Home » Linux Support » Algorithms in Africa

Algorithms in Africa

Wayne Marshall

Issue #86, June 2001

Maybe the rush to market for spreading internet access across the globe isn’t in anyone’s best interest—a report from the front.

Eleven years ago I installed a computer system at a vocational training and development center in Tutume, Botswana. Tutume is a rural village on the northeastern edge of the Kgalagadi desert in southern Africa. The computer was intended to help this organization, known as Tutume Brigades, catch up on its bookkeeping for several business units crucial to the local economy. Businesses included a brick-making unit, carpentry workshop, auto repair garage, sorghum mill, school uniform production unit, tractor hire and vegetable garden. For the local village and the surrounding catchment era, the Brigades were literally the only game in the bush for commodities, trade skills, training and employment opportunities.

When I arrived in Tutume, I was a pure novice in the field of foreign assistance. I was also a mid-career financial professional, with several years of experience in nonprofit and health-care management in the United States. And like most aid workers new on the ground in Africa, I knew what was best. In my assessment of the center, I believed a computer was essential to get a handle on the Brigades’ financial position, which otherwise consisted of eight separate sets of badly maintained manual ledgers, over nine months in arrears. Except for the bank statements of eight separate checking accounts (and even the bank statements proved unreliable), we had no way of knowing if the center had any money. Every time we had to make payroll or buy another truckload of cement, we were in the heart of fiscal darkness.

Over the course of the next several months, I proceeded to computerize the records and train local staff in basic operation of the system. By the end of the first year, the financial records of the center were timely and accurate. Moreover, other staff members were beginning to use the computer for tasks such as word processing and spreadsheets. Many of these employees had never even used a typewriter before.

If I were to tell no more of this story and fade here to one of the glorious Kgalagadi sunsets, this might be called a win. Although set in the predawn (and pre-Linux) history of the Internet era, today this would be described as a small success story of “bridging the digital divide” in Africa—like I was a regular Albert Schweitzer of the Information Age or something.

But the truth is not so simple, and the issues of foreign assistance are not so trivial. The fact is, I am not proud of this story. Because as my time in Tutume went on, I realized I had blundered badly, to the point of putting the Brigades in serious jeopardy. I began to ask myself such basic questions as: What would happen to the computer after I left? Was the staff fully capable of operating the system independently? Would backups be maintained and performed rigorously? Were skills sufficient to troubleshoot problems and reinstall the system if necessary? If the equipment failed or was stolen, could the center afford to replace it? And what would the center do when the staff I had trained for so long were lured away by more lucrative jobs in the big city?

These questions all led to the same answer: the Brigades would be left in even worse shape than I found them. Rather than gaining empowerment, independence and enablement, they would more than likely be left powerless, dependent and possibly ruined. And all because of my own cultural myopia, despite my good intentions.

It is axiomatic in the field of foreign assistance that the aid program will take credit for the successes, while failures are blamed on the host country. The psychology of failure can then be even more severe and long-lasting than the loss of the project. While I was working in Tutume, for example, a friend of mine was working in the village of Lobatse in southern Botswana. Seven years earlier, an aid organization from northern Europe had decided a wool sweater factory would be just the ticket for the economic development of the village. Of course, northern Europeans are fond of nice wool sweaters and very likely have great need for them, particularly in the colder climes of northern Europe. The market for wool sweaters is less extensive in the sweltering and sparsely populated Kgalagadi desert, however. After seven years of subsidizing the losses of the operation, the aid organization finally decided it was never going to be sustainable, and they pulled the plug on the effort. My friend’s unenviable assignment was to put all the women out of work, sell the facility and liquidate the equipment. It was hard for many of the women not to feel that the fault was somehow their own.

Fortunately for Brigades in Tutume, such failure was averted. As the story there continues, once I realized the risks, I spent the next several months converting the accounting system back to manual ledgers, hiring and training additional staff in bookkeeping procedures and enabling them to use the computer primarily as a support system, rather than as the central financial database.

But what do these stories from Tutume and Lobatse have to do with Linux and emerging markets? The rest of this article will consider that question.

The Digital Divide

Nine years have passed since I left Botswana. To put the times into perspective, the first thing I bought when I got back to the US was a fax modem, the cheapest, fastest solution to stay connected with the contacts I had made abroad. My modem then was 2,400 baud. I tried out CompuServe and decided on Delphi, and the buzz was just starting about something called PPP.

During the next several years I was in and out of Africa, became a Linux user in 1995, began installing Linux in nonprofit organizations in 1997, spent a year and Y2K transition in the former soviet state of Ukraine and came to the West African country of Guinea in May 2000. At some point during this period the digital divide was invented.

Actually, the digital divide seems to have its origins in a 1995 report from the US Department of Commerce, whose National Telecommunications and Information Administration (NITA) released the first paper in a series titled “Falling through the Net”. This report analyzed telecommunication access by geographic and demographic patterns throughout the United States. One of the conclusions of the report was the gap between the “information rich” and the “information poor” had widened.

In the later years of the Clinton administration, the digital divide broadened beyond US borders to encompass the globe. The issue gained considerable publicity after a G8 economic summit meeting in 1999, where the most powerful nations on earth decided that the growing gap in information technology was one of the most serious problems facing development in the Third World.

Now, as I write this, bridging the digital divide has become one of the hottest trends in foreign assistance, and many aid organizations and corporate philanthropists have found publicity for their efforts. Simplistically, it seems, the gap in information technology has now come to be identified with access to the Internet. Thus, we have such programs as the USAID-funded Leland Initiative, designed to bring internet access to Africa; the Peace Corps announcing an information technology initiative in partnership with AOL; and a recently formed organization called Geekcorps sending its second group of volunteers on three-month stints designing web sites in Accra, the capital of Ghana in West Africa [see LJ April 2001 for more on the Geekcorps]. Naturally, the high-profile publicity given this issue has created an opportunity for many international aid organizations to develop projects and funding appeals for serving the digitally needy.

The New Tech Testament

Delivering the miracle of the Internet is the new zeal of the high-tech missionary. In what seems to be a rush to market—bringing the Internet to the developing world—sometimes projects are announced with only naïve regard to the technical issues and without full consideration of whether such projects are viable, appropriate, relevant and sustainable. Thus, one hears of a women’s cooperative in Central America marketing their handcrafts over the Web; advocates describe the potential of “telemedicine” for delivering virtual health care to isolated areas; and the US State Department Global Technology Corps proclaims, “We have seen farmers in Mexico using [the Internet] to check weather conditions and crop prices.”

Where once Norwegians may have seen wool sweaters, the tech visionary now sees web browsers.

At the extreme, the new economy proselyte promotes the Internet as the solution for everything from education and health care to pollution, inequality and world peace. As though everyone who has access will be able to browse their way to nirvana, as though the path to heaven is paved with bandwidth. The satellite dish is the new icon of the digital evangelist, replacing the holy cross.

One of the implicit beliefs of this testament is that information, in and of itself, is sufficient to promote economy, remedy problems and narrow inequities. A corollary implication, the message from one side of the divide to the other, is that we have information and you don’t, that our information is good and yours is useless. This is the lesson CNN preaches to its international audience when it tells us, “The human without information is nothing.”

It should be clear that in this form, divide rhetoric is simply new raiment for the familiar old taxonomies of prejudice that have long sought to divide the world between believers and heathens, the enlightened and the savage. From a historical perspective, rather than helping, these kinds of belief systems have generally been devastating to their targets.

More importantly, the belief in the sufficiency of information and information technology is simply wrong. Information alone doesn’t help people. If only this were true, doctors would be made from medical textbooks and entrepreneurs would be born from accounting manuals.

In fact, the developing world is littered with unused X-ray equipment, broken-down tractors and empty schoolrooms contributed over the years by well-intentioned and simpleminded donors. These resources are made useless not from missing user manuals or lack of web access, but by the lack of trained technicians, mechanics and teachers.

In short, what empowers people are skills.

Even in the US, this kind of awareness is emerging. In “How Does the Empty Glass Fill? A Modern Philosophy of the Digital Divide” (Educause Review, Nov/Dec 2000), Solveig Singleton and Lucas Mast write: “From the standpoint of higher education, students who leave high school without exposure to digital learning tools such as the Internet will prove a much less serious problem than students who leave high school with inadequate reading or math skills.”

And the leading journal of free-market capitalism, the Economist, recently observed:

The poor are not shunning the Internet because they cannot afford it: the problem is that they lack the skills to exploit it effectively. So it is difficult to see how connecting the poor to the Internet will improve their finances. It would make more sense to aim for universal literacy than universal Internet access.

It may be that, with the recent outbreak of dot-com bankruptcy and declines in the stock market, the tenets of the digital religion could be losing their currency. At a time when the mega-billion, IPO-funded ebiz stars like Amazon and Yahoo are having a tough go across the US and Europe, it’s hard not to wonder how the promises of e-commerce could possibly prove viable and sustainable elsewhere, particularly in places where there aren’t even good banking and credit systems. And for someone like me who has lived several years of the past decade in both rural and urban parts of the developing world—where most of the population still cook with firewood and carry water in buckets—the practical value of focusing foreign assistance on IT projects would seem negligible, if not ludicrous entirely. Given the more serious fundamental issues facing developing nations—health care (AIDS, TB and malaria), nutrition, sanitation, education, poverty, pollution and political corruption—providing the means to surf the Web should probably fall fairly low on any reasonable scale of human priorities.

So is there any way to make a difference, a real difference that improves people’s lives? Is there any role for Linux and open-source advocacy in emerging markets? Are there ways of using technology for solving human problems in places like Africa, without trying to sell wool sweaters in the desert? I wouldn’t be writing this article if there weren’t.

Algorithms in Africa

When it comes to Africa, the so-called digital divide is just a divide; there isn’t anything especially digital about it. The divide is geographic, because Africa is a long way away, and cultural, because the traditions and histories of Africans developed independently from those of Europeans and Americans. Almost incidentally the divide is economic, from the standpoint of cash resources and differing perceptions of wealth, though the natural resources of this continent are vast. The divide ends up being mostly one of ignorance, and this gap is at its widest in America.

Americans in general know very little about Africa, and what little they do know or think they know is usually prejudiced and fallacious. If I were to know the state of Florida only from news reports, I would think it was a large mobile-home park of fat pink people constantly flattened by hurricanes. Similarly, most Americans probably only know Africa as a disaster zone of epidemic, starvation and genocide. The principal media image Americans hold of African assistance is usually the one of the brave young (white) woman, a nurse or volunteer, holding a helpless black infant, center stage among a group of grateful and admiring Africans in the background.

Of course Africa is nothing like this image at all, and the first step in crossing the divide here is to banish these offensive stereotypes and learn all one can about what Africa is really like. It would be a disservice to the many peoples of the continent to generalize and describe the essence of Africa as though it were a single place. But I would just like to say: Africa is such a joy! Whenever I am in the streets of Conakry or an upcountry village, I am overwhelmed with the pure bandwidth of humanity, of color and vitality and life. So much more than can ever be expressed on even your largest CRT, with even the fastest DSL connection; Africa is the ultimate realization of broadband in culture and diversity, natural and human content. Maybe a virtual, flat-screened reality over the Internet is meaningful in the pitifully dreary cubicle of the US office worker, but Africa is all about face time in real time.

Open-source advocates can be sure that Africans get community; Africans get bazaar. These are concepts intrinsic to the cultures and traditions throughout the continent, where African societies had mastered networking long before the invention of the RJ45 jack. Africans have historically been quite receptive, often at their ultimate peril, to ideas and innovations flowing between cultures and brought in by outsiders. And in general Africa has been early and enthusiastic about adopting new communication technologies, particularly when they are practical and affordable. So in Botswana I was astonished at the number of fax machines per capita ten years ago, and now find a thriving trade in cell phones, both legitimate and black market, in Guinea. On a recent visit to a mosque in the interior of the country, a wizened old muezzin took me up into the minaret specifically to show me their solar-powered amplifier and loudspeaker system, used to call the village to prayers.

As one learns to develop an appreciation of what Africa is really like, it will then help if one can develop a sensitivity to the pitfalls of foreign aid and the havoc such programs have brought to this continent. The subject of other narrations, it is sufficient to observe here that the official assistance programs of foreign governments are usually a foul brew of political hegemony, economic imperialism, cultural ethnocentrism, personal avarice and, too rarely, genuine altruism. Too often the implementation of foreign aid is all about developing market share and spheres of influence, instead of improving lives. Proponents of foreign assistance may even argue that these are synonymous, as though markets for American soft drinks, snack foods and beauty products result in happiness and prosperity for the consumer. The sad fact is, whether intentional or merely consequential, foreign assistance has often had devastating effects on communities, local markets, traditional cultures and environmental conditions throughout Africa.

Finally, it is helpful to bring an honest perspective of one’s own history and culture into focus. For example, the United States represents less than 6% of the world’s total population and has existed for less than a blink of an eye in the span of all human history. So, what makes us think we’ve got it right? What evidence is there to suggest this brief record is proof that our way of life and cultural adaptations will be viable in the long run?

For example, it may be surprising to learn that, due to the predations of infectious illness, urban population levels were not even sustainable until about 100 years ago and required steady migration from rural areas. And it was less than 90 years ago, Gina Kolata writes in Flu, when “Ladies Home Journal proudly declared that the parlor, where the dead had been laid out for viewing, was now to be called the living room, a room for the living, not the dead.”

Shortly after this proclamation, a global flu of epidemic proportion—the origin of which is still not understood—killed 1.5 million Americans and 40 million worldwide. This was not in the murky history of the Dark Ages; this was 1918. Today, with the modern plague of HIV/AIDS, the re-emergence of tuberculosis and new mysteries like the relationship of human CJD to Mad Cow Disease, will our mastery of medicine prove all that enduring, even for the world’s most fortunate few?

In any case, those who would help others should at least try to learn from past failures and have the humility to ask if the modern model of urbanization, congestion, resource utilization and environmental depletion are sustainable, even desirable, let alone worthy of export to others in the world.

Then we may be able to accept that the Internet may not be the solution to all problems of humankind and have the patience to realize that working through the major challenges in Africa will take time and understanding measured in generations. Now it becomes clear that Linux and open-source developers are helping Africa best by what they have been doing already. People who are programming and installing the world-class, free software at the soul of internet technology are helping others around the world in profound and important ways, no matter what license they are using. GNU and open-source software are the perfect fit for the emerging nations of Africa—as for the rest of the world—not only for the superior technical quality of these systems, but for the values embodied in their development.

The mere existence of Linux and open-source systems give people the chance to use these powerful technologies for low-cost, grassroots level applications, an opportunity not possible just ten years ago. The pages of this magazine have described many of these self-directed success stories, everywhere from Mexico to Pakistan, where Linux solutions enabled people to make the difference. Such examples are to be found among African communities as well, from South Africa to Kenya to Nigeria. And Africans like Katim Touray are using Linux servers to connect other Africans in dialogue around the world.

Beyond the software itself, though, it is the culture of Linux and Open Source communities that provides the model for meaningful outcomes. This is the culture of sharing and empowerment, of the thousands of Linux users’ groups throughout the world, of the Linux Documentation Project and the general willingness of one user to selflessly help another. Participating as a Linux user is all about developing crucial skills and passing them on. Often users’ groups hold regular installation clinics, giving new users personal, one-on-one support from an enthusiastic peer. And these users’ groups are often active in other community projects, such as helping schools install servers and network connectivity, while transferring the skills necessary to maintain them. Each of these connections is essentially more human than technical, linking people together more than their machines, and can lead anywhere. Each of these personal connections sows the seeds of others, and the spread of the Linux bloom is now reaching to every corner of the earth. For example, even though the use of internet technology in Guinea is nascent, Linux certainly preceded my own arrival here. One finds Linux books in French in bookstores and Guineans eager to learn more about this “true” operating system.

And there are other instances of Linux and open source helping to solve problems in Africa. One of the most inspiring and hopeful to me involves no computers at all.

Vim in Uganda

The emergence and spread of AIDS has been devastating to sub-Saharan Africa. Sure, you are probably tired of hearing about it. For one thing, it is so hard to come to grips with the scale of the problem. In the short time since I left Botswana—when AIDS was just beginning to emerge as an issue there—life expectancy has plummeted, from nearly 60 years to barely 40. It is now estimated that as many as 40% of the adults in Zimbabwe are HIV positive. This has been a debilitating setback to the emerging countries of the region, where public health efforts had previously been making remarkable gains.

The epicenter of AIDS in Africa has been Uganda, which was hit first and perhaps hardest. The government of Uganda is considered to have mounted an effective and ongoing public health campaign for its people, and there is hope that the incidence of HIV/AIDS is decreasing. Nevertheless, the consequences of the disease have been severe. One of the biggest problems is the large numbers of children left without parents. In a society where children are traditionally treasured and raised with the supportive assistance of extended families, there are simply too few adults left to care for growing numbers of orphans.

Bram Moolenaar is the author of Vim, one of the most popular open-source text editors, with ports available for just about any platform in existence. Bram had already started Vim when he first went to Uganda in 1994, volunteering to work as a water and sanitation engineer for the Kibaale Children’s Centre (KCC).

The center, located in a rural village of southern Uganda, provides food, medical care and education to about 600 children, most of whom have been orphaned by AIDS. The conditions are austere: one book for ten children, a tiny blackboard and a roof with holes.

Bram found that his skills could help at Kibaale, his help made a difference. After a year spent working with the Centre, he wanted to find ways he could continue helping the project while also letting other people know of its existence.

That’s when Bram hit on the idea of “charityware” for Vim. The license for Vim says simply: “Vim is Charityware. You can use and copy it as much as you like, but you are encouraged to make a donation to orphans in Uganda. Please read the file doc/uganda.txt for details.”

While using Vim, type :help uganda to get the complete text of the license and a description of the Kibaale Children’s Centre.

Beyond this, though, Bram is fairly modest about the project. Although he asks for copies of CD distributions that include Vim, he doesn’t appeal to distribution vendors directly for any additional financial support. Bram prefers to remain low key rather than risk annoying people and turning them away from supporting the Uganda project.

Knowing that Linux distributions in use are now in the billions, one may wonder how successful the charityware license has been as a fund-raising method for the Centre. Vim users are asked to make contributions to the International Child Care Fund that Bram and his colleagues have set up specifically to support the KCC project, and the ICCF web site provides annual financial reports. For 1999, donation income totaled about $7,000 US (17,800 Dutch Guilders), up from about $3,500 US in 1998.

These figures may seem rather underwhelming and suggest that the conscience of open-source users and vendors is not as evolved as one may like to think. But the bottom line for Bram is, even at such a modest level, these contributions make a huge difference in what the KCC can accomplish. The funds raised by Vim donors are used to keep the Centre running, maintain and improve the facilities and recently purchased rainwater tanks so that more people have access to clean water.

Bram continues his personal involvement with Kibaale to this day, having made return trips in 1996, 1998 and 2000. This experience gives Bram a thorough grounding in the realities of life in Africa, as well as an understanding of the means of effecting meaningful change. When I asked for his opinions about the digital divide, he said, “I’m afraid I don’t know what the digital divide is. Is it about bringing computer-related stuff to Third World countries? Well, the area around Kibaale first needs a good water supply and a phone.”

When asked if he could give any suggestions to those interested in projects supportive of African information technology, Bram replied, “The best suggestion I can make is to work in small groups. A hundred small projects bring more benefit than one project that’s a hundred times bigger. The strategy and planning done by people in head offices is a waste of time and money.” The message here is that the strength of any bridge depends upon its integrity.

In the end, Bram is doing what the Open Source movement has been all about from the beginning: working with personal conviction, making a difference where one can and sharing the work one loves with others. These are the ideals of a world seeking connections, the values that can link Linux and the Internet with an orphanage in Uganda. The human connections of these efforts empower people, improve lives and build the solid bridges of understanding among diverse global communities, digital and otherwise.


Wayne Marshall (guinix@yahoo.com) is a UNIX programmer and technical consultant living in Guinea, West Africa.