i don't agree with a lot of rms' ideals. he's far too dogmatic, which has driven away a lot of would-be supporters. but i have to admire the guy anyway. it's very rare to find someone who believes in things as strongly as rms does, and is willing to go so far out of his way to practice what he preaches.
Stallman stated that "the plans aimed for low cost, enabling many children to use the machines, and free software, so they would have freedom while using them." From what I understood, the goal of the project was to get computers into the hands of children in third-world countries, so they could simply get experience with computers -- period. It was my belief that the fact that free software was used was simply out of convenience; since free software was also free (as in beer), it was another step in cutting the costs of the laptop.
When Microsoft stepped in and "offered up" its OS at the extremely cut costs, Negroponte decided that would be a more useful option; I don't think his intention is to force-feed proprietary software to children. Either he believes Windows is a better option, or he is an economically self-serving individual. Regardless, I sense that the ultimate purpose of this project was unclear to certain people.
On a less direct note, I can't stand Stallman's zealotry. It's great that he is able to stand so firm on one particular point, but it is also a terrible trait, considering that he isn't willing to even give a moments consideration to alternative options on a scope/level _outside_ his "cause."
"XO embodies the theories of constructionism first developed by MIT Media Lab Professor Seymour Papert in the 1960s, and later elaborated upon by Alan Kay, complemented by the principles articulated by Nicholas Negroponte in his book, Being Digital." --- http://laptop.org/vision/mission/
If you read more about constructionism at http://www.papert.org/articles/SituatingConstructionism.html, you will see that it is consonant with the freedom "to improve the program" and all the rest of the Free Software freedoms.
"Our commitment to software freedom gives children the opportunity to use their laptops on their own terms. [...] we do not want any ceiling imposed on those children who choose to modify their machines." --- http://www.laptop.org/en/laptop/software/
This is his curse: and many people reproach him this. His standard reply is, that if he would have ever given way to "compromises" by being less "dogmatic", GNU/Linux would have never happened: it would be infested with proprietary blobs all the way through giving control to others, rather than you, the proprietor, to the detriment of your freedom. Maybe I am wrong, but I tend to agree, if you compare the popularity of GNU/Linux vs. any other Unix, notably the FreeBSD and its variants.
Finally, I like the door knob analogy of Jeremy Allison:
to which e.g. Mark Shuttleworth would also subscribe: this is the only reason for the popularity and explosive growth of Ubuntu. The day this changes (like it did for say RedHat), it will also decline in popularity (completely different than material profit of stakeholders), something a few people already start seeing signs of.
rms has a lot of geek cred due to emacs, gcc, etc. if he'd spent that cred wisely, i think it's reasonable to assume he could have gotten a lot of other programmers to do his bidding, and gnu hurd would now be the kernel of choice, rather than linux.
Protest marches, animal testing lab attacks, anti-abortion marches, religious pilgrimages, cultish mass suicide so aliens can save you, contributing hours of work to open source software because you believe in it, charity/volunteer workers, the extra religious (nuns, monks, priests, rabbis, imams, and similar in every religion), people who started businesses because they believe in a product or lifestyle, people who blog to evangelise a lifestyle or dietary choice, bodybuilders and exercise fanatics, people who travel around the world to see one particular sports team play, people who travel to see The Sound of Music over and over again because they believe it's the best entertainment...
and it's not just his beliefs that make rms admirable. he started the gcc project, for example. i think you could make a pretty good case that gcc is the most important software project of all time.
As a bonus, Windows does not run on the MIPS processor; it never has.
Actually, it is promised to have hardware support for boosting Qemu x86.