I think differently to this, and here's why.
- It's not just for the "third world".
- It will produce a generation of free software users.
- Development of software for resource-constrained devices is good for everyone.
Even in the affluent societies of the "first world" there are plenty of poor people. Most people who have thought about social equity and justice have come to the conclusion that the way to help people who are blighted by generational low income is through education. The problem with this is that education costs money. Access to information is expensive. The OLPC project, combined with the power of the Internet, drastically reduces this cost.
If OLPC machines are widespread in society, in five or ten years, a significant proportion of the people entering the workforce will be used to Free Software, and the whole ethos of cooperation and sharing in the software world. Using proprietary, closed software will seem, simply, weird to these people.
I work at a university in New Zealand. I own my own house and car. I have a big TV and stereo, I eat out a lot, ... in short: I am not poor. But I can't see the value in buying a new computer every year (or every three years). Neither can I see the value in investing in ASDL for $50 a month when I can have dial-up for $7 a month. So my home computer is old and slow (2.4GHz Athlon with 1GB Ram) and my Internet connection is slow. I can tell you that using GNOME software is a pain in the ass with such a setup. (Compared to using a one-year-old laptop at work with a high-speed connection.)
I am looking forward to the technologies that have been developed to slow, low memory, slow-and-intermittent network connected devices filtering down to the main GNU/Linux and GNOME stack. I think that this would help a significant proportion of GNOME users, i.e. those who don't have the luxury of modern setups.