At the IADB seminar on ICT in the classroom, I asked Nicholas Negroponte why not sell the XO laptop -- at or near cost -- to anyone who wanted one? This gets beyond the hassle of having to convince bureaucrats of the value of the laptop without running pilot programs and delaying the eventual adoption. It (hopefully) creates some side markets in support, software development for non-educational uses of the laptop like rural healthcare, and could enable educational uses without going through the schools themselves, even.
Granted, there are some concerns. OLPC has thus far maintained a clarity of focus by working towards their mission of universal access, rather than having to worry (like Intel and Microsoft) about capturing an emerging market. Working at the ministry level potentially could reduce the transaction costs of each "deal," but more importantly, it guarantees some level of equitable distribution of the laptops, ensuring not just those with money will get access.
And this equity is important - for a education project within a school; you have to have all the students with laptops, or you by definition don't have a 1:1 program and you don't have a good shared computing setup either. Lack of computer saturation also opens it up to higher risk for theft.
This work is licensed under Creative Commons Attribution-Noncommercial-Share Alike 3.0 US.