Internet in Africa

INTRODUCTION

The following article about telecommunications in Africa originally appeared in the 25 November 1995 issue of the Dutch weekly `Vrij Nederland', in an abbreviated version of about 7700 words, 3000 less than this one.

Telecommunications in Africa is gaining a lot of momentum these days - especially now that telecommunication is no longer seen as a luxury for developing countries, but as a prerequisite for their economic growth. This way of looking at telecommunication in the developing world started in the beginnings of the 1980's, when the Internet was still in its infancy; it owes much of it's present popularity to the rapid growth of datacommunications, via the Internet in particular.

While Africa is bustling with plans for new telecoms applications using the existing infrastructure, and for grand new designs, I hope that this article will serve as a general introduction. Telecommunications is bound to profoundly alter the economic, social and political landscapes of Africa - probably even more so than in area's where a good telecommunications infrastructure was already in place when the current, exiting developments began to change the world as a whole.

Publication of this article, or parts of it, is subject to my prior consent, but any recipient should feel free to forward it as a datafile.

A 75 KB www-version is available at http://www.toolnet.org/~hege

Michiel Hegener

(mh@nrc.nl),

24 March 1996

===================================================

TELECOMMUNICATIONS IN AFRICA

- via Internet in particular

"Hahaha," roars Professor Donald Ekong in this spacious villa on the outskirts of Accra, Ghana, that is home to the Association of African Universities. The Nigerian Secretary-General [1] has a keen eye for the funny side of life, including some of the consequences of Africa's poor telecoms infrastructure. "Haha! Fortunately, we could just read the Internet address at the top of the fax you sent us. Otherwise, you would never have got a reply."

Two weeks earlier and a few hours after sending my fax, I had received an e-mail reply from Accra. Interference on the line had made my fax illegible except for one or two fragments that luckily included my e-mail address and allowed Ekong to find out who I was and what my fax was about. I e-mailed him back that I had heard from various sources how the AAU had been trying to encourage the use of electronic mail among its members, and that I wanted to talk to him about it.

Of course, the spread of the Internet in African higher education cannot be considered in isolation from the development of the Internet on the continent as a whole. And that inevitably raises the subject of its telecoms infrastructure, because Internet traffic almost always travels down ordinary telephone lines, usually those of the public telephone network. This partly explains why the Internet has mushroomed in regions of the world with high telephone density: the network was already in place. It also goes some way to explaining why less than one percent of the world's Internet traffic currently reaches Africa: the telephone network hardly exists. Compare Sweden (with 68 telephone connections per 100 inhabitants), the USA (with 57), and the Netherlands (49) on the one hand to Zimbabwe (with 1.22), Ghana (0.3), and Chad (0.07) on the other. Not all Internet traffic travels via the public telephone network. Very busy routes are served by lines with very large throughput capacity, which are dedicated to Internet traffic and usually leased from the public network. But these "backbones" of the Internet are found nowhere in Africa except South Africa. To obtain a fast Internet connection in Kenya or Cameroon, you need a leased line to the UK, France, or South Africa. Unfortunately, leased lines are so expensive that most African Internet traffic travels the cheap "store-and-forward" way. This is the method used by the AAU, which sends and receives all its Internet communications via a computer in South Africa where they are temporarily stored. Every eight hours, the central computer at the AAU-headquarters in Accra automatically dials the computer in South Africa, picks up its incoming e-mail messages, and fires off outgoing ones all over the world.

In Africa, e-mail is not a luxury but a bare necessity - much more so than in Europe. The main reason, says Ekong, is that other modes of telecommunication are too costly for African budgets. It takes ten minutes to read 2,000 words aloud, and a ten-minute voice phone call from the Netherlands to Ghana costs 34 USD. And although the fax machine is faster and cheaper (2,000 words in a compact font will take two minutes to send), the same operation will still cost 7 USD. By contrast, an e-mail message of 2,000 words (around 12 kilobytes or 96 kilobits in digital terms), sent via a modem with a throughput speed of 14 kilobits per second, will take seven seconds to reach Accra from Amsterdam, and cost 0,40 USD. What is more, if the telephone line is good, a 28 k/s modem working at full speed will further halve the transmission charge - making it 175 times cheaper than a voice phone call across the same distance. No wonder e-mail is so important to the developing world - not least for the planet's poorest continent. It is the only mode of international telecommunication that Africa can afford on any reasonable scale.

The AAU now unites 119 universities in 42 African countries. Its main task is to promote cooperation and communication among its members. Africa did not fail to notice the Internet's breakthrough as a medium of communication among Western universities in the early 1990s. Ekong explains: "We conducted a feasibility study, and our conclusion was that, for African universities, e-mail was the mode of telecommunication of the future. In June 1995 we linked up all the PCs at our headquarters in a Novell Network over ethernet cabling, essentially to improve our own internal capacity. By doing so we also developed a more reliable data communication infrastructure in order to support full Internet connectivity in the future - and to gain experience now. Since then, our technicians have given seminars to university staff on setting up a store-and-forward e-mail system. The AAU is not in the business of installing e-mail for its members - that's not part of our mandate. We do however give information and advice."

In addition to a fully-functional telephone network, e-mail requires PCs. The first thing most Dutch people ask when they hear about the Internet in Africa is: "But they hardly have any PCs there, do they?" This is a fallacy, says Professor Ekong: "PCs are widespread in African universities - that's not the problem. The greatest factor limiting the spread of e-mail in Africa is the lack of technical knowhow, followed by the faulty telecommunications infrastructure." The AAU's Internet connection will continue experimentally until the end of 1995. That is as long as the AAU can pay the bill for phoning South Africa three times a day. External users such as the University of Ghana and the British Council will provide experimental support up to then. In return, the AAU will allow them free e-mail access, making them pay only the telephone charges between their PCs and the AAU's server connection to South Africa. Even though things are still at the experimental stage, the advantages of datacommunications are strikingly obvious. Ekong has rarely lost an e-mail message, and he is favourably impressed with his two-year experience with the modem. This impression is echoed by the 40-odd AAU members who now have e-mail. They receive messages within a day after their transmission - many times faster than members who still use "snail mail".

When you walk through the main door of the AAU library, at first glance nothing seems amiss - until you notice the publication years of its books and magazines. The dog-eared Encyclopedia Britannica, for instance, dates from 1974; and the magazines are often months, sometimes years, out of date. In a back office, AAU network manager John Bart-Plange is giving telephone advice to computer users with problems. At the end of a call, he says: "Yesterday, I received a postal query from a lady in Nigeria. She sent it four months ago! The postal services in Africa are extremely slow, the delays are endless, and many letters never even reach their destination. Whenever I'm outside Africa and I see the Internet working - as at a recent conference of the Internet Society in Hawaii - I know that this is what we need! It's as clear as day. For education in Africa, you need the Internet - it'll give you all the information you can possibly desire." The fact that the AAU has only a store-and-forward connection to the Internet (also often called a dial-up link) does not mean that the Internet's tens of thousands of databases and hundreds of thousands of "homepages" are entirely out of reach; they are just harder to access. The information that Dutch Internet subscribers dig up using the World Wide Web, Gopher, and Telnet is available to Bart-Plange by means of automatic search keys that he sends as e-mail. The next time he contacts South Africa, they return with the information he has requested, like a dog fetching the newspaper. "For example, some time ago I needed a software program for the people here who work with Apple Macintoshes, but it was unobtainable in Ghana," says Bart-Plange. "I could have ordered it in America, sent a dollar cheque, and then had to wait for ages. Instead of that, I did an Archie search. I then received a list in my e-mail of all the ftp sites where that software could be obtained free. I ordered it from one of them yesterday, and this morning it was here. I've already installed it on the Macintoshes."

Requesting database directories is another tactic for exploring the Internet using a store-and-forward system. Database hosts will e-mail complete menu structures to Accra, where Bart-Plange then makes a choice offline. "Database menus also give the size of available files. So I can decide if the cost justifies downloading a particular file."

Since recently, such decisions have no longer been a daily necessity for William Tevie, chief systems engineer at Network Computer Systems and a leading figure among Ghanaian computer professionals [2]. There are ten telephone lines in Ghana permanently dedicated to datacoms traffic with Europe. The lines are leased from the Ghanaian PTT; and some of the datacoms hardware and software was installed by Network Computer Systems. One of these lines is leased by the Society for Worldwide Interbank Financial Telecommunication (SWIFT), which allows a Ghanaian bank clerk to see in a matter of seconds whether your credit card is valid, even if the information is stored in San Francisco or Zurich. Another dedicated line is leased by SITA, the worldwide network that lets you book a seat in Beijing on a plane flying from Rio de Janeiro to Montevideo. Network Computer Systems is primarily a computer importer. But you get more out of a computer if you use it for datacoms, too. So the company's acquisition of a leased line to the Internet was a major step forward. The line went into operation on 21 August 1995, making Ghana the first country in West Africa to have a permanent Internet connection. Sitting in a villa even more spacious than the AAU headquarters, Tevie says: "It costs a lot of money: 7,500 dollars a month for a 14 k/s connection to Cambridge, England - which is our gateway to the Internet - plus the actual Internet linkup charges and the necessary hardware." A lot of money, but the investment is already proving profitable. The Ghanaian linkup now has more than 100 users who each pay 100 dollars per month: more than four times as much as the average Dutch Internet user. "Unaffordable for the average Ghanaian," Tevie points out. "Our customers are expats, large companies, and a few Ghanaian researchers." A maximum of seven Internet users with '.gh' at the end of their e-mail address can call Network Computer Systems at one time - that's as many modems as they have right now. There wouldn't be any point in getting any more, because the 14 k/s line fills up as soon as that many Ghanaians all start downloading World Wide Web pages.

This state of affairs is bound to change. Tevie has big plans. Negotiations are already under way for Network Computer Systems to provide Internet connections for Ivory Coast, Nigeria, and Cameroon. And in 1996, Ghana will probably acquire a 64 k/s line dedicated to datacoms - with new access nodes in the towns of Kumasi and Takoradi if enough demand builds up. Tevie: "People here still don't realise how much use and pleasure you can get from the Internet. Mind you, this realisation only hit the West a couple of years ago. I predict an explosion in the sale of computers in Ghana. And now that we can offer the Internet, more and more buyers will be wanting a modem and an Internet course with it." Network Computer Systems also delivers complete homepages to order. An Accra-based construction company already has one, so customers in London or Amsterdam can start deciding on their new Ghanaian headquarters from home. And a Ghanaian fabric printing company wants a homepage, so that its worldwide customers can make choices without leaving the house and see the ordered goods before they arrive.

In a continent where most roads are terrible, where the postal services will often deliver only to post office boxes, and where there is a serious information deficit in almost all fields, a large number of connections to the Internet would be very useful.

According to Donald Ekong, African scientists often hardly know what their colleagues at other universities are doing, and they lack the capability to disseminate their own research results throughout Africa. The extreme shortage of up-to-date academic publications at practically all African universities could be remedied with a robust Internet link, a powerful printer, and a plentiful supply of paper.

Wide, reliable channels for international datacoms traffic (not necessarily via the Internet) could also clear the way for telemedicine. A doctor in a bush hospital might insert an X-ray photo of a complicated fracture into a scanner, and the picture would flit down the telephone line to a PC screen in the West, where a specialist colleague would give assistance with diagnosis and treatment. Telemedicine already has a central place in the plans of the World Health Organisation. The technology, money, and organisation are all available - all that is missing is the telecoms link to the places where telemedicine could be really useful. To contact a fellow doctor in the West, the average doctor in the developing world first needs to drive at least half a day by jeep to reach a city with a working telephone.

Problem-free datacoms connections between the West and the developing world would soon be paying their own way because thousands of development workers could stay at home. This would save a fortune in air tickets, tropical salary supplements, all-terrain vehicles, and rented homes with security staff.

Human rights and democracy also thrive on a good telecommunications infrastructure. Empirical research by American academic Tom Stonier has demonstrated that totalitarian regimes are incompatible with a telephone density of more than twenty lines per 100 inhabitants. [3]

These are just one or two examples. Much longer and more detailed accounts of this type have been available in the past few years in the publications of UNESCO, The World Bank, and the International Telecommunications Union (ITU).

In addition, more and more development cooperation organisations and ministries are opening their eyes to two new gaps that divide the world, more or less parallel with the old gap between rich and poor: the telecommunications gap and the information gap. Maybe it is just one new gap, now that telecommunications and electronic information are converging. Whatever the truth, this new global problem was growing for many years before it was recognised. In the development cooperation climate of the 1980s and before, a motorbike was considered of more use to an eight-year-old boy than a good telecoms infrastructure to Africa. That would come later - it was thought - after the water wells and women's projects. In the early 1980s, there was a change in Western thinking on the role of telecommunications in developing countries. In late 1982, at a Nairobi session of the ITU General Assembly, general support was given to "the fundamental importance of communications infrastructures as an essential element in the economic and social development of all countries" - including the poorest.

It is striking that at that time telecommunications were regarded essentially as a luxury: one of the pleasant side-effects of progress to which developing countries are also entitled. The ideal was a society in which everyone had a moped, a television, a fridge, and a telephone. In May 1983, the ITU set up an international committee of world-class telecoms experts to indicate how this goal might be achieved. The 123-page report of the Maitland Commission, named after its British chairman Sir Donald Maitland, was published in January 1985. According to the report, entitled The Missing Link, telecommunications were not the fruit of economic development - quite the reverse, in fact. The report claimed that a good telecoms infrastructure was a condition for economic progress. However, its conclusions had little impact on Dutch development cooperation policy. Until 1994, there was no cover for "telecommunications" in its yearly four billion dollar budget. Holland was not a big exception, however. The Missing Link disappeared into drawers all over the world. But with the advent of the information-based society in the West - accompanied by the breakthrough of the Internet and multimedia desktop computers - the Maitland Commission's work has gained new relevance. All kind of organisations have been dusting it down and re-reading it - and the ITU has come up with a major follow-up study. In March 1994, the World Telecommunication Development Conference met in Buenos Aires - not to draw up a follow-up report, but to plan a systematic attack on the communications deficit in developing countries. The Conference produced 50,000 words of resolutions, recommendations, terms of reference for working groups, and agendas for follow-up conferences.

The World Bank too expressed deep concern, at least enough for it to set up a brand new programme. The Information for Development Program (or InfoDev) started early 1995 on the considerable task of "helping developing countries integrate fully into the information economy by benefitting from the promise of information technology".

In June 1995, an InfoDev delegation toured Western Europe to put The World Bank's new-found calling to audiences concerned with telecommunications in the developing world. In the The Netherlands the audience was rather small. With some difficulty the Directorate-General for International Cooperation managed to get ten interested people around a table, about half of whom had the desired audience profile. Nevertheless, Elkyn Chaparro, senior adviser to the Vice-President on finance and private sector development, still managed to make a very impassioned speech. InfoDev's vision for telecommunications in the developing world rests on the expectation that telecommunications charges will fall much faster than they have done up to now so that geographical distance will cease to be of any great importance within the next ten years. Anyone wanting to phone Zaire still thinks twice before doing so - but that will change. InfoDev also expects the borderline between computer technology and telecommunications to become more and more blurred, and possibly even disappear. The result will be an information-based society whose telecoms infrastructure is a carrier of more and more economically relevant activity. Large-scale worldwide teleworking by individuals and companies will lead to nothing less than `the end of geography'. Once the global information infrastructure has become the binding factor in the world economy and everyone is equally able to get access to it - whether in an office in New York or in a village in the Congo - opportunities will blossom in areas now gripped by scarcity and isolation.

Chaparro's sparse audience was already in jubilant mood when he started to consider the other side of the coin. The blessings of the worldwide information-based society would be unlikely to accrue equally to every part of the world: some developing countries would use the new opportunities to optimum benefit, but others would do so hardly at all. Differences between countries would be determined not by climate, terrain, or location, but by the mentality of the inhabitants, especially of those in power. In March 1995, The World Bank published a report entitled Increasing Internet Connectivity in Sub-Saharan Africa [4], which made this point sharply and succinctly: "The information revolution offers Africa a dramatic opportunity to leapfrog into the future, breaking out of decades of stagnation or decline. Africa must seize this opportunity, quickly. If African countries cannot make advantage of the information revolution and surf this great wave of technological change, they may be crushed by it. In that case, they are likely to be even more marginalized and economically stagnant in the future than they are today."

Recent ITU reports have shown how great the danger really is. In the 48 least developed countries (LDCs), telephone density (the number of telephone lines per 100 inhabitants) has increased from 0.19 to 0.29 in the past ten years. Over the same period, telephone density in all the other countries with low incomes has quadrupled. 29 LDCs are in Sub-Saharan Africa. The problem is so great that development aid can only make any sort of mark in specific, clearly defined cases. Most insiders seem to agree that telecommunications in Africa can only be improved by the free flow of capital, on which countries with a heavily regulated telecoms sector cannot rely. The Telecommunications and Informatics Division of The World Bank stopped giving aid for infrastructural improvements a few years ago. It now offers aid only when PTTs are being privatised and telecoms monopolies deregulated. The private sector is expected to do the rest.

For Sam Pitroda, former chairman of the Telecom Commission of India and holder of more than fifty world patents for digital communication, the key question is not who invests in the developing world's telecoms infrastructure, but whether such investment takes place. Since May 1995, Pitroda has been chairman of WorldTel, hitherto the ITU's most ambitious initiative to close the telecommunications gap. The Maitland Commission had already proposed setting up a body such as WorldTel, to coordinate the development of telecommunications worldwide. In 1989, the ITU member states - practically every country in the world - decided to subject the old plan to a new study. Early in 1995, this resulted in a favourable report from the Australian division of the McKinsey consultancy, and shortly thereafter in the establishment of WorldTel.

Pitroda and I met at the largest telecommunications exhibition of all time: Telecom '95, last October in Geneva. On show was the latest in electronics from AT&T, Microsoft, IBM, Sprint, Unisource, Alcatel, Intel, Intelsat, Nortel, British Telecom, Belgacom, Informcosmos, Motorola, Beijing Telecommunications Equipment Factory, Ericsson Telecommunicazioni, and hundreds of other companies - not to mention the Indonesia Pavilion, the Ireland Pavilion, the Israel Pavilion, the Portugal Pavilion, and the Iran Pavilion. The exhibitors spent more than a billion dollars and the best part of two months making it all look good. Unfortunately, the seven halls of Geneva's exhibition centre measure only ten hectares in total. The result is a serious lack of space. France, for instance, had requested 12,000 square metres and got only 4,800. And several large telecoms operators had to spend well over five million USD constructing three or more floors, sometimes right up to the 18-metre-high ceiling. All this despite the fact that Telecom 95's organiser, the ITU, had asked the exhibitors to avoid such excess this time - at the previous Telecom exhibition, the German Pavilion had cost more than 30 million dollars.

More modest exhibition stands would certainly have been more in keeping with Telecom 95's theme of telecommunications in the developing world, but would such modesty have detracted from the cause? Probably not. As the chief executives of the world's telecoms giants all know, where telephone lines are sparse, there will always be a market for new ones.

This brings us back to Sam Pitroda and WorldTel, the ITU's response to the fact that 70% of the world's population have to make do with 3% of the world's telephone lines and that three-quarters of Africans have never made a telephone call. Anyone prepared to invest 50 million dollars of risk capital in the developing world's telecoms sector can apply to WorldTel for guidance, advice, and project proposals. That is roughly the idea - and roughly how candidate countries are approached: "Tell us what you want - and we'll say what we want," says Pitroda. "If there's no chance of at least a 20% profit per year, we won't even think about it. Otherwise, the investors will just go elsewhere. To the oil sector or somewhere. We are talking about an investment deficit of 50 billion dollars a year. There's no way you can make up for that with development aid."

WorldTel is not concerned with the profit - which goes to the investors - but with the construction of a basic telecoms infrastructure. Asked why WorldTel literature makes hardly any mention of the Internet, Petroda replies categorically: "Because the Internet comes after the telecoms network."

He is no advocate of interim solutions for obtaining the Internet or other forms of telecommunication. In recent years, for instance, cellular telephone networks have sprung up all over the developing world. In the big cities of poor countries such as Zaire and Sudan, the rich are already talking to one another on portable phones. The number of users of cellular phones rose in Africa from 25,000 in late 1992 to more than 250,000 in late 1994, and this exponential growth is continuing. On the other hand, however, it is often impossible to make a phone call to another city (because the only transmission equipment is an old half-rotting PTT copper wire). Of course, cellular networks hardly extend into the coutryside. Even poor suburbs are often left out. And in the sparsely populated countryside, no one is in a hurry to replace them. Pitroda has the same objection to small-scale satellite connections for datacoms using VSATs (very small aperture terminals), which have been selling like hot cakes in recent years, particularly in the USA and southeast Asia. "VSATs are good for solving specific problems, but they are not much use to a whole country. The point is that, if everyone is running around seeking different solutions, you will never get a proper infrastructure. Cellular phones and VSATs came to the west at a time when a basic infrastructure was already in place - thanks to state telecoms monopolies, don't forget! There is a time to privatise - but what, when, and how are complex questions." WorldTel's only concern is to obtain large profits for the investors and a good basic infrastructure for the developing countries. "If these aims can be achieved only by restructuring the telecoms sector and adjusting the charges, that's the way it'll have to be," explains Pitroda in one of the many nooks and crannies of the ITU stand. And what if a country doesn't want it? "Then WorldTel will go somewhere else. And we will say: you aren't ready for it yet, let us know when you are."

Ivory Coast is one of the African countries at the forefront of telecoms privatisation. The necessary legislation was passed in 1995, and is expected to enter into force in March 1996. The two key actors in the privatisation strategy are the Agence du Telecom and the Conseil du Telecom. The secretary-general of the Council, Samson Brou Yapo, sits in his spacious office on one of the top floors of one of Abidjan's highest buildings, headquarters of the soon-to-be-privatised state concern CI-TELCOM. Dozens of metres below, the Atlantic Ocean washes around Le Plateau, the most prosperous part of the capital. Only the dense masses of small traders on practically every pavement remind the passer-by of the poverty of Africa. The view from Brou Yapo's office window confirms Ivory Coast's reputation as one of the richest countries in Africa: wide arterial roads and a sea of high-rise office buildings dominate the scene. Nevertheless, the 13.5 million inhabitants have to make do with only 114,000 telephone lines, a waiting list of 80,000, and 8,400 villages with no telephone.

"For the villages, we've made a separate arrangement," says Brou Yapo. "After all, once CI-TELCOM is privatised, it is unlikely to do much, since the connection is very expensive and not very profitable. The plan is for the government to go in search of soft loans and guarantee them. We now have a detailed plan to install telephones in 379 villages, using a

...

William Paris,Jr/Current Pana Corp

117 W.Harrison Bldg,Ste C-637;Chicago,IL 60605-1709

Voice:312-667-9257;Fax:312-667-9093

*************************************************************

Date: Sun, 1 Sep 1996 08:28:37 -0700 (PDT) Subject: Internet in Africa From: mh@nrc.nl (Michiel Hegener) Message-ID: <199609090347.WAA08211@zeus.anet-chi.com>