficult as those confronting public operators. Telecom regulators (whether in a ministry or independent agency) have long considered themselves to be sovereign--the gatekeepers of the national market and the arbiter of the boundary lines between one communica- tions service and another. And for much of the 20th centu- ry these boundary lines were respected. The quite different legal regimes (and industries) which exist almost every- where for telephony, television and cable TV testify to regu- lators’ historical power. Today, however, technology and economics are rapidly undercutting the public interest ratio- nale for geography and service-specific regulation. Technology holds out the promises of abundance. And, in the absence of scarcity most consumers (when asked) seem to prefer having the market rather than government decide the players and the products, and where such products can be offered. Regulators have just begun to transform our relatively closed set of compartmentalized national communication markets into a more open global market for digital services. In many cases international regulatory policy has been sub- ordinated to domestic political considerations (e.g., sorting out market access disputes between cable television, tele- Approximately 35% of the world’s inter- national traffic--over 16.5 billion min- utes in 1993ioriginated or terminated in the U.S. American policies regarding the terms on which international carriers can pick up and deliver this traffic accordingly have a large impact on the structure of the global market. Second, the U.S. Government, under President Clinton, has made the promotion of an electronic infrastructure--an informa- tion superhighway--a national and international priority and has agreed upon a set of guiding principles for implementing this agen- da. The vision was articulated in March 1994 by Vice President Gore at the ITU World Telecommunication Development Conference in Buenos Aires: "In this decade ... we now have at hand the technological breakthroughs and economic means to ... create a planetary information network that transmits messages and images with the speed of light from the largest city to the smallest village in every continent ... President Clinton and I believe that an essential prerequisite to sustainable development, for all members of the human family, is the creation of the network of networks ... a Global Information Infrastructure [GIll circl[ing] the globe with information superhighways on which all people can trav- el.,,2] In the United States, Gore continued, "we aspire to build our information highways [the National Information Infrastructure or Nil] according to ... five principles: first, encourage private investment; second, promote competi- tion; third, create a flexible regulatory framework that can keep pace with rapid technological and market changes; fourth, promote open access to the network for all informa- tion providers; and fifth ensure universal service.’’22 These five principles have been seconded by Clinton’s new FCC Chairman, Reed Hundt, an experienced anti-trust lawyer and friend of Clinton.23 Finally, the U.S. is an instructive test bed for regulation in the age of convergence because U.S. companies--AT&T, TeleGeography 1994 MCI, Lotus, Novell, Microsoft, TCI, Time-Warner--are at the center of the new multinational and multi-media alliances. Thus, it is U.S. regulators who are often the first to address the market entry and competition issues raised by these new alliances. Indeed, the importance of these issues to America’s economy has led U.S. law makers to consider a major "rewrite" of the 1934 Communications Act. Although legislative reform eventually stalled in the Senate in September 1994, new reform measures will be reintroduced in 1995 when a new Congress takes office. These bills are likely to promote direct competition between local tele- phone and cable television companies for voice and video services. Future legislation is also expected to phase out the provisions of the 1984 AT&T antitrust consent decree (the MFJ or Modified Final Judgment) which bar the Regional Bell Operating Companies (RBOCs) from providing long distance and international service.24 This new legal blueprint for the U.S. communication indus- try could have a decided effect on the strategies of many North American carriers by the year 2000. See Box 3, On a national level, the FCC and, in certain cases, the Antitrust Division of the U.S. Department of Justice (DO J), will be the primary managers of the U.S. transition. The FCC’s Current Agenda Before looking more closely at the FCC’s current interna- tional policy agenda, some background is helpful. The Hundt Commission has a large policy inheritance; some parts are of continuing value, some might best be revised or scrapped. On the plus side is the FCC’s decade-old decision not to reg- ulate international value-added or enhanced services. In 1985, following the agency’s landmark Computer II order, drawing a regulatory line between telephone (basic common carrier) and computer (enhanced or value added) services, the FCC confirmed that the Computer II rules applied to international telecom services.25 The regulatory "safe haven" created by this FCC order has enabled scores of U.S. Box 3. Who Will Carry North America’s International Traffic In 2001? The North American market--Canada, the U S and Mexico--accounted for almost 40% of the world’s international telephone traffic in 1993. By the year 2000, this traffic base is likely to double to over 35 billion minutes. Trade agreements notwithstanding (basic telecommunication services fall outside of NAFTA, the North American Free Trade Agreement) the market is now divided along national hnes Until at least March 1997, Teleglobe will be Canada’s only overseas carrier Stentor {formerly Telecom Canada) and Unitel, Canada’s second facilities-based long distance carrier, compete with each other to pro- vide service to the U S Telmex has a monopoly over Mexico’s international trafhc until at least August 1996. Thus, astde from calling cards and cross-bor- der private lines, U.S, Canadian and Mexican carriers must now rely primarily upon their own national cus- tomers By 2001, however, there are several factors which may lead North American calls to become more segmented by carrier and product, rather than nation- a/ polnt of origin One factor driving the market realignment/s the new "strategic alliances" between major long distance car- hers In 1993, MCI, which has approximately 20% of the U.S. long distance market, formed an alliance with Stentor, which has over 85% of the Canadian long dis- tance market And in 1994, MCI formed a jo~nt ven- ture with Grupo Financlero Banamex-Accival (Banacct) to b/d for a second long distance and international ser- vices license in Mexico, AT&T, in turn, has acquired a 20% interest (the maximum legally permitted) ~n Umtel and/Tas formed a jo~nt venture with Grupo Alfa to bid for a new long distance license/n Mexico Sprint also has cross-border alliances It has a 25% interest in Call-Net, Canada’s largest long-distance reseller, and a 33% proposed interest in Mexico’s lusacell, a regional cellular Zelephone carrier, which also plans to hie for a long-distance license. These regional alliances have already led to joint cross border (U.S -Canada) busi- ness services and calling plans, more are expected A second driver of change is the rapid growth of mobile phones The largest U.S cellular telephone carrier, McCaw Communications, recently acquired by AT&T, Rogers Communications in Canada and lusacell, have all forged national mobile networks. Each is also party to roaming arrangements for cross-border mobile traffic, Nextel Commumcatlons, Inc, which plans com- peting national mobile service in the U.S has entered into partnerships with CorporaclOn Mobflecom S A. ~n Mexico and Clearnet Communications, Inc, in Canada. Further, these services will soon be supplemented in the U.S by satellite mobile serwces from the American Mobile Satellite Corporation (AMSC) AMSC may also compete with other "global" satellite mobile systems in Canada and Mexico. There are now over 20 million cellular telephones in North America and even if the new U.S Personal Communication Serwce (PCS) grows slowly, there are likely to be over 50 mllhon mobile phones in the region by the end of the 1990s The Challenge of Convergence and foreign companies to resell the capacity of internation- al private lines and offer a wider-and-wider array of "enhanced" services outside the telephone world’s half-cir- cuit, accounting rate regime. Such services generally are subiect to bilateral International Value Added Network (IVAN) agreements.26 And, while these IVAN agreements may become of less value once unlimited resale of interna- tional private lines occurs, they have often let new carriers establish a foreign market presence which may be broad- ened as local service rules are liberalized. IVAN agreements also provide a barrier against re-regulation of value-added services. Second, since the early 1980’s the FCC has permitted unlimited facilities-based competition for voice services. There are now over twenty facilities-based international car- riers in the U.S. including several carriers with maiority for- eign ownership, such as Cable & Wireless Communication, Inc. (U.K.), Telefonica Larga Distancia de Puerto Rico (Spain), and fONOROLA (Canada).27 The FCC’s decision to grant foreign carriers authority to own and operate U.S. based international carriers, and thus to correspond with themselves (i.e., to establish a whole circuit service), has not been without controversy. Such authorizations have all been subject to "competitive safeguards" and reporting (transparency) rules.28 Further, as discussed below, the Hundt Commission has recently revisited this critical issue in two cases, one involving the BT investment in MCI and a second involving an application by a subsidiary of ENTEL- Chile to acquire a controlling interest in a U.S. start-up car- rier, AmericaTel. One of the principal issues raised by the BT/MCI joint ven- ture and the AmericaTel case is the terms and conditions on which a foreign owned U.S. carrier may interconnect with its non-U.S, affiliate when the affiliate has a dominant market position or controls "bottleneck" local exchange facilities. This issue implicates a third inheritance--the FCC’s International Settlement Policy (ISP).29 In simple terms (although the policy has been anything but that to adminis- A third factor reakgnmg the market is competition for value added services MCI, AT&T, Spnnt and other carriers already provide electronic mall and many data services across the three countries These services have been almost entirely deregulated and NAFTA ensures that they will stay so Thus, as new computer- based and interactive (e.g, multimedia) services take a larger slice of consumer telecom budgets, competl- tion along service rather than national lines will inten- sify Finally, the region’s future market will be influenced by new players Additional international licenses for facil- ities-based carriers are likely Lo be granted by Mexico (1996 or 199 7) and Canada (199 7 or 1998) Further, within the next two years new U.S. telecoms legislation likely will permit the seven U.S RBOCs (Regional Bell Operating Companies) to provide inter- national service by the year 2000. Several future scenarios deserve consideration, One sees the evolution of the current MCI, Sprint, and AT&T strategic alliances, with equity holdings enlarged as foreign ownership limitations in Canada and the U.S are relaxed. Another scenario sees the growth of a new "carriers" carrier" for North America, such as Teleglobe or Comsat or LDDS, which would bid to pro- vide low cost international connections for second-tier carriers, cellular providers and resellers. Similarly other "’carriers" carriers" might emerge with service specialties (e g, for enhanced fax, picture phones, television relay service). A third scenario sees the development of two or three new international carriers, possibly based on one or more of the RBOCs Such carriers might have a regional focus (Latin America or eastern Russia and China) and logically might team with independent U S. operators (BC Tel, Hawaiian Telephone, Alascom) or foreign carriers (France Telecom, Telefonica, NTT) which already have invest- ments in the region These scenarios may not be mutually exclusive The volume of international traffic to and from North America probably is large enough to support several more international carriers, especially if they bring with them a share of a region’s long distance calls, typically eight to nine times the size of its international traffic But every scenario is likely to involve some tough regulatory hghts. Long distance tariffs still vary markedly across the region, and the U S, Canada and Mexico still negotiate foreign accounting rates sepa- rately. Further, Canada has a Canada-carriers only policy for routing all domestic long distance and inter- national traffic. Similarly, Telmex is not anxious to share its $ 750 million annual surplus of foreign settle- ment revenues (largely from the U S ) with other regional carriers /n sum, despite commercial and regulatory friction, two or more continent-w/de international carriers are likely to arise by the year 2000, As that occurs, the geo- graphzcal divisions within the North American market will give way to market segments which more closely track the region’s ma]or operators and services ©TeleGeography, Inc 1994 27 TeleGeographg 1994 ter) the ISP requires all U.S.-based international carriers offering switched services to accept: (1) the same (parallel) accounting rate settlements with a given foreign carrier; and (2) the proportionate return of traffic--that is, a carrier may only land inbound U.S. traffic from a given country in pro- portion to its outbound traffic stream on the same route. This policy was mainly adopted to ensure that a monopoly foreign carrier did not obtain more favorable terms by "whipsawing" [playing off) one competing U.S. carrier against another; the policy also ensured that any new U.S. carrier would obtain a "fair share" of inbound (typically more profitable) traffic based upon its U.S. market share. The FCC may waive the ISP if it serves the public interest to do so. But to date, waivers have been of limited scope. Even on routes where there are competing carriers at both ends (approximately 40% of outbound U.S. traffic now terminates on routes where two or more privately owned interna- tional carriers are authorized at the for- eign end) parallel accounting rates and proportional return--that is, managed competition--are still the rule. Communications Act.31 The Act bars foreign companies and individuals from owning more than 20% directly or, absent an FCC waiver, more than 25% indirectly of an entity hold- ing a radio license; the Act does not apply to non-radio authorizations such as those granted under Section 214. Since the mid-1980s, however, the FCC has applied these foreign ownership rules liberally, permitting several U.S. communication companies to sell a majority interest to for- eigners so long as the company’s Section 310 radio assets are first spun-off to a separate U.S. controlled operating company. THE HUNDT FCC HAS A LARGE POLICY INHERITANCE; SOME PARTS ARE OF CONTINU- MIGHT BEST ING VALUE~ SOME BE REVISED OR SCRAPPED, The future of ISP in a competitive global market is, in turn, linked to the future of a fourth policy inheritance: International Simple Resale (ISR). In 1991, after an exten- sive proceeding, the FCC decided that unlimited resale of international private lines was not in the public interest because it might permit carriers, including foreign operators, to bypass inbound U.S. accounting rates, whereas U.S. car- riers might not have similar opportunities in the foreign operator’s home country.3° Thus, even though one-ended resale might help to lower accounting rates and tariffs, for- eign callers and carriers would be the primary beneficiaries. Accordingly, the FCC held that ISR would only be permitted on routes where the foreign country provided resale oppor- tunities "equivalent" to those available under U.S. law. And, to prevent the policy being undermined, ISR carriers would not be able to refile traffic via the U.S. to or from third countries not deemed "equivalent" by the FCC absent the consent of the countries affected. In determining where "equivalence" exists, the FCC stated that it would consider various factors, including local licensing, tariffing and net- work access conditions. Under this standard, by October 1994, the FCC had only approved ISR between the U.S. and Canada and the U.K.; ISR applications to Australia, Sweden, Finland, and Denmark are pending. See Figure 2. Finally, the Hundt Commission inherited the FCC’s alien ownership policies which, unlike the other policies discussed above, have a statutory basis in Section 310(b) of the At the end of 1993, these were the main FCC policies governing the struc- ture of the international services indus- try. Taken together, the policies straddle the old world and the new, seeking to balance the interests of carriers and consumers both in the U.S. and over- seas. On the one hand, the policies firmly commit the U.S. to facilities- based competition for basic internation- al services by U.S. and foreign opera- tors; international private lines may be freely resold for value added services: Yet, on the other hand, the parallel settlement and propor- tional return requirements of the ISP, the limits on resale of international private lines for basic services, including third country refile (hubbing), all insure that competitors primar- ily adhere to the old, half-circuit, accounting rate model where traffic routing and prices are fixed bilaterally. Similarly, although the ISP inspired "competitive safe- guards" placed on international facilities authorizations granted to foreign owned U.S. carriers are designed to pre- clude favoritism among affiliated carriers, where neither car- rier has a dominant position in its home market (i.e., con- trois "bottleneck" local exchange facilities) such policies amount to a form of managed competition or market shar- ing. This raises the question of whether the FCC should now view the ISP and similar "competitive safeguards" as a tran- sitional regime which can be dismantled when competition exists at both ends of an international route or whether such rules have more enduring value. What Happens Next? In September 1993, the tensions between the FCC’s various international policies and the uncertain market access stan- dard for foreign owned carriers led AT&T to petition the FCC for a comprehensive international policy review.32 Despite AT&T’s own global service initiatives, the company’s petition did not ask the Commission to revisit the rationale for the ISP or the related anti-hubbing rules in view of current mar- ket conditions. AT&T took the opposite approach; it urged the FCC to bar access to the U.S. market by any foreign- The Challenge of Convergence owned or "affiliated" carrier ( which AT&T defined as a U.S. carrier having 5% foreign interest) unless the carrier demon- strated that "comparable market access conditions" existed in the carrier’s home market. Further, AT&T argued that no FCC authorization should be granted to such a carrier unless the carrier agreed to vari- ous operating conditions which would: (a) forbid "exclusive arrangements for provision of basic or enhanced services [emphasis supplied]"; (b) implement the current ISP rules regarding parallel settlements and proportional return; (c) prohibit the refiling of U.S. originated or terminated traffic without the consent of the originating and terminating carrier;33 (d) forbid preferential transit rights or other spe- cial concession; and (e) make available to all U.S. carriers tariffed interconnec- tion and distribution arrangements in the foreign carrier’s home market on the same terms as the foreign carrier pro- vides to its U.S. affiliate. Despite its largely one-sided approach to the problems of convergence, the issues flagged by the AT&T petition-- exclusive end-to-end service arrange- ments, traffic refiling, transit terms, the conditions for domestic interconnec- tion--are central to the industry’s transition from one ser- vice paradigm to another. However, it will probably take the FCC and other regulators years to decide whether AT&T’s proposals offer the right solution. THE AMERICATEL APPLICA- TION GAVE THE HUNDT COM- MISSION ITS FIRST OPPORTUNI- TY TO APPLY THE AGENCY’S PRIOR MARKET ENTRY STAN- DARDS FOR FOREIGN CARRIERS. The first indication of how the Hundt Commission might resolve some of these issues was provided in an April 1994 order regarding the legality of U.S.-based "call back" ser- vices.34 The case involved applications by VIAUSA, Ltd. and another carrier to resell the international switched telephone services of U.S. carriers for "call back" and other services. In granting the application, over the objections of AT&T and several South American operators, the FCC ruled that uncompleted international call signalling {sometimes known as "code calling") was not an unreasonable practice under the Communications Act. Further, the FCC stated that inter- national resale, including call back resale, is in the public interest because it places downward pressure on foreign col- lection rates, thereby benefiting U.S. consumers. The FCC’s affirmation of the consumer benefits flowing from call back, arguably pointed the Commission toward weigh- ing user interests more heavily than carrier interests in reviewing the call routing restrictions at the core of the agency’s ISP and ISR rules. Yet, the Commission’s next major international policy decisions--the AmericaTel and MCI orders--appear to reaffirm the status quo, at least pending a more considered review of recent changes in the international marketplace. The AmericaTel Case The AmericaTel application to transfer a controlling interest in a facilities-based U.S. international carrier to ENTEL- Chile, pending since early 1993, gave the Hundt commis- sion its first opportunity to apply the agency’s prior market entry standards for foreign carriers. In June 1994 the FCC approved AmericaTel’s application.35 In so doing, however, the FCC made it clear that even carriers based in a nomi- nally competitive market such as Chile, where five facilities based international carriers have been authorized, may not provide service to and from the U.S. without agreeing to numerous "competitive safeguards" and disclosure requirements--requirements which go beyond past FCC orders. In its AmericaTel order the FCC reiterat- ed that applications by a foreign carriers will be reviewed on a case-by-case basis: the public interest favoring open entry will be balanced against "the potential for discrimination by a foreign carrier against unaffiliated U.S. carri- ers." Discrimination may exist, said the agency, absent market entry opportunities for U.S. carriers on the foreign end, in light of demands by U.S. multination- al firms for end-to-end telecommunication services. "[F]oreign carriers that are permitted to offer end-to-end service on a U.S. international route could obtain an unfair advantage unless U.S. carriers are permitted to do the same." Hence, the FCC said that it would consider, as one factor in its public interest balance, the degree to which Chilean regulation provided U.S. firms with "effective oppor- tunities to compete" with ENTEL-Chile. After a careful review of the Chilean market, the FCC held that U.S. carriers could compete with ENTEL-Chile and that although ENTEL-Chile did not control key international gate- way or long distance facilities, its local affiliate, CTC, did own "bottleneck local telephone facilities" and thus "may have an incentive to discriminate" in favor of ENTEL-Chite and AmericaTel. Consequently, AmericaTel’s application was granted by the FCC only upon AmericaTel’s acceptance of numerous condi- tions. They require that AmericaTel: (a) abide by the ISP; (b) accept "no special concessions" from any other foreign carriers--that is, no "preferential or exclusive operating agreements or marketing arrangements for ... basic services or transit traffic not available on a nondiscriminatory basis to all competing U.S. carriers"; (c) obtain route-by-route and, for Chile, circuit-by-circuit authorization from the FCC; TeleGeography 1994 (d) file copies of any operating agreements that affect traf- fic or revenue flows between the U.S. and Chile; (e) file tar- iffs for all U.S. international services with full cost support data for the U.S.-Chile service (including domestic network access costs); and (f) file annual regulatory progress reports for five years which detail implementation of Chile’s multi- carrier system, provide ENTEL Chile’s market share for tele- phone service on its 15 largest international routes and compare ENTEUChile’s accounting rates on these routes with the Chile-U.S. rate. The BT/MCI Case The "competitive safeguards" and reporting conditions adopted in the AmericaTel case were also reflected in the FCC’s July 1994 decision approving BT’s 20% investment in MCI and the Concert joint venture for end-to-end global service.36 However, the FCC’s action in the BT/MCI case was also affected by the activities of the Antitrust Division of the U.S. Department of Justice (DO J). In July 1994, the DOJ also concluded its year-long investiga- tion of the BT-MCI deal by filing simul- taneously an antitrust complaint in U.S. District Court, Washington, D.C. and a proposed Final Judgment or consent Under U.S. law the proposed Final Judgment is not self-exe- cuting; following public comment, the Judgment must be reviewed by a District Court Judge (in this case Thomas E Hogan) to determine whether it is in the public interest--a process which may not be completed until 1995, following which further court appeals are possible. The FCC’s Order regarding the BT/MCI transaction also breaks new ground. The agency first ruled that BT’s 20% investment in MCI, even when combined with the existing 8% foreign ownership in MCI, was consistent with the pub- lic interest under Section 310 of the Communications Act. IS ALSO SIGNIFICANT. IT PRO- HIBITS MCI FROM ACCEPTING ANY SPECIAL CONCESSIONS FROM BT OR ANY OTHER FOR- EIGN CARRIER. decree (negotiated in advance with MCI and Concert) to resolve the DOJ’s concerns.37 The complaint alleges that MCI’s sale of 20% of its stock to BT, the creation of Concert and related agreements "may substantially reduce competi- tion" in basic and enhanced international telecommunica- tion markets because BT’s home market power and "bottle- neck" local facilities gives BT the ability {and incentive) to favor its U.S. affiliate over other carriers. It is alleged that MCI’s proposed transaction is thus unlawful under Section 7 of the Clayton Act, which bars agreements between compa- nies where "in any line of commerce ... the effect ... may be substantially to lessen competition." The proposed MCI Final Judgment would remedy these alleged violations by (a) requiring MCI and Concert to dis- close to the DOJ and U.S. carriers competing on the U.S.- U.K. route (but not others) the rates, terms and conditions under which they would gain access to BT’s home network; (b) barring MCI and Concert from misusing proprietary information obtained from MCI or BT correspondents; and (c) preventing MCI from providing Concert or BT with U.S. facilities for ISR until the U.K. grants ISR licenses to seven U.S. carriers, including AT&T and Sprint, and BT offers said carriers the "opportunity to interconnect with [its] network on standard, nondiscriminatory and published terms ... without limitation on the amount of traffic carried." The FCC went on to hold that its con- cerns regarding BT’s incentive to prefer MCI and Concert over other carriers would be satisfied, in part, by MCI’s vol- untary agreement to incorporate a new "no special concessions" clause in all of its Section 214 authorizations. The clause would prohibit MCI from accept- ing concessions from any foreign carrier not available to similar U.S. carriers regarding traffic or settlement flows between the U.S. and any foreign coun- try. According to the FCC, the "no special concession" clause would preclude MCI from accepting "preferential or exclusive operating agreements or marketing arrangements for the provision of basic telecommunication services ... including new basic services." The clause would also prohibit MCI and BT from entering into "any arrangement ... for the joint han- dling of basic traffic originating or terminating in third coun- tries" (e.g., transit agreements, hubbing) on terms not available to all competing U.S. carriers. Third, in line with the AmericaTel order, the FCC required MCI to adopt several "competitive safeguards" and report- ing rules. MCI must also obtain BT’s agreement that it will not offer special concessions to Concert regarding the provi- sion of basic services. MCI must disclose publicly its inter- national facilities, revenue and traffic on the U.S.-U.K. routes quarterly and file with the FCC copies of all agree- ments relating to the routing of traffic and settlement of accounts on the U.S.-U.K. route. MCI must also obtain prior FCC approval for BT to change its ownership or voting inter- est in MCI. Finally, at the request of MCI and BT, the FCC did not rule on whether Concert is subject to Title 11 as a common carri- er. This is a crucial point; the parties consider Concert to be a non-carrier furnishing facilities and proprietary services solely to BT and MCl. The FCC decision not to second guess Concert’s status, at least temporarily, thus preserves the structure of the parties’ joint venture and precludes corn- The Chaflenge of Convergence peting carriers from gaining direct access to Concert, as opposed to the basic common carrier services which BT or MCI might furnish to Concert. In sum, read together, the FCC’s AmericaTel and BT/MCI orders leave the FCC with one foot planted in each world. The old international regime, based on the ISP’s parallel accounting rate, non-discrimination and no special conces- sion rules remain in place and, moreover have been but- tressed with new reporting rules. Where a U.S. carrier wish- es to correspond with a foreign affiliate having any market power or "bottleneck" local exchange facilities, the FCC’s view is that local (foreign) regulation and competition laws cannot be relied upon to protect U.S. interests. But the FCC finds some promise in the new world: foreign owned U.S. carriers and new multi-carrier joint ventures should be authorized provided they agree to compete on the same terms open to non-affiliated and non-foreign owned U.S. carriers. Which brings us to the last portion of this essay--the FCC’s approach to the convergence between telecommunica- tion and other services. From Telephony To Video Dial- tones Current FCC regulation, as described above, assumes the communications market is primarily segmented by geog- raphy and services. There is one set of rules for basic common carrier services; world of video dial-tones? Is a "no-special concessions" pol- icy in the public interest between foreign affiliated comput- er game distributors or on-line software vendors or video on demand entertainment producers? Or is such a policy rele- vant only if the service provider is vertically integrated with a common carrier? DOES PROPORTIONAL RETURN OF TRAFFIC MAKE SENSE IN A WORLD OF VIDEO DIAL-TONES? Is A "NO-SPECIAL CONCES- SIONS" POLICY IN THE PUBLIC INTEREST BETWEEN FOREIGN AFFILIATED COMPUTER GAME DISTRIBUTORS OR ON-LINE SOFTWARE VENDORS OR VIDEO ON DEMAND ENTERTAINMENT PRODUCERS? another for enhanced services (over which the FCC has juris- diction but chooses not to exercise it); and a third regime exists for non-common carrier services, notably broadcast- ing. This last regime has no ISP, no anti-hubbing policy, no requirement for non-discriminatory foreign access, no bar to special concessions for international services. Likewise, whereas international common carrier services must be pro- vided on a non-discriminatory basis, under tariff to all--and extensive rate averaging and tariff-based contribution mechanisms make such service almost universally avail- able--broadcasters and other video program distributors, like on-line computer services and e-mail providers, may pick and choose their customers and have extensive pricing flexibility. How is the FCC to reconcile these three or more regimes when the international switched telephone network becomes an end-to-end digital pipe simultaneously providing dial-up voice, video and data services to millions of international users? Does proportional return of traffic make sense in a And what about IVANs; will they provide a regulatory safe- haven for the next generation of international digital video and data services or has their day passed? Can (or should) a level of playing field be established solely by providing non-discriminatory tariffed access to basic transmission facilities? And is this approach viable (or even desirable) when there are mul- tiple basic service providers or when network technologies--e.g., ATM-- make it possible to separate basic net- work control, routing and service func- tions from raw transmission capacity? Not surprisingly, the FCC, like other reg- ulators, has just begun to address these issues. So far as video services are con- cerned, the focus has been almost exclusively domestic. Since August 1992, when then FCC Chairman Alfred Sikes persuaded the agency to authorize telephone companies to provide video dial-tone services, the agency has strug- gled to implement this new policy.38 Yet, video dial-tone services are primar- ily viewed as a one-way offering (e.g., a local exchange carrier downloads movies resident on a central video file server to local subscribers) with limited interactivity. The provision of a two-way, switched video platform which would be functionally the same as today’s switched telephone service has received more limited atten- tion. One reason is that, in theory, most U.S. telephone companies already have a platform for switched video; it is called ISDN and users who want dial-up video links between different national (or international) points are urged to sub- scribe. However, the carriers’ current ISDN offerings are far from ubiquitous and are designed for dedicated site-to-site services--subscribers without an ISDN link are (literally) out of the picture. Even a casual review of the delays which have plagued the FCC’s video dial-tone initiative demonstrate the significant regulatory hurdles which must be crossed to create a more broadly-based switched video platform in the U.S. Since 1992, only five video dial-tone trials have been approved and it was not until July 1994 that the agency finally grant- ed the first commercial video dial-tone application filed by a Bell Atlantic subsidiary, New Jersey Bell (NJB).39 TeleGeography 1994 The NJB authorization reflects the FCCs two tiered basic/enhanced regulatory model. NJB was permitted to establish a basic common carrier platform to distribute video programming to local exchange customers and to offer its own video services on deregulated (enhanced) terms subject to certain safeguards to prevent cross-subsidies (from basic ratepayers) and to ensure that sufficient distri- bution capacity is available to accommodate foreseeable demand from non-affiliated entities. Thus, the FCC required NJB to tariff its basic video dial-tone platform; to create separate accounts for the service (to be disclosed quarterly to the FCC); and to ensure a minimum of 300 video channels for service begin- ning in 1995. The Commission also reserved the right to change these terms and conditions following its review of NJB’s tariff when the agency must resolve the details of allocating NJB’s local exchange costs among the video dial-tone and other regulated services. In short, the FCC’s initial video dial-tone order--which involved service to but 38,000 NJB subscribers--suggests that Likewise, although regulators are most likely to hear repeat- edly from the carriers directly affected by their decisions, these provider interests must be balanced against the con- cerns of the hundreds of millions of telecommunications users whose interests may be only occasionally voiced. And today this balance between provider and user interests also requires taking into account companies and consumers in the broadcasting, computer services and publishing fields as well. THE OLD MODEL OF REGULA- TION IS AS MUCH IN NEED OF A THOROUGH OVERHAUL AS THE BUSINESS PRACTICES OF THE CARRIERS U THEMSELVES. Doing the right thing also means understanding technology --advice strangely absent from the FCC’s T-shirt. For tech- nology has been the great driver of change in the telecommunications mar- ket, changes whose consequences have become ever more far reaching with the deployment of advanced semi-conduc- tors and fiber optic cables. As Vice President Core told the Buenos Aires conference, "the economics of networks have changed so radically that the ... competitive, private market can build much of the GII ... [However], hand in hand with the need for private invest- the domestic task of creating a multimedia regulatory regime (even without new legislation) will be complex and protracted. To some observers, in fact, the disputes between the U.S. telephone and broadcasting/cable TV industries regarding video services seem to mirror those of the 1970s and 1980s between the data services and tele- phone industries. For international carriers, this is not encouraging. Notably, it was not until 1985--almost 15 years after the FCC began its proceeding to deregulate computer communication ser- vices-that the agency confirmed that its the new domestic U.S. regime would apply to international enhanced services. It took another 5 years before IVAN agreements were in place to ensure that these services could actually be pro- vided to such countries as Japan, Hong Kong and the U.K. Thus, if history repeats itself, U.S. companies--whether for- eign affiliated or noticould well find themselves waiting out another 15 to 20 year set of regulatory and court proceed- ings to provide trans-national video services. Part V. Conclusion Shortly after FCC Chairman Reed Hundt assumed office, several FCC staffers were seen wearing T-shirts with the fol- lowing message: "Read the law. Study the economics. Do the right thing." This advice may sound simplistic but it points in the right direction; sound public policy in the telecommunications area involves a question of balance. Purely legalistic or economic considerations may ignore political and commercial realities. ment and competition is the necessity of appropriate and flexible regulation .... This is especially so in the interna- tional arena. In barely a decade, the industry has changed from a supply constrained business with trans-oceanic tele- phone circuits costing $]00,000 or more to an industry with substantial excess capacity on major routes and circuit costs less than that of a wide screen TV. And even before the widespread adoption of ATM, innovative digital trans- mission, s~gnatling and switching technologies will permit international networks to be subdivided on an end-to-end basis into tens of virtual networks for competing carriers and different services, only a small portion of which may be real- time voice communications. In these circumstances, regulators as much as industry face a turning point. The old model of regulation, designed as it was for a monopoly world of nationalistic state-owned "Heavy Carriers" corresponding solely with one another, is as much in need of a thorough overhaul as the business practices of the carriers themselves. However, industry has been quicker off the mark. Spurred by the demands of users, a new breed of "Light Carriers" and the recent explo- sion of international traffic on the Internet, the world’s major carriers have forged a series of new joint ventures which have the potential for delivering substantial price and ser- vice benefits to users far beyond the multinational compa- nies which have been initially targeted. The question of the day is whether regulators will meet this challenge largely by falling back on traditional rules; by making marginal changes, primarily to ensure that one car- The Challenge of Convergence rier is in no better position to exploit the future than anoth~ er--in short, on managed competition. Or will regulators give carriers greater freedom of action to pursue new service models even though this may disadvantage some carriers or ~et other national champions fa~ by the way. Similarly, even if "competitive safeguards" have a role so long as affiliated foreign carriers have the capacity to leverage bottleneck facilities, will regulators have the courage to rethink these rules when affiliated foreign carriers lack market power or bottleneck facilities in any market? Beyond that--and perhaps most importantly in the longer run--will regulators take a fresh look at the relevant inter- national service rules when the operators involved provide solely enhanced services which are indistinguishable, bit for bit, from computer-based or video-based services which have long been deregulated? The answers to these ques~ tions are far from academic. They affect tens of billions of dollars of investment and an equal level of consumer expen~ diture. There are also very large employment consequences which have become of pressing concern to almost every government. It is time, in short, for a reappraisal of prevailing interna~ tional telecommunication regulation--a reappraisal that is at least as bold as that undertaken by the industry~s users and carriers. The one billion telecommunication subscribers who will be connected to the world’s networks by the year 2000 deserve no less. TeleGeography 1994 ix 1: Trends in the global information economy owth trends, 1984-1993 and projections, 1994-2000 Indicator Calls (Bn) Estimated call length (rains) Minutes (Bn) Per subscriber Revenue (USSbn) Price per MIFF ($) Main lines (M) Mobile subscribers (M) Historical trend Base case CAGR CAGR 1984 1993 1984-93 2000 1994-2000 2.7 12.3 18.4% 26.9 11.8% 4.9 4.0 -2.2% 3.0 -4.0% 12.9 47.5 15.5% 80.6 7.8% 33.2 78.3 10.0% 91.0 0.2% 19.5 50.2 11.1% 63.9 3.5% 1.51 1.06 -3.9% 0.79 -4.0% 390 607 5.0% 885 5.5% 0.2 32.7 73.7% 168 26.4% 10% growth 12% growth CAGR CAGR 2000 1994-2000 2000 1994-2000 30.9 14.0% 34.9 16.1% 3.0 -4.0% 3.0 -4.0% 92.6 10.0% 104.8 12.0% 104.6 4.2% 118.4 6.1% 83.0 7.5% 94.0 9,4% 0.79 -4.0% 0.79 -4.0% 885 5.5% 885 5.5% 168.4 26.4% 168.4 26.4% Expansion (MITT) due to: 1984 1993 1984-93 2000 1994-2000 Network expansion 570 2,441 30.7% 4,101 69.6% Organic growth 832 2,670 69.3% 620 30.4% Total 1,402 5,111 100.0% 4,721 100.0% 2000 1993-2000 2000 1993-2000 4,406 54.0% 4,775 45.1% 6,670 46.0% 11,691a 55.3% 11,077 100.0% 16,465 100.0% Note: 1984-1993 based on reported data. 1994-2000 based on ITU forecasts. Traffic growth due to network expansion implies extra traffzc generated by new subscribers. Orgamc growth implies extra traffic generated by existing subscribers. alncludes estimate for traffic generated by new mobile subscribers. Source: ITU/BDT telecommunication indicator database and ITU estimates. © ITUFFGI 1994 3OO 250 200 150 100 5O Growth Trends in Indices Calls 1992 = 100 tes I I 1 I I I 1 I I I I 1 I 1 1 I 1 I 1984 1986 1988 1990 1992 1994 1996 1998 2000 Note: 1983-92, reported data; 1993 ITU estimates; 1994-2000 ITU projections for the "base case" (above). Source: ITU/BDT telecommunication indicator database plus ITU estimates. © International Telecommunication Union/TeleGeography, Inc. 1994 d The Challenge of Convergence TeleGeogrophy 1994 Notes 1. Gregory C. Staple, "Winning The Global Telecoms Market: The Old Service Paradigm And The Next One," TeleGeography 1992 (IIC, London, 1992) p. 32. The Heavy Carrier/Light Carrier dichotomy has been adopted elsewhere. See e.g., Keith E. Bernard, "New Global Network Arrangements," Telecommunications Policy, Vol. 18, No. 5, July 1994, p. 378; US Telecommunications Serwces ~n European Markets, U.S. Office of Technology Assessment (GPO, Washington, D.C. 1993) p. 37. 2. See e.g., "Refile and Alternate Calling Procedures: Their Impact On Accounting Rates And Collection Charges," Organization For Economic Co-operation And Development Working Party On Telecommunication And Information Services Policies, DSTI/ICCP/TISP/AH(94) 1 (4 May 1994). "The net effect of refile and alternate calling procedures [e.g., call back] is to impose downward pressure on the prices relevant to the provision of international telephone services (collection charges and accounting rates) by arbitraging away market distor~ tions which result from the lack of competition. These services introduce an element of competition for international telephone service. As such, refile and other alternate calling procedures play a useful role in the market place and need to be encouraged." Id. at pp. 16- 17. See also M. Cave, "The Economic Consequence of the Introduction And Regulation of International Resale of Telecommunications Services," DSTI/ICCP/TISP/AH (94) 2. 3. The term GII is derived from the U.S. National Information Infrastructure (Nil) program. See generally, "Remarks of Vice President AI Gore," international Telecommunication Union, World Telecommunications Development Conference, Buenos Aires, March 21, 1994. See also "The National Information Infrastructure: Agenda For Action," U.S. Department of Commerce, Information Infrastructure Task Force, September 15, 1993; "Public Hearing and Request for Comments on the International Aspects of the National Information Infrastructure," 59 Fed. Reg. 31979 (June 21, 1994). 4. Technological convergence has also brought cellular telephone and other "mobile" services directly into competition with wireline offer- ings. In many countries consumers can now choose between wire and wireless service for at least part of their local and long distance telecommunications. The development of transnational mobile services--satellite phone service, cross-border cellular roaming--is of spe- cial significance. Such services may bypass the existing facilities and settlement arrangements among wireline carriers, thus pointing the way toward alternative settlement and access charge regimes for the international telephone industry as a whole. See, e.g., Tim Kelly, Michael Minges and Gregory Staple, D~rect~on of Traffic, (ITU/TeleGeography, Inc., Geneva, 1994), on pp. 30-31. See also "Green Paper on a common approach in the field of mobile and personal communications in the European Union," COM (94) 145 Final, 27 April 1994; Gregory C. Staple and Rob Frieden, "The New Space Race," Infrastructure Finance, June-July 1994, pp. 53~55. 5. A review of global telecommunications appears in World Telecommunication Development Report 1994 (ITU, Geneva, 1994). 6. See Revised CCITT Recommendation D. 1, "General Principles For the Lease of International (Continental and Intercontinental) Private Telecommunication Circuits and Networks (ITU, Geneva, May 1991). 7. For an introduction to ISDN, see A.M Rutkowski, Integrated D~g~tal Serwces Network (Artech House, Norwich, MA, 1985) and Jacques Arlandis, "ISDN A European Prospect", in Charles Steinfeld et al. Eds., Telecommunications in Transition (Sage Publications, Inc., London, 1994). Major ISDN standards can be found in CCITT Blue Book Volume It, Fascicles 11.2 and 11 3 and Volume III Fascicles 111.8 and 111.9 (ITU, Geneva, 1989). 8. For an introduction to ATM, see Stephen J. Downs, "Asynchronous Transfer Mode and Public Broadband Networks, the Policy Opportunities." Telecommunications Policy, Vol. 18, No. 2, March 1994, p. 114; Stewart Fist, "Getting to Grips With the Broadband Future," Austrahan Communications, September 1991, p 61. 9. World Telecommunications Development Report, supra, note 4, p. 34 10. See Stephen J. Downs "Asynchronous Transfer Mode Telecommunications Policy, Vol. 18, supra, note 8. 11. See CCITT Circular No. 169, COM Ill/ST (October 7, 1992). 12. See CCITT, Report TEUREM-R3, (October 1991). 13. AT&T, for example, has recently proposed that U.S. carriers be granted the same facilities-based rights as U.K. carners to gain the benefit of TEUREM rates for U.K. originated traffic. AT&T "Comments" In re ACC Global Corp and MFS International, Inc., File Nos. ITC- 93-035, ITC-94-049 and 051, filed August 3, 1994. 14. These figures are based on industry interviews and In re Request of MCl and BT, Declaratory Ruling and Order, File No. INS-P-95- 015, FCC 94-188, released July 25, 1994; NewSource, Unisource customer magazine, Spring 1994, No. 2; and Telecommunications Reports, May 31, 1993, p. 23 and June 20, 1994, p. 9. 15. For details on the U.S. ISR regime, see fONOROLA/EMI Order, 7 FCC Rcd. 7312 (1992); for Canada, see Teleglobe Canada Inc. Regulation after the Transitional Period, Telecom Decision 91-21 ( 1991 ) (CRTC Decision 91-21 ); for the U.K., see Competition and Choice: Telecommunications Policy For the 1990s, CM 1460 (HMSO, London, 1991) p. 15; for Australia, see "Austel’s Investigation Into International Resale, Final Report." Austel, 28 March 1994. On September 19, 1994, the FCC granted the apphcation of two carriers to provide ISR services between the U.S. and U.K. subject to the U.K. designation of the U.S. as equivalent for 1SR purposes. 16. See e.g., International Telephone Cards, Issue 13, July 1994. 17. In re VIA USA Ltd. et al., Order, Authorization and Certificate, FCC 94-86; released May 11, 1994. 18. See Peter H. Lewis, "On The Internet, Dissident Shots Heard Round The World," The New York Times, June 5, 1994, Section 4, p. 1 ; tDT’s D~g~tal Freedom Network can be accessed via gopher.iia.org. 19. The video analogue of IDT’s "human rights" file server ~s suggested by a review of the new Karaoke "video jukebox" produced by JVC. The "jukebox" holds up to 100 CDs with 74 minutes of full motion videos. "Imagine for a moment that the system’s video serv- er--a high speed robot that takes just five seconds to go from one song on one disc to another--is at the end of a phone line ..." Bob The Challenge of Convergence Johnstone, "They Feel Like Singing," Wired, July 1994, p. 30. See also Nicholas Negroponte, "Prime Time Is My Time." Wired, August 1994, p. 134. "In the near future, individuals will be able to run video servers in the same way that 57,000 Americans run computer bulletin boards today. That’s a television landscape ... like the Internet. [Not] [plaint to multipoint ... [but] multipoint to multipoint. 20. For an insightful review of the challenges facing U.K. regulators, see lan DT Vallance, "Managing A Hybrid," Financial Times World Telecommunications Conference, London, 7 December, 1993. See also Marcel Haag and Hans School, "Telecommumcations Regulation and Cable TV Infrastructure in the European Umon," Telecommunications Pohcy, Vol. 28, No. 5, July 1994, p. 367. 21. "Remarks of Vice President AI Gore" supra, note 3. 22. Ibid. 23. See e g., Testimony of Reed M. Hundt, Chairman, FCC, before the House Subcommittee on Telecommunications, "On the Global Information Infrastructure And the Role of Satellites." July 28, !994. 24. In June t994 the U.S. House of Representatives passed H.R. 3626, the "Antritrust and Communications Reform Act of 1994." It permits any Regional Bell Operating Company (RBQC) to provide international services once the FCC finds that it is in the public interM est and the U.S. Department of Justice rules that the RBQC would not use its power over local exchange facilities to ~mpede competition. See Congressional Record--House, p. H5189, June 28, 1994. The U.S. Senate’s proposed reform bill, S. 1822, made it more difficult for an RBQC to provide international and other long distance services. Among other things, S. 1822 delayed an RBQC’s entry into these markets for 2 years or unbl the FCC adopts new universal service regulations; the FCC must also find that the RBQC prowdes nondisM criminatory network access for competitors, including dialing panty and number portability. See Senate Report, No. 103-367 on S. 1822, September 14, 1994. In late September, however, following a bitter division of opinion between the RBOCs, the principal sponsor of S. 1822 decided not to introduce the bill to the full Senate, thus halting legistative reform until at least 1995. 25. The FCC’s definition of enhanced services is codified at 47 Code of Federal Regulation (CFR) Section 64.702. The application of this definition to international services was confirmed in GTE Telenet Communications Corp., 100 FCC 2d 776 (1985). 26. See generally, "International Value-Added Network Services--An Introduction", NTIA pubhcation TM-90-256 (U.S. Dept. of Commerce, Washington, D.C., 1990) 27. See e.g., Telefonica Larga Distancia de Puerto Rico, 8 FCC Rcd 107 (1992). 28. Beyond that,the FCC has yet to act on a 1991 application by U.S.-based Cable & Wireless Communications, Inc., to acquire facili- ties for the provision of switched voice and private line service with affiliated companies in the U.K. and Hang Kong. See FCC Public Notice, Report No. 1-6642, released December 26, 1991, re File Nos. I~%C-92-065 and -066. An application by the U S. affiliate of Telefonica for connecting undersea cable circuits to Spain has also been deferred. See In the Matter of Telefonica Larga Distancia de Puerto Rico, Inc.; Rle. No. I-T-C 92-242 et al., released August 8, 1994. 29. See e.g., Implementation and Scope of Unlform Settlements Policy for Parallel International Communications Routes, Report and Order, CC Docket No. 85-204, 51 Fed. Reg. 4736 (Feb. 7, 1986); 47 CFR Section 43.51. 30. International Resale Order, 7 FCC Rcd 559 (1992); Order on Reconsideration, 7 FCC Red 7927 (1992). 31.47 United States Code (USC) Section 310(b) 32 "AT&T Petition for Rulemaking, Market Entry and Regulation of International Common Carriers W~th Foreign Affiliation," RM-8355, filed September 22, 1993. 33. AT&T later advised the FCC, however, that it does not favor placing routing conditions on U.S. resale carriers. "[M]arket forces have already shown routing conditions to be ineffective and unenforceable in practice." AT&T "Comments," supra, note 13 at 6. 34. See VIA USA Ltd., et al., supra, note 17. Notably, the FCC’s Order did not rule on the legality of "call back" under international law (i.e., the ITU’s International Telecommunications Regulations (ITR)) because the issue had not been properly raised by AT&T. See e.g., International Telecommunications Regulations, Final Acts of WATC-88, Articles 1.1, 1.5 and 3.3 and Appendix 1. (ITU, Geneva, 1989). For this and other reasons, on September 2, 1994, the FCC sa~d it would reconsider the VIA USA order and invited a new round of pub- lic comments. 35. In the Matter of AmericaTel Corporation, File No. I-%C-93-160-TC, released July 11, 1994. 36. Declaratory Ruling and Order, supra, note 14. 37. U.S.v. MCI Communications Corp. et aL, Case No. 94CV01317, (D.D.C., filed June 15, 1994). For further details, see "Proposed Final Judgment and Competitive Impact Statement United States v. MCI Communications Corporation and BT Forty-Eight Company (’NewCo’)," 59 Fed. Reg 33009 (June 27, 1994). European competition rules are also relevant to the BT/MCI joint venture and simi- lar alliances. See "Notice Pursuant to Article 19(3) of Council Regulation No. 17 and Article 3 of Protocol 21 of the EEA Agreement Concerning a request for negative clearance or an exemption pursuant to Article 83(3) of BT-MCI," (94/C 93/03) and "1991 Guidelines On The Application of EEC Competition Rules In The Telecommunications Sector," 91/C233/02, Official Journal of the European Communities, C233/2, 06.09.91. 38. The FCC adopted ~ts video dial-tone policy in Telephone Company-Cable Television Cross-Ownership Rules, 7 FCC Rcd 781 (1992}. Petitions for reconsideration of this order and court appeals are still pending. However, the U.S. Court of Appeals recently upheld the FCC’s determination that neither a telephone company providing "video dial-tone" service nor a video programmer that uses the service to reach subscribers is required to have a local cable television franchise as a cable operator. National Cable Televisions Association, Inc,, et al. v. FCC, D.C. Cir. Nos 91-1656 et al. Slip Opinion dated August 26, 1994. 39 See "In re New Jersey Bell Telephone Company," File No. W-P-C-6840, FCC 94-180, released July 18, 1994, TeleGeography 1994 © TeleGeography, Inc. 1994 COMPETITION IN EUROPE By Graham Finnie B is to catch up with other regions, y European standards, the past year was an annus mirabilis for the telecommunications services mar- ket. Decisive action changed the terms on which most services wilt be provided across Europe, whether by established carriers or newcomers. However, much still remains to be resolved if Europe [] Sweden: Parliament passed a new telecommunica- tions act in June 1995, allowing across-the-board competi- tion with virtually no restrictions of any kind. [] Denmark: The government agreed to allow simple resale of domestic voice telephone services from mid-1994, and began negotiations to allow international simple resale to selected countries. Change came on two fronts: regulatory and commercial. On the regulatory side, the European Commission, the policy- making arm of the European Union, won a series of impor- tant victories, paving the way for across-the-board compe- tition before the decade is out. [] Finland: Facilties-based competition in international services began in July 1994 with the launch of Oy Firmer International AB, owned by Finland’s local telephone com- panies, and Telivo, a subsidiary of a local power company. Domestic services are already liberalized. First, in the summer of 993, it won the agreement of the Council of Ministers for its plan to end existing state monop- olies over basic telephone service.1 The agreement calls on all member states to liberalize all voice services by January 1, 998. As of this writing, only Greece was expected to ask for an extension, to 2003. Though the European Parliament rejected the plan in July 1994, its rejection was widely seen as part of a long-standing power struggle between it and the European Commission and is unlikely to prevent liberaliza- tion from taking place. Second, in mid-1994, the Council backed a plan outlined in an earlier Green (discussion) Paper for complete liberaliza- tion of mobile communications services.2 Ministers did not (as many expected) resist a clause that would allow mobile operators to install their own infrastructure. Thirdly, the Commission succeeded in getting a higher prior- ity for its final goal, the liberalization of physical telecom- munications infrastructure (transmission equipment and wires). A committee of industrialists set up by European Commissioner Martin Bangemann and led by Olivetti chair- man Carlo de Benedetti called publicly in June 1994 for fur- ther liberalization, including the freedom to build infrastruc- ture.3 A study on the issue is likely to be released in late 1994 or early 1995, and the odds are high that limited lib- eralization of infrastucture will begin within two years. Parallel with these EU-wide developments, there was much national activity, mostly suggesting that the tide has turned decisively in favor of greater competition: [] Germany: At least six compa- nies, including IDB Worldcom, BT and Sprint are providing both voice and data services to corporate cus- tomers. [] Netherlands: The government agreed to allow a sec- ond provider of infrastructure (based around the electric and gas utilities, railway companies and cable TV providers). [] United Kingdom: The government licensed more than 10 providers of public telecommunications services and resellers in the wake of its 1991 decision to end the BT/Mercury domestic duopoly; [] France: Piecemeal liberalization continued apace; the government permitted Compagnie Generale des Eaux ICGE) and Societe Francaise du Radiotelephone (SFR) to respec- tively run trial services and build their own infrastructure for services. Meanwhile, long-forecast changes to the industry’s supply structure began to take shape. Apart from the various super-carrier alliances, guerrilla action around the edges of the existing regulatory rules resulted in a number of impor- tant breakthroughs. Graham Finnie is a freelance writer and researcher based in London, England, and consultant editor to Communications Week International. He can be reached by e-mail at 1002 72.372 @compuserve. corn, or by fax at (+44) 322 433733. The most highly-publicized was the proposed agreement between the self-styled European VPN Users Association (EVUA) and AT&T/Unisource. The EVUA, a loose grouping of large users such as Rank Xerox, Du Pont and ICI, sought large discounts on PSTN tariffs (up to 40%) in return for routing traffic via designated carriers. Key to that and other deals was the European Commission’s narrow definition of voice telephony: "the commercial provision for the public... of real-time speech via the public switched network...".4 This meant that closed-user groups could be offered voice telephone service. Though no formal definition of closed-user groups exists in EU law, the Commission’s competition directorate (DG4) told the EVUA that its plans were legal.5 ...continued on page 42. © TeleGeography, Inc. 1994 New Carriers in Asia NEW CARRIERS iN ASIA By Stuart Corner telecommunications market in the Asia Pacific region is short on common denominators. The range of infrastructures, regulatory regimes, service providers and user sophistication levels is as diverse as the cultures. Vietnam, for example, with less than 1 telephone per 100 people relies on Australia’s Telstra to provide its international services. More devel- oped, totally digital networks like those in Singapore and New Zealand provide telephone services to 40% and 44% of citizens respectively. New Zealand has probably the world’s most laissez faire policy on regulating telecommuni- cations. Singapore Telecom still enjoys a monopoly over both fixed and cellular services. The region’s enormous demand for services, from the most basic to the most sophisticated, has produced a broad spec- trum of service providers. They include alliances of the mainstream international carriers, the traditional "Heavy Carriers," through managed network and value added set~ vice operators to the "Lightest" carriers of all: callback operators, using a minimum of hardware to exploit the anomalies in international call rates. Callback operators are primarily meeting consumers’ sim- ple needs for cheaper communications. Corporations have more complex demands, such as uniformity of service in every country where they operate. These multinational cus- tomers want to be able to run many applications at once, including video and large file transfers. They require band- width on demand, reliability, security, and, above all, coor- dinated and centralized ordering, network management and billing. Australia’ s liberal but carefully structured regulatory regime has given Australian customers exposure to a wider range of carriers than almost any country in the region. In Australia only transmission links are reserved to the two licensed fixed network carriers, Telstra and Qptus. Any organization can lease capacity on these links, install its own switches and provide services. Alternatively it can buy end user services from the licensed carriers and repackage and resell these in any way that is profitable. Any international service is per- mitted, including international simple resale ISR with circuits connected to the network at one or both ends. ~’ Concert in Sydney BT has already established a con- siderable presence in Australia as a quasi carrier. In late 1992 it won a contract from the government of the state of New South Wales (NSW) to provide communi- cations for all government enterprises and has now installed a network of three Nortel DMS central office switches and about a dozen Nortel Meridian PABXes interconnected by lines leased from the two licensed carriers; the network serves over 100,000 users. This strong presence and Australia’s liberal legislation made the NSW system the nat- ural choice to be a key network management node in the BT/MCI Concert network, and the first in the region in which Concert’s global managed virtual private network (VPN) services are available. The one major carrier that has so far stood aloof from the new global alliances is Cable and Wireless (C&W); Chairman Lord Young disparaged the concept at the company’s annu- al meeting this year, saying that forming an international alliance was becoming "almost a test of the chairman’s viril- ity." C&W’s subsidiaries think otherwise. Mercury Communications in the U.K. (majority owned by C&W) wanted to join the AT&T led WorldPartners but was dis- suaded by Australia’s second carrier Optus (25% owned by C&W) and IDC, the Japanese international carrier in which C&W is a leading shareholder. Both were concerned at the added advantage it would give their WorldPartners member competitors, Telstra and KDD. Yet, in the summer of 1994, Hongkong Telecom, another C&W affiliate, became a WorldPartner distributor. BT and MCI are not free from such possible conflicts either, Prior to their alliance MCI owned 25 percent of an Australian "light carrier," AAP Telecommunications. MCI’s shares are now being sold back to AAPT. AAPT provides standard telephone and virtual private network services to corporate clients by using its own switches and lines leased from the carriers. Although BT does not compete for busi- ness in the corporate market at present, it is expected to do so soon, and certainly after Australia’s fixed network duop- oly ends in 1997. Stuart Comer, the publisher of Exchange, Austrafia’s leading communications news weekly and Telenews Asia, is based in Rozelle, New South Wales. He can be reached by fax at (+61) 2-818-2294 Moving towards the "lighter" end of the scale, Australia saw for a while one of the first examples of a company providing international simple resale using leased lines. The company, MIG International, targeted the domestic market, mostly South American nationals resident in Australia wanting to call friends and relatives at home. Its calls were delivered to South America via a U.S. based company, International Telecommunications Corporation (ITC). MIG ran into difficulties, however, perhaps because Australia’s IDD 39 TeleGeography 1994 rates are already amongst the lowest in the world. For other Light Carriers the finance sector with its demands for sophis- ticated reliable and secure worldwide communications rep- resents a much better opportunity to exploit relaxed regula- tions and add some real value to basic carrier capacity instead of simply cutting costs. One such company, Saturn Global Networks, also had its genesis in Australia. The com- pany is a wholly owned subsidiary of U.K. money brokers MW Marshall and was set up in 1992 to take over Marshall’s existing international network and expand it as a profitable business in its own right. This background enabled Saturn to gain the business of major Australian customers in the finan- cial market sector. Saturn Global Networks provides ser- vices to multiple international destina- tions at speeds up to 64 kbits/sec. In Australia, the company operates under an international service provider class license. It has recently expanded in to Hong Kong, Europe, and the U.S., where it has partnered with WorldCom. Singcom in Melbourne Singcom is another international service provider operating under a service provider’s license. The company is majority owned by Singapore Telecom International, and is able to offer an end to end service between Melbourne and Singapore by virtue of ST’s monopoly over international services out of Singapore; Telstra has no reciprocal rights in Singapore. Singcom offers store and forward fax services and worldwide packet switched service. It acquired an Australia wide X.25 network which provides services to anywhere in the world via Telepac, the Singapore packet-switched network. Telecommunication Affairs (Ofta), has said the services are legal. City Tetecom’s network enables it to undercut HKTI’s IDD rate to Canada, the USA, Australia and some European nations. But because of the stricter regulations and less advanced networks in China and other developing Asian countries, the company has been unable to offer its services to most other countries. [] Japan: International simple resale is now under con- sideration but is unlikely to be permitted before 1996. However, a recent decision to allow Japan’s domestic satel- lite operators to provide international leased line services is expected to give a IN COUNTRIES WHERE boost to Special Type II international carriers (international value added ser- DEMAND IS STRONG BUT REGU- vice providers) and to greatly expand their range of opportunities. At present LATION STRONGER, CANNY 27 such companies are registered with OPERATORS ARE FINDING WAYS the Ministry of Posts and Telecommunications (MPT) to provide AROUND THE REGULATIONS voice, data and image services. WHICH ALLOW THEM TO OFFER [] India: India’s bureaucracy, is sti- fling liberalization of the country’s INTERNATIONAL SERVICES AT telecommunications. Although the Indian government announced in May CHEAPER RATES. 1994 that at least one additional license would be granted for local tele- phone networks on a regional basis, with up to 49% foreign ownership permitted, no grants have been finalized yet, and international service will remain a monopoly of Videsh Sanchar. Over 60 bids for private sec- tor lease financing of new exchange lines are also pending. Further, India is still without cellular service (though license awards have been announced) and has only recently grant- ed paging licenses. In other countries where demand is strong but regulation stronger, canny operators are finding ways around the regu- lations which allow them to offer international services at cheaper rates. A brief overview of recent developments fol- lows: [] Hong Kong: In theory Hongkong Telecom (HKT) has a monopoly on international telephone calls but is never- theless facing competition. City Telecom, which began in September 1992, has become the major long distance call competitor to Hongkong Telecom and, by the end of 1993, had 61,000 customers. It circumvents Hongkong Telecom’s monopoly on facility-based international services by using a toll-free number out of Canada and using "an indirectly leased line" from HKT. Similar services are offered by AT&T and others. The local regulator, the Office of Despite the hurdles facing new facilities-based telephone networks, "Light Carrier" services are showing signs of emerging. In August 1994 a local company, Compton Graves, announced plans to launch a fax store and forward in conjunction with SDI of the U.S. and fax broadcast service in conjunction with Faxcast of the U.K.. Faxcast uses satel- lite capacity or capacity in the vertical blanking interval of a normal TV broadcast. Meanwhile U.S. carriers AT&T and Sprint are capitalizing on India’s high international call rates by offering cut price col- lect call rates to the U.S., and have waged a small price war in recent months. Sprint is also involved in electronic mail services with local company RPG Telecom. Private Vsat networks are coming into operation, but the Government has imposed very stringent guidelines on the services, including requiring operators to get prior approval New Careers in Asia for tariffs. Seven potential operators of services have emerged, but as of August 1994 only two had reached for- mal agreements with the government: Comsat Max, 50 per- cent owned by Comsat International and Hughes Escorts, part owned by Hughes Corporation. [] Malaysia: Most value added services are open to provision by private operators, subject to grant of a license. However the telecommunications minister indicated in July ]994 that licenses could be restricted to operators offering new services for fear of overcrowding the market. When the country’s first EDI service, Dangangnet, was launched in August 1993 the government said no others would be per- mitted because the country was "too small for more than one." EDIM, which operates Dangangnet, is a consortium formed under the auspices of the national Chamber of Commerce and has a 5 year license. There is no audiotex (pay per call) type service in Malaysia at present but in June Telekom Malaysia announced the for- mation of a joint venture with PS Ultgevrijbedrijven BV of the Netherlands to develop audiotex in Malaysia. The cel- lular operator Celcom had held an audiotex license for over a year but could offer services only to cellular callers because of difficulties in negotiating a fixed network inter- connect agreement with Telekom Malaysia. International packet switched data services are offered through a joint venture between BT’s Global Data Service and Telekom Malaysia but are accessible only from Kuala Lumpur and Petaling Jaya. [] Philippines: To improve its inadequate telecommuni- cations infrastructure (1 million lines for 65 million people) the Philippines Government has licensed seven internation- al gateways and three additional license applications are pending. The Government has also required international gateway operators to become full service carriers by installing local networks in both profitable and non-prof- itable areas. Similar requirements are imposed on cellular operators and the result is that several consortia have announced ambitious plans for international gateways or cellular networks and fibre backbone networks to link these with their local networks. As a result, by 1997, the Philippines should have at least a half dozen full service (local, long distance, and international) carriers. Other net- work services are open to restricted competition. There are five licensed paging operators and five Vsat operators. Box 1. World Bank Report on Telecom Sector Reform Resistance to multiple service providers will only pro- long the "telecommunications crisis in Asia," accord- ing to Telecommumcat~ons Sector Reform m Asia. Towards a New Pragmatism (D~scuss~on Paper #232) a World Bank report by Peter Smith, a senior telecom specialist at the Bank, and Gregory C Staple, a legal consultant The report predicts that the cost to national economies of providing only limited or no service to millions of its citizens will be greater than any loss of economies of scale from franchising new operators It brushes aside arguments that multiple service providers are inefficient because, it says, the current gap between supply and demand in low income countries is so great that concern about "duplicative investments" is misplaced To support its case, the World Bank says that more than 2.8 billion people in lower income countries in Asia must "make do" wlth just over 25 million lines, and that despite recent expansion plans, average annual growth in the number of lines in Asia between 1982- 1991 was only seven percent, increasing wait- ing hsts m most countries However, the report warns that merely pnvabz~ng and divesting a state-owned operator is not necessarily a solution, citing the Philippines as an example of substandard private sec- tor performance. The report suggests that in low income countries w~th less than two rnam lines per hundred people (Bangladesh, Cambodia, China, India, Indonesia, Laos, Pakistan, Sn Lanka, Vietnam) the overriding issue is not pnvatlsatlon but how to attract capital from the private sector to accelerate network growth The report recommends a "bottom up" approach to telecommunications reform, financed by a mixture of public and private capital, which would entail the franchlsmg of multiple service providers, build-oper- ate-transfer concessions, ]omt ventures and revenue shanng schemes. It also warns against capital-inten- sive programs that are currently in vogue and sug- gests that cheaper options, such as increasing public payphones, and using wireless, trunked mobile, pag- ing and volcemafl services could provide a more pragmatic alternative However, the report says that even the most ambi- tious expansion plans will be compromised by ineffec- tive regulation which it describes as "the weakest part of sector reform in poor countries". It suggests that an independent, publicly accountable regulatory body w~th a clear legislative mandate is crucial to reform and that the "no regulation option" adopted m New Zealand is "illusory" and prone to disadvantage new operators J 1994, p Adapted from Telenews Asia. 10 February 2 41 TeleGeography ! 994 continued from page 38... Similarly, the fact that separate virtual private networks (VPNs] can be provided over a common infrastructure allows for economies of scale. On international links, this traffic can be "refiled" via the networks of national carriers with the lowest international tariffs, then forwarded over highly-loaded leased circuits. Such arrangements bypass local tariffs and the accounting rate structure, allowing sub- stantial (20%+) discounts against PSTN rates. The EVUA deal was only one of many special deals; most European carriers now offer published discounts of ]0-30% to large cus- tomers. Some customers have won dis- counts of 50% or more--hidden from prying regulators by non-disclosure agreements. The more aggressive European carriers are all offering new calling card services. And the massive branding campaigns by AT&T (WorldPlus) and MCI (WorldPhone) show that the new super-carriers are targeting residential customers as well as the large corporate market. Both products allow users to make calls and receive voice messages, and switch calls via a European hub (the U.K. or Sweden) rather than through the U.S., as previously. Competition has also intensified in the data communications services market. The Services Directive, passed by the European Commission in 990 and upheld by the European Court of Justice in November 1992, has freed up this mar- ket, encouraging new entrants.7 For example, AT&T, BT, Infonet, Reuters/Wiltel, SITA (through its Scitor subsidiary), Sprint and Transpac (a France Telecom affiliate) are all offer- ing cross-border public frame relay services, Most of these companies also intend to provide Asynchronous Transfer Mode services, probably within the next two years, raising the possibility that these operations may become the vehi- cle for corporate multimedia traffic that includes voice and video as well as data. ",~HOW ME A COMPANY WITH A FIVE-YEAR STRATEGY FOR EUROPEAN TELECOMMUNICA- TIONS~ AND I’LL SHOW YOU A DOESN’T UNDER- COMPANY THAT J STAND WHAT’SGOING ON, Brash new entrants are also putting pressure on prices. They include several "Light Carriers," the most important of whom are four U.S.-based carriers: [] IDB Worldcom. Much the best-established, and to some degree part of the "establishment" (owning the U.S. half-circuits in several transatlantic cables, for example), IDB affiliates are especially active in the U.K. and Germany, where IDB is the largest international reseller. [] International Discount Telecommunications (IDT). Now three years old, IDT is investing in new technologies (especially compression technologies) which it hopes will keep it ahead of competitors. The company has attracted funding from Mario Gabelli, a well known U.S. fund manag- er. [] Viatel. Part-owned by financier George Soros and Comsat, the largest U.S. provider of international satellite capacity, Viatel claimed ]992 sales of over $20 million, a three-fold increase over the previous year. [] Esprit Telecom is targetting larger businesses and says it will build infrastructure as soon as it is allowed to. Each of these carriers benefits from lean operations, aggres- sive sales teams and ambitions that extend well beyond simple callback services and price arbitrage.6 The European Commission, meanwhile, has placed much faith in cable television companies, encouraged by experience in the U.K., where half the homes cabled for TV are also taking telephone service from the cable TV provider (400,000 by mid-1994). Despite Europe’s extraordi- nary regulatory, commercial and techni- cal patchwork, competitive provision of telecommunications services over cable TV infrastructure is likely to emerge in at least the Netherlands, France and Spain. Initially, provision of real-time voice telephony will probably be excluded in most of those countries--though not where it is part of a multimedia service. Yet for all these advances, fear, uncertainty and doubt con- tinue to stalk the European telecommunications services landscape. "Show me a company with a five-year strategy for European telecommunications," said one cynical execu- tive recently, "and I’ll show you a company that doesn’t understand what’s going on." One reason is that many key regulatory issues remain unre- solved. In particular, Europe must decide how and when to liberalize infrastructure, introduce an effective policy on interconnection, and decide how to control the dominant players as they seek out new business across borders. At this writing, so far as infrastructure competition is con- cerned, the most likely outcome is a compromise in which existing infrastructure owners will be allowed beginning in 1995 to offer data communications services and services to closed user groups, but not voice services to the general public. More substantial liberalization might follow in 1998. At the Corfu meeting in June 1994, European leaders agreed to push ahead with some form of infrastructure lib- eralization and bring forward publication of a Green Paper to late 1994. But on many issues--e.g., defining infrastruc- ture, picking providers, protecting universal service and so on--there remain significant differences of opinion. Compettdon in Europe The European Commission also lacks an effective policy on interconnection of dominant and competing networks once services are liberalized in 1998. As competition develops and the accounting rate mechanism atrophies (few believe it wilt survive the decade in Europe), this issue will come alive. For new entrants without foreign affiliates to terminate traf- fic, accounting rates are likely to be replaced by some kind of access or termination fee--an interconnection charge by any other name. When a foreign affiliate exists, there still needs to be a workable regime for handing off traffic between competing local networks, something the U.K. experience suggests will be hard to implement. The European Commission has already indicated that local access fees do not have to be based on cost. It has bowed to PTO (Post, Telecommunications Operators) arguments that unless cross-subsidies are continued, they may no longer be able to provide universal service. This position may gain additional support if the principle of universal ser- vice is extended to include access to the "information high- way" and its multimedia services. Until now, the main vehicle for interconnection policy has been Open Network Provision (ONP), a set of rules that establishes the general conditions under which dominant or monopoly operators must make resources available to users and third-party service providers. However, the new emphasis on "subsidiarity," an EU principle that states that decisions should be taken at the lowest possible govern- mental level, means that the European Commission is dele- gating much of the detailed enforcement to national telecommunications regulators, who have very different views about interconnection. ONP rules, among other things, anticipate that most inter- connection agreements can be reached by straightforward commercial negotiation, though history suggests that the parties inevitably end up in the regulator’s office, or in court. Indeed, if the U.K.’s battle over interconnection (ten- years-old and still running) is any guide, such disputes will be a perennial feature of the new landscape. Other regulatory problems are likely to impede unfettered competition. For example, many ambitious new providers need high bandwidth pipes: 8 Mbit/s, 34 Mbit/s, 140/155 Mbit/s or even more. At present, however, PTQs are under no obligation to provide such services; most have no pub- lished tariffs for services above 2 Mbit/s, and are under- standably reluctant to provide them to competitors. In June 1994, one competitor, Metropolitan Fiber Systems, was quoted an 18-month lead time on a circuit between Frankfurt and London, at a price per bit ten times the pre- vailing rate in the U.S. Similarly, national and pan-European rules to prevent predatory and uncompetitive behavior by PTOs in other markets are muddled and inconsistent. In Sweden, for example, BT took Transpac to the regulators, complaining that its contract price for a corporate network was below cost and cross-subsidized from monopoly income earned in France. Yet BT itself stood (informally) accused by the two Swedish operators, Telia and Tele2, of predatory pricing; it allegedly undercut the Swedes by up to 80%, prices that BT defends as a necessary entry fee to a new market. Meanwhile, Telia itself charges customers :30% less to make calls from London to Paris via Stockholm than it does for a direct call from Stockholm to Paris. Such anomalies and dis- putes are likely to spread throughout Europe as regulatory barriers fall and PTOs seek to make up for loss of market share at home. For Europe’s PTOs, rapidly escalating competition--both real and virtua!--will have two certain effects: loss of mar- ket share and falling prices for basic services such as voice telephony and private lines. This will mean some tough- minded decisions over the next two or three years. Most PTOs will face politically difficult large-scale staff lay-offs, divestiture and separation of some businesses, further alliance-building and market segmentation. Some may choose to go all-out for a broadband local infrastructure, leaving partners to bear the risk in global networking; oth- ers, conversely, may place heavy emphasis on the construc- tion of global businesses and look for partners or even buy- ers for Iocat infrastructure. But all PTOs are likely to lobby national and European reg- ulators to give them time to adjust--in particular, by enforc- ing the payment of access and interconnection charges that reflect universal service obligations, fully distributed (his- toric) costs and only gradual removal of historic cross-sub- sidies. The regulators’ response may set the pace of change in European telecommunications over the next few years. Notes I Council Resolution of 22 July 1993 on the review of the s~tuation in the telecommumcat~ons sector (93/C213/01 ). 2. "Green Paper on a common approach m the field of mobile and personal communicabons ~n the European Un~on," COM (94) 145 Final 127 April 1994. 3. "Europe and the global information society," Recommendation to the European Council H~gh Level Group on the Information Society, Brussels, 26 May 1994. 4. See Council Directive 90/387/EEC, OJL 19211 24 07.90, Arbcle 2, Clause 7. 5 Communications Week Internatlonal, April 1994, p 1 6 See Communications Week International, 28 March 1994, Network Review secbon, for further details. 7. Commission Directive of 28 June 1990 on competition in the markets for telecommunications services (90/538/EEC), OJL 192/10, 27 7 90. TeleGeography 1994 © TeleGeography, Inc. 1994 NAVIGATING CYBERSPACE--MAPS AND AGENTS Different Uses of Computer Networks Call for Different Interfaces By Zachary M. $chrag I s geography the future of cyberspace? Or will software agents take the "space" out of cyberspace altogether? Today’s computer networks--the lnternet, the commer- cial online services, and others--are made of words, lots and lots of words, and words are slow and ungainly. While verbs like "online cruising" and "net-surfing" imply exciting encounters, for the most part logging onto a com- puter network means reading page after page of text, then typing a command to get some more text. Even where icons and graphics have made some headway, such as on the rapidly growing service known as the World Wide Web, they are often used just to break up the text, rather than as the dominant elements on the screen. Text has its fans, but few people think that the status quo, with its endless stream of menus and command lines, addresses and prompts, will last long into the next century. Given the ability of computers to produce, transmit, and reproduce drawings, photographs, sounds, and even mov- ing videos, a text-based cyberspace seems anachronistic, old-fashioned. It resembles a news broadcast from the early days of television, when a stationary camera was pointed at an announcer in a studio, reading from a prepared script just as in the days of radio. It was television, all right, but it had not begun to exploit the possibilities of the medium. What will replace text as the dominant element of comput- er communications? One possible answer is geography. Though we call our present-day computer networks "cyber- space," they are nothing like the world envisioned by sci- ence fiction author William Gibson when he coined the word ten years ago. His cyberspace was a simulated city, where one saw not text and icons but computer-generated, three- dimensional images of buildings, streets, and walls, and objects that represented information and software. Though the technology needed to achieve Gibson’s vision does not yet exist, advances in virtual reality {VR} make it seem not far away. Already programmers, authors, and architects--such as the contributors to Cyberspace: First Steps {ed. Michael Benedikt, MIT Press, 1991}--are work- ing hard to plan this new sphere of human existence. They imagine the technology of virtual reality being grafted onto the present data communications network, so that virtual office towers rise into the sky and the electronic city is born. Instead of abstractly "telnetting" to a dis- tant computer site, we would walk or fly there. And once there, we 44 could navigate through rooms, down hallways, and into gar- dens to find whatever it is we were looking for. A new geog- raphy would be established to complement the old. At the same time, however, there is another movement afoot. While one set of programmers is laboring with head- mounted displays and rendering algorithms to make soft- ware "objects" visible, another group is working on "intelli- gent agents," designed to make the computer networks invisible. Like macros and scripts, agents let computer users auto- mate repetitive tasks such as logging onto a computer sys- tem or downloading information from a database. Unlike macros, agents have the potential to go beyond their mas- ters’ computers to the network itself, retrieving information, sending messages, and perhaps buying goods. Already sub- scribers to Hoover, an agent produced by the Sandpoint Corporation of Cambridge, Mass., can skip the drudgery of scanning database after database or asking around for the best university or corporate computer to raid for files. Instead, for about $250 a month they can ask Hoover to go fetch, and it will assemble material on whatever subject is needed. Similarly, Mead Data Central’s Eclipse service will search hundreds of newspapers for articles including terms specified by the user, then send him the results as often as he likes. Thus, the agent acts like a trustworthy chauffeur, sparing its master the need to ever consult a map himself. Which will it be: a cyberspace future in which everything is geography or nothing is? Each scenario has a definite appeal, and its own uses, so I expect that both will emerge. The challenge will be to match each function of cyberspace with the appropriate form, to use geography when it can be used and to abolish it when it interferes. The movement toward geography There is no denying that the analogies between data com- munications and geographical space come naturally and in large number. For just as a city is more than a collection of buildings, and a forest is more than a collection of tress, cyberspace is more than a collection of linked computers, and geographical terms can help us understand its nature. The terms "information F Zachary M. Schrag I 72 722.1214@compuserve.corn) is Deputy Editor of TeleGeography superhighway" and "infobahn" have been become ridiculed cliches, with weak jokes about rest stops, road- kill, and speeding tickets. Yet those very jokes indicate the power of the metaphor. Highways are something that people in industrialized countries know, are comfort- able with, even love, and whose value they recognize. Americans understand highways as things that the federal government pays for and that bring them food, goods, vacations, reunions, and even produce their own culture, from Jack Kerouac’s 1957 novel, On the Road to Ridley Scott’s 1991 film Thelma and Louise. Europeans might see data networks as the counterpart to the autobahns and autostradas that link the diverse components of the European Union. At its best, cyberspace could do all of this, so the highway metaphor is optimistic and appropriate. Other spatial and geographical metaphors abound: gate- way, forum, room, Electronic Mall, dungeon, library, tour. One New York Times article ran a good thousand words on the subject of the popularization of cyberspace, with almost every paragraph containing a metaphor drawn from real estate or homesteading. And several of the major online services--CompuServe, America Online, eWorld--use globes as icons in their front-end software, as if claiming to be substitutes for the 4-billion-year-old planet we know so well. Now some of these online services are taking geography a step further, using geographical interfaces to let users explore their resources. A geographical interface is essen- tially a GUI (graphical user interface, pronounced "gooey") Geography and Cyberspace writ large. Whereas the GUI, popularized in 1984 by the Macintosh computer, tries to make a computer screen resemble an office where a user can find his "desktop," "folders," and "trash," a geographical interface (GeoGUI?) makes the screen into a city (or forest, or ocean), where many users can move between buildings, rooms, and, one supposes, a landfill. The designers of some new online environments, such-as Apple’s eWorld, General Magic’s Magic Cap, and the ImagiNation Network, have begun adding geographical objects to their system. Logging onto one of these systems, the user sees a view of the "Town Square," or "Downtown." He just clicks on a shop, a post office, or a casino and sud- denly he is ready to do whatever transactions are appropri- ate. And since he learned what to expect from a post office (if not a casino) at age five, his years of experience in the real world are all the training he needs. The GeoGUIs in use today are quite shallow, often only one layer deep. Click on the post office, for example, and you are likely to be confronted not with a deeper geography (the interior of a post office, complete with counters, RO. boxes, stamp machines, and so forth) but with a conventional text- based or GUI menu whose elements could be arbitrarily rearranged without changing their meaning. And the towns themselves aren’t towns at all, for only one person can be Whatever your interest, eWorld Sports flies with fast scores, stats and engrossing detail. Redgate compares 31 color printers for speed, accuracy, detail and easy maintenance. Let MacUser’s "Window Washer" brighten and clean your grungy screen with no physical effort. Join the Community Center’s "Growing Small Businesses~ real time conference at 7 p.m. Figure 1: eWorld Top Screen. Apple’s new online service, eWorld, uses a picture of a "town square" to orient users. But those fig- ures milling around are just for show. In a truly geographic interface, each figure would represent another user, who could be engaged in conversation, ogled, or ignored. TeleGeography 1994 there at a time. (Each Magic Cap user gets to build her own Downtown, so Jt would be impossible for two users to inhab- it the same Downtown at once without getting into a shout- ing match over whether a building is there or not.l But it is easy to imagine these early interfaces becoming more and more complex and encompassing--first adding more layers, then going to 3D, then improving virtual reality until an e- mail message is something that one can hold, something to be posted with a stamp whose virtual mucilage leaves a sour taste on the tongue. No more cryptic commands to memo- rize. No more vain attempts to craft menus and pictures out of typographical characters. Just space, as we have always known it, with tasks accomplished by speaking, walking, and touching. Programmers are also considering ways to make the Internet more geographical. Mark P. McCahill, project leader of the team that produced gopher, one of the most popular software tools for finding information on the Internet, and col- leagues have proposed a "5-D Gopher" system to work with more powerful computers. Gopher now uses text based menus, though there are front- ends available which add icons as well. The 5-D Gopher proposal, presented in April 1994, argues that a more geographical approach will make cyberspace more welcoming, since it will allow the dis- play of more data (such as abstracts of articles, links between files, and the popularity of some databases/ while drawing on the knowledge that users have acquired through years of experience in the real world. "We have spent our lives learning to recognize spatial cues: what entrances look like, what a bulletin board is, where you are likely to find a you-are-here map, and so on." Why not apply that knowl- edge to cyberspace? WHILE THE 3D GOPHER IS STILL IN THE VISIONARY STAGE, ONE PARTICULAR CORNER OF While the 3D gopher is still in the visionary stage, one par- ticular corner of cyberspace has been somewhat geographi- cally organized for over a decade. First created around 1980, MUDs (variously expanded as multi-user dimensions or multi-user dungeons) are network games that allow many players to control "characters," who can hunt for treasure, fight, build, and solve puzzles. What sets MUDs apart from the regular computer games one buys in the store is that players from all over the world can interact simultaneous- ly--talking to each other, fighting alongside or against one another, even simulating sex. Though MUDs have only text interfaces--the user sees only a string of ASCII characters running across her screen--the game is organized geo- graphically, with a series of "rooms," which can be anything from taverns to streets to forests to space stations. MOOs (MUDs, Object-Oriented) have "objects" that characters can move from one room to another. CYBERSPACE--THE WORLD OF The interface McCahill et al. propose, still in the very early planning stages, would transform today’s gopher servers (e.g., the card catalog at a university library) into circular "neighborhoods." Descending into a neighborhood, one would be able to view several representations of files at once, learning at a glance something of their contents, their relevance to one’s search, how many other people had used them, and other information. Going from one neighborhood to another would be like being the passenger in an airplane. One would glimpse the origin and destination neighborhoods from above, but in between one would see only "an anonymous looking grid." Because neighborhoods would have no fixed relation to one another (one can hop quite easily from gopher to gopher), there would be no real geography beyond the local level. MUDS--HAS USED GEOGRAPHY FOR OVER A DECADE. In many MUDs, the attraction of social- izing outweighs that of playing games, so the MUD becomes mainly a site for conversation. In this respect, they resemble the chat rooms of some online services and the "lnternet Relay Chat" service of the Internet. Users of these services often spend hours at the keyboard, engaged in real- time conversation with total strangers or familiar voices, known only by their often cryptic user names. Many MUDders and chatters long for a technology that would eliminate all that typing. A virtual-reality environ- ment would allow them to don fantastic bodies made of whatever human and animal parts strike them, to visit each other in virtual houses that resemble French chateaux or Thai pagodas, to hit when they want to hit and to kiss when they want to kiss. Experiments have already been made adding graphics and sound to MUDs, and clearly there is a vast potential market for a multi-user, VR-GeoGUI. The perils of geography The great advantages, then, of VR-GeoCUIs are that they are intuitive and fun. But while it is very attractive at first glance, geography has its drawbacks as well. First, geography can be very expensive. Virtual reality requires architects to design how the structures should appear and what physical laws--optics, gravity, sound, etc.--should operate. Then computer programmers must translate the architects’ vision into thousands of lines of code. This costs money. While most institutions--govern- ments, libraries, universities--can spare some change to scan their seal onto a World-Wide-Web server or put some Geography and Cyberspace Mozart onto the voice-mail hold message, it’s not clear how much they will be willing to invest in flashy, inviting virtual lobbies. Major university and national libraries certainly have gorgeous main buildings, but for the most part a library is dearly a library and not a gilded, marble-encrust- ed bank. Virtual reality also consumes bandwidth. Chip Morningstar and F. Randall Farmer, designers of Lucasfilm’s Habitat (essentially a visual MUD using somewhat cartoonish graph- ics) have written, "Communications bandwidth is a scarce resource.., when communications technology advances to the point where we all have multigiga- baud fiber-optic connections into our homes, computational technology will have advanced to match." This phe- nomenon haunts Mosaic, the graphical browser for the the World Wide Web. Journalists often become wildly enthusi- astic about Mosaic’s ability to display color graphics and even photographs along with text. But it takes a long time, even on a high-speed line with a state- of-the-art modem, to download a color picture. And on many servers, graphics have been used only for ornament, so that rather than organizing the space they merely slow down one’s search to the point where older technology, like a 2400 bps modem and a text-only front- end, is more efficient. can be quite disorienting, and it is easy to lose track of where one found an interesting archive. But while eliminat- ing the ability to jump from any site to any other and replac- ing it with a system of streets or landmarks would improve one’s sense of place, it would also slow down a search ter- ribly. If getting where I need to go in cyberspace, to pick up a product review or download a new font, requires me to cruise down some virtual boulevard, ask directions, make a left, descend into a neighborhood, and find an address, tfien I may as well get on my bicycle and go to the real library. NO MATTER HOW ASTOUND- ING THE THROUGHPUT~ SERIOUS RESEARCHERS AND PEOPLE IN A HURRY WILL FIND BETTER WAYS TO USE IT THAN WATCHING ALL THE OTHER DENIZENS OF CYBERSPACE WHIZ BY IN THEIR VIRTUAL BODIES. Such strains on one’s patience would onty increase in a shared, public cyberspace where the presence of other users was also displayed graphically. No matter how astounding the throughput, serious researchers and people in a hurry will find better ways to use it (faster transmission, higher fidelity music, higher resolution photographs, more frames per second} than watching all the other denizens of cyber- space whiz by in their virtual bodies. The 3-D gopher team hopes to lessen the strain on bandwidth by having the user’s computer and its software generate images, rather than having those images transmitted from the network. But many of the advantages they claim for a 3-D cyber- space (such as allowing organizations to customize their databases with pictures and sound} can onty be achieved if these bulky elements are sent over the network. And geographical interfaces are, by their nature, time inten- sive. Current interfaces, like gopher, Mosaic, and most com- mercial online service software, allow users to jump from site to site. In the middle of a gopher search, for example, one can hop from a gopher at a university in Germany to one in New Zealand to one in Costa Rica without having to go back to some top screen or master page. This hopping MUD players have some experience with this problem. Most MUDs allow any room to be con- nected to any other room, providing the builder of each room is willing. Moreover, users can usually "teleport" from one room to another, skipping all the distance in between. The result is that familiar principles, like the shortest distance between two points being a straight line, collapse, along with all the two- and three-dimensional geography we know. Jennifer Smith, a seasoned MUD veteran, recalls an experiment with making cyberspace resemble real- world geography: lslandia--a TinyMUD which ran from roughly February 1990 until May 1990--tried the "strictly topological" model, for the first time. A strict grid- work of streets was laid out, and people were not allowed to build silly things like a complete jungle in the space that was no larger than a city block. It didn’t work. First thing that happened was that people com- plained that if they were trying to be realistic, that tele- porting should be turned off. So it was. That was the big mistake; people then started complaining that it took too long to walk anywhere (each "mite" was five rooms or so long). Despite "Stepstones"--exits that would "jump" you from corner to corner, skipping the rooms in between--it still took too long to walk any- where, and they’d more or less destroyed their original reason for denying teleportation. In the end, it was decided that only under very good circumstances can you get away with enforcing rigid "real-world" topology, and that for the sake of sanity, it ought not to be so rigidly defined. MUDders, then, have decided that the convenience of tele- porting outweighs the comfort of familiar geography. And they generally have accounts that are free or paid by the month. On a system that charged by the minute (as do most commercial services and some lnternet providers), any geographical conceit that forced the user to spend time trudging from place to place would be doubly unwelcome. All of these costs are especially significant given the Internet’s present dependence on the willingness of univer- sity and corporation computer departments to donate some TeleGeography 1994 extra disk space and computing capacity to relatively frivo- lous uses like recipe databases and MUDs. With require- ments of design, bandwidth, and time, a VR-GeoGUI envi- ronment would put a much greater strain on these institu- tions’ generosity. Perhaps money will not be such an issue. Electronics do have a way of becoming very cheap after a few years. Non- profit institutions could set up shop in some sort of mass- produced, pre-fab virtual quonset huts, and perhaps every- one could own a computer powerful enough to run a small VR-MUD, along with a fiber-optic hookup as thick as a tree- trunk. But even if virtual geography were dirt cheap to set up and use, territoriality is a grave threat. From birdsongs to border wars, territory has always gone hand in hand with conflict. Being an intangible mass of binary data, a virtual geography will escape some of the tra- ditional causes of envy and strife: water, soil, gold, strategic mountain passes and harbors. But it would unavoidably import one of the chief principles of geography, that some locations are more desirable than others. What makes one cyberplace better than others? The same thing that puts office towers downtown: centrality and agglomeration. In the nineteenth century city, centrality meant proximity to the transportation hub: the railroad ter- minal or the ferry pier. In cyberspace, it means proximity to wherever or whatever it is that the user enters--the top screen, the "town square," or the gateway. Agglomeration is the force that makes dozens of carpet stores cluster on a single street in Istanbul. If several similar shops huddle together, they will draw more customers than each could on its own. A user cruising around cyberspace will go where the action is, and any institution in the boondocks will suffer. These principles will hold true no matter how things are arranged. Gibson describes the basic layout of Cyberspace as "a transparent 3D chessboard extending to infinity," rem- iniscent of both an urban grid like Manhattan or a rural grid like the sections of the American Midwest. Other authors, architects, and thinkers on the subject have also taken a grid as their starting point, showing that ideas of planning present in America since the eighteenth century die hard. But while a grid has much to recommend it, it will not solve all the problems of cyberspatial planning. Despite the apparent equality of all squares in a grid, some squares are bound to be more equal than others. There is no perfect way to allocate lots in a geographical cyberspace. Cyberspace could be settled by the pioneers, the first to write the software that structures the world. This is, to a degree, what happened after the University of Minnesota released "gopher," one of the most popular tools (~ I)ownlown "~:7ednesday, June 15, 1994 Hallway Figure 2: Magic Cap Downtown. Magic Cap’s "Downtown," still a fairly primitive geography, nonetheless illustrates apotential~ problem with all downtowns: they can become dominated by corporations, not people. J 48 Geography and Cyberspace to navigate the Internet. Because they had invented the protocol, the gopher team had a great deal of say in deter- mining how "gopherspace," the sum of all the world’s gophers, would be organized. It is worth noting that only after universities and other institutions all over the world had moved into gopherspace did the University of Minnesota float the idea of charging royalties for gopher, i.e., exacting rent from its tenants. The plan was defeated when outraged programmers threatened to release public domain versions of gopher, but that tactic might not work with a more complex, virtual or geographical interface-- harder to imitate and better protected by copyright or patent. Perhaps governments will assign the lots. Alexander the Great, William the Conqueror, and Abraham Lincoln all had experience parceling out newly subdued territory. Even today, governments are allocating, Iotterying, and auctioning slices of the radio spectrum. Government control might frighten the more ardent civil libertarians who enjoy the present anarchic state of the Internet, and one country’s government will have limited authority to control a network already global. But such a sce- nario is far less scary than the prospect of a corporation creating a working standard for cyberspace and inviting others to join up. A pioneering corporation could police cyberspace like a retirement village or an elite resort, pun- ishing offenders and booting out the riffraff without any form of due process. And it could portion out the best spots by The conquest of geography So maybe the answer is to do away with territoriality by doing away with territory, using automated software agents instead. After all, isn’t communications supposed to free us from geography? This is the principle behind the popularity of cellular phones, services that will activate a pocket pager with stock market news, and pay phones that will keep ring- ing a busy line so you don’t have to stand by the phone, dialing over and over again. ~NTELLIGENT AGENTS RESOLVE MANY OF THE PROBLEMS RAISED BY GEOGRAPHICAL INTERFACES. ~N TERMS OF COST~ AGENTS ARE TO WHAT UNMANNED SPACE PROBES ARE TO THE SPACE SHUTTLE. auction. After all, since some forgotten Sumerian staked out a cool spot just uphill from the river’s floodplain, people have been arranged in space according to wealth. Yet, worst of all would be an lnternet-like cyberspace with- out central planning--letting a UN Agency set the standards and then letting everyone build (first come, first served) to create a sprawling, tangled maze like a thousand or ten thousand medieval cities piled on top of one another. "French wine? Well, this is the European Wine Center, where people come to discuss German, Italian, Spanish, and Bulgarian wine. But if you want French wine, why, that’s over in the Cuisine Section of the French Pavilion way the heck over in Virtual World’s Fair. To get there, you have to bear sort of around, well, to the right, and through a few things.., it’s hard to explain. There used to be a highway that would take you right to it, but then this pair of lawyers from Phoenix built their office right on top of the road, and no one has been able to make them budge." The most recent flowering of this drive toward automation is Telescript, an intelligent agent program released by General Magic. Though produced by the same company, TeleScript is the opposite of Downtown. The idea is that a user can program his agent to go shopping, sort through his messages, negotiate with colleagues, and research information all without the user’s supervision. Properly developed, this technology could make cyberspace not a place to be visited with a head-mounted display, but a place to send one’s agent to do one’s bidding with a few voice commands. Why push your way through the tangled bazaar of cyberspace when you can relax in the sunshine and have the rum- maging done for you? Think of voice mail systems, for instance. Going through a geographical interface is like hav- ing to listen to all the menu choices each time you call, while using an agent is like bypassing the menu dialing the exten- sion of the right person. Only with a good agent, you will never need to listen to the menu because the agent will find out who that person is for you. Intelligent agents resolve many of the problems raised by geographical interfaces. In terms of cost, agents are to VR what unmanned space probes are to the space shuttle. Just as the space probe Voyager 2, now somewhere past Neptune, does not need oxygen or food or water or much in the way of temperature control, an intelligent agent could dispense with all the bells and whistles--the lights, sign- posts, and horizons--of a geographical cyberspace. Cyberspace would be data, just plain data, and to humans it would appear invisible. The science fiction version of this is the computers in Star Trek; ask them a question, and they tell you the answer, with no fuss about how they found it. Let’s say you wanted to retrieve a file from a computer at the Massachusetts Institute of Technology (MIT). Today, if you try using standard Internet tools, you are likely to get a message saying that there are only 50 slots open for file transfers (ftp) by people not affiliated with MIT, and that TeleGeography 1994 they are all taken, even late at night on a weekend. This sit- uation is only going to get worse, with the number of lnternet users growing exponentially and MIT unlikely to devote more resources to total strangers. With a virtual reality cyberspace, it would be worse still, having to wait on a virtual line, browsing back issues of the alumni magazine or hearing a loop of MIT football songs (if there are such things), while people ahead of you rummaged slowly through the files. With an agent-based cyberspace, however, you could simply dispatch an agent to the site and if it had to wait until 3 a.m. to get access, no problem. Moreover, because agents can operate far faster than humans (limited by the mechan- ics of reading, digesting, and typing), so the queue would be much shorter in the first place. As for hierarchy, agents don’t notice it, or at least they make it disappear. Today, any computerized library catalog can quickly assemble a list of books by Isaac Asimov, no matter how scattered they are across the subject classification sys- tem. A fully realized computer catalog, or agent search, could just as easily arrange materials by publisher, illustra- tor, language, date, popularity, or medium, allowing a user to pul! up a list of 1980s action movies with gross revenues over $50 million, for example, and then to download clips from those films. It is the user who defines what informa- tion is important, eliminating battles over rival hierarchies. Agents do have their problems. By sending an agent straight for a precise chunk of information, a user skips the browsing stage of research and thus misses some opportu- nities to stumble across a related document, the sort that might cause a flash of inspiration and suggest a new approach to one’s project. Agents could be annoying if they were allowed to directly contact human beings, like those nasty computer voices that telephone during dinner. In a worst-case scenario, agents could combine everything that is hateful about impersonal communication, from junk mail to answering machines that your friends never check. And no matter how cute and obsequious the little cartoon agent appears as he prepares to do your bidding, using intelligent agents is not going to be as much fun as the psychedelic experience of virtual reality. A Two-Lane Information Highway Since both geoGUls and intelligent agents have their pluses and minuses, there may well be room for both of them on the Information Highway. The massive and growing network of computers could handle two forms of traffic, one designed for real-time browsing and another for automated exploration by software. A user would find each service use- ful depending on her needs of the moment. And a corpora- ~~ [] Compact I-~,L4 User ID, Phone number, Modem settings ": ; CompuServe "ldhat’s ~’~ :~ Ne~/s about CompuServe GO HOHE NE~ i CompuServe Ha~] System E]ectromc Ha~] G0 HOHE NAVHAIL V~rtual reahtu G0 CYBE£FORUH The Electronic Frontier Forum The E]ectromc Frontier Foundation G0 EFFSIG Adobe Forum ~]]ustrator, ATH, Photoshop GO ADOBE Desktop Publishing Forum Professional DTP, TUpe G0 DTPFORUH Hac Hard.are Forum Hacmto~h hardvare d~cuss]ons G0 HACH~ Hicrosoft Exce] Forum Database, Spreadsheet GO HSEXCEL Hac Communications Forum Navigator, Te]eoom, Fax, Lan GO HACCOHH Navigator Support Forum (FREE) Navigator Assistance 60 NAVSUPPO£T Aot]Ye Serv]ce/Vets/Hlstoru GO HILIT&RY Outdoors Forum Camping, Hunting, F~shmg, Boating G0 OUTDOORS ~,,;~ Hanua] operabon, custom scripts :~ Session script for "ZMS" Your name Zachary M Sehrag User ID 72722,1214 Selected modem set "Standard" Tone dlalmg will be used The speaker will be off wh~le d~almgl Imtlallzatlon string "mO atEF" The modem port elll be used Selected phone set "CIS" Phone 5470061 (CompuServe) at 2400 baud Phone 3014031600 (CompuServe) at 2400 baud Phone 7035910506 (CompuSerue) at 2400 baud In the CompuSerue Mall System Delete message "291-lg34" Read the mall Send a message To Tom Bppleton-OeMoulm 75140,1137 Sb Kern/Track editor~ Send a message To Gregory Staple IMTERMET~gstapleemc~mall cam Sb" My ortlcle In the The Eleotron~c Frontier Forum Get all mg messages Get message 20675 and all repl~es Summarize new messages in sections 1,2,4,8,10,13,14 In the Telecommunications Forum Get all my messages Get message 886g5 and all repl~es In the Desktop Publlsh~ng Forum Set thread w~th messaqe 50g581 2 Figure 3: The CompuServe Navigator. Set up a list of tasks for the program, then go for a cup of coffee. By the time you come back the computer will have logged on, performed all its tasks, and logged off again, leaving you time to read your messages offline, free of connect charges. Geography and Cyberspace tion might use both services at once, just as today compa- nies advertise the same products Jn print, by broadcast, and by direct mail. The possibilities of a two-track cyberspace are illustrated by the two types of front-end programs available for CompuServe Information Service, the oldest of the major commercial online services. (Other online services allow similar choices, and off-line mail readers like Eudora offer some of the same possibilities to Internet users.) One can, if one desires, log on directly to CompuServe, typing one’s name and password in each time, and using a number of three-letter commands to access different functions of the system. It is relatively slow, and takes a lot of typing. So most CompuServe members bypass the tedious command- line interface by using a front-end. This is a piece of soft- ware that presents an easier set of menus or icons, and then translates the user’s button clicking into the commands that CompuServe understands. The most popular front-end is CompuServe Information Manager {CIM], which CompuServe more or less gives away (it costs S10 to download, but comes with a $10 usage credit). CIM is a GUI, with icons representing categories of information, like travel or computers, and functions, like reading one’s mail. Though CIM is not geographical, it is fairly close to the eWorld interface, so it is easy to imagine the icons being replaced with buildings. Already, some VR enthusiasts are at work on VISCIS, a virtual reality interface something along the lines of the 3-D Gopher proposal. The advantages of these systems is that they allow easy brows- ing, encouraging users to explore new areas of CompuServe while warning them when they are about to enter a more expensive service. Everything is very up-front, and could even be more so with a 3-dimensional display. Alternatively, CompuServe subscribers can use an "offline" front-ends, like TapCIS and Navigator, which function more like agents. Before logging on, a CompuServe user tells the program what it is he wants, whether he wants it to check the mail, check forum messages, erase old messages, send messages, or download files. Then he sets his computer running, and it logs in and performs all the tasks, much faster than a human operator could, thus saving connect charges. They can even be set to run late at night, when connect charges are reduced. And the programs can remember settings from one session to another, so that the user can just punch in his favorite forums once and the pro- gram will get lists of new messages with only a few key- strokes or mousedicks. The fact that CompuServe itself sells both CIM and Navigator suggests that both products, and their descen- dants, have a future. It all depends on what you want out i Figure 4: The ImagiNation Network: There is no question that much of the impetus toward geographic interfaces comes from a love of computer games and their special effects. Geographic interfaces and games go together like grassy fields and soccer balls; after the first screen, you can lust hike on up the mountain to begin your adventure. 51 TeleGeography 1994 of cyberspace: slow wandering where the fun lies in the chase as much as in the kill, or quick raids for specific facts. A geographical/VR cybermall would be a fine place to chat with new people, admire the newest sportscar in the show- room, and play spectacular reattime games against human opponents. A joyful place, full of glitz and fireworks, it would provide its commercial tenants a steady stream of customers, who would buy tangible goods, like mail-order clothing, or services that could be transmitted on the net- work itself: movies, games, and entrance to virtual dis- cotheques and theme parks. Financed by entrance fees, the sale of advertising, and rent charged to the merchants, it would also be a place where money talks. But it would not be much of a place for research, or for a non-profit or government institution to disseminate free information cheaply. The 3-D Gopher proposal frankly admits that "eventually we would like to allow.., a real Las Vegas-style user experience." I have been to Las Vegas, and I don’t remember seeing a public library. I am sure there is one, but what 1 noticed were the flashing electric lights and artificial volcanos of the casinos. A geographical cyberspace would likely turn out the same way, with money-making ven- tures prominent and noisy, and non-profit enterprises bare- ly visible. The Las Vegas Strip is a fine place for gambling, gluttony, voyeurism, and big-name singers, but scholarly research (other than anthropology] is best done elsewhere. A Vegas-like geographical cyberspace would not preclude the existence of another cyberspace, one dominated by software agents. This would be a very different place, more like the New York Public Library than Caesar’s Palace. People would not log onto it in real time, but would program their agents to perform tasks for them as quickly as possi- ble. This agentspace would be a place for serious research--where scientists could exchange ideas, citizens could register to vote, and victims of a new illness could hunt down medical information and form support networks. It would also be a place to shop, but for things that are no fun to shop for, like toner cartridges and mortgages. And instead of meeting strangers there, one would send quick messages to friends, who could read them at leisure. And because this "place" would not be a place at all, but an invisible network with no fixed hierarchy, where the order of the searches would be defined by the user, this cyberspace would be far more democratic. It would not be the first time that a communications tech- nology has succeeded in more than one guise. Conceived as a means of two-way communication, the radio achieved its most widespread use as a way to broadcast music, news, and other programs to millions of receivers. But it has never lost its original function, and airplanes, police forces, and armies, not to mention trucks, taxi cabs, and ham enthusi- asts, still depend on two-way transmission. If the radio, a single technology, could blossom as two radically different social phenomena, so could cyberspace. If designers give some sober consideration to how computer networks are being used, and what interfaces will best suit the purposes of each type of user, two very complementary cyberspaces could grow side by side. For More Information... For a thorough examination of the possibilities of using virtual reality to construct computer network interfaces, see Cyberspace: First Steps, edited by Michael Benedikt (Cambridge, Mass.: MIT Press, 19911. Benedikt, an architect, has drawn together contributors from the fields of design, computer science, literature, and elsewhere to dream about making cyberpunk science fiction into a reality. Carried away with the glittering opportunities, the writers sometimes forget to take into account real-world limitations, such as cost and crowding. Nevertheless, their thoughtful essays raise many of the issues that will be confronted as computer networks truly enter everyday life. "NCSA Mosaic and the World Wlde Web: Global Hypermedia Protocols for the Internet," by Bruce R. Schatz and Joseph B. Hardin, Science, vol. 265, 12 August 1994, provides a very thorough discussion of the capabilities of Mosaic and the World Wlde Web and examines the reasons behind their extraordinary popularity. For information on some of the specific services mentioned in this article, please contact: CompuServe: CompuServe World Headquarters; 5000 Arlington Centre Boulevard; P.O. Box 20212; Columbus, OH 43220. Tel. (614) 45-/-8600 eWorld: Apple Computer, Inc., 20525 Mariani Avenue; Cupertino, CA 95014. Tel. (408) 996-1010. The ImagiNation Network: The ImagiNation Network; 41486 Old Barn Way; P.O. Box 1550; Oakhurst, CA 93644. Tel. (209) 642-0700. See also Elliot Gold’s article on page 3. Magic Cap and Telescript: General Magic, Inc., 2465 Latham Street; Mountain View, CA 94040. Tel. (415) 261-3860. Mosaic: Mosaic can be downloaded from the ftp server ftp.ncsa.uiuc.edu. For information, write mosaic@ncsa.uiuc.edu. Maps TeleGeography 1994 CARTOGRAMS Maps CANADA 05% UNITED STATES ~:~ CARIBBEAN 46% ~’.,..~ ~0.6 % CENTRAL "~ :~ AMERICA ~L~ A~ERICA~ 5.7%~ World Population, 1994 EUROPE AFRICA 11,1% Ex-USSR 4 9% INDIA 162% 67% CHINA 22,3% SE ASIA 74% AUSTRALIA and NEW ZEALAND 0 3% 2,3% Source: United Nations @ Te~eGeegraphv, Inc, 1994 d Source Slerfler~s Distribution of telephone main lines, 1992 EASTERN EUROPE AND U.S,S.R, 9.7% ASIA WESTERN j 8 8% EUROPE 31 9% IAPAN 105% AMERICA 38% ’~ MIDDLE EAST AFRI 1 6% 17% AUSTRALIA AND NEW ZEALAND (OCEANIA) © TeleGeographV, {nc, 1994 ~5 TeleGeographg 1994 TARIFF IV APS The Geography of International Telephone Charges F 150 RUSSIA K~ nhi United States 100 USA _- Mexicn City 100 150 ~Copenhag~ Prague ~.- Budapest ~Ku~ait ~ahrai~ Tehran ~mman getown ~il Buenas Aires Tirana .. Bermuda Mogadishu Maps These four maps show the cost of making a telephone call from the country in the center of the map to the rest of the world. Countries are arranged according to the cost per minute of a three-minute, dayhme, direct-dialed international call made from the center country. The scale of the concentric circles (isobars) is stated in U.S. cents per minute. A city which costs $1.40 per minute to call would appear between the $1.25 and $1.50 isobars. Call charges based upon June 1994 MCl (USA), KDD (Japan), BT (U.K.), and DBP Telekom (6ermany) tariffs and June 1994 exchange rates. United Kingdom 1o0 75 --, 5[} 5o 75 ]00 125. Graphics by Michael Roscoe © TeleGeography, Inc+ 1994 Germany INDIA ®’~.~os Dhaka Delhi Lusaka ~-r°b~ Khartoum Kabul tND]AN OCt’AN TeleGeography 1994 INTERCONTINENTAL TELECOMMUNICATIONS FLOWS, 1993 Maps EUROPEAN TELECOMMUNICATIONS FLOWS, 1993 TeleGeography 1994 ASIAN TELECOMMUNICATIONS FLOWS, © TeleGeography, Inc. 1994 CHINA 1993 AUSTRALIA Key All figures are given in millions of Minutes of Teleeomlaumcat~ons Traffic (mMITTs), for the public telephone network The map shews all mtra-As~an routes with an annual volume of more than 20 mM~Trs The total volume of these routes ,s 3 6 bdhon o M¢I-Fs, a pprox~mately 8 of all global international traffic Traffic Flows 2OO 100 50 Each band is proportional to the total annual traffic on the public telephone network m both directions between each pair of countries -~.~ ;°~:..- Balance of Traffic On routes where traffic m one d~rect~on accounts for more than 60% of the total, an arrow shows the direction most of the traffic flows (~)Total Outgoing Traffic (~ 250 500 1,000 The area of each circle is proportional to the volume of the total annual outgoing traffic from each country, shown as a number Maps SOUTH AMERICAN TELECOMMUNICATIONS FLOWS, 1993 ¯ GUATEMALA ,NICARAGUA COSTA RICA PANAMA ELSALVADOR CHILE BRAZIL JSUGUAY © TeleGeography, Inc. 1994 Key W Traffic Flows 12 6 3 Each band is proportional to the total annual traff=c on public networks =n both directions between each pa~r of countries. All figures are g=ven ~n millions of Minutes of Telecommunications Traffic (rnM=TTs) The map shows all routes w=thln South and Central America with an annual volume of more than 2.5 mflhon M~]-i-s. The total volume of these routes is 258 m=lhon M=TTs, approximately 0.5% of global international MiTTs (~Total Outgoing Traffic lo0 Each circle ~s proportional to the volume of the total annual outgoing traffic from each country, shown bythe number. --->~ Balance of Traffic On routes where traffic in one direction accounts for more than 60% of the total, an arrow shows wNch way most of the traffic flows TeleGeography 1994 Maps TP, AFF C BALANCES, 1993; These maps show the balance of international switched telephone circuit traffic in 1993. Countries with more than 50% outgoing traffic generally had a traffic deficit, sending out more traffic than they received. In some cases, such as the U.S., the actual traffic deficit is less than it appears here because the data is based on the billing point of calls, not the physical point of origin. (See Methodology, pp. 148-149.) By and large, how- ever, countries with a traffic deficit also had a settlements or trade in services deficit because carriers in such countries paid out more to terminating country carriers than they received for landing inbound traffic. Countries with the lowest percentage of outbound traffic (e.g., Poland, Turkey, Pakistan, Uruguay, the Dominican Republic) likely profited most from the traffic imbalance in the past year. For each country’s precise outgoing and incoming traffic and its surplus or deficit, see pages 164-167. DEN~ Outgoing Traffic as Percentage of Total Based on caf] m~nutes 60%+ 55-60% 45-55% 4O-45% 40% No data ©TeleGeography, Inc 1994 TeleGeography 1994 ©TeleGeography, Inc 1994 Maps MAJOR SUBMARINE CABLES I Planned bber-apt~c cables .... I ©Tele~eography, Inc 19~4 TeleGeographg 1994 MEN (MOSTLY) CALLING! INTERNATIONAL PHONE SEX Major Providers and Consumers of Adult Audiotex, 1994 .CZECH & SLOVAK REP&(2~,5) NORWAY (1.5)~ 6~.~\ CANADA GERMANY\ ST PIERRE& V DOMINICAN REP ITALY ~ETHER~NDS ’ ’ S~O TOME & PRINCtPE (2.5) ANORTHERN MARIANAS (2) FHILIPPINES (4.5) CHILE(2)~ (5)A Incoming Traffic (Audiotex providers) (5)V Outgoing Traffic (Audiotex consumers) All figures in milions of MiTTs per month A U STRAUA (3) A Inc. 1994 ©TeleGeography, J S ince 1990, stricter national regulation of "adult" audiotex and live, sexually oriented call-in ser- vices has led many phone sex businesses to study telegeography. By relocating a sex-talk business in a relatively distant country (e.g., Guyana), the services may stimulate millions of additional incoming calls (e.g., from Germany) and substantial for- eign settlements. The host carrier typically shares these settlements (paid by the originating carrier) with the audiotex entrepreneur. The higher the local settlement charge, the better off the audiotex service is likely to be. (See chart). Offshore phone sex services thus do not need to bill their customers directly and also bypass local obscenity taws, Although not all international audiotex services are pornographic, international phone-sex calls now total over 65 million minutes a month--1.5% of all interna- tional traffic. The largest participants are mapped above. However, some of the termination countries may not actually host as many sex-talk services as the map suggests. Some services use the termination country pri- marily as a collection vehicle for reward traffic which is then rerouted (by private line) to an unidentified third country or in some cases back to the country from which most of the incoming calls originate. Sources: Industry interviews; 3 Net Communications (London); Voice International Settlement Rate per minute (US$) Country Country Code U.S (6/94) W Europe (est.) Australia 61 .38 .40-.50 Canada (B.C. and Ontario) 1-604/416 .14 .40-.60 Chile 56 .60 .70-1.00 Domimcan Republic 1-809 .55 .60-.80 Guyana 592 .85 .60-.70 Hong Kong 852 .50 .60-.70 Moldova 373 1.25 .40-.70 Netherlands Antilles 597 .38 .60-.80 Nor[hem Marianas (Saipan) 670 .75 .80-1.00 phil.ipp!~e_s 63 .~4/.64" .75-1..00 Portugal 351 .65/.35" .40-.60 S~o Tome 239 n.a. .75-1.00 St. Pierre and M~quelon 508 .53 .60-.80 United States (NY, Calif.) 1-212/415/213 :-. _ .35-.80 *Peak/0ff-Peak The settlement rate is the payment made by the orginat~ng car- rier to the terminating carrier. TNs payment is typically split with the audiotex service bureau. ©TeleGeography, Inc. 1994 Audiotex Traffic Growth Year Annual MiTT(millions) 1991 100-150 1992 200-250 1993 500-600 1994 750-850 Tables I TeleGeography 1994 Table 1: Capacity and Cost per Voice Path of Selected Trans-Oceanic Cable Systems, 1956-2000 Tables Trans-Atlanfic Cable Cost (US$) Capacity Year in Service System per voice path (voice paths) 1956 TAT- 1" 1965 TAT-4* 1970 TAT-5* 1983 TAT-7* 1988 TAT-8 1989 PTAT 1991 TAT-9 1993 TAT-10 1994 CANTAT-3 1996-97 TAT-12/13 557,000 365,000 49,000 23,000 9,000 6,000 5 500 4,000 1,000 1,000 89 138 1,440 8,400 37,800 85,000 75,600 75,600 302,000 600,000 Trans-Pacific 1957 Hawaii1* 378,000 91 1964 TPC-I* 406,000 167 1974 Hawaii 2* 41,000 1,690 1975 TPC-2* 73,000 1,690 1988 TPC-3 16,000 37,800 1991 Noah Pacific Cable 5,000 85,000 1992 TPC-4 5,500 75,600 1996 TPC-5/6 2,000 605,000 Japan/Saufli Arabia/U. K. 1997 FLAG 1,500 605,000 No longer in service. Notes: Costs are capital and construction costs only, stated in US$ to the nearest $500, unadjusted for infla- tion. Current technology permits approximately 5 virtual voice paths to be derived from a digital channel oper- ating at 64,000 bits per second (64 kbit/s). Fiber optic 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 connects 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 is generally excluded. Source: FCC and carriers. ©Tele6eography, Inc. 1994 TeleGeography 1994 Table 2: Cable and Satellite Capacity on Trans-Atlantic and Trans-Pacific Routes, 1986-2000 ~ear 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997-2000" Trans-A tlantic (North America-Europe) Voice Paths Trans-Pacific (North America-East Asia) Voice Paths Cable Satellite Cable Satellite 1 264,000 1 864,600 1 1,464,600 22,000 22,000 60,000 145,000 145 000 221 000 296 600 372 200 664 000 ,264 000 ,264 000 ,264,000 78 000 78 000 78000 93000 283000 283000 496,000 620,800 620,800 710,800 710,800 737,500 2,000 39 000 37,800 39 000 37,800 39 000 37,800 39 000 37,800 39 000 114,200 27 000 190,500 27 000 264,000 83 300 264,000 234 000 234 000 234 000 424 500 *minimum available Netes: Estimates of cable voice paths assume that 5 virtual voice paths can be derived from one 64 kbit/s digital circuit; cable estimates do not include circuits held in reserve for cable/satellite restoration ser- vices. Estimates of trans-Atlantic capacity exclude proposed PTAT-2 cable in 1997-2000 timeframe. Estimates of trans Pacific cable circuits are based on capacity from North America to Japan via Hawaii or Guam and exclude proposed SE-ME-WE-3, CANPAC-1, and Trans-Siberian Link (TSL) cables, all scheduled for 1996-97 timeframe. Estimates of satellite voice paths are based on Intelsat 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 Intelsat Vl series with 24,000 channels or 120,000 voice paths using Digital Code Multiplication Equipment (DCME). The Intelsat VII series, deployed in 1992, has 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-3, 0rion-1 and TDRS-4; Trans-Pacific PAS-2; RimsatJExpress (2 satellites) and TDRS-174. In the near term, some additional telecommunications circuits are likely to be available from Intersputnik, Hispansat and, after 1997, from the Iridium, Globalstar, and Inmarsat proposed mobile satellite systems. Currently, non- Intelsat satellites are limited to 1250 64 kbit/s circuits per satellite for public switched telephony. This limit is expected to be raised to 8000 64 kbit/s circuits in 1995 and phased out by 1998. Additionally, the capac- ity of the following "national" satellite systems may provide some trans-Pacific telecommunications ser- vice: Optus (Australia), Palapa Pacific (Indonesia), and, in the North Pacific, Aurora (U.S.). Regional capacity estimates do not necessarily imply that full capacity is available to satisfy demand on any given bilateral route. Source: F0C and carriers. ©TeleGeography, Inc. 1994 Table 3: The Top 40 International Carriers, 1988-93 Tables Rank 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13, 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27, 28. 29. 30. 31. 32. 33. 34. 35. 37, 38. 39. 40. Outgoing Traffic (millions of MiTTs) Company Country 1993 1992 Change 92-93 1991 1988 AT&T (a,b) United States 7129 6984 2.1% 6557 4778 DBP Telekom (c) Germany 4680 4087 14.5% 3605 2479 MCI (a,b) United States 2839 2083 36.3% 1600 262 France Telecom (a) France 2576 2449 5.2% 2295 1570 BT (d) United Kingdom 2310 2188 5.6% 2105 1654 Telecom Italia (f) Italy 1610 1473 9.3% 1220 n.a. Swiss PTI Switzerland 1572 1551 1.4% 1429 1014 Stentor (b,g) Canada 1430 1520 -5.9% 1425 1054 Hongkong Telecom (d,e) Hong Kong 1377 1137 21.1% 913 441 Netherlands PTT (a) Netherlands 1238 1134 9.2% 1018 706 Sprint (a,b) United States 1175 940 25.0% 723 131 Belgacom (a) Belgium 979 911 7.5% 823 561 KDD (d) Japan 944 893 5.7% 850 529 China MPT (e) China 895 635 40.9% 440 170 Telefonica ....... Spain 847 804 5.3% 719 330 Mercury (d) United Kingdom 820 661 24.1% 493 75 Teleglobe (a) Canada 808 722 11.9% 647 358 Austrian PTT (a) Austria 767 713 7.6% 642 401 Telia AB (h) Sweden 685 691 -0.9% 672 485 Telstra (i) Australia 640 659 -2.9% 610 415 Telmex (a) Mexico 625 684 -8.6% 500 211 Singapore Telecom (d,j) Singapore 480 412 16.5% 344 142 TeleDanmark Denmark 452 425 6.4% 395 296 Saudi Com. Ministry Saudi Arabia 448 465 -3.7% 411 288 DGT Taiwan (a) Taiwan 441 369 19.5% 255 143 Telecom Eireann (d,k) Ireland Norwegian Telecom Norway Etisalat U.A.E OTE (a) Greece Videsh Sanchar (d,l) India Telekomunikacja Polska Poland Korea Telecom Rep. of Korea Turkish PTT Turkey Telekom Malaysia Malaysia Telkom South Africa South Africa Telecom Finland Finland IDC (d) Japan ITJ (d) Japan HTC Hungary Rostelecom Russia 384 297 29.4% 273 202 376 349 7.8% 309 246 342 299 14.2% 262 177 336 299 12.5% 245 179 284 260 9.4% 186 91 273 213 28.2% 125 28 265 245 8.2% 229 104 265 227 16.7% 198 118 258 217 19.2% 176 95 255 222 14.9% 203 125 253 235 7.7% 215 149 239 197 21.3% 154 0 228 193 18.1% 156 0 213 184 16.0% 151 62 201 176 14.4% n.a. MiTT a. 1993 traffic based on bdhng point of call, not originating point. b. Data for North American carriers include continental traffic. c. For DBP Telekom, all data include outgoing traffic from the former East Germany. d. Data are for the Fiscal Year (April 1993 to March 1994). HKT and Mercury are majority owned by Cable & Wireless (U.K.). e. Includes Hong Kong-China traffic. f. Combined totals for Iritel and Italcable. Prior to 1994, tritel (formerly ASST) handled intra-continental traffic only, and Italcable carried overseas traffic. is Minutes of Telecommunications Traffic. Data are for public voice circuits only rounded to the nearest million MITT. g. Stentor was formerly Telecom Canada; Stentor traffic ~s for U.S. only of which approximately 70% is originated by Bell Canada. h. Telia AB was formerly Televerket. i. Telstra was formerlyAOTC. l- Singapore Telecom data ~nclude traffic to Malaysia (except local border traffic). k. Telecom Eireann data include traffic to the U.K I V~desh Sanchar data excludes traffic to Pakistan. © TeleGeography, Inc. 1994 TeleGeography 1994 Table 4a: Market Share of Competing International Carriers, 1986-94 Country/Carrier 1986 United States AT&T 94.3 MCI 4.0 Sprint 1.6 IDB U.K. (F.Y.) BT 99 8 Mercury 0.2 IPL Resellers Japan (F. Y.) ITJ New Zealand (F. Y.) TNZ ClearCom Korea, Republic of Korea Telecom Dacom Chile Entel Chile Chflesat VTR Telecom Philippines PLDT Philippine Global Com. Eastern Telecom Sweden Teha AB Tele-2 Australia Telstra Optus Canada (Canada-U.S. route only) Stentor Unltel Resellers Pemen~ge ofOu~oingMiTT 1987 1988 1989 1990 1991 1992 1993 93.0 89.1 83.3 78.4 74 8 70.3 62.2 4.7 7.0 10.2 14.6 17 8 21.2 24.8 2.3 3.5 5.8 64 6.3 7.3 10.3 n.a. 0.6 98.5 95.5 91 0 86.0 81 0 76.8 1.5 4.5 9 0 14.0 19 0 23.2 72.4 25.8 1.8 93.3 88.0 73 3 69.7 66.9 37 6.5 133 15.3 16.9 3.0 55 13.4 !5.0 162 92 0 82.0 80.0 78.4 8.0 18.0 20.0 21.6 79.9 20.1 80.0 20.0 1.0 91 6 84 98.0 2O 74.5 25.5 55.0 20.0 5.0 84.2 15.8 92.6 7.4 87.0 13.0 857 22 121 Notes: MiTT is Minutes of Telecommunications Traffic. Data based on international outgoing traffic for the public switched net- work only. Unless stated, data exclude traffic and market share of carriers reselling international private line services (IPL resellers). Market shares are for the full year, beginning in the first year of competition. Market shares for U.S. carriers prior to 1993 exclude resellers and traffic to Canada and Mexico; minor U.S. carriers are not listed. For U.K. carriers traffic to Ireland ~s excluded. For the Phihppines, market share data exclude Eastern Telecom. For Chile and the Phihppmes, start-up carriers are listed in Table 4b, opposite. Chilean shares to not add up to 100% because Chilesat reportedly acted as an international gateway in 1993. This gateway handled an additional 20% of outbound traffic originated by CTC, the largest local exchange company, prior to the establishment of CTC Mundo. ©TeleGeography, Inc. 1994 Table 4b: Other Competing International Carriers Tab/es Additional facilities-based international telephone carriers have also been authorized in the fol- lowing markets, but were not operational or had limited international traffic as of 1 July 1994: Entel Chile T.D.I. (affiliate of BellAtlantic, U.S.) CIDCOM Larga Distancia (100% owned by BellSouth, U.S.) CTC Mundo Dominican Republic ¢ CODATEL Tricom Finland € Telecom Finland Telivo (owned by Finnish state electric company, Imatran Voima) Finnet International (owned by regional Finnish telephone companies) Arlands Mobil Telefon Ltd. Indonesia PT Indosat PT Satelindo Israel Bezeq DARCOM Communications Ltd. Malaysia =P Telekom Malaysia Technology Resource Industries (TRI) Philippines PLDT Isla Communications, Inc. (30% owned by Shinawatra group, Thailand) Smart Communications, Inc. Capital Wireless Inc. (20% owned by Korea Telecom) Globe Telecom (38% owned by Singapore Telecom) International Communication Corp. Russia Intertelecom Aerocom Ltd. Astelit Baltic Communication Comstar Telecom Leningrad Trunk Line Nakhodka Telecom Sakhalin Island Tatin Corn Teleport-TP Vostoktelecom Former monopoly operator in the market. ©TeleGeography, Inc. 1994 TeleGeography 1994 Table 5: International Carriers: Company Share symbol AT&T U:T- T.N BCE (a) C:B- B.TO Bezeq (b) N/A- BEZQ.TA BT BT- BT.L C&W (c) CABL- CW.L CPRM MVNB- MARN.IN CTC U:CTC CTCpa.SN Dacom KO:DAC 84310.KS Entel (Chile) CH:ENT ENT.SN GTE (d) U:GTE GTEN Globe Telecom (’A’ share) PH:GLA N/A Hong Kong Telecom K:TELE 0008.HK IDB Comm IDBX IDBX.O KDD J:LZ@N 9431 .T Matav (Hungary)(e) Not listed MCI MCIC- MCIC.O OTE (Greece)(f) Not listed PLDT PH:PLD N/A PTT Netherlands (KPN) H:KPN KPN.AS Singapore Telecom T:TELC TELE.SI Sprint U:FON FON.N Tele Danmark N:TELE TLDa.C0 Telebras (’PN’ share) 8R:TL4 TEL.SA Telecom Argentina (g) AG:TEC TEC2.BA Telecom Corp of New Zealand Z:TELN TEL.NZ Telecom Italia (h) I.SIP TELE.MI Telefonica de Argentina AG:TEA TEA2.BA Telefonica de Espana E:TEF TEF.MC Teleglobe 745251 TLO.TO Telekom Malaysia L:TKOM TLMM.KL Telmex (’L’ share) MX:TML TMXL.MX VSNL ID.VSN N/A Share price Principal Exchange 8/31/94 EPS (Ic) PER (x) New York 55.25 US$ 3.45 16.0 Toronto 48.13 C$ 2.75 17.5 Tel Aviv 5.05 NIS 0.08 31.5 London 391.00 £ 29.00 13.5 London 473.00 £ 22.30 21.2 Lisbon 5,050.00 Esc 287.00 17.6 Santiago 2,210.00 P 117.77 18.8 Seoul 83,900 00 Won 1,728.00 48.6 Santiago 5,220.00 P 366.78 14.2 New York 31.75 US$ 2.20 14.4 Manila 28.00 P -0.57 -49.1 Hong Kong 16.90 HK$ 0.65 26.0 New York 9.50 US$ 0.43 22.1 Tokyo 11.70 Y 000 243.68 0.0 N/A 2,200.00 Ft N/A New York 24.56 US$ 1.30 18.9 N/A Drs N/A Manila 1,755.00 P 101.50 17 3 Amsterdam 53.90 FI 4.20 12.8 Singapore 3.46 S$ 8.00 0.4 New York 39.63 US$ 1.99 19.9 Copenhagen 135.00 DKK 119.20 1.1 S~o Paulo 57.69 US$/000 shares 1.52 38.0 Buenos Aires 7.23 P 0.20 37.1 Wellington 5.20 NZ$ 27.60 0.2 Milan 4,560.00 Lit I82.00 25.1 Buenos Aires 7.30 P 0.25 29.3 Madrid 1,840.00 Pta 93.00 19.8 Toronto 18.63 C$ 1 20 15.5 Kuala Lumpur 21.40 RM 0.55 38.9 Mexico City 10.80 NP 0.85 12.7 Bombay 1,100.00 Rp 15.00 73.3 Share symbol is Datastream mnemonic and Reuter code where available. Ic = local currency N/A = Not Applicable. n.a. = not available. a. BeE is the parent of Bell Canada, which serves the U.S. It also holds a 30% interest in TeleGIobe. b. Bezeq’s stock prices are listed as of 6/30/94 and 6/30/93 c. C&W’s principal international carriers are Hong Kong Telecom, 58%-owned, and Mercury Communications (U.K.), 80%,-owned. d. GTE provides international service through its wholly owned U.S. subsidiary Hawaiian Telephone Co. GTE also owns CODATEL in the Dominican Republic, 50% of BCTel and Quebec Telephone and 20% of CANTV (Venezuela). e. Matav price is for the ’grey’ market for c.5% of stock f. OTE: latest estimate for proceeds of IP0 equal to Drs250-400bn (US$1.065 1.672bn). g. Telecom Argentina revenue figures correspond to first two quarters of FY ending 9/94. h. Telecom Italia has resulted from the amalgamation of SIP, Iritel, Italcable and Telespazio on 18/8/94. The share price is that of SiP before that date. Tables Stock Prices and Market Capitalization International Service Exchange Capitali- Share price (Ic) % ch on 8/31/93 Revenues % of total Number of Mkt. cap. rate zation Cap, as % 8/31193 Ic US$ US$m revenue shares (m) (US$bn) 8/31194 (USSbn) total mkt. 62.88 -12.1 -12.1 5,206 7.8 1,352.40 74.72 1.00 4070 1.8% 43.25 11.3 6.2 n.a. n.a. 305.35 10.74 1.37 259 4.1% 6.31 -20.0 -25.4 468 34.2 720.52 1.20 3.06 45 2.7% 428.00 -8.6 -5.8 2,925 14.3 6,207.00 37.21 1.53 1129 3.3% 431.00 9.7 9.7 3,463 48.5 2,185.00 15.84 1.53 1129 1.4% 5,550.00 -9.0 -1.2 204 62.1 15.60 0.49 160.50 15 3.3% 1,823.00 21.2 n/a 135 14.8 859.02 4.55 417.24 55 8.3% 90,000.00 -6.8 n/a 220 54.3 7.80 0.82 800.75 162 0.5% 5,740.00 -9.1 n/a 146 58.5 92.60 1.16 417.24 55 2.!% 36.63 -13.3 -13.3 2,277 14.3 944.90 30.00 1.00 4070 0.7% 22.00 27.3 31.0 N/A N/A 93.50 0.10 26.45 38 0.3% 12.40 36.3 34.6 1,958 62.8 11,153.00 24.39 7.73 273 8.9% 15.80 -39.9 -39.9 138 44.4 74.03 0.70 1.00 4070 0.0% 12.80 -8.6 -4.9 2,196 91.2 64.30 7.55 99.67 3766 0.2% N/A N/A N/A N/A 2.7 E 108.16 1.7 N/M 28.00 -12.3 -12.3 2,003 16.8 540.20 13.27 1.00 4070 0.3% N/A n.a. n.a. n.a. n.a. 239.20 8.7 N/A 1,200.00 46.3 54.4 519 67.0 52.40 3.48 26.45 38 9.1% N/A N/A N/A 1,296 14.8 460.47 14.01 1.77 248 5.6% N/A N/A N/A 917 49.9 5,200.00 12.00 1.50 124 9.7% 37.00 7.1 7.1 764 6.7 341.00 13.51 1.00 4070 0.3% 170.00 -20.6 -12.2 508 18.0 6.77 0.15 6.24 43 0.3% 35.00 64.8 n/a 575 11.2 285,024.00 16.44 1.00 147 11.2% 3.81 89.8 89.2 90 10.8 984.40 7.12 1.00 45 15.8% 4.75 9.5 22.2 273 18.0 1,889.60 5.91 1.66 25 23.7% 3,605.00 26.5 29.3 n.a. n.a. 8,177.00 23.43 1591.75 170 13.8% 4.27 71 69.8 214 12.0 1,178.80 8.61 1.00 45 19.1% 1,635.00 13 13.0 1,013 10.2 927.00 13.02 131.03 123 10.6% 16.38 14 10.7 290 68.7 56.80 0.77 1.37 259 0.3% 16.50 30 30.0 188 12.2 1,980.40 16.61 2.55 152 10.9% 8.24 31 18.9 1,435 19.7 10,603.20 33.89 3.38 192 17.6% t,250.00 -12 n/a 214 91.0 80.00 2.81 31.37 150 1.9% International Service Fievenue figures include all international services traffic (telephone, data, private line, etc.) but does not include equipment sales or sales of foreign affiliates unless stated. Data for North American carriers is net of settlements with foreign carriers. Source: Baring Securities, Datastream, company reports, Reuters, Financial Times. This table was compiled with the assistance of Baring Securities, Ltd; 1 America Square, London EC3 N2L7, Tel. +44-71-522-6000, Fax +44-71-702-0008. The table is provided for information purposes only and under no circumstances should be considered as an offer to sell or as a solicitation of any offer to buy. © Tele6eography, Inc., and Baring Securities, Ltd., 1994 TeleGeography 1994 Table 6: lnternet Statistics The internet Society The data on the following four pages were supplied by the Internet Society: [] The global international organization for open systems internetworking and the Internet. [] A common mechanism for standards making, operational administration and coordination, research coordination and education, global cooperation among national, regional, and other inter- national bodies. [] An international means for sharing information and encouraging development of internet-relat- ed infrastructure and use around the world. Members consist of individuals and organizations--commercial, governmental, and non-profit. For more information, contact the Internet Society; Suite 270; 12020 Sunrise Valley Drive; Reston, VA 22091 USA. isoc@isoc.org> Tel: 1-703-648-9888. Fax: 1-703-648-9887. internet Networks July 1993 Government (7%) Educational (8%) Defense (10%) Research (including commercial) (44%) Commercial (31%) An internet network consists of one or more Wide Area Networks (WANs) or Local Area Networks (LANs) capable of supporting the connection of almost any kind of computer, which is called a host. The size of networks vary from a few dozen computers to tens of thousands. Network Growth, 1990-94 [] non-USA [] USA 15O0O 1O000 © Internet Society/TeleGeography, Inc. 1994 Tables Host growth An Internet host is a term used for a computer of any kind attached to the Internet capable of supporting the TCP/IP protocol and possessing a unique global address. Nov/92 Mar/93 Au~94 Argentina n.a. 1 248 Australia 287 346 1780 Brazil n.a. 68 5896 Chile n.a. 9 3703 Colombia n.a. 0 144 Czech Republic 80 93 304 Finland 72 98 421 France 384 457 1648 Germany 377 516 1512 _Hong Kong 9 9 62 Indonesia 0 0 49 Italy 203 233 435 Japan 206 216 1564 Korea, Rep. of 37 38 348 Mexico n.a. 23 5164 Netherlands 107 127 291 New Zealand 58 60 334 Peru n.a. 42 Russia n.a. n.a. 259 Singapore 28 28 67 _Spain 44 68 246 Sweden 87 111 260 Switzerland 70 81 289 Taiwan 95 165 376 Thailand 2 12 73 © Internet Society/TeleGeography, Inc. 1994 United Kinqdom 308 368 1449 Uruguay n.a. 0 101 Venezuela n.a. 6 399 TeleGeography 1994 Traffic Growth, 1991-94 Traffic on the U.S. National Science Foundation (NSF) Backbone, by category of service The NSF Backbone Network is the most heavily used, large- scale Internet interconnection facility inthe world. In early 1995, the traffic will be transferred to commercial public Internet back- bones, and the NSFNet will be decommissioned. TCP/UDP are basic Internet stan- dards that allow up to 128,000 dif- ferent network applications and services to be provided over the Internet. Non-TCP/UDP stan- dards are for special network services and Open Systems Interconnection. [] Non-TCP/UDP [] Name lookup [] Interactive ¯ Other TCP/UDP (including gopher, WWW) [] Networked mail [] File Exchange United States International Traffic Traffic to and from the NSFNet Backbone, August 1994 This graph shows inbound and out- bound traffic from 25 of the top 27 countries to the NSFNet backbone. The top two countries, the United States and Canada, are excluded for reasons of scale. For those figures and others, see page 79. Traffic to and from CERN (the European Center for Nuclear Research), the ~argest European backbone, as of March 1993 is detailed in TeleGeography 1993. From U.S. NSFNet Megabytes To U.S NSFNet 200000 100000 0 100000 200000 300000 United Kingdom F r a n c e .......... Sweden .............. Netherlands ......... Swi~erland .......... J a p a n .................... Finland .................. Korea, Rep. of .................. Noway ....................... Taiwan ...................... Italy ........................... Singapore ...................... Mexico ........................ Hong Kong ..................... Israel ......................... Austria ........................ New Zealand ................... Denmark ....................... Spain .......................... Belgium ......................... South Africa ..................... B razil ........................... Poland ......................... © Internet Society/ TeleGeography, Inc. 1994 lnternet Networks and Traffic by Country, 1991-94 Tables United States Canada Australia United K~ngdom Networks Inbound Traffic to Aug 94 Jun 93 Sep 91 Aug 94 Jun 93 12900 5576 1758 13501.9 6057.3 1028 429 144 402.5 116.7 482 189 96 141.8 59.3 748 420 44 137.7 69.8 Germany ........ 822 443 France 780 453 Sweden 164 87 Netherlands 207 131 Switzerland 172 87 Japan 696 257 Finland 231 102 Korea, Rep. of 117 35 Norway 138 52 Taiwan 171 78 Ita_[y___ ........... 266 169 Singapore 41 16 Mexico 65 34 Hong Kong 28 8 Israel 123 48 Austria 153 84 New Zealand 163 Denmark 24 Spain 94 Belgium 57 South Africa 141 Brazil 108 Poland 75 Chile 51 Czech Republic 105 Russ=a 180 Ireland 77 Turkey 45 Thailand 38 Greece 38 NSFNet Outbound Traffic from NSFNet Sep 91 Aug 94 Jun 93 Sep 91 1034.4 12371,0 5446.0 1108.0 48.0 799,4 277.4 34.1 22.6 285.8 93.8 11.0 12.6 272.0 169.4 4.2 144 138.5 47.5 19.3 244.1 112.3 8.1 95 150,4 89.8 8.8 105,7 80.7 5,5 25 106.0 32.2 8.9 103.3 39.3 7.4 52 122 8 52 2 10.7 85.9 48 9 5.4 35 112.3 40.9 9.4 85.5 45.2 7.7 73 62.9 19.5 4.6 125.2 41.8 2.2 12 73.6 50.8 4.2 67.5 27.2 9.9 9 21.6 4.3 3.9 62.3 17.7 0.6 8 25.3 15.9 4.6 51.2 23.7 2.5 0 34.5 33.2 0 40.7 40.6 0.0 30 14.8 13.5 5,5 44.5 47.7 4.8 2 12,2 3.4 2.4 44,5 19.0 0.3 5 16.4 5.9 2.4 38,9 52.3 1.5 1 15.9 1.7 0.3 36.6 14.0 0.0 16 18.4 6.7 5.6 30.9 17.5 1.9 14 14.7 9.1 3.9 30,9 23.1 1.0 50 15 10.6 2.0 1.5 33,0 9.6 0.0 8 3 8.1 7.2 2.8 35.1 13.6 1.1 39 7 11.0 2.4 0.9 16.0 23.1 0.4 14 n.a. 3.9 1.2 n.a. 20.7 7.2 n.a. 42 0 3.6 1.7 0 19.8 9.7 0.0 52 8 3.5 1.9 0.2 18.7 17.5 1.9 45 n.a. 10.7 0.8 n.a. 10.2 3.1 n.a. 14 1 5.1 3.4 0.1 15.5 7.5 0.1 51 n.a. 9.5 1.1 n.a. 10.4 4.2 n.a. n.a. n.a. 2.6 n.a. n.a. 11.7 n.a. n.a. 24 n.a. 3.3 0.8 n.a. 9.0 4.4 n.a. 9 n.a. 2.2 n.a. n.a. 8.8 n.a. n.a. 13 n.a. 1.8 1.4 n.a. 9.2 1.5 n.a. 11 n.a. 3.2 0.6 n.a. 7.4 3.7 n.a. I::lungary 107 Slovakia 27 China 4 Kuwait 4 Portugal 70 Puerto Rico 6 India 7 Malaysia 6 SIovenia 31 Iceland 30 Costa Rica 5 13 n.a. 4.4 0.5 n.a. 3.4 1.5 n.a. n.a. n.a. 1.7 n.a. n.a. 4.7 n.a. n.a. 1 n.a. 0.7 0.0 n.a. 5.5 0.3 n.a. 35 n.a. 1,4 1.0 n.a. 4,6 3.6 n.a. n,a. n.a. 1.8 n.a. n.a. 4.1 n.a. n.a. 3 n.a. 1.2 1.4 n.a. 4.0 1.5 n.a. 3 n.a. 0.7 0.1 n.a. 4.6 0.4 n.a. n.a. n.a. 1.4 n.a. n.a. 3.7 n.a. n.a. 13 n.a. 0.9 0.4 n.a. 3.6 1.4 n.a. 2 n.a. 0.5 0.0 n.a. 3.7 1.1 n.a. Countries listed in descending order of total traffic to and from the NSFNet. Traffic data in gigabytes (109 bytes) per month. TeleGeography 1994 Table 7: The Top 50 International Routes, 993 C Countries 1. 2 3. 4 USA/Canada USA/Mexico USA/United Kingdom Hong Kong/Chma .5 .. USA/Germany 6 Switzerland/Germany 7 Germany/Austria 8. USA/Japan 9 Germany/France 10. Germany_AU_nit~ed Kingdom 11. Germany/Netherlands 12 United Kingdom/France 13. Germany/Italy 14. United Kingdom/ireland 15. Belgl~m_/Fran~ ce 16 Germany/Turkey 17 USA/France 18 Italy/France 19. Switzerland/France 20. Netherlands/Belgium 21. Switzerland/Italy 22 USA/Korea, Rep of 23 USA/Italy 24 USA/Dominican Republic MiTT eachway Total MiTT 2493.1/1602.3 4095 4 1398 8/456.1 1854 9 799 8/499 9 1299 7 688 4/535 0 1223 4 572.4/263.2 835 6 363.0/340 0 703.0 368 0/322 2 690.2 397 2/287.8 685.0 348 0/285.0 633.0 343.0/28~.0 624.0 290.0/284.8 574 8 287.0/285 0 572.0 320 0/239 8 559.8 266 0/207 1 473 ! 234 9/201.0 435 9 330.0/97.6 427 6 263 6/161 7 425.3 212 2/212 0 424.2 271.0/143 0 414 0 206.4/202 2 408 6 232.0/162.8 394 8 237.3/101.4 338 7 229 6/98.7 328.3 253 3/57 6 310.9 25. Franc~[~pain ._ 155~[1.43.Q .... 298.0 26. Netherlands/Unlte~ Kingdom 155 4/131 0 286 4 27 USA/Ta~wan 184 3/88 5 272 8 28 Umted Kingdom/Italy 139.0/124 5 263 5 29 Germany/Belgium 132 0/127 6 259.6 30 Germany/Spain 137 0/121 2 258.2 31 Germany/Poland 164 0/93 2 257 2 32 USA/Phlhppmes 219 1/35 9 255 0 33 USA/Colombia 200.2/51 8 252 0 34 USA/Austraha 1261/1182 2443 35 Australia/New Zealand 124.0/117 0 241 0 36 Spain/United Kingdom 123 6/115.0 238.6 37 Singapore/Malaysia 121 0/116.0 237 0 38. USA/Hong Kong 142 8/894 232 2 39 Japan/Korea 139 8/89 4 229 2 40. _Jap~n/_Cl~!~ ........................ 138 3/85~0 __ 22~3 41 Canada/United Kingdom 123 0/96 0 219.0 42. USA/Brazil 171 4/46 9 218.3 43. Australia/United Kingdom 117 0/101.0 218 0 44 USA/Israel 162.6/53.2 215 8 45. Sweden!Finland 105 0/98.0 203 0 46 Sweden/Norway 102.0/92 4 194 4 47 USA/India 134.1/55 5 189.6 48 USA/Netherlands 107 3/76 7 184.0 49 USA/Jamaica 144 8/37 4 182.2 50 Germany/Denmark 90 6/86.5 177 ! All data in millions of minutes of telecommunications traffic (MiFr). The country which generates more traffic on each route is listed first. The routes listed above total 24.8 billion mintues, 52% of all international traffic. For routes to and from the United States, calls are measured by point of billing in both directions. See Methodology, page 149. Americas International Traffic TeleGeography 1994 International Telephone Traffic--Americas Argentina Largest Telecommunications Routes, 1993 Destination MiTT Percentage of Outgoing Traffic 1. United States .......... 29.4 2. Uruguay .............. 26.1 3. Brazil ................ 14.2 4. Chile ................. 11.6 5. Spain ................. 8.9 6. Italy .................. 7.6 7. Paraguay .............. 6.7 8. Peru .................. 3.4 9. France ................ 3.0 10. Germany .............. 2.8 11. Bolivia ................ 2.7 12. United Kingdom ........ 2.3 13. Mexico ................ 2.3 14. Colombia .............. 1.5 15. Venezuela ............. 1.4 16. Switzerland ............ 1.4 17. Canada ................ 1.3 18. Israel ................. 1.1 19. Netherlands ........... 0.7 20. Ecuador ............... 0.6 Other ................. 7.9 2.5% 2.2% 2.0% 2.0% 1.7% 1.7% t1% 6.5% 5.6% 4.9% 1.0% 1.0% 0.9% 0 8% 0.5% 0.5% 5.8% MiTT is Minutes of Telecommunications Traffic. 8.5% 10.4% 21.5% 19.1% Data are in millions of minutes for public voice circuits. © Tele6eography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 132.5 159.7 192.3 Outgoing 89.4 124.3 137.1 Surplus (Deficit) 43.1 35.4 55.4 Total Volume 221.9 284.1 329.4 Note: Surplus/deficit and total volume data may appear inconsistent by +0.2 due to rounding. TeleGeography 1994 Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. United States .......... 63.9 2. Argentina ............. 20.2 11.1% 3. Italy .................. 9.4 5.2% 4. Germany .............. 8.9 4.9% 5. Portugal ............... 8.1 4.4% 6. France ................ 6.2 3.4% 7. United Kingdom ........ 6.1 3.3% 8. Uruguay ............... 5.2 2,9% 9, Japan ................. 4.5 2.5% 10. Spain ................. 4.4 2.4% 1993 35.O% 11. Paraguay .............. 4.4 2.4% 12. Chile .................. 4.1 2.2% 13. Switzerland ............ 3.7 2.0% 14. Sweden ............... 2.9 1.6% 15. Canada ................ 2.8 1.5% 16. Mexico ................ 2.4 1 3% 17. Venezuela ............. 2.0 1.1% 18. Netherlands ........... 1.9 1.0% 19. Bolivia ................ 1.8 1.o% 20. Colombia .............. 1.8 1 o% Other ................ 17.7 9 7% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. @ TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 261.6 330.6 373.8 Outgoing 171.2 169.9 182.4 Surplus (Deficit) 90.4 160.7 191.4 Total Volume 432.8 500.5 556.2 Note: Data based on billing point of traffic. International Telephone Traffic--Americas Canada Largest Telecommunications Routes, 1993 Destination MiTr Percentage of Outgoing Traffic 1, United States ........ 1669 2. United Kingdom ....... 132 53% 3. Hong Kong ............ 51 21% 4. France ............... 45 1.8% 5. Germany ............. 43 1.7% 6, Italy ................. 35 14% 7. Australia ............. 22 o.9% 8. Jamaica .............. 20 08% 9, India ................. 19 o.8°/o 10. Japan ................ 18 07% 11. Philippines ............ 18 07% 12. Netherlands ........... 15 o,6% 13. Mexico ............... 14 0.6% 14. Portugal .............. 13 0.5% 15. Switzerland ........... 12 05% Other ............... 351 142% 674% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits, © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 398.2 441.2 503 Outgoing 647.3 722.3 808 Surplus (Deficit) (249.1) (281.1) (305) Total Volume 1045.5 1163.5 1311 Notes: Incoming and outgoing totals are for Teleglobe only and exclude all Canada-U.S. traffic. Teleglebe data based on billing point of traffic. U.S. route traffic includes IPL resellers. Route data are rounded to the nearest million minutes. TeleGeography 1994 Largest Destination Mi’l-r 1. United States .......... 17.0 2. Argentina ............. 11,0 3. Brazil ................. 3.0 4 8% 4. Spain ................. 3.0 4.9% 5. Peru .................. 2.5 4 1% 6. Germany .............. 2.1 34% 7. Canada ................ 1.6 26% 8. United Kingdom ........ 1.5 25% 9. France ................ 1.5 2.5% 10. Bolivia ................ 1.3 21% 11. Italy .................. 1.4 2.3% 12. Mexico ................ 1.2 2.0% 13. Venezuela ............. 1.0 16% 14. Colombia .............. 0.9 1.5% 15. Uruguay ............... 0.8 13% Other ................ 11.2 Telecommunications Routes, Percentage of Outgoing Traffic 1993 18.0% 18,4% 27 9% MiTT is Minutes of Telecommunications Traffic. Data are in m=llions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 61.8 85.6 N.A. Outgoing 49.0 55.0 61.0 Surplus (Deficit) 12.8 30.6 N.A. Total Volume 110,8 140.6 NoA. Notes: Data are rounded to the nearest million minutes for routes above 3 million minutes. Data include Entet Chile, ChileSat, and VTR only. International Telephone Traffic--Americas Colombia Destination Largest Mi’i-r Telecommunications Routes, Percentage o| Outgoing Traffic 1. United States .......... 51.0 2. Venezuela ............ 11.1 10,8% 3. Panama ............... 4.3 4.2% 4. Ecuador ............... 4.3 4.2% 5. Spain ................. 3.1 3,1% 6. Mexico ................ 3.1 3.0% 7. United Kingdom ........ 2.1 2.1% 8. Germany .............. 1.9 1.9% 9. Italy .................. 1.9 1.8% 10. Brazil ................. 1.8 1 8% 11. Peru .................. 1.7 1.7% 12. France ................ 1.7 1.7% 13. Argentina .............. 1.6 1.5% 14. Canada ................ 1.2 1.2% 15. Costa Rica ............. 1.1 1.1% 16. Chile .................. 1.1 1.1% 17. Puerto Rico ............ 0.8 0.7% 18. Switzerland ............ 0.7 o,7% 19. Japan ................. 0.6 0.6% 20. Netherlands ........... 0.6 Other ................. 6.7 6.5% 1993 49.8% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 203.4 238.0 278.7 Outgoing 83.6 94.5 102.4 Surplus (Deficit) 115.8 143.5 176.3 Total Volume 287.0 332.5 381.1 Note: Data based on bilhng point of traffic. TeleGeography 1994 Cuba Largest Telecommunications Routes, 1992 Destination Mirr 1. United States .......... 6.6 2. Mexico ................ 0.9 83~/o 3. Spain ................. 0.9 8OO/o 4. Canada ................ 0.6 56~/o 5. Italy .................. 0.4 3.9% 6. Venezuela ............. 0.3 24% 7. Panama ............... 0.2 18O/o 8. Argentina .............. 0.2 1 4% 9. Dominica .............. 0.1 3~/o 10. Dominican Republic ..... 0.1 12% 11. Brazil ................. 0.1 0°/o 12. Chile .................. 0.1 o~/o 13. Colombia .............. 0.1 08% 14. Ecuador ............... 0.1 °7°/° 15. Nicaragua ............. 0.0 Other ................. 0.3 26% Percentage of Outgoing Traffic 59 7% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 18.6 N.A. Outgoing 12.0 12.4 N.A. Surplus (Deficit) N.A. 6.2 N.A. Total Volume N.A. 31.0 N.A. International Telephone Traffic--Americas Ecuador Destination Largest Telecommunications Routes, Mi’l-[ Percentage of Outgoing Traffic 1. United States .......... 18.1 2. Colombia .............. 4.2 12.6% 3. Chile .................. 1.0 3.1% 4. Peru .................. 1.0 3.1% 5. Venezuela ............. 1.0 3.o% 1993 54.6% 6. Brazil ................. 0.8 2.4% 7. Argentina .............. 0.7 2 1% 8. Spain ................. 0,7 2.1% 9. Panama ............... 0.6 1.9% 10. Germany .............. 0.6 1.7% 11. Mexico ................ 0.5 1.6% 12. Italy .................. 0.5 1.5% 13. Canada ................ 0.5 1.5% 14. France ................ 0.4 1.1% 15. United Kingdom ........ 0.4 1 1% 16. Costa Rica ............. 0.3 0.8% 17. Switzerland ............ 0.2 o.7% 18. Japan ................. 0.2 o.6% 19. Bolivia ................ 0.2 o5% 20. Netherlands ........... 0.2 o.5% Other ................. 1.2 3.7% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits, © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 78.9 87.4 101.4 Outgoing 23.6 28.6 33.1 Surplus (Deficit) 55.3 58.8 68.3 Total Volume 102.5 116.0 134.5 TeleGeography 1994 CO Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. United States ......... 542.1 2. Canada ............... 11.1 1.8% 3. Spain ................. 8.4 1.3% 4. France ................ 6.3 1.o% 5. United Kingdom ........ 4.8 o8% 1993 86.7% 6. Germany .............. 4.5 7. Colombia .............. 4.4 o.7% 8. Cuba .................. 3.8 o.6% 9. Italy .................. 3.5 10. Argentina .............. 3.2 0.5% 11. Guatemala ............. 3.1 o.5% 12. Brazil ................. 2.7 13. Costa Rica ............. 2.0 o.3% 14. Peru .................. 1.9 o.3% 15. Japan ................. 1.8 o.3% 16. Chile .................. 1.7 0.3% 17. El Salvador ............ 1.7 03% 18. Switzerland ............ 1.6 0.3% 19. Venezuela ............. 1.6 o.3% 20. Panama ............... 1.5 Other ................ 13.5 2.2% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 1115.0 1370.6 Outgoing 500.2 683.5 625.4 Surplus (Deficit) N.A. 431.5 745.2 Total Volume N.A. 1798.5 1996.0 Note: Data based on billing point of traffic. International Telephone Traffic--Americas Peru Destination Largest Telecommunications Routes, 1993 Mi’l-r Percentage of Outgoing Traffic 1. United States .......... 15.8 2. Chile .................. 3.2 8.2% 3. Spain ................. 2.7 6.9% 4. Argentina .............. 2.4 6.2% 5. Venezuela ............. 1.5 38% 6. Italy .................. 1.5 3.8% 7. Brazil ................. 1.5 3.8% 8. Colombia .............. 1.3 3.3% 9, Bolivia ................ 1.0 2.6% 10. Mexico ................ 0.9 2.3% 40.5% 11. Japan ................. 0.9 2.3% 12. Ecuador ............... 0.8 2.1% 13. Germany .............. 0.8 2.1% 14. Canada ................ 0.7 1.8% 15. United Kingdom ........ 0.7 1.8% 16. France ................ 0.7 1.8% 17. Panama ............... 0.4 1.0% 18. Switzerland ............ 0.4 1.0% 19. Costa Rica ............. 0.2 o5% 20. Netherlands ........... 0.2 o.5% Other ................. 1.4 3.6% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © Tete6eography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 105.9 128.1 152.4 Outgoing 29.3 32.1 39.0 Surplus (Deficit) 76.6 96.0 113.4 Total Volume 135.2 160.2 191.4 TeleGeography 1994 United States (Outgoing) Destination Largest Telecommunications Routes, MiTT 1992 MiTT 1993 1993 Percentage of Outgoing Traffic 1993 1. Canada ............. 2226.4 ..... 2493.1 2. Mexico ............. 1277.2 ..... 1398.8 12.2% 3. United Kingdom ....... 733.4 ...... 799.8 7.o% 4. Germany ............ 562.9 ...... 572.4 5.o% 5. Japan .............. 363 ...... 397.2 3.5% 21.8% 6. France .............. 239.8 ...... 263.6 2.3% 7. Dominican Republic ...249.4 ...... 253.3 2.1% 8. Korea, Rep. of ........ 206.4 ...... 237.3 2.0% 9. Italy ................ 207.2 ...... 229.6 1 9% 10. Philippines ........... 195.2 ...... 219.1 22% 11. Colombia ............ 172.4 ...... 200.2 1.7% 12. Taiwan .............. 162.5 ...... 184.3 1.6% 13. Brazil ............... 144.7 ...... 171.4 1.5% 14. Israel ............... 137.6 ...... 162.6 1.4% 15. Jamaica ............. 124.3 ...... 144.5 1.3% 16. Hong Kong ........... 113.9 ...... 142.8 1 2% 17. India ................. 97.8 ...... 134.1 1.2% 18. Australia ............ 113.2 ...... 126.1 1.1% 19. Netherlands ........... 92.7 ...... 107.3 o.9% 20. El Salvador ........... 99.9 ...... 107.0 o9% Other ......................... 3116.7 27.2% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 In c o m in g 2829.8 3149.4 3284.4 0 utg o in g 5984.5 6670.4 7500.3 Surplus (Deficit) (3154.7) (3521.0) (4125.9) Total Volume 8814.3 9819.8 10784.7 Notes: Incom=ng and outgoing traffic totals exclude Canada and Mexico traffic and traffic from non-Continental U.S. territories (Puerto Rico, Virgin Islands, and Guam). Route-by-route data also exclude these territories. Canada traffic excludes IPL resellers. All data based on the bilhng point of the traffic. See Methodology. International Telephone Traffic--Americas United States (Incoming) 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Largest Telecommunications Routes, 1993 Origin Mi]’r 1992 Mi]-I 1993 Canada ....... 1512.1 ...1602.3 United Kingdom .501.1 ...499.9 Mexico ......... 608.6 ...456.1 Japan .......... 277.9 ...287.8 Germany ....... 235.7 ...263,2 France ......... 156.6 ...161.7 Australia ....... 103.6 ...118.2 Korea, Rep. of .... 95.4 ...101.4 Italy ........... 117.9 ..... 98.7 Hong Kong ...... 82.3 ..... 89.4 Taiwan .......... 82.2 ..... 88.5 Netherlands ..... 66.7 ..... 76.7 Switzerland ...... 66 ....... 66.0 Dominican Rep .... 50.2 ..... 57.6 India ............ 60.1 ..... 55.5 Sweden ......... 53.4 ..... 53.5 Israel ........... 48.1 ..... 53.2 Colombia ........ 50.6 ..... 51.8 Brazil ........... 49.3 ..... 46.9 Venezuela ....... 53 ....... 45.6 Other .................. 1118.8 MiTT is Minutes of TelecommunicationsTraffic. Percentage of Incoming Traffic 1993 9.3% 8.5% 5.3% 4.9% 3.0% 2.2% 1,9% 1.8% 1.7% 1.6% 1.4% 1.2% 1.1% 1.0% 1.0% 1.0% 1.0% 0.9% 0.8% 29.7% 20.7% Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 TeleGeography 1994 USA: Other Correspondents Other Telecommunications Routes, 1993 Outgoing MiTr Incoming Mi’n" Country 1992 1993 1992 1993 Argentina 66.8 84.8 25.7 26.6 Austria 33,9 37.4 23.3 21.7 Bangladesh 17.7 20.1 1.5 1.9 Belgium 63.6 75,3 32.9 35.2 Bu~lg~.~!a .......... 4.6 5.7 1.1 1.3 Chile 38,3 49.7 16.2 17.8 China 68.4 110.1 25.4 41.1 Croatia n.a. 10.5 n.a. 3.4 Cuba 21.1 8.4 3.6 3,3 Cyprus ...... 9.6 6.3 3.9 3.8 Czech and Slovak Republics 16.2 14.9 10.8 8.4 Denmark 32.3 36.4 22.0 23.6 Ecuador 73.3 86.4 11.6 12.1 El Salvador 99.9 107.0 8.5 7.8 Finland 17.2 19.1 14.6 15.9 Greece 70.0 75.5 25.7 25.3 Guatemala 88.0 93.8 18.5 14.1 Hungary 17.6 20.4 12.9 14.0 Iceland 6.9 7.3 5.5 6.4 Indonesia 37.5 44.8 12.8 17.5 Ireland 66.8 75.7 27.6 31.4 Jamaica 124.4 144.8 42.4 37.4 Luxembourg 5.4 6.6 4.2 4.7 Macau 1.5 1.7 0.8 1.0 Malays!a 28.7 32.7 10.2 12.7 New Zealand 21.3 24.7 16.1 19.2 Norway 32.6 34.7 26.0 28.3 Pakistan 64.2 68.7 7.2 7.5 Peru 80.1 89.6 12.5 14.7 Philippines 195.2 219.1 22.3 35.9 Poland 73.9 81.6 25.0 28.7 Portugal 35.1 47.1 9.6 10.4 Russia n.a. 45.6 n.a. 12.8 Saudi Arabia 61.5 67.6 29.9 29.7 ~Si~n~g~a p o r e 47.1_ 56~.6 ........ ~34._0_ ..... 37.4 Slovenia n.a. 2.1 n.a. 0.0 South Africa 29.1 36.3 18.7 20.8 Spain 105.6 109.5 53.1 45.4 Sri Lanka 5.4 6.1 1.2 1.7 Sweden 57.4 62.0 53.4 53.5 Switzerland 93.5 104.4 65.3 66.0 Thailand 53.1 62.8 15.7 18.8 Tu rkey 37.6 40.1 16.0 16.8 United Arab Emirates 22.9 27.9 16.1 17.4 Ur_uguay 13.6 15.1 2.2 3.1 Venezuela 76.0 87.9 53.0 45.6 Yugoslavia n.a. 18.7 n.a. 1.1 Note: All data based on the billing point of the traffic. See Methodology. International Telephone Traffic--Americas Uruguay Largest Destination Mil"r 1. Argentina ............. 22.3 2. Brazil ................. 4.7 3. United States .......... 3.1 4. Spain ................. 1.1 2.9% 5. Chile .................. 0.9 2.4% 6. Paraguay .............. 0.6 1.7% 7. Italy .................. 0.6 1.5% 8. France ................ 0.4 1.0% 9. Germany .............. 0.4 1.0% 10. United Kingdom ........ 0.3 o.8% 11. Mexico ................ 0,3 o.7% 12. Canada ................ 0.2 o.7% 13. Israel ................. 0.2 0.]% 14, Venezuela ............. 0.2 o.6% 15. Switzerland ............ 0.2 o.6% 16. Peru .................. 0.1 o4% 17. Colombia .............. 0.1 0.4% 18. Australia .............. 0.1 o.3% 19. Ecuador ............... 0.1 o.3% 20. Netherlands ........... 0.1 o3% Other ................. 1,3 3.4% Telecommunications Routes, Percentage of Outgoing Traffic 12.7% 8.2% 1993 59.7% MiTT is Minutes of Telecommunications Traffic, Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 48.0 53.0 58.0 Outgoing 29.3 30.2 37.4 Surplus (Deficit) 18.7 22.8 20.6 Total Volume 77.3 83.2 95.4 Note: Data based on billing point of traffic. TeleGeography 1994 :ii Venezuela Destination Largest Telecommunications Routes, Mi’l’r Percentage of Outgoing Traffic 1. United States .......... 55.9 2. Colombia ............. 17.1 12.8% 3. Spa~n ................. 8.0 6.0% 4. Italy .................. 6.9 51% 5. Dominican Republic ..... 3.7 2.8% 1993 42.0% 6. Mexico ................ 3.5 2.6% 7. Peru .................. 3.3 2.5% 8. Canada ................ 2.6 2.o% 9. Portugal ............... 2.4 1 8% 10. Brazil ................. 2.3 1.7% 11. Chile ................. 2.2 1.6% 12. France ................ 2.2 16% 13. Germany .............. 2.1 1.6% 14. Argentina .............. 2.0 1.5% 15. Ecuador ............... 1.8 1.4% 16, United Kingdom ........ 1.8 1.3% 17. Netherlands Antilles ..... 1.5 12% 18. Puerto Rico ............ 1,3 1.0% 19. Panama ............... 1.2 o.9% 20. Switzerland ............ 1.0 o.7% Other ................ 10.5 7.9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 103.9 128.6 148.3 Outgoing 91.7 115.5 133.3 Surplus (Deficit) 12.2 13.1 15.0 Total Volume 195.6 244.1 281.6 Note: Data based on billing point of traffic. Europe International Traffic TeleGeography 1994 Largest Telecommunications Routes, 1993 MiTT Percentage of Outgoing Traffic 1. Germany ............ 322.2 2. Switzerland ........... 58.4 7 6% 3. Italy ................. 39.3 5.1% 4. Yugoslavia ............ 30.8 4.o% 5. Hungary .............. 26.5 3 4% 6. Turkey ............... 23.9 3 1% 7. United States ......... 23.7 3.1% 8. United Kingdom ....... 23.3 3.0% 9. Croatia ............... 21.2 2.8% 10. France ............... 20.2 2.6% 11. Poland ............... 18.2 2.4% 12. Netherlands ........... 18.1 2.4% 13. Czech Republic ........ 14.1 1.8% 14. Slovenia .............. 13.3 1 7% 15. Slovak Republic ........ 9.7 1.3% 16. Belgium ............... 8.1 1.1% 17. Romania ............... 7.8 1.o% 18. Russia ................ 7.3 o.9% 19. Sweden ............... 7.1 o.9% 20. Spain ................. 6.5 0.8% Other ................ 68.0 8 9% MiTT is Minutes of Telecommunications Traffic. 42.0% Data are in millions of m~nutes for public voice circuits. © Tele6eography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 636.9 692.3 751.0 0 utg o in g 641.9 713.4 767.4 Surplus (Deficit) (5.0) (21.1) (16.4) Total Votu me 1278.8 1405.7 1518.4 Note: Data based on billing point of traffic. International Telephone Traffic--Europe Belgium Destination Largest Telecommunications Routes, MiTr Percentage of Outgoing Traffic 1. France .............. 234.9 2. Netherlands .......... 202.2 3. Germany ............ 127.6 4. United Kingdom ....... 87.2 5. Italy ............... 50.7 8 9% 5 2% t3.0% 1993 20.6% 24.0% 6. Luxembourg ........... 35.7 3.6% 7. United States .......... 35.4 3.6% 8. Spain ................ 28.6 2.9% 9. Switzerland ........... 24.4 2.5% 10. Sweden .............. 11.8 1.2% 11. Portugal .............. 10.1 1 0% 12. Greece ................ 9.8 1.o% 13. Turkey ................ 9.2 0.9% 14. Denmark .............. 9.0 o.9% 15. Austria ................ 7.8 o.8% 16. Morocco .............. 7.8 0.8% 17. Ireland ................ 5.5 0.6% 18. Canada ................ 4.8 0.5% 19. Norway ............... 4.4 05% 20. Israel ................. 3.8 0.4% Other ................ 68.7 7.o%o MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 845.0 952.7 1025.3 Outgoing 822.7 911.1 979.4 Surplus (Deficit) 22.3 41.6 45.9 Total Volume 1667.7 1863.8 2004.7 Note: Data based on billing point of traffic. TeleGeography 1994 Largest Telecommunications Routes, 1992 Destination MiTT 1. Germany ............. 13.6 2. Turkey ............... 10.3 3. Greece ................ 8.8 4. Yugoslavia ............. 4.6 5. Italy .................. 4.2 6. Austria ................ 3.4 7. France ................ 3.2 8. United Kingdom ........ 3.1 9. Switzerland ............ 1.6 10. Cyprus ................ 1.5 11. Czech & Slovak Reps .... 1.4 12. Poland ................ 1.2 13. Netherlands ........... 1.1 14. Sweden ............... 1.0 15. Belgium ............... 1.0 16. Hungary .............. 0.8 Other ................ 20.2 Percentage of Outgoing Traffic 2.0% 1.9% 1.7% 1.5% 1.4% 1.2% 10% 57% 52% 42% 4.0% 38% 127% 10.9% 16.8% 24.6% MiTT is Minutes of Telecommunications Traffic. Data are ~n millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. N.A. Outgoing 73 81 91 Surplus (Deficit) N.A. N.A. N.A. Total Volume N.A. N.A. N.A. Note: Route data exclude Romania and Russia. Intemotional Telephone Traf~c--Europe Croatia Destination Largest MiTT 1. 6ermany ............. 41.0 2. Austria ............... 13.2 3. Italy ................. 11.6 4. SIovenia ............... 6.8 5. Switzerland ............ 6.1 Telecommunications Routes, Percentage of Outgoing Traffic 11.2% 9.9% 58% 5.2% 1993 35.0% 6. United Kingdom ........ 4.7 4.0% 7. France ................ 4.5 3,8% 8. United States .......... 3.4 2,9% 9. Netherlands ........... 2.6 2,3% 10. Macedonia ............ 2.1 1.8% 11. Canada ................ 1.7 1.5% 12. Sweden ............... 1.6 1.4% 13. Spain ................. 1.6 1.4% 14, Hungary ............... 1.6 1.4~/~ 15. Denmark .............. 1.5 1.3% 16. Czech & Slovak Reps .... 1.4 1.2% 17. Russia ................ 1.2 1.o% 18. Belgium ............... 1.2 t.o% 19. Australia .............. 1.1 1.o% 20. Norway ............... 0.6 o.5% 21. Other ................. 7.7 6.6% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 156.7 189.1 170.3 Outgoing 93.4 104.7 117.2 Surplus (Deficit) 63.3 84.4 53.0 Total Volu m e 250.1 293.8 287.4 101 and Slovak Reps. Largest Destination MiTT 1. Germany ............. 52.8 2. Austria ............... 31.4 3. Italy ................. 10.0 55O/o 4. United Kingdom ........ 9.9 5.4% 5. United States .......... 8.8 49% 6. France ................ 7.1 3.9% 7. Poland ................ 6.9 38% 8. Russia ................ 6.7 37% 9. Switzerland ............ 6.4 3.5% 10. Hungary ............... 6.2 34% 11. Netherlands ........... 5.2 2.8% 12. Canada ................ 4.8 26% 13. Belgium ............... 3.1 1.7% 14. Ukraine ............... 3.1 1.7% 15. Yugoslavia ............. 3.0 17% 16. Hong Kong ............. 2.9 1.6% 17. Croatia ................ 2.1 !.2% 18. Spain ................. 2.1 12% 19. Sweden ............... 2.1 1 2% Other ................ 17.0 Telecommunications Routes, Percentage of Outgoing Traffic 9.4% 11.8% 1993 29.1% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 179.5 236.0 253.2 Outgoing 128.1 178.0 181.4 Surplus (Deficit) 51.4 58.0 71.8 Total Volume 307.6 414.0 434.6 Note: Though the Czech and Slovak Republics have been independent countries since January 1, 1993, separate traffic data has only been calculated since September 1, 1993. The above data reflect combined traffic from both countries and do not include traffic from the two countries to each other. International Telephone Traffic--Europe Denmark Destination Largest Mi’n" Telecommunications Routes, Percentage of Outgoing Traffic 1. Germany ............. 86.5 2. Sweden .............. 75.8 3. United Kingdom ....... 49.0 4. Norway .............. 46.5 5. United States .......... 22.3 4.9% 6. France ............... 20.2 4.5% 7. Netherlands ........... 17.8 3.9% 8. Italy ................. 11.3 2.5% 9. Switzerland ........... 10.3 2.3% 10. Belgium ............... 9.2 2.o% 11. Spain ................. 9.0 2.o% 12. Finland ................ 8.9 2.o% 13. Faroe Islands .......... 8.1 1.8% 14. Poland ................ 7.5 1.6% 15. Turkey ................ 4.8 1 1% 16. Greenland ............. 4.2 0.9% 17. Austria ................ 4.1 o.9% 18. Iceland ................ 3.1 o.7% 19. Canada ................ 3.0 o.7% 20. Greece ................ 2.8 0.6% Other ................ 47,9 1993 10.8% 10.3% 10.6% 16.8% 19.1% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 397.2 425.2 460.0 Outgoing 394.5 424.5 452.3 Surplus (Deficit) 2.7 0.7 7.6 Total Volume 791.7 849.7 912.3 Note: Surplus/deficit and total volume data may appear inconsistent by +0.1 due to rounding. 103 TeleGeography 1994 ,,,,,Finland Largest Telecommunications Routes, 1993 Destination Mi’l-r Percentage of Outgoing Traffic 1. Sweden .............. 98 2. Germany ............. 23 9.1% 3. Russia ............... 20 7.9% 4. United States .......... 16 6.3% 5. United Kingdom ....... 15 5.9% 6. Denmark .............. 9 36% 7. Norway ............... 9 36% 8. France ................ 7 2.8% 9. Netherlands ........... 6 24% 10. Switzerland ............ 5 2.o% 11. Italy .................. 5 2.o% 12. Spain ................. 5 2.o% Other ................ 35 13.8% 38.7% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. @ TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. N.A. Outgoing 215 235 253 Surplus (Deficit) N.A. N.A. N.A. Total Volume N.A. N.A. N.A. Nete: Data are rounded to the nearest million minutes for Telecom Finland only. Traffic to Estonia is excluded. International Telephone Traffic--Europe France Largest Telecommunications Routes, 1993 Destination Mil-r Percentage of Outgoing Traffic 1. Germany ............ 285.0 2. United Kingdom ....... 285.0 3. Italy ................ 212.0 4. Belgium ............. 201.0 5, Spain ............... 155.0 6.0% 6. United States ........ 153.0 5.9% 7. Switzerland .......... 143.0 5.6% 8. Portugal ............. 119.0 4.6% 9. Netherlands ........... 81.0 3.1% 10. Morocco ............. 71.0 2.8% 11. Algeria ............... 56.0 2.2% 12. Tunisia ............... 34.0 1.3% 13. Canada ............... 32.0 1.2% 14. Turkey ............... 29.0 1 1% 15. Sweden .............. 22.0 0.9% 16. Poland ............... 20.0 0.8% 17. Denmark ............. 19.0 0.7% 18. Austria ............... 18.0 0.7% 19. Greece ............... 18.0 0.7% 20. Luxembourg ........... 18.0 0.7% Other ............... 605.0 MiTT is Minutes of Telecommunications Traffic. 11.1% 11.1% 8.2% 7.8% 23.5% Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 2355 2540 2710 0 utg o in g 2295 2449 2576 Surplus (Deficit) 60 91 64 Total Volume 4650 4989 5356 Note: Data based on billing point of traffic. 105 TeleGeography 1994 Germany Destination Largest Telecommunications Routes, MiTT 1. United States ......... 425 2. Austria .............. 368 3. France .............. 348 4. United Kingdom ....... 343 5. Switzerland .......... 340 Percentage of Outgoing Traffic 7.9% 7.4% 7.3% 7.3% 9.1% 1993 6. Turkey .............. 327 7. Italy ................ 320 8. Netherlands .......... 290 9. Poland .............. 164 3.5% 10. Spain ............... 137 29% 11. Belgium ............. 132 2.8% 12. Croatia .............. 107 2.3% 13. Greece .............. 105 2,2% 14. Denmark ............. 91 1.9% 15. Czech & Slovak Reps...90 1.9% 16. Sweden .............. 66 1 4% 17. Yugoslavia ............ 65 1.4% 18. Hungary .............. 55 1 2% 19. Canada ............... 49 1.o% 20. Russia ............... 48 lO% Other ............... 810 7.0% 6.8% 6.2% 17.3% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 2900 3100 3707.8 0 utg o in g 3605 4087 4679.6 Surplus (Deficit) (705) (987) (971.8) Total Volume 6505 7187 8387.4 Notes: Incoming totals for 1991 and 1992 are rounded to the nearest 100 million minutes. Other data for those years are rounded to the nearest million minutes. Data for 1993 include 159 million MiTT of foreign-billed Country Direct traffic, of which 145 million MiTT were on the United States route. 106 International Telephone Traffic--Europe Greece Largest Telecommunications Routes, Destination Mi’l-r Percentage of Outgoing Traffic 1. Germany ............. 68.1 2. United Kingdom ....... 45.7 3. Italy ................. 30.7 9.1% 4. United States .......... 23.8 7.1% 5. France ............... 19.2 5.7% 6. Cyprus ............... 13.6 4.1% 7. Canada ................ 9.8 2.9% 8. Australia .............. 9.4 2.8% 9. Belgium ............... 9.1 2.7% 10. Netherlands ........... 9.0 2.7% 11. Switzerland ............ 8.7 2.6% 12. Bulgaria ............... 7.5 2.2% 13. Albania ................ 6.9 2.1% 14. Romania ............... 5.7 1.7% 15. Russia ................ 5.5 1.6% 16. Yugoslavia ............. 5.5 1.6% 17. Turkey ................ 5.3 1.6% 18. Austria ................ 5.1 1.5% 19. Sweden ............... 5.1 1 5% 20. Poland ................ 4.0 1.2% Other ................ 38.5 MiTT is Minutes of Telecommunications Traffic. 1993 13.6% 20.3% 11.5% Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 320.1 359.7 406.1 Outgoing 245.0 298.9 336.2 Surplus (Deficit) 75.1 60.8 70.0 Total Volume 565.1 658.5 742.3 Note: Surplus/deficit and total volume data may appear inconsistent by +0.1 due to rounding. Data based on billing point of traffic. 107 TeleGeography 1994 Hungary Largest Telecommunications Routes, Bestination MiTT Percentage of Outgoing Traffic 1. Germany ............. 51.0 2. Austria ............... 2B.8 3. United States ......... 12.8 6.0% 4. Italy ................. 11.6 54% 5. Romania .............. 10,5 4.9% 6. United Kingdom ........ 9.0 42% 7. France ................ 7.0 33% 8. Russia ................ 6,9 3 2% 9. Switzerland ............ 6.7 31% 10. Czech & Slovak Reps .... 5.4 25%0 11. Netherlands ........... 5.3 2.5% 12. Yugoslavia ............. 4.9 23% 13. Canada ................ 4.1 1,9% 14. Sweden ............... 3.9 1,8% 15, Ukraine ............... 3.7 1.7% 16. Belgium ............... 3.4 1,6% 17. Hong Kong ............. 2,7 1.2% 18. Croatia ................ 2.6 1.2% 19. Poland ................ 2.4 1.1% 20. Other ................ 32.7 126% 1993 153% 23.9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice c~rcuits, © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 137.4 150.5 192.8 0 utgoing 150.6 183.8 213.2 Surplus (Deficit) (13.2) (33.3) (20.4) Total Volume 288.0 334.3 406.0 108 International Telephone Traffic--Europe iceland Largest Telecommunications Routes, Destination Mil"r Percentage of Outgoing Traffic 1. United States .......... 6.4 2. Denmark .............. 3.3 13.6% 3. Sweden ............... 2.6 11.0% 4. United Kingdom ........ 2.6 lO.6% 5. Norway ............... 2.0 8.4% 6. Germany .............. 1.8 7.6% 7. France ................ 0.7 3 0% 8. Netherlands ........... 0.7 2.8% 9. Faroe Islands .......... 0.6 2.4% 10. Spain ................. 0.3 1.4% 11. Italy .................. 0.3 1.3% 12. Belgium ............... 0.3 1.2% 13. Finland ................ 0.3 1.1% 14. Switzerland ............ 0.2 1.0% 15. Luxembourg ........... 0.2 o7% 16. Austria ................ 0.1 0.6% 17. USSR (former) .......... 0.1 0.4% 18. Portugal ............... 0.1 0.4% 19. Poland ................ 0.1 0.4% 20. Ireland ................ 0.1 0.3% Other ................. 1.4 5.7% MiTT is Minutes of Telecommunications Traffic. 1993 26.4% Data are in millions of minutes for public voice circuits. © Tele6eography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 20.8 21.7 23.4 Outgoing 20.8 22.1 24.1 Surplus (Deficit) 0.0 (0.4) (0.7) Total Volume 41.6 43.8 47.5 Note: Data based on billing point of traffic. 109 TeleGeography 1994 Destination Largest Telecommunications Routes, FY 1993/94 MiTT Percentage of Outgoing Traffic 1. United Kingdom ....... 207.1 2. Northern Ireland ....... 68.1 17.8% 3. Umted States .......... 37.0 9.6% 4. Germany ............. 13.3 3.5% 5. France ............... 10.9 28% 6. Netherlands ........... 6.2 1 6% 7. Spain ................. 5.0 3% 8. Italy .................. 4.7 1.2% 9. Belgium ............... 4.0 1.o% 10. Canada ................ 3,8 1,o% 11. Australia .............. 3,2 08% 12. Switzerland ............ 2.4 o.6% 13. Sweden ............... 1.9 o 5% 14. Denmark .............. 1.9 o5% 15. Japan ................. 0.9 o.2% 16, Austria ................ 0.8 o.2% 17. Norway ............... 0,7 o.2% 18. Hong Kong ............. 0.7 o2% !9. South Africa ........... 0.6 0.2% 20. Portugal ............... 0.6 o.2% Other ................ 10.2 27% MiTT is Minutes of Telecommunications Traffic. 53.9% Data are in millions of minutes for public voice circuits. National Traffic Balance MiTT FY 1991/92 FY 1992/93 FY 1993/94 I n c o m in g 140.4 383.0 423.0 Outgoing 83.4 296.6 383.9 Surplus (Deficit) 57.0 86.4 39.0 Total Volume 223.8 679.6 806.9 Notes: FY ends on 31 March. Traffic balances for FY 1991/92 exclude the Umted Kingdom. Surplus/deficit and total volume data may appear inconsistent by _+0.1 due to rounding. 110 International Telephone Traffic--Europe ltaly Largest Telecommunications Routes, Destination MiTT Percentage of Outgoing Traffic 1. Germany ............ 239.8 2. France .............. 212.2 3. Switzerland .......... 162.8 4. United States ........ 154.5 5. United Kingdom ....... 124.5 7.7% 6. Canada ............... 56.7 3.5% 7. Spain ................ 55.7 3.5% 8. Belgium .............. 43.1 2.7% 9. Austria ............... 38.8 2.4% 10. Netherlands ........... 29.4 1.8% 11. Morocco ............. 27.4 1.7% 12. Greece ............... 25.4 1.6% 13. Poland ............... 19.5 1.2% 14. Hong Kong ............ 19.0 1.2% 15. Tunisia ............... 18.4 1.1% 16. Romania .............. 17.8 1.1% 17. Slovenia ............. 15.8 1.o% 18. Dominican Republic .... 14.9 o9% 19. Brazil ................ 14.2 0.9% 20. Russia ............... 13.1 0.8% Other ............... 306.8 10.1% 9.6% 1993 13.2% 14.9% 19.1% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeteGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 1541.0 1672.7 Outgoing 1219.5 1473.4 1609.7 Surplus (Deficit) N.A. 67.6 63.0 Total Volume N.A. 3014.4 3282.4 Note: Data reflect combined totals for Iritel and Italcable. Prior to 1994, Iritel (formerly ASST) handled intra- European traffic only and Italcable carried overseas traffic. TeleGeography 1994 Destination 1. Belgium 2. Germany ............. 41.5 3. France ............... 38.6 4. Portugal .............. 12.7 6.4% 5. United Kingdom ....... 11.9 6.o% 6. Italy .................. 9.7 4.9% 7. Netherlands ........... 7.2 3.6% 8. Switzerland ............ 6.1 3.1% 9. United States .......... 4.7 2.3% 10. Denmark .............. 3.4 1.7% 11. Spain ................. 2.9 1.5% 12. Austria ................ 1.8 o.9% 13. Sweden ............... 1.7 0.8% 14. Greece ................ 1.2 o.6% 15. Ireland ................ 0.8 o.4% 16. Japan ................. 0.7 o3% 17. Norway ............... 0.6 0.3% 18. Hong Kong ............. 0.6 0.3% 19. Canada ................ 0.6 o.3% 20. Finland ................ 0.5 0.3% Other ................. 6.8 3.4% Largest Telecommunications Routes, Mi’l-r Percentage of Outgoing Traffic .............. 45.3 1993 22.8% 2O.8% 19.4% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © Tele6eography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 102.4 107.5 131.7 Outgoing 165.3 181.0 199.3 Surplus (Deficit) (62.9) (73.5) (67.6) Total Volume 267.7 288.5 331.0 International Telephone Traffic--Europe Netherlands Destination Largest MiTT 1. Germany ............ 284.8 2. Belgium ............. 206.4 3. United Kingdom ...... 155.4 4. France ............... 93.4 5. United States ......... 76.6 Telecommunications Routes, Percentage ef Outgoing Traffic 6. Italy ................. 39.6 3 2% 7. Switzerland ........... 36.7 3.0% 8. Spain ................ 34.0 2.7% 9. Turkey ............... 28.1 2.3% 10. Sweden .............. 19.3 1 6% 11. Denmark ............. 18.1 1.5% 12. Austria ............... 17.9 1.4% 13. Canada ............... 14.0 1.1o/o 14. Poland ............... 11.2 o9% 15. Norway .............. 10.8 0.9% 16. Morocco ............. 10.0 o.8% 17. Greece ............... 10.0 o.8% 18. Portugal ............... 9.2 0.7% 19. Suriname .............. 9.0 07% 20. Ireland ................ 8.2 07% Other ............... 145.5 12.6% 11 8% 1993 16.7% 23.0% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 i.qc.£rni.ng ..... NIA: ..-.~_-‘ _1~0~9.0" .... 1~6310-i 0Utgpi£g 1!:1!~7:7 ....... 1133.# ..... 12##:2 Surplus (Deficit) N.A. (94.9). (75.2). Total Volume N.A. 2172.9 2401.2 Nete: Data based on billing point of traffic. Largest Telecommunications Routes, 1992 Destination MiTT Percentage of Outgoing Traffic 1. Sweden .............. 92.4 2. Denmark ............. 50.4 3. United Kingdom ....... 45.5 4. Germany ............. 25.4 7.3% 5. United States .......... 19.3 5.5% 6. France ............... 11.7 34% 7. Netherlands ........... 9.6 2.8% 8. Finland ................ 9.5 2.7% 9. Hong Kong ............. 7.6 2.2% 10. Spain ................. 6.4 1.8% 11. Switzerland ............ 5.4 1.5% 12. Italy .................. 5.1 1.5% 13. Belgium ............... 4.6 1.3% 14. Poland ................ 3.8 1.1% 15, Canada ................ 3.7 lO% 16. Netherlands AnNles ..... 3.4 1 o% 17. Yugoslavia ............. 2.3 o.7% 18. Austria ................ 2,1 o6% 19. Turkey ................ 2.1 20. Iceland ................ 1.8 0.5% Other ................ 37.0 14.5% 13.1% MiTT is Minutes of Telecommumcations Traffic. 26.5% 106% Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance 1992 1993 314 322.5 349 376.2 (35) (53.7) 663 698.7 M iTT 1991 Incoming N.A. Outgoing 308.5 Surplus (Deficit) N.A. Total Volume N.A. 114 International Telephone Traffic--Europe Poland Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. Germany ............. 93.2 2. United States .......... 28.7 3. United Kingdom ....... 14.5 4. France ............... 14.3 5. Italy ................. 11.9 10.5% 5.3% 5.2% 4.4% 1993 34.2% 6. Austria ............... 10.4 3.8% 7. Sweden ............... 9.6 3.5% 8. Ukraine ............... 8.4 3.1% 9. Netherlands ........... 8.2 3.o% 10. Canada ................ 7.0 2.6% 11. Belgium ............... 5.5 2.o% 12. Australia .............. 5.3 1.9% 13. Denmark .............. 4.9 1.8% 14. Czech & Slovak Reps .... 4.6 1.7% 15. Switzerland ............ 4.1 1.5% !6. Belarus ............... 3.6 1.3% 17. Lithuania .............. 2.5 o.9% 18. Greece ................ 2.3 o.8% 19. Spain ................. 2.2 o.8% 20. Norway ............... 2.0 o.7% Other ................ 29.5 10.8% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 366.6 431.5 Outgoing 125.0 212.7 272.7 Surplus (Deficit) N.A. 153.9 158.8 Total Volume N.A. 579.3 704.2 TeleGeography 994 Largest Telecommunications Routes, 1993 Destination MiTT 1. France ............... 49.0 2. Spain ................ 34.6 3. United Kingdom ....... 25.5 4. Germany ............. 24.1 5. Switzerland ........... 12.4 6. United States .......... 10.3 7. Italy .................. 9.3 8. Brazil ................. 8.9 9. Belgium ............... 7.7 10. Netherlands ........... 7.4 11. Angola ................ 6.2 12. Canada ................ 4.4 13. Sweden ............... 2.7 14. Luxembourg ........... 2.4 15. Denmark .............. 2.1 16. (3uinea-Bissau ......... 2.0 17. Cape Verde ............ 2.0 18. Mozambique ........... 1.9 19. South Africa ........... 1.9 20. Austria ................ 1.3 Other ................ 16.4 Percentage of Outgoing Traffic 5.3% 44% 4.0% 3.8% 33% 3.2% 2.7% 1,9% 1.2% 1.0% 0.9% 0.9% 0.9% 0.8% 0.8% 0.6% MiTT is Minutes of Telecommunications Traffic. 11.0% 10.4% 149% 21.1% 7.0% Data are in millions of minutes for public voice circuits. @ TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 384.0 N.A. 438.2 Outgoing 186.1 212.0 232.6 Surplus (Deficit) 197.9 N.A. 205.6 Total Volume 570.1 N.A. 670.8 Note: Data include traffic for Telecom Portugal, which carries intra-European traffic only, and CPRM, which car- ries overseas traffic. CPRM had 49.3 million MiTT outbound and 84.3 million MiTT inbound in 1993. Data based on billing point of traffic. International Telephone Traffic--Europe Russia Largest Telecommunications Routes, Destination MiTr Percentage of Outgoing Traffic 1. Germany ............. 30.1 2. United States .......... 22.5 3. China ................ 10.7 5.3% 4. United Kingdom ........ 8.6 4.3% 5. Finland ................ 8.6 4.3% 6. Italy .................. 8.5 4.2% 7. Poland ................ 8.0 4.0% 8. France ................ 6.7 3.3% 9. Yugoslavia ............. 6.7 3.3% 10. Israel ................. 6.1 3.0% 11. Bulgaria ............... 6.1 3.0% 12. India .................. 5.3 2.6% 13. Czech & Slovak Reps .... 5.2 2.6% 14. Turkey ................ 5.1 2 5% 15. Hungary ............... 4.9 24% 16. Austria ................ 4.1 2.o% 17. Sweden ............... 3.5 1.7% 18. Netherlands ........... 3.2 1.6% 19. Switzerland ............ 3.0 1.5% 20. Spain ................. 3.0 1.5% Other ................ 41.! MiTT is Minutes of Telecommunications Traffic. 11.2% 1993 15.0% 20.4% Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 230.7 268.0 Outgoing N.A. 175.6 201.0 Surplus (Deficit) N.A. 55.1 67.0 Total Volume N.A. 406.3 469.0 Note: Data do not include traffic to or from other former Soviet republics. For traffic within the Commonwealth of Independent States, see page 124. !17 TeleGeography 1994 Largest Telecommunications Routes, Destination MiTT 1. Germany ............. 12.7 2. Austria ................ 9.4 3. Italy .................. 8.7 4. Croatia ................ 7.6 5. Yugoslavia ............. 6.3 6. Switzerland ............ 2.3 37% 7. Macedonia ............ 1.8 2.9% 8. France ................ 1.7 26% 9. United Kingdom ........ 1.6 2.5% 10. United States .......... 1.3 21% 11, Hungary ............... 0,9 1.5% 12. Netherlands ........... 0.8 t.3% 13. Czech & Slovak Reps .... 0.8 1.2% 14. USSR (former) .......... 0.6 1.o% 15. Sweden ............... 0.6 1 o% 16. Australia .............. 0.6 o9% 17. Belgium ............... 0.5 0.8% 18. Canada ................ 0.3 0,5% 19. Spain ................. 0.3 04% Other ................. 4.1 Percentage of Outgoing Traffic 1993 lo.o% 14.9% 13.8% 12.1% 20.1% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. 55.8 Outgoing 34.6 46.2 62.8 Surplus (Deficit) N.A. N.A. (7.0) Total Volume N.A. N.A. 118.7 Note: Surplus/deficit and total volume data may appear inconsistent by _+0.1 due to rounding. International Telephone Traffic--Europe Spain Destination Largest Telecommunications Routes, MiT[ Percentage of Outgoing Traffic 1. France .............. 143.0 2. United Kingdom ...... 123.6 3. Germany ............ 121.2 4. United States ......... 70.4 5. Italy ................. 58.9 8.3% .... 70% 1993 14.6% 143% 169% 6. Portugal .............. 34.6 4 1% 7. Netherlands ........... 32.7 3.9% 8. Switzerland ........... 30.6 3.6%0 9. Belgium .............. 28.7 3.4% 10. Morocco ............. 19,3 2.3% 11. Argentina ............. 15.2 1.8% 12. Sweden .............. 12.0 t.4% 13. Denmark .............. 8.1 1,0% 14. Ireland ................ 7.9 0.9% 15. Mexico ................ 7.8 09% 16. Venezuela ............. 7.8 0.9% 17. Colombia .............. 7.5 o,9% 18. Hong Kong ............. 6.9 0.8% 19. Austria ................ 6.4 o.8% 20. Brazil ................. 6.! 0.7% Other ................ 98.3 11.6% MiTT is Minutes of Telecommunications Traffic, Oata are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 736.9 847.2 908.4 Outgoing 718.7 804.5 846.9 Surplus (Deficit) 18.2 42.7 61.5 Total Volume 1455.6 1651.7 1755.3 119 TeleGeography 1994 Sweden Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. Finland .............. 105 2. Norway ............. 102 3. Denmark ............. 89 4. Germany ............. 77 5. United Kingdom ....... 66 6. United States ......... 55 7. France ............... 28 8. Spain ................ 18 24% 9. Italy ................. 17 23% 10, Netherlands ........... 16 22°/o 11. Switzerland ........... 15 20% 12. Belgium .............. 12 16% 13. Yugoslavia ............ 10 14O/o 14. Austria ................ 9 12% 15. Turkey ................ 7 o9% 16. Canada ................ 7 oo°/o 17, Other ............... 107 38% 74% 89% 104% 1993 120% 142% 138% 145% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. @ TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 571 595 630 Outgoing 672 691 740 Surplus (Deficit) (101) (96) (110) Total Volume 1243 1286 1370 Note: Data are rounded to the nearest million minutes for Teha and Tele-2 only. 120 International Telephone Traffic--Europe Switzerland Destination Largest Telecommunications Routes, Mrl-r Percentage of Outgoing Traffic 1993 1. Germany ............ 363 2. France .............. 271 17.2% 3. Italy ................ 232 14.7% 4. United Kingdom ....... 84 5.3% 5. Austria ............... 70 4.4% 6. United States .......... 67 4.2% 7. Portugal .............. 60 3.8% 8. Spain ................ 55 3.5% 9. Netherlands ........... 38 2.4% 10. Turkey ............... 31 2.0% 11. Belgium .............. 28 1 8% 12. Yugoslavia ............ 24 1.5% 13. Sweden .............. 16 1.o% 14. Canada ............... 12 o.8% Other ............... 221 13 9% 23 1% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 1124.0 1191.5 1258.7 Outgoing 1429.4 1551.0 1572.0 Surplus (Deficit) (305.4) (359.5) (313.3) Total Volume 2553.4 2742.5 2830.7 Note: Route data are rounded to the nearest million minutes. TeleGeography 1994 United Kingdom Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. United States ......... 530 2. France .............. 287 9.2% 3. Germany ............ 281 9.o% 4. Ireland .............. 266 8.5% 5 Italy ............... 139 4.4% FY 1993/94 16.9% 6. Netherlands .......... 131 4.2% 7. Spain ............... 115 3.7% 8. Australia ............ 101 3.2% 9. Canada ............... 96 3.1% 10. Belgium .............. 84 2.7% 11. Switzerland ........... 82 2.6% 12. Sweden .............. 52 1.7% 13. Denmark ............. 46 15% 14. Greece ............... 45 1.4% 15. India ................. 42 1.3% 16. South Africa .......... 41 1.3% 17. Hong Kong ............ 41 13% 18. Japan ................ 39 19. Pakistan .............. 38 1.2% 20. Norway .............. 37 1 2% Other ............... 637 20.4% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT FY 1991/92 FY 1992/93 FY 1993/94 Incoming 2692 2789 3086 0 utgoing 2598 2849 3130 Surplus (Deficit) 94 (60) (44) Total Volume 5290 5638 6216 Notes: Data are rounded to the nearest million minutes. Data are for BT and Mercury only; traffic of IPL resellers is excluded. IPL resellers had approximately 60 million outbound MiTT in FY 1993. See Methodology. 122 International Telephone Traffic--Europe Yugoslavia Destination Largest Telecommunications Routes, MiT[ Percentage of Outgoing Traffic 1. Germany ............. 51.7 2. Austria ............... 21.9 12.1% 3. Switzerland ........... 17.0 9,4% 4. Greece ................ 9.6 5,3% 5. Italy .................. 9.3 6.1% 6. France ................ 8.2 4.5% 7. Slovenia ............... 7.2 4.0% 8. United States .......... 7.1 3.9% 9. United Kingdom ........ 5.4 3.0% 10. Sweden ............... 5.2 2.9% 11. Hungary ............... 4.7 2.6% 12. Russia ................ 4.1 2.3% 13. Turkey ................ 4.1 2.3% 14. Bulgaria ............... 2.7 1.5% 15. Macedonia ............ 2.6 1.4% 16. Canada ................ 2.5 1.4% 17. Netherlands ........... 2.1 1.2% 18. Australia .............. 2.0 1.1% 19. Czech Republic ......... 1.8 1.o% 20. Cyprus ................ 1.3 o.7% Other ................ 11.0 6.1% 1993 28.5% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. 223.5 Outgoing 106.6 142.8 181.5 Surplus (Deficit) N.A. N.A. 42.0 Total Volume N.A. N.A. 405.0 Note: Includes only traffic from present-day Yugoslavia (Serbia and Montenegro). For traffic from Croatia and Slovenia, see pages 101 and 118. TeleGeography 1994 Traffic within the Commonwealth of Independent States September-December 1993(thousands of minutes) From: Armenia To: Azerbaijan 1.0 63.8 72.7 8,217.0 203.3 Armenia -- 22.5 164.6 11,947.5 242.3 Bela rus 438.9 111.4 965.3 22,603.5 385.6 Georgia 572.0 5.4 0.2 8,192.7 113.5 Kazakh_s_t~E ......... _138.6 2,977.6 200.5 .... 21,79_7__.9 2,493.8 Kyrgyzsta n 14.6 -- 25.8 4,077.8 1,031.6 Moldova 98.4 18.6 -- 6,514.1 78.5 Russia 11,797.1 6,296.7 8,964.8 -- 13,409.9 Tajikistan 9.8 184.3 10.4 2,624.9 1,114.2 Turkmenistan 257.3 Ukraine 1,825.5 Uzbekistan 237.6 Kyrgyzstan Moldova Russia Uzbekistan .......... 10__ 5.6_ .3.8 _ 1,#.7_0~.~~ .................... 515.3 333.9 6,781.4 70,635.5 1,445.3 1,031.9 85.9 11,615.9 -- Source: Regional Commonwealth in the Field of Communications (RCC) © TeleGeography, Inc. 1994 Asia, Middle East, & Africa International Traffic TeleGeography 1994 Australia F Largest Telecommunications Routes, Percentage of Outgoing Traffic 1993 MiTT New Zealand ......... 124 2. United Kingdom ....... 120 3, United States ......... 119 4. Hong Kong ............ 32 43% 5. Japan ................ 28 3,8% 6. Singapore ............ 22 3.0% 7. Germany ............. 19 2,6% 8. Canada ............... 19 2.6% 9. Italy ................. 18 2.4% 10. Malaysia ............. 16 21% 11. Philippines ............ 15 2.0% 12. Indonesia ............. 13 17%0 13. Greece ............... 11 1.5% 14. Papua New Guinea .... 10 13% 15. China ................ 10 13% 16. Taiwan ................ 9 1 2% 17. France ................ 9 12% 18. Thailand ............... 9 12% 19. Fiji .................... 8 11% 20. South Africa ........... 7 o.9% Other ............... 127 MiTT is 16.6% 16.1% 160% 17.0% Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. National Traffic Balance © TeleGeography, Inc. 1994 MiTT 1991 1992 1993 incoming N.A.N.A.- N.A. Outgoing ~_ 610 670 745 SU[P!us(Beficit) N.AI .... .~ .B.A. _ B.A.s- ’.~ Total Volume N.A. N.A. N.A. Notes: Data rounded to the nearest million minutes, for Telstra and Optus only. Data include Country Direct traffic, but exclude traffic of IPL resellers. See Methodology. 126 Bestination 1. India .................. 1.9 2. United Kingdom ........ 1.9 3. United States .......... 1.7 4. Singapore ............. 1.2 5. Hong Kong ............. 1.1 6. SaudiArabia ........... 1,0 7, Japan ................. 0.9 8. Korea, Rep. of .......... 0.8 9. Malaysia .............. 0.7 10. Pakistan ............... 0.7 11. United Arab Emirates .... 0.4 12. Germany .............. 0.4 13. Italy .................. 0.4 14. Thailand ............... 0.3 15. China ................. 0.3 16. Kuwait ................ 0.3 17. Canada ................ 0.2 18. France ................ 0.2 19. Netherlands ........... 0.2 20. Australia .............. 0.2 Other ................. 2.4 International Telephone Traffic--Asia, Middle East, & Africa Bangladesh Largest Telecommunications Routes, MiTI" Percentage of Outgoing Traffic 2.4% 2.2% 2.0% 2.0% 1.7% 15% !.4% 11% 0.9% 7.2% 63% 5.5% 5.4% 4.4% 4.2% 4.0% 13.7% 1993 11.2% 11.2% 9.7% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 56.1. 66.8 83.9 Outgoing 10.3 14.3 17.2 Surplus (Deficit) 45.8 52.5 66.7 Total Volume 66.4 81.1 101.1 TeleGeography 1994 Largest Telecommunications Routes, 1993 Destination MiT[ 1. Hong Kong ........... 535 2. Japan ................ 85 9.5% 3. Taiwan ............... 65 7.3% 4. United States .......... 45 5.o% 5. Macau ............... 30 34% 6. Korea, Rep. of ......... 15 1.7% 7. Singapore ............ 10 1.1% 8. Germany ............. 10 1.1% 9. Australia ............. 10 1.1% 10. Canada ............... 10 t.1% 11. Russia ............... 10 1.1% 12. Thailand ............... 5 0.6% 13. France ................ 5 o.6% 14. United Kingdom ........ 4 o.4% 15. Italy .................. 3 0.3% Other ................ 53 5.9% Percentage of Outgoing Traffic 59.8% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. N.A. 0utgoing 440.0 635.1 895 Surplus (Deficit) N.A. N.A. N.A. Total Volume N.A. N.A. N.A. Note: Data are rounded to the nearest 5 million minutes. Routes below 5 million minutes are rounded to the near- est million. 128 International Telephone Traffic--Asia, Middle East, & Africa Cyprus Destination Largest Telecommunications Routes, Mi’l-I" Percentage of Outgoing Traffic 1. United Kingdom ....... 27.6 2. Greece ............... 19.3 3. United States .......... 4.2 4. Germany .............. 4.0 5. USSR (former) .......... 3.2 4.5% 4.3% 3.4% 1993 20.6% 29.5% 6. Lebanon ............... 2.5 2.7% 7. France ................ 2.3 2.5% 8. Yugoslavia ............. 2.0 2.1% 9. Italy .................. 2.0 2.1% 10. Israel ................. 1.8 2.o% 11. Romania ............... 1.7 1.9% 12. Bulgaria ............... 1.6 1.7% 13. Egypt ................. 1.4 1.4o/0 14. Switzerland ............ 1.3 1.4% 15. Syria .................. 1.2 13% 16. Sweden ............... 1.2 1,3% 17. Canada ................ 1.1 1.2% 18. Australia .............. 1.1 1.1% 19. Austria ................ 1.0 1.1% 20. Jordan ................ 1.0 1.1% Other ................ 12.1 12.9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 71.7 73,7 72.2 Outgoing 70.1 85.3 93.8 Surplus (Deficit) 1.6 (11.6) (21.6) Total Volume 141.8 159.0 166.0 129 TeleGeography 1994 ~Largest Telecommunications Routes, FY 1993/94 Destination Percentage of Outgoing Traffic MiTt China ............... 688 2. United States .......... 96 7% 3. Taiwan ............... 83 6%° 4. Canada ............... 55 5. Japan ................ 55 6. United Kingdom ....... 55 4% 7. Macau ............... 41 8. Singapore ............ 41 3% 9. Australia ............. 41 3% 10. Philippines ............ 41 3% Other ............... 179 13% 5O% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT FY 1991/92 FY 1 992/93 FY 1993/94 Incoming 783.0 1009.4 1260.3 0 utgoing 913.2 1136.6 1376.9 Surplus (Deficit) (130.2) (127.2) (116.5) Total Vol u me 1696.2 2146.0 2637.2 Note: Fiscal year ends on 31 March. Route data are rounded to the nearest million minutes. Surplus/deficit and total volume data may appear inconsistent by _+0.1 due to rounding. ......... i%" International Telephone Traffic--Asia, Middle East, & Africa lndia Destination Largest Telecommunications Routes, FY 1993/94 aiT1" Percentage of Outgoing Traffic 1. SaudiArabia .......... 51.9 2. United States .......... 51.6 3. United Arab Emirates ...31.9 4. United Kingdom ....... 26.5 5. Germany .............. 9.6 3.4% 9.3% 11.2% 18.3% 18.2% 6. Singapore ............. 9.6 3.4% 7. Hong Kong ............. 8.8 3.1% 8. Canada ................ 6.6 2.3% 9. 0man ................. 5.7 2.0% 10. Australia .............. 5.0 1.8% 11. Japan ................. 5.0 1.7% 12. France ................ 4.9 ¯ 1.7% 13. Qatar ................. 4.0 1 4% 14. Italy .................. 3.4 1.2% 15. Malaysia .............. 2.8 1 0% 16. Bahrain ............... 2.5 0.9% 17. Thailand ............... 2.3 o 8% 18. Switzerland ............ 2.3 0.8% 19. Netherlands ........... 2.3 08% 20. Russia ................ 2.1 0.8% Other ................ 45.3 16.o% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © Tele6eography, Inc. 1994 National Traffic Balance MiTT FY 1991/92 FY 1 992/93 FY 1993/94 Incoming N,A. 354.6 441.0 Outgoing 185,8 259.6 283.9 Surplus (Deficit) N,A. 95.0 157.1 Total Volume N,A. 614.2 724.8 Note: Outgoing totals and route data do not include India-Pakistan traffic. Fiscal year ends on 31 March. Surplusldeficit and total volume data may appear inconsistent by +_0.1 due to rounding. 131 TeleGeography 1994 Destination Largest Telecommunications Routes, Mi’l-r Percentage of Outgoing Traffic 1. Singapore ............ 35.0 2. United States ......... 18.1 3. Japan ................ 14.0 4. Australia ............. 11.4 5. Hong Kong ............ 10.3 12.6% 9.7% 7.9% 7.2% 1993 24 4% 6. Taiwan ................ 7.1 4.9% 7. Malaysia .............. 6.7 4.7% 8. Korea, Rep. of .......... 5.9 4.1% 9. United Kingdom ........ 4.9 3.4% 10. Germany .............. 4.4 3.1% 11. Netherlands ........... 4.0 2.8% 12. France ................ 2.9 2.o% 13. Philippines ............. 2.4 1.7% 14. China ................. 2,1 1.5% 15. Thailand ............... 1.8 ].3% 16. Canada ................ 1.8 13% 17. Italy .................. 1.7 1.2% 18. Saudi Arabia ........... 1.5 I.O% 19. India .................. 1.4 ~.0% 20. Switzerland ............ 0.9 0 6% Other ................. 5.5 3.8% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 135.7 165.5 201.8 Outgoing 100.0 118.1 143.8 Surplus (Deficit) 35.7 47.4 58.0 Total Volume 235.7 283.6 345.6 Note: Data based on billing point of traffic. Data are for Indosat only. International Telephone Traffic--Asia, Middle East, & Africa Israel Largest Telecommunications Routes, 1992 Destination MiT[ Percentage of Outgoing Traffic 1. United States .......... 48.3 2. United Kingdom ....... 15.1 9.9% 3. France ............... 11.8 7.7% 4. Germany ............. 10.8 7.1% 5. USSR (former) .......... 6.2 4.0% 6. Italy .................. 5.6 3.7% 7. Canada ................ 4.6 3.0% 8. Switzerland ............ 4.5 2.9% 9. Netherlands ........... 4.0 2.6% 10. Belgium ............... 3.8 2.5% 11. South Africa ........... 2.6 1.7% 12. Australia .............. 2.4 1 6% 13. Romania ............... 2.1 1.4% 14. Turkey ................ 1.8 12% 15. Argentina .............. 1.7 1.1% Other ................ 27.8 18.2% 31.5% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 247.0 266.0 N.A. Outgoing 139.4 153.1 175.5 Surplus (Deficit) 107.6 112.9 N.A. Total Volume 386.4 419.1 N.A. Note: Data are for Bezeq only. 133 TeleGeography 1994 Japan Largest Telecommunications Routes, FY 199S/94 Destination MiTT Percentage of Outgoing Traffic 1. United States ........ 273.1 2. Korea ............... 139.8 3. China ............... 138.3 4, Philippines ........... 103.6 7.8% 5. Taiwan ............... 79.2 6.0% 6. Thailand .............. 59.8 4.5% 7. Brazil ................ 51.0 3.9% 8. Hong Kong ............ 49.2 3.7% 9. United Kingdom ....... 43.0 3.3% 10. Iran .................. 34.2 2.6% 11. Malaysia ............. 33.5 2.5% 12. Singapore ............ 33.0 2.5% 13. Australia ............. 28.4 2.1% 14. Germany ............. 26.2 2.0% 15, Peru ................. 23.2 1.8% 16. Indonesia ............ 19.2 1.5% 17. France ............... 18.8 1.4% 18. Canada .............. 18.2 1.4% 19. Pakistan .............. 17.5 1.3% 20. Italy .................. 9.4 O.7% Other ............... 121.5 10.6% 10.5% 9.2% 20.7% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT FY 1991/92 FY 1992/93 FY 1993/94 in c omin g 836.8 891.5. 981.2 Outgoing .... 1!60.5 .1283.5,, 141.1.2 ...... Su[p!~s~ (Deficit) (323.7) _ (392.0) (~29.8) Total Volume 1997.3 2174.8 2392.4 Note: Route-by-route data include only IDD calls, while total data include operator assisted calls as welt. Fiscal year ends on 31 March. Data are for KDD, IT J, and IDC combined. International Telephone Traffic--Asia, Middle East, & Africa Republic of Korea Largest Destination MiTT 1. United States ......... 100.2 2. Japan ................ 89,4 3. China ................ 23.5 4. Hong Kong ............ 16.9 4.8°1o 5. Philippines .......... 11.0 31°/o 6. Germany .............. 9.3 26% 7. Canada ................ 8.0 23% 8. United Kingdom ........ 7.4 2.1°/° 9. Taiwan ................ 7.3 21% 10. Indonesia .............. 7.0 2.0% 11. Australia .............. 7.0 20% 12. Singapore ............. 6.5 1.8% 13. France ................ 5.2 1 5% 14. Thailand ............... 4.2 1 2% Other ................ 52.3 Telecommunications Routes, Percentage of Outgoing Traffic 66% 14.7% 1993 25.2% 28 2% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © Tete6eography, Inc, 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 425.1 453.9 510.5 Outgoing 229.2 305.9 355.4 Surplus (Deficit) 195.9 148.0 155.1 Total Volume 654.3 759.8 865.9 Note: 1991 data are for Korea Telecom only. 1992 and 1993 data are for Korea Telecom and Dacom combined. TeleGeography 1994 lVlacau Largest Telecommunications Routes, 199,3 Destination MiTT 1. Hong Kong ............ 41.8 2. China ................ 38.2 3, Portugal ............... 2.3 2.6% 4. Taiwan ................ 1.3 ~.5% 5. Thailand ............... 1.1 1.2°/o 6. United States .......... 1.0 1.1°/o 7. Canada ................ 0.7 0.8% 8. Australia .............. 0.5 0.5% 9. Philippines ............. 0.4 0.5% 10. Japan ................. 0.4 0.4°/o 11. Singapore ............. 0.3 0-3% 12. United Kingdom ........ 0.3 o.3°/o 13. Malaysia .............. 0.2 0.3% 14. France ................ 0.2 0,2% 15. Korea, Rep. of .......... 0.2 °.2°/° 16. Indonesia .............. 0.1 o~% 17. Germany .............. 0,1 o.1% 18. Belgium ............... 0.1 19. Italy .................. 0.0 00% Other ................. 0.7 0.8% Percentage of Outgoing Traffic 42.4% 46.5% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice c~rcuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 57.1 68.6 71.3 Outgoing 63.2 76.9 89.9 Surplus (Deficit) (6.1) (8,3) (18,6) Total Volume 120.3 145.5 161.2 136 International Telephone Traffic--Asia, I~hddle East, & Africa Malaysia Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1, Singapore ........... 116.0 2. Japan ................ 18.4 7.1% 3. Australia ............. 14.0 54% 4. United States ......... 13.9 54% 5. United Kingdom ....... 13.4 5.2% 1993 44.9% 6. Hong Kong ............ 11.6 4.5% 7. Taiwan ............... 10.1 3.9% 8. Thailand ............... 8.6 3.3% 9. Indonesia .............. 8.1 3.2% 10. India .................. 6.6 2.5% 11. Philippines ............. 4.1 1.6% 12. China ................. 3.5 1.4% 13. Germany .............. 3.1 1.2%, 14. Korea, Rep. of .......... 3.1 1.2% 15. Canada ................ 2.4 o.9% 16. Bangladesh ............ 2.1 o.8% 17. Brunei ................ 2.1 0.8% 18. New Zealand ........... 1.8 o.7% 19. Pakistan ............... 1.7 o.7% 20. France ................ 1.7 07% Other ................ 11.8 4.6% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © Tele6eography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 249.0 304.2 Outgoing 176.2 216.5 258.1 Surplus (Deficit) N.A. 32.5 46.1 Total Volume N.A. 465.5 562.3 Note: Above totals do not include local Malaysia-Singapore border traffic. Data are for Telekom Malaysia only. 137 TeleGeography 1994 New Zealand Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic I. Australia ............ 117 2. United Kingdom ....... 24 11.o% 3. United States .......... 19 8.7% 4. Hong Kong ............. 7 3.2% 5. Japan ................. 5 23% 6. Fiji .................... 4 1.8% 7. Canada ................ 4 1.8% 8. Western Samoa ........ 4 1.8% 9. Singapore ............. 3 1.4% 10. Taiwan ................ 2 o.9% 1993 53 4% 11. Germany .............. 2 0.9% 12. Malaysia .............. 2 0.9% 13. Netherlands ........... 1 o.5% 14. China ................. 1 o5% 15. Indonesia .............. 1 o5% 16. South Africa ........... 1 o5% 17. India .................. 1 05% 18. France ................ 1 o5% 19. Papua New Guinea ..... 1 o5% 20. Cook Islands ........... I 05% Other ................ 18 82% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 0~gging ......... 165 ......... !89 .... 2!9 .... Su!pl#§ {D#ficit) ............ N:A. _B.A. N.A. Total Volume N.A. N.A. N.A. Note: Data rounded to the nearest million minutes for Telecom New Zealand and Clear Communications Ltd. Data based on billing point of traffic. International Telephone Traffic--Asla, I~hddle East, & Africa Pakistan Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. United Kingdom ........ 8.4 2. United States .......... 7.7 3. United Arab Emirates .... 6.1 4. Saudi Arabia ........... 6.0 5. Germany .............. 2.5 4.4% 1993 14.8% 13.5% 10.7% t0.6% 6. India .................. 2.0 3.5% 7. Japan ................. 1.5 27% 8. Iran ................... 1.4 2.4% 9. France ................ 1.2 2.2% 10. Singapore ............. 1.2 2.2% 11. Hong Kong ............. 1.1 2.0% 12. Kuwait ................ 1.1 2.o% 13. Canada ................ 0.9 1.7% 14. China ................. 0.9 1.6% 15. Netherlands ........... 0.8 1.3% 16, Italy .................. 0.8 1.3% 17. Switzerland ............ 0.7 1.2% 18. Korea, Rep. of .......... 0.6 1,o% 19. Qatar ................. 0.5 1.o% 20. Turkey ................ 0.5 o.9% Other ................ 10.7 18.9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeteGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. 305.7 Outgoing 34.3 44.2 56.5 Surplus (Deficit) N.A. N.A. 249.2 Total Volume N.A. N.A. 362.2 139 TeleGeography 1994 Philippines Largest Telecommunications Routes, Destination MiTT Percentage of Outgoing Traffic 1. United States .......... 67 2. Japan ................ 23 14.0% 3. Hong Kong ............ 13 7.9% 4. Canada ................ 9 5.5% 5. Australia ............ 7 4.3% 6. Singapore ............. 5 3.0% 7. Taiwan ................ 5 3.o% 8. Korea, Rep. of .......... 4 2.4% 9. Saudi Arabia ........... 4 2 4% 10. United Kingdom ........ 3 1.8% 11. Italy .................. 2 1.2% 12. Malaysia .............. 2 1.2% Other ................ 20 11.o% 1993 40.9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 385.4 462.1 N.A. Outgoing 128.5 135.8 164 Surplus (Deficit) 256.9 326.3 N.A. Total Volume 513.9 597.9 N.A. Note: Traffic to the United States ~ncludes traffic to Guam. Data rounded to the nearest million minutes and include PLDT and Philippine Global Com. only. 140 International Telephone Traffic--Asia, Middle East, & Africa Saudi Arabia Destination Largest Mi’n" Telecommunications Routes, Percentage of Outgoing Traffic 1. Egypt ............... 116.9 2. Pakistan .............. 44.4 9.9% 3. United States ......... 29.7 6.6% 4. United Kingdom ....... 22.2 5.0% 5. India ................. 22.0 4 9% 1993 26.1 6. United Arab Emirates .21.7 4.8% 7. Kuwait ............... 17.9 4.0% 8. Yemen ............... 15.7 3.5% 9. Jordan ............... 15.6 3.5% 10. Syria ................. 14.7 3.3% 11. Bahrain .............. 13,0 2.9% 12. Philippines ............ 11.6 2.6% 13. Turkey ................ 8.9 2.0% 14. Sudan ................. 8.7 1 9% 15. Fra,qce ................ 8.6 1.9% 16. Bangladesh ............ 8.5 1 9% 17. Morocco .............. 7.4 1.6% 18. Lebanon ............... 6.8 15% 19. Qatar ................. 6,1 !4% 20. Germany .............. 5.9 1 3% Other ................ 41.2 9 2% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 266.7 292.1 N.A. Outgoing 410.6 464.6 447.5 Surplus (Deficit) (143.9) (172.5) N.A. Total Volume 677.3 756.7 N.A. 141 TeleGeography 1994 Singapore Largest Telecommunications Routes, 1993 Destination MiTT Percentage of Outgoing Traffic 1. Malaysia ............ 121 2. Indonesia ............. 45 9 5% 3. Hong Kong ............ 42 8.9% 4. United States .......... 38 8.1% 5. Japan ................ 36 7.6% 6. Thailand .............. 26 55% 7. Australia ............. 22 4.7% 8. Taiwan ............... 22 47% 9. United Kingdom ....... 21 44% 10. India ................. 21 4.4% 11. China ................ 18 38% 12. Philippines ............ 17 36% 13. Germany ............. 10 2.1% 14. Korea, Rep. of .......... 9 1 9% 15. France ................ 6 13% 16. Brunei ................ 6 13% 17. Other ................ 13 28% 25.6% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT FY 1991/92 FY 1992/93 FY 1993/94 Incoming N.A. N.A. N.A. Outgoing 344 412 480 Surplus (Deficit) N.A. N.A. N.A. Total Volume N.A. N.A. N.A. Note: Data rounded to the nearest milhon minutes. Fiscal year ends on 31 March. Totals exclude local Malaysia- Singapore border traffic. Route-by-route data are for 1993 calendar year. 142 Destination 1. United Kingdom ....... 37.7 2. Namibia ............. 27.8 3. Zimbabwe ............ 19.5 4. United States ......... 18.3 5. Botswana ........... 14.0 6. Germany ............. 10.8 7. Swaziland ............ 10.0 8. Mozambique .......... 9.5 9. Lesotho ............... 7.3 10. Australia .............. 5.5 11. Portugal .............. 4.4 12. France ................ 4.0 13. Israel ................. 3.8 14. Italy .................. 3.7 15. Canada ............... 3.6 16. Switzerland ........... 3.4 17. Netherlands ........... 3.3 18. Zambia ............... 2.9 19. Malawi ............... 2.2 20. Belgium ............... 2.1 Other ................ 27,9 International Telephone Traffic--Asia, Middle East, & Africa South Africa Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 8.8% 8.3% 6.3% 4.9% 4.5% 4.3% 3,3% 2.5% 2.0% 1.8% 1.7% 17% 1.6% 1.5% 1.5% 1 3% 1.0% 0.9% 1992 17.0% 12.5% 12.6% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. @ TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. 250.7 N.A. 0 utg o in g 202.6 221.7 255 Surplus (Deficit) N.A. 29.0 N.A. Total Volume N.A. 472.4 N.A. 143 TeleGeography 1994 Destination Largest Telecommunications Routes, MiTT Percentage of Outgoing Traffic 1. India .................. 2.2 2. United Kingdom ........ 2.2 3. Singapore ............. 1.6 4. United States .......... 1.5 5. Japan ................. 1.2 6. Hong Kong ............. 1.1 7. Austraha .............. 0.9 8. Germany .............. 0.8 9. Korea, Rep. of .......... 10. United ArabEmirates ...0.6 6.2% 5.6% 4.6% 4.1% 3.1% 3.1% 8.2% 7.7% 1993 11.3% 11.3% 11. France ................ 0.5 2.6% 12. Canada ................ 0.5 26% 13. Italy .................. 0.5 2.6% 14. Saudi Arabia ........... 0.4 2.1% 15. Pakistan ............... 0.3 1 5% 16. Maldives .............. 0.3 1.5% 17. Switzerland ............ 0.3 1 5% 18. Thailand ............... 0.3 15% 19. Kuwait ................ 0.3 ! 5% 20. Malaysia .............. 0.3 1 5% Other ................. 3.1 15,9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. 65.0 Outgoing 15.6 17.8. 19.5 Surplus (Deficit) N.A. N.A. 45.5 Total Volume N.A. N.A. 84.5 International Telephone Traffic--Asia, Middle East, & Africa Taiwan Destination Largest Telecommunications Routes, FY 1993/94 MiTT Percentage of Outgoing Traffic 1. United States .......... 89.0 2. Hong Kong ............ 87.3 3. China ................ 71.6 4. Japan ................ 57.1 5. Singapore ............ 14.7 3.3% 13.0% 16.2% 20.2% 19,8% 6. Thailand .............. 11.4 26% 7. Canada ............... 10.9 2.5% 8. Philippines ............ 10.4 2.4% 9. Malaysia .............. 9.4 2.t% 10. Australia .............. 8.2 1.9% 11. Germany .............. 7.7 1.7% 12. Indonesia .............. 7.2 1.6% 13. Korea, Rep. of .......... 6.3 1.4% 14. United Kingdom ........ 6.3 1.4% 15. France ................ 4.6 1.0% 16. Vietnam ............... 3.7 o.8% 17. Italy .................. 2.5 o.6% 18. South Africa ........... 2.2 0.5% 19. Netherlands ........... 2.2 o5% 20. New Zealand ........... 2.1 0.5% Other ................ 25.9 5.9% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT FY 1991/92 FY 1992/93 FY 1993/94 Incoming 415.0 484.4 490.8 0 utg oi n g 254.7 368.7 440.7 Surplus (Deficit) 160.3 115.7 50.1 Total Volume 669.7 853.1 931.5 Note: Fiscal year ends 31 March. Data based on billing point of traffic. 145 TeleGeography 1994 Largest Telecommunications Routes, Destination MiTT Percentage of Outgoing Traffic 1. United States .......... 26.9 2. Japan ................ 24.6 3. Singapore ............ 16.5 4. Hong Kong ............ 12.8 5. Taiwan ............... 10.8 6.7% 6. United Kingdom ........ 8.2 5.1% 7. Australia .............. 6.3 3.9% 8. Germany .............. 5.8 3.6% 9. China ................. 5.4 3.3% 10, Korea, Rep. of .......... 4.1 2.5% 11. France ................ 4.1 2.5% 12. India .................. 3.0 1.9% 13. Italy .................. 2.8 1 7% 14. Canada ................ 2.2 1 4°/0 15. Switzerland ............ 2.0 1.2% 16. Netherlands ........... 2.0 1 2% 17. Indonesia .............. 1.9 1.2% 18. Myanmar .............. 1.8 11% 19. Sweden ............... 1.4 0.9% 20. Vietnam ............... 1.4 o.9% Other ................ 17.8 MiTT is Minutes of Telecommunications Traffic. 7.9% 1993 10.2% 16.6% 15.2% 11.o% Data are in millions of minutes for public voice circuits, © TeteGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 175.2 212.7 218.7 Outgoing 115.3 132.4 161.8 Surplus (Deficit) 59.9 80.3 56.9 Total Volume 290.5 345.1 380.5 Note: Data based on billing point of traffic. 146 International Telephone Traffic--Asia, Middle East, & Africa Turkey Destination Largest Telecommunications Routes, 1993 iiTr Percentage of Outgoing Traffic 1. Germany ............. 97.6 2. United Kingdom ....... 23.3 8.8% 3. United States .......... 16.6 6.3% 4. France ............... 16.3 6.2% 5. Netherlands ........... 11.8 4.5% 36.9% 6. Switzerland ............ 9.2 3.5% 7. Italy .................. 9.0 3.4% 8. Austria ................ 7.4 2.8% 9. Russia ................ 6.8 2.6% 10. Bulgaria ............... 5.7 2.2% 11. Romania ............... 5.5 2.1% 12. Belgium ............... 5.3 2.0% 13. Saudi Arabia ........... 4.1 1.5% 14. Israel ................. 4.1 1.5% 15. Greece ................ 4.0 1.5% 16. Iran ................... 3.8 1 4% 17. Sweden ............... 2.9 1.1% 18. Denmark .............. 2.1 0.8% 19. Spain ................. 2.0 0.8% 20. Canada ................ 1.8 0.7% Other ................ 25.3 9.6% MiTT is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeleGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming 470.0 560.0 605.0 Outgoing 198.1 226.8 264.6 Surplus (Deficit) 271.9 333.2 340.4 Total Volume 668.1 786.8 869.6 147 TeleGeography 1994 United Arab Em rates Destination Largest Telecommunications Routes, 1993 MiTT Percentage of Out~loing Traffic 1. India ................. 61.4 2. Pakistan .............. 32.5 3. Saudi Arabia .......... 27.5 4. Egypt ................ 26.1 5. United Kingdom ....... 24.0 9.5% 8.0% 7.7% 7.0% 18.0% 6. Oman ................ 18.8 7. Iran ................. 14.3 4.2% 8. United States ......... 12.9 3.8% 9. Syria ................. 11.7 3,4% 10. Qatar ................ 11.1 3.3% 11. Jordan ............... 10.4 3.o% 12. Bahrain .............. 10.3 3.o% 13. Kuwait ............... 10.0 2.9% 14. Philippines ............. 6.0 1.8% 15, Lebanon ............... 5.9 1.7% 16. Bangladesh ............ 5.4 1,6% 17. Sudan ................ 5.2 1 5% 18. Yemen ................ 4.6 1 3% 19. Germany .............. 4.5 1.3% 20. France ................ 4.4 1,3% Other ................ 34.7 5.5% 10,1% Mi]-I is Minutes of Telecommunications Traffic. Data are in millions of minutes for public voice circuits. © TeteGeography, Inc. 1994 National Traffic Balance MiTT 1991 1992 1993 Incoming N.A. N.A. N.A. 0utgoing 262.3 299.0 341.6 Surplus (Deficit) N.A. N.A. N.A. Total Volume N.A. N.A. N.A. International Telephone Traffic--Methodology and Sources Methodology and Sources The traffic statistics in 1994 TeleGeography were compiled primarily from an indepen- dent survey of telecommunications service providers. For some countries and carri- ers, traffic data have been estimated based upon annual reports, government publications and industry interviews. See the footnotes to each table for further information. The following publications were also consulted: Direction of Traffic: International Telephone Traffic 1983-1992 (TeleGeography, lnc./ITU, Geneva, 1994), Internationale Fernmeldestatistik (Siemens, Munich, 1994); The World’s Telephones A Statistical Compilation as of January 1991-1992, (AT&T, Indianapolis, IN, 1993}. A common accounting unit known as MiTT (Minutes of Telecommunications Traffic) is used throughout the report. MiTT generally refers to paid minutes of traffic on public switched voice circuits and thus includes voice as well as non- voice (facsimile, data) traffic. For the origins of MiTT and its various applications (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. Calling card traffic Historically, most international calls were billed at the point of origination. The number of billed minutes thus coincided with the volume of outgo- ing traffic. Billed minutes also included collect or reverse charge calls because the calls were set up by an operator in the originating country. However, the recent use of credit and debit cards has shiffed the billing point for many internation- al calls. Calls from Country A to Country B (or Country C) may now be set up and billed in Country B. Unless otherwise stated in the notes to a table, the outbound MiTT reported for countries in TeleGeography 1994 refers to outbound traffic originated in the stated country even if it is billed in another country. That is, unless stated, traffic originated in a foreign country but billed to a call- ing card or credit card in the reporting country is not counted as outbound traffic. However, some countries (including the U.S.) report international traffic data based solely on the location where the traffic is billed; consequently, "outbound" traffic data for these countries include Home Country direct traffic originating in third countries. For these and other reasons, the national statistics in TeleGeographg are not directly comparable, and incoming MiTT reported for one country may not match the outgoing MiTT on the same route by the correspondent country. Some double count- ing may also occur. For example, a Country Direct call from Malaysia to.the U.S. which is billed to a U.S. calling card is reported here as outbound U.S. MITT; the same call also is reported as out- bound MiTT by Malaysia. Third-country routint~ The growing volume of traffic routed via a third country using Home Country Beyond and "call back" services is also making national traffic sta- tistics harder to interpret. A Home Country Beyond call may originate in Country A, be billed to a calling card in Country B and terminate in Country C. Similarly, a call routed via a "call back" service may be placed by a subscriber in Country A, but originate in Country B and termi- nate in Country C. In both cases, the calls from Country A to Country C generally will not be counted in Country A’s outbound MiTT but may be reflected by an increased volume of MiTT from Country B to Country C. Thus, in countries where Home Country Beyond and call back services are widely used, a year-to- year comparison of national MiTT also requires examining the statistics of countries where the calls are being refiled or hubbed. For example, in 149 TeleGeography 1994 993 Argentina’s national carrier, Telintar, report- ed a growth in outbound MiTT of only 10%; traf- fic to the U.S. grew by 6.5% and traffic to Italy also rose by 6.5%, significantly below the histori- cal (1988-1992) growth rate on this route of 15%. However, U.S. carriers reported that U.S. billed traffic on the U.S.-Argentina route in ]993 grew by 26.9% as compared to a 4.3% growth in Argentina-billed traffic on this route. U.S.-billed traffic to Italy in 1993 grew by 10.9%, whereas Italy-billed traffic to the U.S. declined by 16.2%. These figures suggest the magnitude of calls from Argentina (and other countries) which are now routed via the U.S. to Italy To assist readers in making similar comparisons, the U.S. tables in TeleGeography 1994 have been expanded to provide t992 and 1993 route-by- route statistics for over sixty countries. For fur- ther discussion of the methodological issues raised by the growth of alternative call routing arrangements, see Direction of Traffic op. cit. Chapter 7. Resellers Traffic carried by resetlers of international switched voice services generally is included in the MiTT for the facilities-based carrier whose facili- ties are resold. However, unless otherwise stated, MiTT data exclude traffic carried by resellers of international private line (IPLs) connected to the public switched network at one or both ends. Transit traffic is also excluded. Other factors There may also be other reasons (beyond those referred to above) which cause inbound traffic data on a given route to differ from the outbound traffic data for the originating country (e.g., cal- endar 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. Some differences exist between the his- torical statistics (1992 or earlier) reported in TeleGeography 1994 and data stated in prior reports or Direction of Traffic. The variations reflect corrections and/or revised data subse- quently provided to TeleGeography. In addition, rounding may cause the figures on total national traffic and surpluses and deficits to appear incon- sistent with other national data by +0.2. Blue Pages TeleGeography 1994 152 International Dialing Codes, by Number Blue Pages 1 Canada United States |-81)9 Anguilla, Antigua & Barbuda, Bahamas, Barbados, Bermuda, British Virgin Islands, Cayman Is- lands, Dominic& Dominican Republic, Grenada, Jamaica, Montserrat, Nevis Islands, Puerto Rico, St. Kitts, St. Lucia, St. Vincent & the Grenadines, Trinidad & Tobago, Turks & Caicos, U.S. Virgin Islands :21) Egypt :21:2 Morocco :213 Algeria :216 Tunisia :218 Libya 221) Gambia 221 Senegal 22:2 Mauritania 223 Mall 224 Guinea 225 Ivory Coast 226 Burkina Faso 22? Niger 228 Togo 229 Benin 231) Mauritius 231 Liberia :232 Sierra Leone :233 Ghana 234 Nigeria 235 Chad 236 Central African Republic 23? Cameroon 238 Cape Verde Islands 239 Sac Tome and Principe 241) Equatorial Guinea 241 Gabon 24:2 Congo :243 Zaire 244 Angola 245 Guinea-Bissau 246 Diego Garcia 24? Ascension Island 248 Seychelles 249 Sudan 251) Rwanda 251 Ethiopia 252 Somalia 253 Djibouti 254 Kenya 255 Tanzania 256 Uganda 25? Burundi 258 Mozambique 259 Zanzibar 46 Sweden 261) Zambia 47 Norway 261 Madagascar 48 Poland 262 Reunion Island 49 Germany :263 Zimbabwe 581) Falkland Islands 264 Namibia 51)1 Belize 265 Malawi 51)2 Guatemala :266 Lesotho 51)3 El Salvador 26? Botswana 51)4 Honduras 268 Swaziland 51)5 Nicaragua 269 Comoros & Mayotte 506 Costa Rica 2? South Africa 507 Panama 291 Eritrea 508 St. Pierre & Miquelon :297 Aruba 509 Haiti 298 Farce Islands 51 Peru 299 Greenland 52 Mexico 30 Greece 53 Cuba 31 Netherlands 54 Argentina 32 Belgium 55 Brazil 33 France 56 Chile 33-628 Andorra 5? Colombia 33-93 Monaco 58 Venezuela 34 Spain 591) Guadeloupe 351) Gibraltar 591 Bolivia 351 Portugal; Azores 592 Guyana 35:2 Luxembourg 593 Ecuador 353 Ireland 594 French Guiana 354 Iceland 595 Paraguay 355 Albania 596 Martinique 356 Malta 597 Suriname 35? Cyprus 598 Uruguay 358 Finland 599 Netherlands Antilles 359 Bulgaria 61) Malaysia 36 Hungary 61 Australia 371) Lithuania 62 Indonesia 371 Latvia 63 Philippines 372 Estonia 64 New Zealand 3?3 Moldova 65 Singapore 378 San Marino 66 Thailand 3?9 Vatican City 671) Northern Marianas 381 Yugoslavia 671 Guam 385 Croatia 6?2 Australian Territories 31)6 Slovenia 6?3 Brunei 31)7 Bosnia-Herzegovina 6?4 Nauru 389 Macedonia (EY.R.) 6?5 Papua New Guinea 39 Italy 6?6 Tonga Islands 41) Romania 6?? Solomon Islands 41 Switzerland 6?8 Vanuatu 41-75 Liechtenstein 6?9 Fiji 4:2 Czech Republic 681) Palau 42 Slovak Republic 681 Wallis & Fortuna 43 Austria 682 Cook Islands 44 United Kingdom 683 Niue 45 Denmark 684 American Samoa 685 Western Samoa 686 Kiribati 687 New Caledonia 688 Tuvalu 689 French Polynesia 690 Tokelau 691 Micronesia 692 Marshall Islands 7 Armenia Belarus Kazakhstan Kyrgyzstan Russia Tajikistan Turkmenistan Ukraine Uzbekistan 81 Japan 82 South Korea 84 Vietnam 850 North Korea 852 Hong Kong 853 Macao 1)55 Cambodia 856 Laos 86 China (PRC) 871/872J873/874 Inmarsat 880 Bangladesh 886 Taiwan 90 Turkey 91 India 92 Pakistan 93 Afghanistan 94 SriLanka 95 Myanmar (Burma) 950 Matdives 961 Lebanon 962 Jordan 963 Syria 964 Iraq 965 Kuwait 966 Saudi Arabia 967 Yemen 968 Oman 971 United Arab Emirates 972 Israel 973 Bahrain 974 Qatar 975 Bhutan 976 Mongolia 977 Nepal 98 Iran 994 Azerbaijan 995 Georgia 153 TeleGeography 1994 International Dialing Codes, by Country Afghanistan ........... 93 Albania .............. 355 ]]rana ............ 42 Algeria .............. 213 Algiers ............ 2 American Samoa ...... 684 Andorra ........... 33-628 Angola .............. 244 Luanda ............ 2 Anguilla ........... 1-809 Antigua & Barbuda .1-809 Argentina ............. 54 Buenos Aires ........ 1 Armenia ............... 7 Yerevan ......... 8852 Aruba ............... 297 Ascension Island ..... 247 Australia ............. 61 Canberra .......... 62 Melbourne .......... 3 Sydney ............. 2 Australian Territories .672 Austria ............... 43 Vienna ............. 1 Azerbaijan ........... 994 Baku ........... 8922 Bahamas ........... 1-809 Bahrain .............. 973 Bangladesh .......... 880 Dhaka ............. 2 Barbados ........... 1-809 Belarus ................ 7 Minsk ........... 172 Belgium .............. 32 Brussels ........... 2 Belize ............... 501 Belmopan .......... 8 Benin ............... 229 Bermuda ........... 1-809 Bhutan .............. 975 Bolivia .............. 591 La Paz ............. 2 Bosnia .............. 387 Sarajevo ........... 71 Botswana ............ 267 Brazil ................ 55 Brasilia ........... 61 Rio de Janeiro ...... 21 S~o Paulo .......... 11 British Virgin Islands 1-809 Brunet ............... 673 Bandar Seri Begawan..2 Bulgaria ............. 359 Sofia .............. 2 Burkina Faso ......... 226 Burundi .............. 257 Cambodia ............ 855 Cameroon ............ 237 Canada ................ 1 Montreal ......... 514 Ottawa ........... 613 Toronto ........... 416 Cape Verde .......... 238 Cayman Islands ..... 1-809 Central African Republic236 Bangui ............ 61 Chad ................ 235 Chile ................. 56 Santiago ........... 2 China, People’s Republic of ........ 86 Belling ............. 1 Guangzhou ......... 20 Shanghai .......... 21 Colombia ............. 57 Bogota ............ 1 Cocos Islands; Norfolk & Christmas Islands .672 Comoros ............. 269 Congo ............... 242 Brazzaville .... 81/82/83 Costa Rica ........... 506 Croatia .............. 385 Zagreb ............ 41 Cuba ................. 53 Havana ............ 7 Cyprus ............... 357 Nicosia ............ 2 Czech Republic ........ 42 Prague ............. 2 Denmark .............. 45 Diego Garcia ......... 246 Djibouti .............. 253 Dominca ........... 1-809 Dominican Republic .1-809 Ecuador ............. 593 Quito .............. 2 Egypt ................. 20 Cairo .............. 2 El Salvador ........... 503 Equitorial Guinea ..... 240 Eritrea ............... 291 Estonia .............. 372 Tallinn ............. 2 I:tbiopia ............. 251 Addis Ababa ........ 1 Fiji ................. 679 Finland .............. 358 Helsinki ............ 0 France ................ 33 Paris .............. 1 French Antilles ....... 596 French Guiana ........ 594 French Polynesia ..... 689 Gabon ............... 241 Gambia .............. 220 Georgia .............. 995 Tbilisi .......... 8832 Germany .............. 49 Berlin ............. 30 Bonn ............ 228 Frankfurt ........... 69 Munich ............ 89 6hana ............... 233 Accra ............. 21 Gibraltar ............. 350 Greece ............... 30 Athens ............ 1 Greenland ........... 299 Grenada ........... 1-809 Guadeloupe .......... 590 Guam ................ 671 Guatemala ........... 502 Guatemala City ...... 2 6uinoa .............. 22a, Guinea-Bissau ........ 245 Guyana .............. 592 Georgetown ........ 2 Haiti ................ 509 Honduras ............ 504 Hong Kong ........... 852 Hungary .............. 36 Budapest ........... 1 Iceland .............. 354 Reykjavik ........... 1 India ................. 91 Bombay ........... 22 Calcutta ........... 33 New Delhi ......... 11 Indonesia ............ 62 Jakarta ........... 21 Inmarsat East Atlantic ....... 871 West Atlantic ...... 874 Pacific ........... 872 Indian ............ 873 Iran .................. 98 Tehran ............ 21 Iraq ................. 964 Baghdad ........... 1 Ireland .............. 353 Dublin ............. 1 Israel ................ 972 Jerusalem .......... 2 Tel Aviv ............ 3 Italy .................. 39 Rome .............. 6 Milan .............. 2 Ivory Coast ........... 225 Jamaica ........... 1-809 Japan ................ 81 Osaka .............. 6 Tokyo .............. 3 Jordan .............. 962 Amman ............ 6 Kazakhstan ............ 7 Alma Ata ........ 3272 Kenya ............... 254 Nairobi ............ 2 Kiribati .............. 686 Kuwait .............. 965 Kyrgyzstan ............. 7 Bishkek ......... 3312 Laos ................ 856 Latvia ............... 371 Riga ............... 2 Lebanon ............. 961 Beirut ............. 1 Lesotho .............. 266 Liberia .............. 231 Libya ................ 218 Tripoli ............ 21 Liechtenstein ....... 41-75 Lithuania ............ 370 Vilinius ............ 2 Luxembourg .......... 352 Macao .............. 853 Macedonia (F.Y.R.) ..... 389 Skopje ............ 91 Madagascar .......... 261 Antananarivo ........ 2 Malawi .............. 265 Malaysia ............. 60 Kuala Lumpur ....... 3 Maldives ............ 960 Malt ................ 223 Malta ............... 356 Marshall Islands ...... 692 Martinique ........... 596 Mauritania ........... 222 Mauritius ............ 230 Mayotte ............. 269 Mexico ............... 52 Guadalajara ........ 36 154 Blue Pages Mexico City ......... 5 Monterrey ......... 83 Micronesia ........... 691 Moldova ............. 373 0hisinau .......... 422 Monaco ............ 33-93 Mongolia ............ 976 Montserrat ......... 1-809 Morocco ............. 212 Casablanca .......... 2 Robot ............. 7 Mozambique ......... 258 Maputo ............ 1 Myanmar (Burma) ..... 95 Namibia ............. 264 Windhoek ......... 61 Nanru ............... 674 Nepal ............... 9?7 Kathmandu .......... 1 Netherlands ........... 31 Amsterdam ........ 20 Netherlands Antilles .599 Nevis Islands ....... 1-809 New Caledonia ....... 687 New Zealand .......... 64 Auckland ........... 9 Wellington ......... 4 Nicaragua ........... 505 Managua ........... 2 Niger ................ 227 Nigeria .............. 234 Lagos .............. 1 Niue ................ 583 North Korea .......... 850 Pyongyang .......... 2 Northern Marianas .... 670 Saipan ........... 322 Norway ............... 47 Oslo ............... 2 Oman ................ 968 Pakistan .............. 92 Islamabad ......... 51 Palau ................ 680 Panama ............. 507 Papua New Guinea .... 675 Paraguay ............ 595 Asuncion .......... 21 Peru ................. 51 Lima ............. 14 Philippines ............ 63 Manila ............ 2 Poland ............... 48 Warsaw .......... 22 Portugal ............. 351 Lisbon ............. 1 Puerto Rico ......... 1-809 Qatar ................ 974 Reunion Island ....... 262 Romania .............. 40 Bucharest .......... 1 Russia ................. 7 Moscow ......... 095 St. Petersburg ...... 812 Rwanda ............. 250 St. Kitts ............ 1-809 St. Lucia ........... 1-809 St. Pierre & Miquelon..508 St, Vincent & ........... the Grenadines..1-809 San Marino .......... 378 S~o Tome and Principe 239 Saudi Arabia ......... 966 Fliyadh ............. 1 Senegal ............. 221 Seychelles ........... 248 Sierra Leone ......... 232 Freetown .......... 22 Singapore ............. 65 Slovakia .............. 42 Bratislava .......... 7 Slovenia ............. 386 Ljubljana .......... 61 Solomon Islands ...... 677 Somalia ............. 252 Mogadishu ......... 1 South Africa ........... 27 Johannesburg ....... 11 Pretoria ........... 12 South Korea ........... 82 Seoul .............. 2 Spain ................ 34 Madrid ............ 1 Barcelona ........... 3 Sri Lanka ............. 94 Colombo ........... 1 Sudan ............... 249 Khartoum .......... 11 Suriname ............ 597 Swaziland ........... 268 Sweden .............. 46 Stockholm .......... 8 Switzerland ........... 41 Berne ............. 31 Zurich .............. 1 Syria ................ 963 Damascus ......... 11 Tahiti ................ 689 Taiwan .............. 886 Taipei ............. 2 fajikistan .............. 7 Bushanbe ....... 3772 Tanzania ............. 255 Dar Es Salaam ...... 51 Thailand .............. 66 Bangkok ........... 2 Togo ................ 228 Tokelau .............. 690 Tonga ............... 676 Trinidad & Tobago ...1-809 Tunisia .............. 216 Tunis .............. 1 Turkey ................ 90 Ankara ............ 4 Istanbul ............ 1 Turkmenistan ........... 7 Ashkhabad ....... 3632 Turks & Caicos ..... 1-809 Tuvalu ............... 688 Uganda .............. 256 Kampala .......... 41 Ukraine ................ Kiev .............. 44 United Arab Emirates .971 Abu Dhabi .......... 2 Dubai .............. 4 United Kingdom ........ 44 London ....... 171/181 Manchester ........ 61 United States ........... 1 Chicago ....... 312/630 Houston .......... 713 Los Angeles ....... 213 Miami ............ 305 New York ..... 212/718 Washington ....... 202 Uruguay ............. 598 Montevideo ......... 2 Uzbekistan ............. 7 Tashkent ........ 3712 Vanuatu ............. 678 Vatican City .......... Venezeela ............ 58 Caracas ............ 2 Vietnam .............. 84 Wallis & Fortuna ...... 681 Western Samoa ....... 685 Yemen ............... 967 Sanaa ............ 51 Yugoslavia ........... 381 Belgrade ........... 11 Zaira ................ 243 Kinshasa .......... 12 Zamhia .............. 260 /usaka ............. 1 Zanzibar (Tanzania) .... 259 Zimbabwe ........... 263 Harare ............. 4 155 TeleGeography 1994 North American Area Codes, by State and Province Alabama Birmingham ....... 205 Montgomery ..... 334* Alaska .............. 907 Alberta ............. 403 Arizona Phoenix .......... 602 Elsewhere ...... 520** Arkansas ............ 501 British Columbia & NW Territories ....... 604 California Anaheim ......... 714 Bakersfield ....... 805 Burbank .......... 818 Fresno ........... 209 Long Beach ....... 310 Los Angeles ....... 213 Oakland .......... 510 Riverside ......... 909 Sacramento ....... 916 San Diego ........ 619 San Francisco ..... 415 San Jose ......... 408 Santa Rosa ....... 707 Colorado Colorado Springs ...719 Denver ........... 303 Connecticut ......... 203 Delaware ........... 302 District of Columbia Washington ....... 202 Florida Jacksonville ....... 904 Miami ........ 305/954 Orlando .......... 407 Tampa ........... 813 Georgia Athens ........... 706 Atlanta .......... 404 Savannah ......... 912 Hawaii ............. 808 Idaho ............... 208 Illinois Cairo ............ 618 Chicago ...... 312/630 Oak Brook ........ 708 Peoria ........... 309 Rockford ......... 815 Springfield ........ 217 Indiana Evansville ........ 812 Gary ............. 219 Indianapolis ....... 317 Iowa Coucit Bluffs ...... 712 Des Moines ....... 515 Dubuque ......... 319 Kansas Topeka ........... 913 Wichita .......... 316 Kentucky Dade Park ........ 812 Lexington ......... 606 Louisville ......... 502 Louisiana New Orleans ...... 504 Shreveport ........ 318 Maine .............. 207 Manitoba ........... 204 Maryland Baltimore ......... 410 Rockville ......... 301 Massachusetts Boston ........... 617 Springfield ........ 413 Worcester ........ 508 Michgan Detroit ........... 313 Flint ............. 810 Grand Rapids ...... 616 Lansing .......... 517 Sault Ste. Marie ...906 Minnesota Duluth ........... 218 Minneapolis ....... 612 Rochester ........ 507 Mississippi .......... 601 Missouri Kansas City ....... 816 St. Louis ......... 314 Springfield ........ 417 Montana ............ 406 Nebraska North Platte ....... 308 Omaha ........... 402 Nevada ............. 702 New Brunswick ...... 506 New Hampshire ...... 603 New Jersey Elizabeth ......... 908 Newark .......... 201 Trenton .......... 609 New Mexico ........ 505 New York Albany ........... 518 Buffalo ........... 716 Hempstead ....... 516 Ithaca ........... 607 Syracuse ......... 315 White Plains ...... 914 New York City Manhattan ........ 212 Bronx, Queens ..... 718 auxilliary code ..... 917 Newfoundland ....... 709 North Carolina Charlotte ......... 704 Greensboro ....... 910 Raleigh .......... 919 North Dakota ........ 701 Nova Scotia & Prince Edward Island .... 902 Ohio Cincinnati ........ 513 Cleveland ......... 216 Columbus ......... 614 Toledo ........... 419 Oklahoma Oklahoma City ..... 405 Tulsa ............ 918 Ontario Ft. William ....... 807 London ........... 519 North Bay ........ 705 Ottawa .......... 613 Toronto .......... 416 Oregon .............. 503 Pennsylvania Altoona .......... 814 Harrisburg ........ 717 Philadelphia ....... 215 Pittsburgh ........ 412 Puerto Rico & Caribbean ..... 809 Quebec Montreal ......... 514 Quebec .......... 418 Sherbrooke ....... 819 Rhode Island ........ 401 Saskatchewan ....... 306 South Carolina ....... 803 South Dakota ........ 605 Tennessee Memphis ......... 901 Nashville ......... 615 Texas Amarillo .......... 806 Austin ........... 512 Dallas ........... 214 El Paso .......... 915 Fort Worth ........ 817 Galveston ........ 409 Houston .......... 713 San Antonio ....... 210 Tyler ............. 903 Utah ................ 801 Vermont ............. 802 Virginia Alexandria ........ 703 Richmond ......... 804 Washington Seattle ........... 206 Spokane ......... 509 West Virginia ........ 304 Wisconisin Madison ......... 608 Milwaukee ....... 414 Eau Claire ........ 715 Wyoming ............ 307 Effective 15 January 1995 ** Effective 19 March 1995 156 North American Area Codes, by Number :~1)1 New Jersey 40;2 Nebraska 60;2 Arizona ;20;2 District of 4.1)3 Alberta 603 New Hampshire Columbia 4,04 (Seorgia 604, British Columbia & ;21)3 Connecticut 406 Oklahoma NW Territories ;21)4, Manitoba 406 Montana 605 South Dakota ;205 Alabama 407 Florida 606 Kentucky 206 Washington 408 California 607 New York ;207 Maine 409 Texas 608 Wisconisin ;21)8 Idaho 4,11) Maryland 61)9 New Jersey ;21)9 California 4,1;2 Pennsylvania 611) Pennsylvania ;211) Texas 4,13 Massachussetts 61;2 Minnesota ;21;2 New York City 4,14, Wisconsin 613 Ontario ;213 California 416 California 614, Ohio ;214 Texas 416 Ontario 615 Tennessee ;216 Pennsylvania 4,17 Missouri 616 Michigan ;216 Ohio 418 Quebec 617 Massachusetts 217 Illinois 419 Ohio 618 Illinois ;218 Minnesota 500 Personal Communication 619 California 219 Indiana Services (PCS) 630 Illinois (1 July 1995) 301 Maryland 501 Arkansas 701 North Dakota 302 Delaware 502 Kentucky 702 Nevada 303 Colorada 503 Oregon 703 Virginia 304 West Virginia 504 Louisiana 704 North Carolina 305 Florida 505 New Mexico 705 Ontario 306 Saskatchewan 506 New Brunswick 706 Georgia 307 Wyoming 507 Minnesota 707 California 308 Nebraska 508 Massachusetts 708 Illinois 309 Illinois 509 Washington 709 Newfoundland 310 California 510 California 710 U.S. 6ovemment Emergency 312 Illinois 512 Texas Telecommunications Service 313 Michgan 513 Ohio 712 Iowa 314 Missouri 514 Quebec 713 Texas 315 New York 515 Iowa 714 California 316 Kansas 516 New York 715 Wisconsin 317 Indiana 517 Michigan 716 New York 318 Louisiana 518 New York 717 Pennsylvania 319 Iowa 519 Ontario 718 New York City 334 Alabama (15 Jan. 1995) 520 Arizona (19 March 1995) 719 Colorado 360 Washington (15Jan. 1995) 566 Public Information Services 800 Toll-free services 401 Rhode Island 601 Mississippi 801 Utah Note: Bates in parentheses indicate when the area code will become active. Blue Pages 802 Vermont 803 South Carolina 804 Virginia 806 California 806 Texas 807 Ontario 808 Hawaii 809 Puerto Rico and Caribbean 810 Michigan 812 Indiana/Kentucky 813 Florida 814 Pennsylvania 815 Illinois 816 Missouri 817 Texas 818 California 819 Quebec 900 Information Services 901 Tennessee 902 Nova Scotia and Prince Edward Island 903 Texas 904 Florida 905 Ontario 906 Michigan 907 Alaska 908 New Jersey 909 California 910 North Carolina 912 Georgia 913 Kansas 914 New York 916 Texas 916 California 917 New York City 918 Oklahoma 919 North Carolina 954 Florida (proposed) 157 TeleGeography 1994 © TeleGeography, /nc, 1994 AII Rights Reserved. Hawaii Inmarsat Mobile Services Pacific Ocean Telephone Numbering Zones of the World Guatemala 12) Managua (2) Quito (2) ,Boron 617 2~ Personal Services Nassa~ Inmarsat Mobile Services Atlantic Ocean ~71, #74 Caribbean [Except as stated] Guadeloupe ¯ ~j 596 Martinique 297 Aruba Georgetown 597 102) Montevideo (2) Rio de Janeiro 121) 158 World Telephone Codes Te/eGeography, Inc. ~9# All Rights Reserved. KEY FOR EUROPE INSET Reykjavik (1) Dublin London Gibraltar Key Telephone Ca~ii~al Ciey Number Country Code Code (~ Czech Republic 42’ Prague 2 ~) SIovak Republic 4~ Bratislava 7 (~) Moldova 373 Chisinau 2 (~) Slovenia 386 Ljubljana 61 (~) Croatia 38~ Zagreb 41 ~) Bosnia-Herzegovina 387 Sarajevo 71 (~) Yugoslavia 381 Belgrade 11 (~ Macedonia 389 Skopje (~) Malta Valletta (~ Cyprus 3,$7 Nicosia 2 Blue Pages 245 (4) Rabat F~eetown Accra (21( Brazzaviile 181.82,83) Luand~ Windhoek (61) (31) Mogadishu (1) Dar Es Salaam (51) (22) Diego Gamia Bangkok (2) PhnomPenh Kuala Lumpur ’ (3) Jaka (21) Inmarsat Mobile Services Indian Ocean 81 Tokyo (3) Nor~heni Marianas Port MoresLh’ Note: No ciW code is required for listed cities unless stated. Wellington For additional copies of this map, contact: TeleGeography, Inc., Suite I000, 1150 Connecticut Avenue, NW, Washington DC 20036 Tel. 202-467-0017; fax 202-467-5915 159 TeleGeography 1994 International Date Line 160 Blue Pages Time Zones 161 TeleGeography 1994 PACIFIC 0 CEA N PANAMA P,ACIFIC O.CEAN ATLANTIC OCEAN FR ~FALKLANB I$ 162 Political Geography Blue Pages L~ONE LIBERIA GHANA =, ~__~_ A00 E ATI,LANTIC OCEAN PA E’tFI C SEYCHELLES COMOROS INDIAN OCEAN MhURITIUS (FRANCE) TropicafCap~icq~n :, Apt~rctic Circle EUROPE 1. NETHERLANDS 2. BELGIUM 3. LUXEMBOURG 4. CZECH REPUBLIC 5. SLOVAK REPUBLIC 6. SWITZERLAND 7. LIECHTENSTEIN 8. AUSTRIA 9. HUNGARY 10. SLOVENIA 11. CROATIA 12. BOSNIA 13. YUGOSLAVIA 14. ALBANIA 15. MACEDONIA (F.Y.R.) 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 163 TeleGeography 1994 National Telecommunications Indicators (A-J) Population Area Main Lines Main Lines Lines/100 Cellular phones Fax machines Internet Hosts 1993 (millions) (Miles2thous.) 1993 (millions) 1990 (millions) people 1993 1993 (thous.) 1993 (thous.) August 1994 Algeria 27.4 920 1.1 0.8 3.9 4.8 8 n.a. Argentina 33.4 1068 4.1 3.5 12.3 115.5 30 248 Australia 17.9 2968 8.5 8.1 47.8 760.0 600 1780 Austria 7.8 32 3.6 3.2 46.1 221.0 140 n.a. Bahrain 0.6 1 0.1 0.1 22.6 11.4 4 n.a. Bangladesh 113.1 56 0.3 0.2 0.2 0.5 2 n.a. Belgium 10.0 12 4.4 3.9 44.0 66.9 165 n.a. Brazil 158.9 3286 11.7 9.4 7.4 180.2 200 5896 Bulgaria 8.7 43 2.3 2.1 26.3 1.0 10 n.a. Canada 27.2 3852 16.3 d 15.3 59.7 1326.4 525 n.a. Chile 13.7 292 1.5 0.8 11.1 85.2 15 3703 China 1183.6 3705 17.3 6.6 1.5 638.0 125 n.a. Colombia 34.0 440 2.8 d 2.4 8.2 0.0 65 144 Croatia 4.8 d 22 1.0 n.a. 20.2 11.4 14 n.a. Cuba 10.8 43 0.3 d n.a. 3.2 0.2 1 n.a. Cyprus 0.7 4 0.3 0.3 43.2 15.3 7 n.a. Czech & Slovak Reps. 10.3 49 2.0 e 2.3 19.0 7.5 46 304 e Denmark 5.2 17 3.1 2.9 58.9 359.3 185 n.a. Dominican Republic 7.4 19 0.6 0.3 7.4 11.0 3 n.a. Ecuador 11.0 109 0.6 0.5 5.4 0.0 30 256 Egypt 55.2 387 2.4 2.1 4.3 7.6 20 n.a. Finland 5.0 131 2.8 2.7 54.9 466.1 115 421 France 57.2 213 30.8 27.7 53.9 579.0 1,000 1648 Germany 79.9 138 36.9 31.9 46.2 1770.0 1,296 1512 Greece 10.1 51 4.7 4.0 46.9 17.0 15 n.a. Hong Kong 6.0 1 3.0 2.5 50.2 285.6 234 62 Hungary 10.4 36 1.5 n.a. 14.4 45.7 30 n.a. Iceland 0.3 40 0.1 0.1 54.7 17.4 4 n.a. India 896.9 1269 8.0 5.0 0.9 0.0 45 n.a. Indonesia 187.6 735 1.8 1.0 1.0 53.4 45 49 Iran 61.9 636 3.6 2.2 5.8 0.0 25 n.a. Ireland 3.5 27 1.2 0.9 33.4 61.1 80 n.a. Israel 5.4 10 2.0 1.6 36.5 50.0 80 n.a. Italy 57.8 116 24.2 22.4 41.8 1207.0 250 435 Japan 124.8 146 57.7 d 55.3 46.2 2008.0 5,750 1564 Sources: World Bank, International Telecommunication Union, Internet Society, TeleGeography, Inc. ©TeleGeography, Inc. 1994 164 Blue Pages International Telephone Traffic Outgoing mMil-I" Incoming mMil"r 1992 1993 Change 92-93 1992 1993 Chanqe 92-93 118.0 121.0 2.5% n.a. n.a. n.a. 124.3 137.1 10.3% 159.7 192.3 20.4% 670.0 745.0 11.2% n.a. n.a. n.a. 713.4 767.4 7.6% a 692.3 751.0 8.5% 74.1 77.0 3.9% n.a. n.a. n.a. 14.3 17.2 20.2% 66.8 83.9 25.6% 911.1 979.4 7.5% a 952.7 1025.3 4.6% 169.9 182.4 7.4% a 330.6 373.8 13.1% 81.0 91.0 12.3% n.a. n.a. n.a. Surplus/(Deficit) 1992 1993 n.a. n.a. 35.4 55.2 n.a. B.a. (21.11 (16.41 n.a. B.a. 52.5 66.7 41.6 45.9 160.7 191.4 n.a. n.a. 2262.0 2477.0 9.5% a 2667.0 3066.0 15.0% 405.0 589.0 55.0 61.0 10.9% 85.6 n.a. n.a. 30.6 n.a. 635.1 895.0 40.9% n.a. n.a. n.a. n.a. n.a. 94.5 102.4 8.4% a 238.0 278.7 17.1% 143.5 176.3 104.7 117.2 12.0% 189.1 170.3 -9.9% 84.4 53.0 12.4 n.a. n.a. 18.6 n.a. n.a. 6.2 n.a. 85.3 93.8 9.9% 73.7 72.2 -2.0% 11.6 (21.6) 178.0 181.4 1.9% 236.0 253.2 7.3% 58.0 71.8 424.5 452.3 6.6% 425.2 460.0 8.2% 0.7 7.6 53.5 58.3 9.0% n.a. 318.4 n .a. n.a. 260.0 28.6 33.1 16.1% 82.4 101.4 23.1% 58.8 68.3 73.2 80.0 9.3% n.a. n.a. n.a. n.a. n.a. 235.0 252.9 7.6% n.a. n.a. n.a. n.a. n.a. 2449.0 2576.0 5.2% a 2540.0 2710.0 6.7% 91.0 134.0 4087.0 4679.6 14.5% a 3100.0 3707.8 19.6% (987.0) (971.8) 298.9 336.2 12.5% a 359.7 406.1 12.9% 60.8 70.0 1136.6 1376.9 21.1% c 1009.4 1260.3 24.9% (127.2) (116.5) 183.8 213.2 16.0% 150.5 192.8 28.1% 22.1 24.1 9.0% a 21.7 23.4 7.9% 259.6 283.9 9.4% c 354.6 441.0 24.4% 118.1 143.8 21.8% a 165.5 201.8 21.9% 131.3 156.5 19.2% n.a. n.a. n.a. 296.6 383.9 29.4% c 383.0 423.0 10.4% 153.1 175.5 14.6% 266.0 n.a. n.a. 1473.4 1609.7 9.3% 1541.0 1672.7 8.6% 1283.5 1411.2 11.0% c 891.5 981.2 11.0% (33.3) (20.4) (0.4) (0.7) 95.0 157.1 47.4 58.0 B.a. n.a. 86.4 39.0 112.9 n.a. 67.6 63.0 (392.0) (429.8) a. MiTT based on billing point of traffic. b. Year ending 30 September. c. Year ending 31 March d. 1992 e. Czech Republic only Source: TeleGeography, Inc. ©TeleGeography, Inc. 1994 TeleGeography 1994 National Telecommunications Indicators (K-Z) Population Area Main Lines Main Lines Lines/100 Cellular phones Fax machines Internet Hosts 1993 (millions) (Miles2thous.) 1993 (millions) 1990(millions) people 1993 1993 (thous.) 1993(thous.) August 1994 Korea, Republic of 44.0 38 16.7 13.5 38.0 384.7 350 348 Kuwait 2.4 7 0.4 0.3 15.2 60.1 25 n.a. Luxembourg 0.4 1 0.2 0.2 56.7 5.1 6 n.a. Macau 0.5 1 0.1 0.1 27.7 18.1 7 n.a. Malaysia 19.2 127 2.4 1.6 12.6 309.0 53 n.a. Mexico 91.3 756 7.6 5.2 8.4 385.3 180 5164 Morocco 27.1 172 0.8 0.4 3.0 6.7 8 n.a. Netherlands 15.2 16 7.6 6.9 50.3 216.0 400 291 New Zealand 3.5 104 1.6 1.6 45.8 143.8 45 334 Norway 4.3 125 2.3 2.1 52.9 375.0 130 n.a. Pakistan 122.3 307 1.6 0,9 1.3 16.0 7 n.a. Peru 23.1 496 0.7 0.8 3.0 36.9 5 42 Philippines 65.2 116 0.9 0.7 1.3 76.9 30 n.a. Poland 38.7 121 4.4 3.3 11.4 10.7 30 n.a. P_ortugal 10.5 36 3.3 2.4 31.1 101.2 30 n.a. Qatar 0.5 4 0.1 0.1 23.2 4.3 5 n.a. Russia 148.7 8592 23.4 n.a. 15.7 10.0 125 259 Saudi Arabia 16.6 830 1.6 1,4 9.5 19.0 75 n.a. Singapore 3.1 1 1.3 1.0 39.8 179.0 60 67 Slovenia 2.0 8 0.5 n.a. 26.0 6.5 11 n.a. South Africa 38.5 471 3.7 3.3 9.5 40.0 60 n.a. Spain 39.2 195 14.3 12.6 36.3 257.3 215 246 Sri Lanka 17.6 25 0.2 0.1 0.9 13.0 10 n.a. Sweden 8.6 174 5.9 5.9 68.3 774.5 325 260 Switzerland 6.8 16 4.3 3.9 62.8 249.2 169 n.a. Taiwan 20.9 14 8.0 6.8 38.1 538.5 275 376 Thailand 58.5 198 2.2 1.3 3.7 436.0 60 73 Tunisia 8.6 63 0.4 0.3 4.9 2.3 12 n.a. Turkey 59.6 301 11.0 6.9 18.5 84.2 70 n.a. United Arab Emirates 1.7 32 0.6 0.4 32.4 70.5 25 n.a. United Kingdom 57.8 94 26.6 4.0 46.1 2215.0 1,200 1449 United States 256.2 3619 148.1 136.3 57.8 16009.5 10,000 n.a. Uruguay 3.1 68 0.5 0.4 18.9 5.0 8 101 Venezuela 21.1 352 2.1 1.5 9.9 177.0 16 399 Yugoslavia 10.5 40 1.9 3.8 18.3 0.0 13 n.a. Sources: World Bank, International Telecommunication Union, Internet Society, TeleGeography, Inc. ©TeleGeography, Inc. 1994 166 Blue Pages International Telephone Traffic Outgoing mMiT]" 1992 1993 Chanqe 92-93 1992 305.9 355.4 16.2% 453.9 112.7 116.8 3.7% n.a. 181.0 199.3 10.1% 107.5 76.9 89.9 16.9% 68.6 216.5 258.1 19.2% 249.0 683.5 625.4 -8.5% a 1115.0 102.6 125.0 21.8% n.a. 1133.9 1238.2 9.2% a 1039.0 189.0 219.0 15.9% a n.a. 349.0 376.2 7.8% 314.0 44.2 56.5 27.9% n.a. 32.1 39.0 21.5% 128.1 135.8 164.0 20.8% 462.1 212.7 272.7 28.2% 366.6 212.0 232.6 9.7% a n.a. 50.7 58.3 15.0% n.a. 175.6 201.0 14.4% 230.7 464.6 447.5 -3.7% 292.1 412.0 480.0 16.5% c n.a. 46.2 62.8 36.0% n.a. 221.7 255.0 15.0% 250.7 804.5 846.9 5.3% 847.2 17.8 19.5 9.6% n.a. 691.0 740.0 7.1% 595.0 1551.0 1572.0 1.4% 1191.5 368.7 440.7 19.5% a,c 484.4 132.4 161.8 22.2% a 212.7 68.0 67.0 -1.5% b n.a. 226.8 264.6 16.7% 560.0 299.0 341.6 14.2% n.a. 2849.0 3130.0 9.9% c 2789.0 10173.6 11461.2 12.7 % a 5153.9 30.2 37.4 23.8% a 53.0 115.5 133.3 15.4% a 128.6 142.8 181.5 27.1% n.a. Incoming mMil"l" Surplus/(Deficit) 1993 Change 92-93 1992 1993 510.5 12.5 148.0 155.1 n.8. ~.a. rl.a. rl.a. 131.7 22.5% (73.5) (67.6) 71.3 3.9% (8.3) (18.6) 304.2 22.2% 32.5 46.1 1370.6 22.9% 431.5 745.2 n.a. n.a. n.a. n.a. 1163.0 11.9% (94.9) (59.2) n.a. n.a. n.a. B.a. 322.5 2.7% (35.0) (53.7) 305.7 n.a. n.a. 249.2 152.4 19.0% 96.0 113.4 n.a. n.a. 326.3 n.a. 431.5 17.7% 153.9 158.8 438.2 n.a. n.a. 205.6 n.a. n.a. B.a. n.a. 268 16.2% 55.1 67.0 n.a. n.a. (172.5) n.a. n.a. B.a. B.a. B.a. B.a. B.a. B.a. B.a. n.a. n.a. 29.0 n.a. 908.4 7.2% 42.7 61.5 65 n.a. n.a. 45.5 630 5.9% (96.0) (110.0) 1258.7 5.6% (359.5) (313.3) 490.8 1.3% 115.7 50.1 218.7 2.8% 80.3 56.9 n.a. n.a. n.a. n.a. 605 8.0% 333.2 340.4 B.a. n.a. n.a. B.a. 3088 10.7% (60.0) (44.0) 5392.8 4.6% (5,019.7) (6,068.4) 58 9.4% 22.8 20.6 148.3 15.3% 13.1 15.0 223.5 n.a. n.a. 42.0 a. MiTr based on billing point of traffic. b. Year ending 30 September. c. Year ending 31 March d. 1992 Source: TeleGeography, Inc. ©TeleGeography, Inc. 1994 TeleGeography 1994 Notes 168 TeleGeography 1994 Notes 169 TeleGeography 1994 The E-Mail Comet File ~_dit Options Navigate onno~ate OO~llSt released that day to the rest of the planetary sci- ~-,~-~-h0e~-~--[-~-~xo~ ~-~i ence community and the public. For Shoemaker- ~,~-~:.~#,~ ............. Levy, that kind of centralization was out of the UEL~lh~tp:(/et~Pr?d~tt~#£l~sa~/s]9/~9:ht~ .............................. question. There was no mission control, no pro- http://newprodu cts.jpl.nasa.gov/s19/s19.html W hile comet Shoemaker-Levy mount- ed its fireworks display at Jupiter, another kind of tour de force was underway on a computer in the astronomy department at the University of Maryland. An Internet "exploder"--a system capable of multiplying any incoming electronic mail message hundreds of times and automatically sending the copies to a predetermined mailing list--had linked hun- dreds of observers in an unprecedented on-line community. "The exploder was one of the great- est things that ever happened to observational astronomy," says planetary scientist Paul Weissman of the Jet Propulsion Laboratory (JPL}. Full participation in the Voyager flybys of the 1980s and in earlier spacecraft encounters, after all, was generally limited to a coterie of scien- tists. Teams operating each Voyager instrument met daily at JPL in Pasadena, California, to review the latest data and decide what would be ject scientist--just hundreds (if not thousands) of astronomers aiming different instruments at Jupiter from around the world. The flow of information was sure to be freer than during earlier events in planetary science, but it also threatened to be slower--because of the lack of coordination. So last January, astronomers planning to observe the impacts set up the exploder, which ultimately linked about 250 observing groups and theorists. One result was that during impact week, theorists could start puzzling over startling observations almost immediately. For example, when David Rabinowitz and Harold Butner of the Carnegie Institution of Washington, observing from Chile, saw Jupiter’s satellite Io redden for 7 minutes during the impact of fragment B, they sent a message to the exploder. In minutes to hours, depending on network traffic, the exploder had dispatched the puzzling news from Chile to everyone else on the mailing list. Similarly, astronomers preparing to make obser- vations could modify their plans depending on what others had seen. Astronomers waiting for sunset in Arizona, for example, were privy to what their colleagues had just seen from a tiny island in the Indian Ocean, including information on when the impacts were occurring and which filters and exposure worked best. [A] few theorists trying to make sense of the stream of messages sounded a little wistful for times when the pace was more deliberate. "Its like trying to drink from a fire hose," says Donald Yeomans of JPL. "There’s too much to digest before the next message comes in." But Yeomans, who like all scientists prefers a data glut to a shortage, adds: "It’s fun." Excerpted with permission from Richard A. Kerr, "The Comet Explodes by E-Mail," Science, vol. 265, 29 July 1994, p. 602. ©AAAS 1994. 170 PREPARATION OF THIS REPORT WAS SUPPORTED, IN PART, BY GRANTS FROM MCI COMMUNICATIONS CORPORATION AND STENTOR RESOURCE CENTRE, INC. STENTOR A THIS REPORT IS PUBLISHED ANNUALLY BY TELEGEOGRAPHY, INC., WASHINGTON, D.C. IN COOPERATION WITH THE INTERNATIONAL INSTITUTE OF COMMUNICATIONS (IIC), LONDON ISBN 1-886142-01-7