Rumblings of discontent within the One Laptop per Child (OLPC) project erupted to the surface last week with the resignation of chief software architect, Walter Bender. A split has formed between those, like Bender, who see the project as primarily educational, and OLPC leader Nicholas Negroponte, who wants to push as many laptops into the hands of as many children as possible, even if it means getting into bed with Microsoft. Or so it’s been characterized.
But having looked at both sides, I think what it essentially boils down to is that geeks really, really don’t like Microsoft. There would be trouble within the organization if Negroponte merely batted his eyelashes towards Redmond, never mind actively blessing Microsoft’s development of a version of XP that will run on the XO. Sure, Microsoft is the devil, and of course it doesn’t want a generation of kids in the developing world exposed to Linux. But when did promoting Linux become one of the OLPC’s goals?
Negroponte isn’t abandoning his commitment to OLPC’s software platform, Sugar. As he wrote on the OLPC Community News listserv:
Sugar needs a wider basis, to run on more Linux platforms and to run under Windows. We have been engaged in discussions with Microsoft for several months, to explore a dual boot version of the XO. Some of you have seen what Microsoft developed on their own for the XO. It works well and now needs Sugar on top of it (so to speak).
Meanwhile, Walter Bender may have resigned from OLPC, but Sugar is released under the GNU General Public License (GPL), which allows anyone to do whatever they like with it, provided credit is given where due. That’s exactly what he intends to do, and not even necessarily on the XO:
Over time there are lots of things that will happen with Sugar in terms of efficiency and platform independence. Already, the community has by and large ported Sugar to Ubuntu [a form of Linux]. You can do an “apt-get Sugar” and if you’ve put the right repositories in place, you can install Sugar on Ubuntu. There is also a live CD that some folks in Austria put together, so you can run Sugar from your CD drive. There’s a lot of discussion on the developer forums about how to make all of that happen more efficiently.
The flip side — it’s been attributed to Steve Jobs, though I never heard him say it — is that if you really care about software you have to work on hardware. Certainly there are a lot of hooks from Sugar into the OLPC hardware, because the hardware itself is pretty special. But while I think that the things that OLPC has done with the hardware are necessary for successful deployment, I think that there are compromises that can be made with other hardware in the short term. So [you could get Sugar running on] other laptops and even other computers.
So, to summarize, one side holds the position that Sugar should be able to run on other platforms, whereas the other side asserts that Sugar should be able to run on other platforms. Not much of a split, it would seem, except that you can google your way through everything said and written by Bender and I doubt you’ll ever find him advocating Windows as one of the platforms on which Sugar should run. In fact, as he also wrote in Xconomy:
I think the culture around free software is actually a powerful culture for learning, and one of my goals from the very beginning of the project was to try to instill in the education industry some of the culture and technology and morals of the open source movement. I think it would greatly enhance the learning and education industry and their ability to engage teachers and students. So many different things are tied up in this concept. It’s both about freedom, and the freedom to be critical. Criticism of ideas is a powerful force in learning, and unleashing that is, I think, an important part of the OLPC mission.
So there you have it. It’s not about the XO, and it’s not about Sugar — both sides are committed to both those manifestations of the OLPC project. But even if you could get Sugar to run as an application suite on Windows, with its own program manager and activities journal, there are those in the OLPC project, and others who felt so strongly about these matters that they had to leave, who would still be opposed. These are the ones who believe that open source software in general is critical to the mission of education, and that closed source software, especially that of a convicted monopolist corporation like Microsoft, is not only undesirable, but detrimental to that mission.
So when Negroponte characterizes his opponents as “open-source fundamentalists,” he’s not entirely wrong. A less inflammatory term would be preferable, though — say, “people uncompromisingly committed to the empowerment of educators and students through the freedom which open-source software provides.” That’s a bit more of a mouthful, but it does sound much nicer, doesn’t it?
I laughed when I saw the title of your article, because the idea of an
“open source fundamentalist” is absurd.
The free software movement is based on a simple ethical idea: software
must respect the users’ freedom. A non-free program restricts you
and/or forbids sharing; to have freedom, you need to reject them and
move to free software. That is why we insist on free software at
every level. Whatever we don’t find, we develop.
People who value freedom and cooperation less than practical
convenience might be tempted to label anyone seriously committed to
freedom as a “fundamentalist” or a “zealot”. Those people generally
avoid ever mentioning “free software”; they may talk about “open
source”, which presents the issue in “sanitized” terms that avoid
rocking anyone’s ethical boat. To link “open source” with
“fundamentalist” is therefore quite ironic. See
http://www.gnu.org/philosophy/open-source-misses-the-point.html.
The software platform of the OLPC Project is the GNU/Linux operating
system. Basically this is the GNU system, which we have developed
since 1984 specifically to be free software (see http://gnu.org/gnu).
However, it also includes Linux, the kernel.
Using GNU/Linux is not the only way that the OLPC could respect the
users’ freedom. This can also be done using the BSD system. (With
either one, one must take care to avoid offering non-free add-ons.)
However, using a non-free system such as Windows (or MacOS) would mean
leading children under the dominion of a callous corporation.
That is not the way to make the world a better place.
From what I see Microsoft have created enough FUD around the project to make potential customers taking up the device, think twice, and potentially opt for the reactionary alternatives. This means the market and profits ( meeting initial costs ) are unlikely to be met for the project, which Mr Negroponte ( quite rightly) does not wish to sacrifice. It would appear that if market share cannot be maintained, the OLPC collapses from not reaching critical mass. But by being willing to negotiate with Microsoft his prospective customers will see that he is willing to negotiate, and not force them to make an either/or decission. If Microsoft can not then deliver comparable technology to what is already being achieved then Microsoft gets it in the face. If they can, then OLPC survives, and remains competative within the global market. A market that OLPC has generated and well done to them.
If, and that should really be a big IF, Microsoft can rebuild their system to provide the same functionality the Linux has already achieved then the consumers win. If they can’t Microsofts reputation slips further. Some times the tree has to bend in the wind to survive. Given the circumstances, I go with Negroponte on this one.
Microsoft will do what they always do, position themselves to be indispensable and leverage that position to push others out and their agenda trough.
Given their fraudulent, anti-competitive and open source hostile stance, that invariably means Suggar out, costs up, proprietary Patented Microsoft technology in.
The OLPC will become yet another market where they’ll be the Supreme dictator, and Negroponte hands it to them on a silver platter.
“But when did promoting Linux become one of the OLPC’s goals?”
It never was. But straight from the OLPC mission statement is: “XO embodies the theories of constructionism first developed by MIT Media Lab Professor Seymour Papert in the 1960s”
IMHO the close source proprietary software model is in direct opposition to many of the principles of constructionism.
The other issue here is that there is no legitimate reason for this change to commercial closed source operating system other than the fact the Microsoft must have gotten to Negroponte somehow. They either paid someone off, threatened or otherwise cajoled their way into this space. Where were they in the beginning? Where were they when all the groundwork was being done on this project? The only excuses Negroponte himself can come up with that I’ve seen are lame arguments regarding Adobe Flash compatibility. He cites problems accessing websites built on Flash 9 like Disney.com. “It’s an education project, not a laptop project.” That quote now rings hollow built on a software platform that the OLPC recipient doesn’t really own…just licenses from Microsoft.
It should be embarrassing for the Microsoft apologists by now considering their track record in playing nicely/competitively with others. To sit there and try to minimize this situation displays one of two things. Either a complete ignorance on the history of Microsoft or an intentional/malicious whitewash of propaganda.
This article deeply, deeply misses the point about OLPC and Sugar. The hackability of the device is a substantive and important part of the pedagogic mission — kids need to have the capacity to hack, break, fix, and rebuild the OS on these machines, otherwise they’re hardly half of what they were supposed to be. So if MS wants to release the source code to a stripped-down version of XP, and run sugar on top of it, that’d be great. but as far as anyone can tell the likelihood of that is very slim.
add to this the fact that building sugar on windows is currently a pipe dream — there are about 3 engineers working on the sugar interface right now, and all of them want to fix bugs and add new features, not port the interface over to another kernel — and this whole conversation is shown to be a bit of a tempest in a teapot, and yet another indication that Negroponte’s bombastic, self-important pronouncements on OLPC are hardly any help to the project.
One foot in the door Microsoft will do what they always do; They will figure out how to position themselves such that they can’t be kicked out, and then leverage their position to push trough their own agenda at all costs.
Given _their_ stance towards OpenSource, that’ll invariably mean their agenda is kicking out opensource from olpc and make it their consulting/proprietary software playground.
And when all is said and done, and the OLPC project is just yet another ultraportable that will be forever out of reach of most 3rd world children, and is beeing used to open up new markets for Microsoft, just remember who handed it to them on a gold platter, and squarely lay the Blame to Negroponte.
What the whole thing is about is that the technicians and volunteers that did most of the grunt work for the current XO made one set of technical choices, and Negroponte thinks he can just decreet a 180° turn from high on and they’ll follow his lead silently.
Hint #1: the technicians may be better placed than Negroponte to make technical choices (and make no mistake Linux and OLPC is a technical choice)
Hint #2: the people complaining now don’t *have* to work for OLPC. A lot of them are not paid and those who are could easily find a better paying job somewhere else.
Hint #3: in a risky endeavour like OLPC, where contributors are asked to make a lot of sacrifices, you *must* have a credible technical vision. It’s difficult enough to make an OLPC-like project succeed on sound technical bases, without them the risks are too high. It’s not that people are not in OLPC for the children, it’s just that they believe a MS OLPC will tank and is a waste of efforts.
So what it all boils down to, is that a large part of the current project was sold on one choice, and turning tables mid-project will just make those people leave. You can’t keep your cake and eat it too.
I sure hope for Negroponte he has budgeted replacements because he’d going to need them soon. Making a project work more like Microsoft also includes relying on people happy to work in a Microsoft-like environment and paying them enough they turn a blind eye to management mistakes.
Welcome to OLPC, the Dilbert edition.
You say that FOSS fundamentalists at OLPC would sooner leave than have *anything* to do with Microsoft, and yet Ivan Kristic (who did later leave) observed some of MS’s initial efforts to port XP to the XO with some cooperation from OLPC folks, and even blogged to calm fears about any impending MS takeover.
So there’s something more than a knee-jerk reaction to MS going on here.
Do we remember how Microsoft and Intel were bashing and criticizing OLPC? They claimed it will never work, it will benefit no one, cell phones is the way to go. etc…
Well what happened? Why is Microsoft so eager and optimistic about OLPC all the sudden?
well, my guts tell me that they have some ulterior motives. You might think they don’t want to missing on the opportunity of doping all the children with XP, I believe there is more to their enthusiasms that that. The are after the Suger OS. Specifically, they need it for embedded XP.
Linux is dominating embedded market and Microsoft can’t stand it. Losing to Linux is devastating and stealing the technology that is in Sugar would help it out tremendously.
Yes I know, Sugar is GPLed, but that doesn’t prevent Microsoft from finding convenient loopholes.
We shall see.
Well, there’s one party who probably does NOT keep Sugar their highest priority, and that’s Microsoft.
I completely agree that the split are not directly related to Sugar, but rather a worry (from my perspective a very relevant worry) that allowing Microsoft to inject their software in the machine will eventually push Sugar away.
Whether Sugar should be pushed away up is a topic on it’s own, but it’s not very strange for a group of people be upset with a decision jeopardising most of their work.
The thing is,
we all hate Microsoft. Sure, try to downplay that any way you want. Really, why would you even want your company associated with something that is such a trainwreck?
The thing is (and has been mentioned many times), if people can’t modify and improve XYZ item for the future then where will things go when the company that has XYZ item decides to stop making/supporting/interacting with them?
Do you really think in 50 years people will be running something with MS? Beyond my own personal distaste I can’t say I see MS surviving more than 5 or 7 years as long as litigation moves forward.
I doesn’t much matter how you phrase it. Your Open Source Advocate is my Open Source Zealot. It’s a matter of perspective. Fortunately, both groups can continue on with Sugar and we’ll see where it goes.