Looking beyond OLPC: The hardware is important, software is important — but there’s more. Photo CC Mike Lee, via Flickr.

Ah, the seasoned OS zealot. Never fear: no actual issues of substance will ever distract them from one-dimensional tirades about how their platform is best. And so, in the last week or so, you may have run across angry Free Software advocates railing against the inclusion of Windows on the OLPC ("One Laptop Per Child") XO laptop — or, in a really surreal turn, people waxing poetic about XP, like this commenter on the Win Supersite: "We get a world wide audience of children who will embrace XP and gain valuable lifetime skills."

All of this is a complete waste of time, not because the OS question is unimportant, but because it’s detracting from the more important question of education, which was supposed to be the point.

Part of why the OLPC mattered — and continues to matter — is it raises questions about what computers mean for learning. That’s a question we haven’t asked enough recently in the US, let alone across the planet. Whether Negroponte and the remaining OLPC project leaders have lost their way or not, that central question of computers and learning seems to be lost in the usual blog banter. Fortunately, it’s a discussion I think will survive after the immediate technologies have faded away.


Tam-Tam, the innovative music app ("activity") built for OLPC’s Sugar educational environment. Here’s why I think the connection between software and learning is getting lost in tired arguments about OS.

I was lucky enough to grow up in a generation that got some exceptional educational training on computers, ironically because I think a certain suspicion of them made people more rigorous about educating with computers instead of just teaching them for their own sake. Show of hands, Reagan-era kids: how many of you learned to program with LOGO ("turtle graphics")? How many of you got to use music software? How many got to work with HyperCard? How many of you then later saw an education that later shifted to basic skills in tools like PowerPoint, instead of understanding real connections to other fields, mathematics, and programming techniques?

Platform does matter — especially given that, currently, the use of Windows breaks the Sugar interface, the educational software written for the OLPC, and critical hardware support for mesh networking, e-book reading, and power management. Maybe Negroponte will keep his word and port those to Windows; there remains reason to believe he won’t.

The question of learning, though, has been lost. I do believe that free software could be powerful for education, but it should be as a means to an end — not an end in itself. It’s one thing to say the software is free, it’s open source — another thing to figure out what it is you’re teaching. Free software opens the doors to the classroom, but it’s only a first step. And, honestly, those questions are important enough that we should be asking them about Windows and Mac software, too, software on proprietary platforms. Getting hung up on the free software question seems to derail that discussion — and allow people to conveniently duck all the real work of developing the tools.


Python programming: you know, for kids.

The only really good analysis of the OLPC situation I’ve seen comes from Ivan Krstić, the head of security architecture for the OLPC before he (like so many recently) left the project:

Sic Transit Gloria Laptopi

Worth reading in full, but this for me is the bottom line:

But really, I digress. The point is that OLPC was supposed to be about learning, not free software. And the most upsetting part of the Windows announcement is not that it exposed the actual agendas of a number of project participants which had nothing to do with learning, but that Nicholas’ misdirection and sleight of hand were allowed to stand.

In fact, I quit when Nicholas told me — and not just me — that learning was never part of the mission. The mission was, in his mind, always getting as many laptops as possible out there; to say anything about learning would be presumptuous, and so he doesn’t want OLPC to have a software team, a hardware team, or a deployment team going forward.

Yeah, I’m not sure what that leaves either.

There are three key problems in one-to-one computer programs: choosing a suitable device, getting it to children, and using it to create sustainable learning and teaching experiences. They’re listed in order of exponentially increasing difficulty.

It’s disturbing stuff — but then again, I’m convinced that there are enough people who really do care about the deeper issues of learning that the issue will be alive — assuming the dunderheads in the blogosphere don’t let this disintegrate into a meaningless Linux vs. Windows debate.

Software and ideas could go well beyond just one piece of hardware — even carrying some of those hardware design principles to other devices, which arguably has begun to occur with the popularity of affordable laptops like the Asus Eee. That’s why I think some of the good news in all of this is former OLPC president of software and content Walter Bender founding the Sugar Labs Foundation. It suggests a future for the free and open-source learning software and unique "activity-based" interface on the XO, one that could work on other inexpensive laptops and your personal computer, too:

Sugar Labs

Sugar’s game-changing UI generated a lot of discussion — and often-warranted criticism. But one thing I noticed is that almost every review mentioned the music applications favorably. Music is one of the major draws of computing. And that’s not only for kids, but the adult reviewers, as well. If you think about how this can be built over time, music is a superb medium for talking about sound, physics, mathematics, aesthetics, time, and fundamental principles of communication, expression, and perception.

Music learning — and learning in general — also benefit from some of the other aspects of Sugar:

  • Focusing on activities: I really love this interface. Everything you do is based on "activities" — files and applications allow you to pick up an actual project where you left off and continue work, logging what you’re doing in a persistent journal. It feels fantastic for creative work, not just for "kids." I expect we could see this interface pop up in other places.
  • Teaching programming: Built-in apps teach Python coding, even to non-programmer children. It brings computing full circle to the days when PCs like the Apple II shipped with BASIC (incidentally, the product that launched Microsoft — otherwise Bill Gates would presumably still be an obscure college dropout). And the ability to code simple tools makes sure that computer users don’t hit walls with their ability to make the machine do what they want.
  • Free, open-source, easy development: Forget about the philosophical aspirations of the free software movement for a moment. The ability to easily extend a computer with free software, and to see lots of source code for what you’re using as an example, has practical benefits. One real-world result: Sugar can live far beyond the OLPC if that project goes away.

Sugar does appear to have a future independent from the OLPC. It’s already included with a couple of major Linux distributions. It’s relatively easy to install on your PC. Activities run on cross-platform, open Python, which could eventually bring their benefits to Mac and Windows — no specific hardware required. (Java is getting added, as well.) The music software is perhaps the deepest and richest, based on Csound as a synthesis engine. I’m also interested in the partly-finished port of the Java-based coding language Processing — or ways in which Processing itself could benefit from

Again, the execution in Sugar may not be perfect. But the point isn’t whether it’s perfect. It’s not the OLPC, or Sugar, or Linux, or even free software as an ends in themselves. It’s figuring out what’s essential to building better educational tools for computers — and that’s a far more interesting question.

Ironically, amidst all this controversy, an OLPC developer XO machine just arrived at my doorstep. So I’ll be working to code for it, and will share what I make and what I learn about the device. I’m also in touch with other music developers working on the XO. Whatever happens to the project, I think there’s plenty to be learned. Stay tuned.

  • deb

    as always, a cool headed thoughtful analysis, peter.

    i am not as up on the latest announcement regarding windows on the OLPC as i should be before i write this post (so will probably get decimated later in the comments) but the idea of actually learning something was a concept that i found lacking in much of the run up to the start of the OLPC project (it was my understanding that this project was aimed at the developing world).

    ivan's note about "using it to create sustainable learning and teaching experiences", and the lack of support for this aim, is what i was worried about all along. i don't think i was alone, either.

    being a fan of technology makes me really really want to embrace the OLPC project and think that it is a real and substantial gesture towards ending some of the disparity between the developed and developing world. but i can't help but feel, and feel newly justfied in feeling (given the growing awareness of our ongoing food and water crisis), that projects that for example create clean water locally are far, far, far more important. i can't quite shake the feeling that we are putting the cart before the horse.

  • Thanks for this great post. It helps us see the OLPC project back with a large angle.

  • I live in Uruguay, a country whose goverment has already introduced the OLPC project in many public schools and plans to have them all covered with OLPCs in one year time. We are a third wourld country, although we do better than most of our neighbours. Thats because we are really small, just 3 million. In this country almost 99% of the people uses PC (not Mac) and from that total around 98% uses cracked software, including Windows. Putting XP on the OLPCs may seem a commercial move from Microsoft, but in our case (and I'm sure in most cases) it is just helping children to deal with the most popular OS in the country, and it won't get MS no money. For sure. Only mid sized to big companies pay for their software here, and really few people own international credit cards, so there's no way to sell them none. You know what? Amazingly, IT is one of our most important exporting products. Lots of software, games and other stuff is being developed here, and it gives the country a really huge income. Ain't the world odd?

  • Heretical as it may be, I believe that computers hinder learning in childhood. There are a set of basic cognitive and psychomotor skills that children master as they grow, and a PC cannot teach those skills. As far as computers go, ask a kid why s/he likes a PC, and the answer will be "Games!"

    The idea that open source software will somehow result in a generation of kernel-tweaking super-programmer kids is bizarre – there are far more important concrete problems for a low income child in a developing nation to tackle than abstract computer skills. Feeding the family, food, shelter are pretty important. A computer can't help with those fundamental socioeconomic issues, especially when programming paradigms and technology itself are so ephemeral.

  • @cooptrol: Thanks for sharing what's going on there! Do you know if Uruguay is going to be part of the Microsoft pilot program?

    I can see the appeal of running Windows for other countries, but then I really do wonder about these other issues (power management, mesh networking support, e-reader support, Sugar software port from Linux to Windows) — and it seems like those are getting ignored in the Microsoft pilot.

    I'll also say, Linux in the US is less than 1% of what most people use (it's closer to 3-4% on CDM), but people still find it appealing on many of the lighter laptops.

  • @James: of course, that's part of the problem of referring generically to the world's "children" — who do you mean, exactly? (As in the Uruguay case, even less-wealthy nations now have a lot of laptops.) There are plenty of people living in disadvantaged situations (in the US included) who may well want the computer as a learning and communication tool, where it's not necessarily going to be an either/or case of "you don't need fresh water / rice / shelter, just have this laptop." But that's the challenge here, is precisely who/what.

    Also, I agree on the kernel hacking thing, but what we're talking about here is simple Python scripting. If someone's doing enough of that that they're writing code to the exclusion cognitive and psychomotor skills, then something's horribly wrong. (But I don't think we're talking children that young, anyway… again, specifics the OLPC communications have often left out.)

  • @PK: I agree that OLPC's target audience is somewhat cloudy – these machines would seem most appropriate for teenagers.

    Overall, I just can't help feeling that funds in a developing nation could be better spent on additional teachers or school infrastructure. After all, in ten years these things will be a pile of outdated broken plastic.

  • Lots of insight. I tend to agree with most of your points and I really hope discussions about tools for learning will continue long after the effects of the debacle have subsided.

    About music on Sugar. As you say, music is an efficient way to draw people into using tools and what's already available on the XO (Jean Piché's TamTam) can be the starting point of something quite interesting. Especially if it isn't imposed on children.

    There's a lot to be said about constructionist learning and music. Learning by doing, playful learning, experimenting with sounds…

    Not only through Orff, Dalcroze, and Samba schools. But through musicking tools. From HyperCard stacks, MetaSynth, MOD trackers, TurboSynth, Alchemy, and Music Mouse to Max/MSP, Bhajis Loops, Guitar Hero, GarageBand, MooCowMusic, Jam Sessions, ElectroPlankton, and Nanoloops.

    It wouldn't be completely absurd to make Sugar into some kind of cross-platform environment to develop cool music "toys" and other learning-friendly applications.

  • Sizzurp Sippa

    I is a good thing that rich white folk in the first world get to "raise questions about learning" with the tax money and children of people in poor countries.

    The XO is what happens when intellectuals from MIT and their favorite electronics manufacturers have more political power in 3rd world countries than the actual citizens of those countries. It is a perfect example of how people can be racist, even when they believe that they aren't racist.

    We get to manipulate poor countries to spend their money being guinea pigs for our educational pet theories. Amazing!

  • @Sizzup:

    Well, wait a minute. What I'm talking about is the software and the educational aspects, which don't involve any hardware investment whatsoever.

    The software's free. "Raising questions about learning" doesn't cost anything, either. Working on new tools for music education and shared educational knowledge doesn't cost anything. And my whole is precisely that no one's really adequately investigating these problems *in the US* — a country which has huge problems with poverty, hunger, illiteracy, and (disproportionately in some circles) unemployment of its own. Some run along racial lines, some don't. I'm not proposing music education as the solution to those problems, let alone music education with little Linux laptops; that'd be silly. But as a musician, I believe in making the tools I use accessible to other people. I don't presume what level of interest other people may have in them, but I'm sure going to share the stuff I care about.

    I have reservations about the way the *hardware* laptop purchases are being financed. And that's part of the point above from Ivan: just dumping lots of laptops on other countries sounds like a disaster.

    But no one's making any money off of Linux, Python, Sugar, and the largely volunteer-created apps for them. No one's forcing anyone to use them, either. If people decide to ignore them, there's no harm done.

  • Peter, I'm no t sure if we are in the MS pilot. Thanks for giving the opportunity to debate about these kind of topics

    James, I agree that money should be invested in teachers and infrastructure. That is prety much the case here. These year for instance lots of class-days have been lost already because of teacher strikes all over the country. They are rightfully demanding a better salary, which indeed is now very very low. And they also demand better buildings, bigger classrooms, more working material (other than pcs). This has always been the situation of education here and in the region. It seems a little bit odd that the government invests in the OLPC, yet no-one argues with that, cos I think the feeling here is that we are always behind when it comes to technology, and most people believe that that delay is what makes us underdeveloped as a nation. The poorest people in Uruguay, are indeed not a subculture, but a different one. They sometimes make lots of money classifying the garbage the middle class produces, but they spend that money differently according to their set of values and priorities. This priorities put the trash-classifying and/or good-stealing skills on top, way above going to school or having healthy food. You will see tin hut villages in the outskirts of the capital, filled with big TV sets, cable connections, even broadband internet, but very few children going to school or reading books.

  • lematt

    i think this OLPC project is pointless. why would a poor kid who is hungry and thirsty want to geek out some python ???!!!

    it's crazy. for me: this project is made by nerds who hide themselves behind the education and learning thing to realize their technological fantasies.

    by the way: learning with/on a computer is good … when you end up in science, business or working for the "computer nerds using" industry: IT, games, software,… if not: it most of the time makes you only a consumer, eating western imperialist culture.

  • M-.-n

    OMG. Fail.

    I never trusted negroponte. Great article Peter, as always.

  • Pingback: OLPC Linkfest « Disparate()

  • I have to disagree with the article. Linux vs. Windows is not a question you can't put aside easily and ignore. First with Windows the software is not free, Microsoft get 3$ for each laptop. Remember that Apple offered a special OSX version for free?

    And while it might be right that the kids don't get Kernel hackers but at least a lot of potential is there plus the extreme costumizability. I think you don't get the idea of open source. On Windows there are always barriers and obstacles, you are not allowed to do this, your computer is in danger etc. I am happy that I didn't grow up with Windows.

    Microsoft fight for sustaining mindshare, one big problem for them is when people recognize that they could do their stuff like browsing, writing a letter or a casual game not only with free open source but also on alternative operating systems.

    And I fear the day when Windows with its endless security problems is out there on millions laptops. With XP soon to be ceased don't expect them to do any support. Say hello to the next big botnet.

  • lematt

    "Overall, I just can’t help feeling that funds in a developing nation could be better spent on additional teachers or school infrastructure. After all, in ten years these things will be a pile of outdated broken plastic."

    i totally agree.

  • Pingback: OLPC Linkfest « Disparate()

  • Hey wait a minute! who told you the kids using OLPC are hungry or badly fed? they are humble working class children but thats it! they watch TV, they go to the cyber coffe shop and are very very aware of the global world. much more than most adults! I don't know if this is bad or good for them, but what I'm sure is that it is the way the world is today, and we must be able to deal with it. They are 1000 times more capable of adapting to ANY OS than us, the regular grown ups. Some opinions in this thread really show a typical 1st world ignorance of how a country like this actually works. I live here and I can clearly see that the OLPC is doing good to our children, without any discussion.

  • There are completely different economic questions here that we're conflating.

    1. Mass-purchased hardware: costs money. Even if the per-unit cost of the OLPC is less, I agree with commenters here that having a government buy up a whole bunch of XO laptops costs taxpayer money, and that it could be spent elsewhere. As for whether that's damaging or not, to me it would depend on the government buying the laptop and the specific case of that country, but I also have reservations. The obvious flipside of this, though, is that you can erase the divide between those with money to buy a laptop on their own and those who can't. And it's not a zero sum game, either; computer investment could be a part of investment in other essential infrastructure, like classroom facilities (which, again, are an issue right here in NYC — so this is not an elite US vs. the rest of the world situation; everyone has needs, the US just happens to have more means at its disposal).

    2. Cheaper commodity laptops: save money — for those who can afford buying them. We're seeing lower-power, lower-cost computers all over the planet. And, incidentally, you don't have to have an imaginary distinction about the "third world" or "developing world." Have you ever been to the rural US? Or poorer urban communities? Talked to Aboriginal communities in Australia? These are supposedly "first world" countries, and we face a lot of the same challenges. (ironically, the difference is that our "first world" governments have the money to spend on these problems and often don't.) It's a global question. And, for better or worse, electronics makers see this as an opportunity. It'll be up to consumers in these markets whether or not to buy them. I'm not necessarily saying this is better or worse than the OLPC model, but it's out there, and we'll see what the benefit of this is. But the decision, at least, is in the hands of the consumer rather than the government. (Railing against capitalism to me misses some of the real damage of the lending regime of the past decades. It's involved the so-called "developed" world funding massive socialist projects like dams and such, ultimately to the detriment of the people affected and the finances of the country "we" loaned the money to. I'm all for a discussion of economic ideology, but in this case it's just plain abuse by certain world powers, and in the long run has hurt everyone.)

    3. Open source software: costs nothing. So, as far as Sugar, Python, all the software that we're discussing, it's simply NOT stealing resources from, say, water programs. It's quite the opposite — open source software is replacing either pirated software, or expensive licenses. In other words, it's saving money. (I don't encourage throwing eggs at Microsoft, because I don't like getting things thrown at me, but this was the point of a Hungarian student angry about money spent on Microsoft licenses in the education system there.)

    To me, #3 is a no-brainer. It costs nothing. It's volunteer sweat going into these projects, and the reason is that the people involved with work on Linux, OpenOffice, open source software, etc. believe there's benefit for them and others. Anyway, since every country in the world has *some* computer infrastructure, saying that open source, you can't say the *software* side of this project is stealing money from other places.

    As far as whether the government-purchase model of the OLPC or the free-market model of the independent electronics companies is better, we'll see. But ironically, I think a lot of the "free-market" companies may go Linux and not Windows.

    Oh, and I didn't say the Linux vs. Windows question was unimportant — I said it was tragic to me that the discussion was OS and not education. If you follow the education argument, then there are terrific reasons to use Linux over Windows in this case (not least being, Windows appears to *break* the OLPC and make it harder for kids to use).

    But that's my whole point: the end should be learning. And speaking as a teacher and someone who works with a lot of teachers, that's not an academic question.

  • Also, I'd point out the Give 1 / Get 1 program — that's a case of people funding the OLPC in which no money comes from the receiving user's country. So there are opportunities and alternatives here; it'd be great to continue the spectrum of what's available, even with a healthy dose of skepticism.

  • To take a slight breather from all the economic and political logistics, I'd just like to heap praise on the OLPC for platforming something like TAM-TAM. We have two XO units in our research lab, and the TAM-TAM program offers great promise/salvation to digital music as I know it. Let me explain. I remember back when Photoshop was new, and a bunch of the rich kids who's parents could afford to buy/run it would brag about becoming "photoshop digital graphic aritsts." Well, over time, such image-manipulation software has both refined into more powerful tools (like photoshop suite), as well as idiot-proof tools for casual usage (like the Kodak EZ-photo programs that let my grandma enjoy removing red-eye from photos).

    Well, nowadays, it's digital music software that's on the cusp between pro-level and casual/toy apps, especially when it comes to youth appeal (…I should know, I teach high-schoolers about GarageBand and PureData).

    I yearn for the day when second- and third-world kids around the globe can use programs like TAM-TAM to digitize their musical inspiration and heritage. And the modern (urban) culture of "making beats" will no longer be such a hip and/or acculterated thing.

    (…wow, where'd this soap box come from?).