Are cheap computers the answer?

A number of papers  are carrying articles about Nicholas Negroponte's $100 computer (on orders of one million or more!). While at the MIT Media Lab he started the Digital Nations project to enlist governments in developing countries to become paying partners with the lab. Judging from the web site, nothing new has been posted for two years. This piece is a good one because it raises the issues I don't think Negroponte and associates have considered: connectivity and cost of electricity--even if it is available. I say this because in one grass roots project in Uganda, men would take the batteries from their radios, thus preventing the women in the family from listening while they were away.  It seems it was a matter of control, of cost, and of ownership.

Nevertheless, a low cost laptop would be very popular. Negroponte uses the analogy of the French Minitel system where the state-designed terminals were very cheap and very popular. In urban areas they effectively replaced phone books.  There have been other schemes for cheap equipment: sending container loads of recycled PCs, but the recipients have vary opinions about the practicality of this.  Some welcome the machines; others find the effort at refurbishing, maintenance, and provision of software for different generations to be a poor use of labor--even at the rates in poor countries.  The Indian Simputer excited many people but it has never gained traction. (photo from the Pittsburgh Post-Gazette.)

Comments

[Marco Figueiredo]

Pushing individuals to own computers in remote areas is not the most efficient way to bridge the digital divide. Those who promote this model more often face challenges such as how to service the machines and train their users. These problems are easily solved in a pilot project with a limited number of users, but the solutions don't show scalability.

More than bridging the digital divide, we are constructing a New Society of Knowledge, where everyone in this planet is connectable to the Internet, if s/he desires. That is only when a computer becomes an essential tool to an individual, a community, a city or a country, when it is connect to the Internet, preferably in a broadband connection.

The current cost and feasibility of having everyone in this planet with a computer connected to the Internet is prohibitive. The shared access model is the optimum but less glamarous way to reach total digital inclusion and phase in to individual ownership. Telecenters, defined as places for public and shared access to information and communication technologies, are the most obvious way to pursue the closure of the digital divide.

More attention should be given by technology developers and manufacturers to the needs of telecenters both in software and hardware. The presence of a telecenter connected to the Internet via broadband enables simplified and low-cost devices to be reconfigured by an Application Server. Telecenters are not being treated as a potential market by software and hardware companies because of their current limited numbers. But the number of telecenters could multiply by orders of magnitude if better solutions are available to make them more sustainable. It is a market waiting to be serviced with innovative products.

Millions of cheap PCs with no training and no service is definetely not the solution to the digital divide. As one adds the costs of training and service, those PCs are no longer cheap and telecenters become more attractive a solution.

[Steve Cisler]

Marco, Thanks for your comments! I have worked on telecenter/community tech centers since the mid-90's.

The international deployment of telecenters has peaked and a lot of aid agencies are pulling away after some successes and many failures. Two exceptions: IDRC in Canada working with Microsoft community affairs to support a telecenter help network. This idea began with somos@telecentros, the Latin American network, and was co-opted by UNESCO. However, they did nothing with the idea, and now years later there will be quite a bit of money divided among regional networks and individual centers. Official rollout will be in Tunis at WSIS.

[Edward Cherlin]

Steve's observations make it clear that if cheap computers are the answer, you asked the wrong question. Computers can be part of a solution to poverty, but only if they are designed to meet the specific, stringent requirements of the poor, and only if they are combined with the other elements of a solution.

The defining characteristic of poverty is lack of disposable income. Further constraints that the poor in many areas operate under include

o lack of an electric grid
o lack of a phone system
o lack of knowledge
o lack of software in their languages

Lack of knowledge is what the computer is supposed to cure, but first you have to know how to use one. Second, you have to know English, or you have to have a lot of friends translating software and creating content for the Web in your local language.

Any proposal to help the poor without specific measures for circumventing these constraints is a fantasy. In principle it can be done using solar power to recharge batteries, using wireless to connect to the Internet, and using microcredit to finance placement of the equipment plus the training to use the computer to make the money to pay back the loan.

The microbanks have not yet put such a training program together. The Grameen Foundation USA and Grameen Communications both have village computing projects, but I haven't heard of results from either. For now, that means that none of this is happening.

It is not that Simputers, for example, are too expensive. It is not that Simputers, for example, are too expensive. They cost more than the the cell phones that Grameen and other microbanks place successfully in villages all over the world, but not vastly more. (And they use rechargeable batteries and wireless, and support some of the target languages.)

It is not that there is a fundamental difficulty in training people to use computers to make money. The ITC e-choupal program is lifting millions of farmers in India out of extreme poverty by giving them free access to computers and offering to buy their crops at world prices. But you see that it is not enough just to provide the computers.

In fact, it will also be necessary to train villagers to service computers, to create local content, and to program computers. Meaning that we have to make the computers operate in their languages.

Given computers in villages, and the initial methods of making money using them, we could obviously design programs in health, education, appropriate technology, sustainable agriculture, and so on and on. But we would still have the obstacle of language. We need to provide software and content in as many languages as possible. This can only be done by speakers of the languages. The rest of us can provide financial, technical, and moral support, and license our content to them at a price they can afford--free.

There are a number of distributions of Linux in languages of Africa and Asia not supported by commercial software, and more being created. The Free Software in use in developing countries is the functional equivalent of billions of dollars of commercial software, although the Free Software movement doesn't get to claim that we have donated billions of dollars worth of software.

What we need to do together (the techies, the NGOs, and the poor, among others) is to create a program that integrates these elements and more, and to test it and make it work for the poor. Then we can begin to talk about taking it around the world and answering some questions in the way that counts, by making it happen. Many of my friends are working on various components and on the way to bring them together. Some have successful development projects in quite poor countries that are ripe for the addition of computers. They could do much of the R&D for the rest of what we need. You are all welcome to join in.

My own efforts right now, apart from working on the overall vision and plan, are going into e-commerce with Africa and Asia and promoting Linux development in more languages. I have some other ideas, for example telemedicine in the villages once they have computers of some sort. But that is for another time.

[cheap computers]

Yes they are. It seems it was a matter of control, of cost, and of ownership.