The division between rich and poor grows exponentially, year upon year. The notion of the digital divide, the view that the poor are disadvantaged by their lack of access to modern technology, is the subject of much genuflection by politicians and commentators, but little changes, and less is done. The plight of the poor enters our consciousness only when we see pictures of poverty on our tv screens or dip into our pockets for charity. But charity is a palliative rather than a solution. The poor want practical solutions that facilitate financial independence.

That this is true has long been recognised by those who work in the field. A computer is of little use to those who have neither an electricity supply nor a telephone line. Better to spend money on practical technologies, which in some cases may be drilling a well for water, and in others may be the introduction of appropriate technologies that encourage sustainable methods of agriculture. The rest can come later.

But free software can and does have a useful role to play. The success of Linux has given access to industrial strength computer technology on commodity hardware without licensing overheads. This has meant that the educated minority in some developing countries have been able to access technologies that aren't available to them in the proprietary world. The lack of affordability of proprietary software is helpful to the dominant industries of the West, but creates barriers of entry for developing industries, which are always playing catch up. Free software provides access to technologies that would otherwise be prohibitively expensive for those living in developing countries, and in so doing, levels the playing field. On a smaller scale, activists will tell you that one computer that provides Internet access for a remote community can transform the possibilities of that community.

Free software has opened up other possibilities, such as that provided by ThinkCycle, a Web-based non-profit industrial-design project that brings together engineers, designers, academics, and professionals from a variety of disciplines, and seeks to create "a culture of open-source design innovation", with a collaborative community of individuals and organizations around the world.

At the heart of this community of designers, academics and professional engineers is "an evolving database of reasonably well-posed problems and ongoing design solutions contributed by universities, Non-Govermental Organizations (NGOs), companies and the general public." The idea is to use the resources of academic and research institutions to solve pressing, practical problems that affect people in the developing world.

Thinkcycle is already providing results that matter. Faced with cholera epidemics in large parts of the world, where the only reliable treatment depends on a computer-regulated saline drip that costs approximately US$2,000, ThinkCycle has come up with a reliable alternative that will cost a mere US$1.25 to manufacture, and will save many lives.

Free software has made a more direct contribution to alleviating poverty in the shape of Bram Moolenaar's "charityware" license for the Vim text editor, which has helped the financing of the Kibaale Children's centre in southeastern Uganda. Despite being relatively rich in natural resources, Uganda has been beset by a miserable history of civil war, disease and dictatorship, epitomised by the oppressive misrule of the ex-African Rifles sergeant, Idi Amin Dada, which led to the deaths of hundreds of thousands of the country's people and the destruction of its infrastructure.

Before and after Amin's downfall the country was torn apart by internal struggles which have been exacerbated by the overspill from the conflicts of its neighbours in the Sudan, the Congo and Rwanda, and the devastating impact of HIV and AIDS, which is estimated to have infected between 10 and 30 per cent of the population, and has left countless thousands of children orphaned and homeless.

In the face of such calamities, what value could free software have? You could, of course, use it to slash the administrative costs of rebuilding the country's infrastructure and facilitate the reconstruction of the nation's industries, but it cannot produce vaccines or treat disease. Nor can it put a roof over an orphan's head or fill an empty stomach.

Such was the dilemma that faced Moolenaar when he first visited the Kibaale Children s Centre in 1993 on a working holiday. Kibaale is a village in south-eastern Uganda next to the Tanzanian border. "It's on the map," Moolenaar has said, "but it's in the wrong position, probably to mislead anybody with 'rebel' in his name." The holiday was spent building a nursery school in a month "from nothing until the roof."

Located on a former army camp, the Centre provides shelter, medication, food and education for over 600 children who have lost one or both parents, usually through AIDS. Resources are limited: there is one book for every ten children, no electricity or running water, and the day-to-day amenities are sparse; but Moolenaar liked the project so much that he returned a few months later to work as a volunteer water and sanitation engineer between the summer of 1994 and the summer of 1995, and has returned several times since.

Moolenaar had already begun work on his software project when he first visited Uganda. Vim took its initial inspiration from the classic Unix text editor, vi, which Bill Joy developed at Berkeley, and has expanded and developed into an all-purpose program editor available for all platforms. His quandary was to find a way in which he could best put to use his skills in developing software to publicise and help the work being done at the Kibaale Children's Centre. Moolenaar's solution was simple. He changed the licence of Vim to read: "You can use and copy (the code) as much as you like, but you are encouraged to make a donation to the orphans of Uganda". It's what he describes as "charityware", a free software licence that wants you to participate in its aims but is non-intrusive.

The licence made a simple request that you show your appreciation of Vim by denoting a small sum to the children in Kibaale: US$15 a month to sponsor a child, US$25 a month to add to the child s environment, anything you can afford to sponsor the health team, or a one-time donation to the project. It was a low-key, non-obligatory but direct contribution to the children of Kibaale, whose immediate plight might otherwise receive little attention from the world outside.

As an example of what this contributes to Kibaale, Moolenaar gives the example of a boy named Boaz whom he met on his first trip to Uganda: "He was an orphan and lived at the centre as he had no relatives to take care of him. All he owned was a sack of old clothes." Thanks to "charityware" sponsorship by users of Vim, Boaz was able to go to school and be given the basic facilities. On a later trip "I met him again. He is studying to become a medical assistant, and was giving an injection to a sick baby at the medical post. It is exciting for me to see how this helpless boy has grown into a man who is able to help others."

Bram Moolenaar lives in the Netherlands, “in between the tulips.” He trained as an electronics engineer and gravitated towards software development through his work at Océ, producing integrated hardware and software solutions for copying machines.

He discovered free software when he built his first computer. “Well, actually, the second. The first one blew up when I accidentally reversed the power.” There was no money to buy software, so he looked around to see what was available for free: “There wasn’t much in those days, and no compilers. I had to buy one. But I spent more time fixing and improving the computer than actually using it.”

In 1988, he bought an Amiga 2000, “a real fancy computer in those days. And expensive too.” Among the free programs he found for the Amiga was a Vi clone called Stevie that came with the source code, but didn’t quite work as he expected, so he began fixing bugs and adding features. “At first I did it just for myself,” he says, “and later on for others.” Distribution was via newsgroups and via the Amiga disks collection made by Fred Fish, and the program provoked some enthusiastic responses. “Since then,” says Moolenaar, “it seems I have never stopped. Working on Vim is just like running a business, but doesn’t pay that well – in fact, not at all.”

Vim (which stands for ‘Vi IMproved’) is a hugely successful free software project that supports more than 30 different operating systems and many languages, including Arabic and Hebrew where text is written from right to left. The discerning user can even mix text that is written from left to right with text that is written from right to left, so that code can be documented natively in the language of choice. It is this level of attention to detail that has led to the widespread adoption of Vim on everything from Vax to Windows, by programmers all over the world. Vim can act as an OLE automation server, accessible from any automation client, such as Visual Basic, Python, or Perl, and is often used as the default editor within Visual Studio.

On Linux, Vim is seen as a replacement for Vi, attractive to many programmers because of the many features that distinguish it from a basic Vi implementation, such as unlimited undo, syntax colouring, split windows, visual selection, multiple buffers on screen at any one time, a graphical user interface that includes menus, mouse control, scroll bars, and text selection, and a thousand and one other features – yet still conforming to the original Posix description of Vi.

It isn't surprising that Vim occupies “too much” of Moolenaar’s time. “It leaves little time to do much else,” he says. “I still have to wash the dishes...!"

"Since Vim has become so popular," he says, "it has given me the responsibility to make sure that it works well. Vim has changed my life. The availability of free software, like the GNU compiler, has made this possible. If I had had to buy all the software I need to work on Vim, I probably wouldn’t have done it.” The children’s centre has provided the extra incentive.

Moolenaar is sceptical about the concept of the digital divide. It's not that he doesn't see a need for computers in their place, but he believes that current thinking reflects the historical failing in the way rich countries give aid. "We see the images of a poor country and we 'know' what the solution is. We see hungry children and send ships full of grain. We see illness and we send a plane filled with medicine. That is a normal reaction, but it's wrong." Rich countries impose their own short-term solutions on the rest of the world, he argues, when they should be looking beyond the immediate circumstances.

Just because the Internet and computer use are at the heart of economic growth in the West, and other countries are clearly lagging behind, "doesn't mean that we should be packing computers and cables in a container and shipping it over." He believes much else needs to be tackled first, such as training and skills, "networks of repair shops, and many other things, such as a mains plug and a phone line."

In Moolenaar's view, free software does have its place in less developed countries, where the cost of proprietary software is prohibitive and a drain on already limited resources. "Linux has the advantage that it can also run on old hardware," he points out. "There are computers in the third world. Don't underestimate that." But their use is more or less restricted to organisations that have money in the cities. "Can you imagine a bank without a computer?" he asks. "Well, it does work, but very slowly."

In the rural areas of Uganda, where most of the population lives, a computer remains a distant priority. Kibaale doesn't have electricity or a running water supply. Water is taken from a local river and there is no filtering system. Electricity is needed for essential facilities in the town, like a welding shop. The nearest phone is 10km away, so modern "necessities" like the Internet are not even a remote possibility, and a computer is worse than useless.

"What is required is the building up of a long-lasting cooperation with the people themselves," says Moolenaar. "Ask what they want, and provide what they need. This means a bit of everything. Some food when there is a drought. Medical help and immunisation where it is needed. Practical training and skills for the real world. And a lot more besides," which inevitably will include computers in their rightful place.

Moolenaar tells the story of his search to find ways to improve the water supply in Kibaale: "The knowledge wasn't there. If I had been able to search the Internet to find similar projects, I would have been able to do a lot more in less time. We can expect to see Free Software helping to provide affordable, reliable access to information, once the enabling infrastructure is in place." But without a mains plug and a phone line, and the necessary training and skills, a computer is no more than a very expensive ornament.

More information on work that is being done to span the digital divide can be found at bridges.org.

Richard Hillesley

This article first appeared in LinuxUser magazine