I am one of the people who has made an attempt to configure computers for use in rural areas of the developing world and have tried to install a system that would work over a long period of time while owned and operated by a village. I did not succeed, but I learned a great deal (the effort has been continued by people who inherited the project.
The most important lesson was the fact that when one sets up a computer there, one must necessarily create an infrastructure which will support the system. This is a much harder job than merely designing and building a piece of equipment, but without it the equipment will fall into disuse and disrepair, and your intentions will fall into disrepute. Word will get around that you have proven that "it can't be done".
My major point of contention with the OLPC project deals with how they are, or rather are not, handling this infrastructure problem. The idea was introduced as a revolutionary piece of hardware that would embody educational advancement and raise the level of education amongst the children who used it.
Great blessings would apparently flow from the mere possession of the equipment, so no effort would be necessary to provide infrastructure supporting its use. The closest thing to a stated goal was that children would "learn learning" without the intercession of adults under a vaguely-defined theory called "constructionism".
When questions were raised, in many cases answers were avoided. The most egregious tactic was to say "it's an education project, not a laptop project" whenever questions were raised about the elements of the laptop, and to say "we're only making a laptop" when questions came up about the educational goals and methods of the project.
The sales effort was directed at top decision makers in the governments of candidate countries, and no one to this day knows what was promised to whom, outside the inner circle of the project.
High governmental officials, being political creatures to the core, live and breathe perceptions - the perceptions of their peers and of their constituents.
While anyone, in my opinion, who has been involved in education or in computer system development would not accept the idea that mere possession of a laptop will by itself bring about an educational revolution, it is surprising that this blatant example of technological determinism gains acceptance amongst the officials.
This is because they are not expert in either area (education or technology) and they know that neither are their peers nor their constituencies. Since the idea of advanced technology (the computer, in this case) as magic totem with wonderful powers has been spread throughout the industrial age, these politicians are keenly aware that they must not be caught on the wrong side of this perception.
OLPC's basic sales pitch can be: "If you don't show that you're a supporter of wonderful new technology you will give your enemies a weapon to use against you - now, are you with us or against us?"
As I pointed out in November 2005, the big problem with the OLPC project was the assumption that there is no need to work with the parents, the community leaders, and the working educators to develop the infrastructure needed to support the intended operation of the laptop.
Without such infrastructure, I contend, the project will not work. Other results (some quite distressing to contemplate) are possible, but not the ones presented as the intended results.
It appears that a few things are changing in OLPC's approach. Constructionism is no longer the watchword - it has been purged from the definition of the project. While the price has increased (not too surprisingly) the project is compensating by dropping the minimum order quantities, very much like a merchant with slightly overpriced stock.
Competition of a sort is emerging from Intel and the voice of Microsoft has been heard, softly as it always is in the beginning, forcing OLPC to claim that Windows will run on the laptop. Of course, the port of Windows must necessarily be done by Microsoft, who will thereby control the perception of the laptop within the Windows usership.
Still, infrastructure is the neglected child of the lot, as one might expect given the immense complexity and variations of culture, societies and expectations within which the OLPC laptop must function. For OLPC to claim an exemption from the infrastructure problem by saying "we are only twelve people" is to concede that they have not done and have no intention of doing the necessary homework. But they do expect a high grade.
Returning to my opening statement, I fear that OLPC's performance without necessary infrastructure will poison the official perceptions of other efforts to spread the use of computer technology in developing countries.
"Computers for those people? No, OLPC has tried it and it didn't work".