OLPC crushed itself. FLOSS fans were clamouring to develop for it, seeing it as a way to bring emerging minds into the OSS mindset. It created a lot of buzz. Then OLPC jumped into bed with Microsoft, and the FLOSS geeks were driven away, with no similar group of developers to replace them. OLPC made a massive miscalculation, and never regained the buzz they had back then.
Microsoft had nothing to do with OLPC's downfall. It had everything to do with failed project management, and at the highest levels, failed leadership. Expectations (on national levels) were not managed correctly, strange omissions in implementation (such as even rudimentary technical support) crippled uptake, and in the end participating nations had to work with volunteers to get anything done. The volunteer effort continues.
Eventually the organization was mismanaged into oblivion; there is currently some while-label vulture in possession of the commercial brand, and the non-profit arm has progressed rapidly into irrelevance.
As bad as things got (and are), OLPC never shipped a single machine running Windows -- if there are any out there, the participating school program installed it. As for FLOSS fans, mostly they made a lot of noise on blogs without ever having actually been involved with the program. If they had, they would have known that OLPC didn't spend a dime on Microsoft support; MS was allowed to send engineers to develop support within OpenFirmware for loading Windows, should the recipient nation desire it.
Strange, I remember going to linux conferences, of which the OLPC talks were heavily attended; competitions to win one were heavily patronised; colleagues going for the G1G1 program. There was a ton of interest in this low-cost, open-source laptop project. I remember the "for the children" argument for why they were partnering with Microsoft. The FLOSS community was pretty vocal in asking please don't do this, but OLPC ignored them. The buzz died, and the media moved on as a result. It didn't matter that MS never shipped in great numbers[1]; the damage was done.
It was mismanaged, absolutely, and the proposed MS partnership wasn't the only nail in the coffin. But with my own eyes I've seen a lot more interest than "noise on blogs". With that silly move, OLPC lost a pool of enthusiastic, free developers and media buzz. It also caused the OLPC chief of software to resign.
Given that you disparage the FLOSS complaints as just "noise on blogs", it's clear you didn't read the complaints. It had little to do with dimes moving from OLPC to MS. The issue was OLPC letting MS use them as a conduit to train new users in 'the windows way' - that MS was effectively going to co-opt OLPC as a loss-leader program. How far would OLPC go with MS? Why bother developing when they're so intent on providing XP on a clearly underspecced machine for it? And how could a sluggish OS actually be a decision 'for the kids'? How else would OLPC break their previously loud promises? The license fees really had nothing to do with the FLOSS community largely abandoning interest in the project.
[1] You're wrong about Windows never shipping. Windows-only machines never shipped, but dual-boot Windows machines did.
And I remember being personally involved in the OLPC project, which was not driven by linux conferences. The G1G1 program was a massive albatross that generated almost nothing but bad PR. Nobody 'partnered with' Microsoft. The project never lacked developers (until it began to run aground due to the mismanagement).
I not only read the complaints, I made many of them myself. I disparaged noise on blogs as noise on blogs; the FLOSS community as a whole is fine but was not a major factor in the development of the project. This was not because they were unwilling; it was because the senior leadership made an executive decision not to engage with that community.
I understand that many people in the world confused OLPC's mission with that of free software. But the cold hard fact is that nobody really ever gave a shit about software licensing; the goal was to get computers to kids, and then try to set up a sustainable pedagogical practice around them. I'm not really interested in the completely tangential issue of whether you think Windows XP is sufficiently performant for this task -- the point is that FLOSS was never the point. It was an era where anyone's use of GNU software was considered to be some kind of philosophical statement on the validity of GNU, sure, but this was never the intention at the executive level. It was gratis software that was easy to customize. The decisions involved were primarily pragmatic. Any 'loud promises' you felt betrayed by are merely further examples of the wildly terrible expectations mismanagement perpetrated by the leadership.
I remember when the FLOSS community abandoned OLPC. It didn't make a damn bit of difference. But I'd be interested in which countries received Windows loadouts -- I don't recall ever seeing a single support issue regarding it, which makes me think you may be mistaken.
Uruguay. This blog article clearly states that every OLPC runs linux, and also clearly states that some shipped with windows. Not many - they went to pilot programs and they weren't taken up largely because of that non-performance you hand-wave away - but they were shipped.
> Nobody 'partnered with' Microsoft.
Getting super-cheap licenses and Microsoft to customise their OS for your hardware is 'partnering with' Microsoft.
> the point is that FLOSS was never the point
I knew this, as did a lot of the FLOSS advocates. The point was that MS was still seen as the Evil Empire at the time. The geeks were interested both in the low-cost laptop and the idea of spreading FLOSS instead of MS's stranglehold. And they were naturally excited about OLPC's strong promises. And when OLPC went back on their promises... as I say above, why bother continuing to work on the machine? What other promises will they break?
While you characterise the FLOSS argument by the more frothy fringe's statements of betrayal, what I saw was more "what's the point?". What's the point of doing work we believe in if they're not going to stick to their statements? Especially for the people who were far more interested in the machine than the kids - the concern that if OLPC switched to MS, hardware might be used that wasn't supported by linux.
> Any 'loud promises' you felt betrayed by are merely further examples of the wildly terrible expectations mismanagement perpetrated by the leadership.
What a strange argument. You chide me for having a particular opinion, and then state the same thing that I'm arguing. You should be a spin doctor.
By the way, I was a Windows guy then, working in Windows support until 2009. I wasn't a FLOSS advocate, though I am now. Colleagues were into FLOSS, and I went along to conferences to hang out with them. I personally didn't feel betrayed, I just thought it was a stupid thing to do, and I noticed the buzz in both the tech and mainstream media evaporate with the FLOSS movement's disillusionment (which is a better description of the overall feeling than 'betrayal', I think).
However, it appears that we both agree that from the GP's original comment, it wasn't Apple or Google that crushed OLPC - OLPC crushed itself.
I agree that OLPC crushed itself, but I think that had more to do with CIA associations (in the minds of possible client nations, even if not in reality) and with the kind of paternalistic "we know what is good for your children" attitude, combined with the "sign up the whole country or we won't sell you a single machine" avoidable problem with adoptability, and plenty of promising what they could never deliver (promises which, unfortunately, were repeated mouth to mouth throughout the FLOSS community). By the time they sold their users down the river to Microsoft, they were already years late and an enormous disappointment.
So, the $100 laptop is real. It just isn't very FLOSS-friendly or made by OLPC. Instead, the mainstream $100 laptop is an Android tablet or cellphone, or maybe an iPad or iPhone. It's Apple and Google's fault that those machines are so user-hostile and dangerous, but it's not their fault that OLPC failed to provide a user-friendly, safe alternative.
