zunguzungu, true, open-source advocates and techies in general are, characteristically, overoptimistic, oversimplifying, underprepared, and egotistical. They get stuff done anyway, but frequently not what was intended… It’s true that the hardware is very top-down, in both production and administration, but the software is trying to be to-each-their-own and also bubble-up and peer-to-peer and community-knowledge. Like cellphones, but more powerful. What’s your take on cellphones, in the places you’ve taught?

It will work better if it isn’t all in English, yo. So I went and looked at the Project Gutenberg Children’s Bookshelf, and of course it’s preponderously North American (strongest non-Anglo influences: French and maybe Navajo). The OLPC Library-bundles are less so, but clearly could use more ‘content’. I’m discussing this on Distributed Proofreading (which gets books ready for Project Gutenberg); I think there would be enthusiasm there for any work that would go out to under-booked schools, on the OLPC or from the Internet Bookmobile. Or printed by the bale for a gov’t, for that matter. We don’t know what the schools want or whether there are public-domain copies of it, though. I assume the OLPC local projects might, so I have a thread there too. Do you happen to know if there are US or UK collections of classic foreign works that are inadequately available in schools? …No reason you particularly should know, but it would simplify the process a lot.

I would have thought that people—even children—who reliably skip a meal and are picky about its social acceptability probably expect a better meal later that day. Maybe not! In any case, that didn’t help either you or them at the time. I join in decrying hunger-based inattentiveness.

Clew,
Well, my experience is limited, and I hope you’re right that they’re not as naive as I think. But the amount and widespreadness of willful and arrogant ignorance that I saw (in my limited experience) was enough to make me tend to think the worst of such projects rather than extend the benefit of the doubt. But hopefully you’re right. It just seems to me that so many of the basic foundational premises are flawed that whether or not they address smaller problems is kind of beside the point. And I’m not sure where these places are that have food security but lack education, and it seems to me like it might be a faulty formulation; I had kids whose parents could afford food in the abstract sense, but if a school for budgetary reasons doesn’t feed the kids between the hours of 8 and 3 (and the kids are at school during that period of time), the question of whether the parent’s have food security doesn’t quite get at the heart of the issue. It might be, in this case, that my experience doesn’t match the reality on the ground where they’ve decided to insert themselves, but I tend to doubt it.

As for the issue of computers versus paper, being on a blog doesn’t compel one to have a particular position on the medium in any particular case (blogs can do what they do here for particular reasons, but tend to work more or less well in other places, for reasons that are just as locally specific). Were the infrastructure of a different nature than it is, computers might well be an excellent way to distribute textbooks to kids. In my judgment, though, it doesn’t seem like this is likely to be the case (nor does it seem to me that these people are asking the right questions). Technology, as such, is not the issue; the problem is that solving problems with technology usually seems to reinforce systems of economic dependence, and when the planning process seems to exclude those considerations from the very beginning, I find myself getting very weary.

re: “boys,” well, I had been thinking about how the boys would play soccer instead of eating lunch, which led them (I suspected) to be more difficult to handle in the afternoon (and many of them told me that they found the uji porridge unappetizing; they thought it was funny that I liked it, since it was sort of the cheapest food available). The girls, socialized into gender roles as they were, did not play soccer and the administrators, being just as gendered, put the girls in charge of doling the stuff out, so it was sort of a female space for consumption. Girls ate more than boys, in short. So, yes, it was creepy in a way, but maybe not the way you were thinking. That said, I’m glad you’ve given me the opportunity to go on the record as denouncing hunger-based inattentiveness in any and all genders: I do so.

Telling the developing world to use the OLPC scheme rather than the O/S the rest of the world use is pushing them into a ghetto.

Okay, but the OS the world really uses is the Internet, and the OLPC has a browser, so this is a wash.

zunguzungu, many of your minor points have been tackled by the OLPC —I won’t be surprised if their solutions fail, but they aren’t quite as naive as you seem to think.

On the other hand, your major point is a good one, the history of failed tech-in-education and tech-in-the-poorer-countries visionary projects. I worry that the OLPC project could be useful but will fail for lack of follow-through, because too many people hoped it would be instantly obviously successful. The development project I go back to took decades, with a bunch of people spending years at a time, and it certainly started with clean water; but they’re now working on Internet access for schools and local government and farmers, and that’s pretty hard but is useful when it works.

I want to put in a word for the idea of every kid getting a slate/book/emulated-physics-lab/math-lab to noodle with. The ‘good old days’ of one shared terminal had a huge problem with letting only the right kind of nerd play; I was the wrong kind, was bullied off the terminals and machine shop, and learned much more as soon as I could do so on my own.

It’s true that noodling about is a particular learning style, but the classroom server speaking to the laptops should be able to support several more directed styles of learning in the way that blackboards and tablets do. zunguzungu, you clearly aren’t impressed by the potential of computers – in general – as small-scale printing presses and newspapers. Mm, okay, but that’s an odd stance to be taking on a blog.

There are small-scale travelling printing presses, of course, which AFAIK are touchy to maintain and depend on all the translation, production, telecoms, and power that the OLPC servers-and-laptops do. If the government already needs to write or commission textbooks, and the OLPC is already there, then one government copy can fairly readily be transmitted to all the laptops; not so with books. (And isn’t the production of local textbooks a useful job in places with inadequate markets for skilled labor? A Sen sort of bootstrap? Failing that, this is up the Distributed Proofreaders’ alley, multiple languages and all.)

Bit creepy, your first reference to maize porridge, when you specify that you don’t want the boys to be hungry. In general I agree with you—especially as there are so many places in the US that should feed all their schoolchildren—but I think the OLPC is aiming for places with food security, just not a lot of education.

Using technology as a magic bullet in a country that lacks the infrastructure to support and maintain that technology (on a necessarily massive scale, no less) would produce a structural dependence, and would make that country’s education system wholly an offshoot of the West’s benevolence. Do you disagree?

No, you’re quite convincing. I haven’t thought very hard about this, I just thought it was a neat idea.

While some developed-world technologies clearly do have useful applications in poor countries—drugs and medical technologies, agricultural methods, transportation, cell phones—it’s also true that post-WWII experience is littered with ill conceived technology driven projects.

I also happen to think that in the case of computers in the class room, this applies to rich-world experience as well: Lots of PCs have been installed in schools without properly thinking through what they’re really good for.

The OLPC project may be similarly flawed. But at least it has gone to great lengths to radically rethink the laptop based on a Ford-like vision (“a car for the multitudes”) rather than just pushing some off-the-shelf thingy. I’d also hope the national education authorities will have the good sense to address your concerns and device good ways to introduce laptops to schools and integrate them in useful ways. After all, it’s their responsibility.

What is sometimes tiresome in discussions of this and similar projects is the frequent appearance of an almost Pavlovian cynicism which I find superficial and unconstructive.

It goes without saying that your comments don’t fall under that heading. Your skepticism is reasoned and passionate and you bring personal experience to bear.

When I was learning computing in 1979 there was only one computer for the whole school – and most shools waited several years to get one. So during break time there were six people crowded round the machine watching the person at the keyboard. It isn’t a bad way to learn by any means.

But no, Negropointy-head must know best. He invented the intelligent fridge.

One computer per village has the potential to change the world. But only if those computers are being supported by a community of content providers. Telling the developing world to use the OLPC scheme rather than the O/S the rest of the world use is pushing them into a ghetto.

Let OLPC die, its already done its work by pushing the prices down.

It’s amazing how quickly an economy of massive scale is presumed, and I think that’s really telling. First of all, is this project really going to provide computers for millions and millions of children on an ongoing basis? If not, then what’s the point of introducing a few computers, which will then lack appropriate reading software and support structure because they won’t be enough by themselves to cover the population? What infrastructure is in place to get textbooks digitized? Repair them? Replace them? If every student doesn’t have access to these computers, then there won’t be incentive to produce the kind of textbooks that you need, and you’ll end up with out of date or inappropriate textbooks being the only thing available, for the handful of students that actually have computers. And this is what I think will happen. Plus, if the kids are supposed to be carrying their incredibly valuable computers home every day, what are you going to do when they show up to school without one? The model presumes that each kid will have his own every day, but if ten of fifty kids come to class without their computers (and the reasons why this might happen are legion), what the heck is a teacher supposed to do? The whole scheme has to presume that the computers will be available in these kinds of numbers, but it seems absolutely inevitable that this presumption will be frustrated and produce all sorts of new problems.

In any case, the fact that the planners are thinking in such numbers is indicative of the market logic they’re using: this isn’t about charity or selfless development, this is about the creation of a market for western technology. They’re ultimately wrong, I think, to imagine that the third world will be a good market for this kind of project, but in order to imagine that such a project would be useful, you have to believe that it would be able to get millions and millions of these computers in the hands of kids, and you have to be wildly optimistic to think that private charitable donations are going to make that happen (or that a charitable organization is going to be able to make it work). The only place I can see that kind of money coming from is massive USAID or World Bank loans that African nations will further impoverish themselves by, as they’ve been doing with other kinds of semi-useless western technology for decades now. And they will, coincidentally, be buying those computers from Westerners who will expect a Western salary for doing the work, even if they’re doing it in the “non-profit” setting.

in comment 13, “bemused” also wrote: “there is room for many paths to aid people in the developing world. If you feel that a contribution to food aid or clean water development is better, then that is what you should give.”

The overwhelming number of development projects out there have not, in fact, aided people in the developing world; it’s still “developing” and shows no sign of reaching the present perfect tense because all the conventional wisdom on how to produce development has been monumentally unsuccessful. There are many ways to address this fact (and I imagine that you and I will have different views on how to do that), but unless you actually address the incredibly broad and rich history of failure in only the last few decades, you can’t possibly have any credibility with me. The lack of humility that development workers still manage to have, in the face of the abysmal track record of development projects, absolutely infuriates me, and you can use that as an excuse to be “bemused” at my “ignorance,” if you’d like. But when you speak of a “world-changing approach” and “changing lives,” I think of how incredibly difficult it was to provide enough food for children that couldn’t concentrate without it; when you gesture towards massive economies of scale, I think of how bad the track record is of even meeting the most humble of benchmarks, precisely because Western interventions get planned in an abstract, airy ignorance of how local conditions will radically shape what it is possible to do. James Ferguson’s first book, for example, is still an excellent illustration of how this works, and very little seems to have changed since then, so far as I can tell. Before Western philanthropists should aspire to change the world on a grand scale, they should demonstrate that they can change the world on the most basic scale, something that (so far) the massive preponderance of evidence suggests it is not capable of.

By the way “most places have access to the internet and computers – even in Africa” is rather like saying “most people have access to Rolex watches”.

Actually it isn’t, and that was my point. Outsider’s views of what is and isn’t available in much of the third world tend to be fairly inaccurate. In most of the world’s urban centres (which is where most people live) internet cafes are available and reasonably cheap when one factors out the infrastructure costs. Even in rural areas, email is often available, and actually used fairly widely as an easy and fairly cheap form of communication. In Africa the main problems are language and the backbone to the rest of the world being very expensive and slow (the OLPC doesn’t address either of these problems, and currently the writing/reading software is hopeless if the child doesn’t speak English, or a major European language). In addition electricity is extremely expensive, which is why in one of the trials in Africa carried out by the education ministry (possibly Nigeria, though that doesn’t sound right) the OLPC was abandoned when they realised that the long term running costs would far exceed the cost of the device and exhaust the education budget. Total Cost of Ownership is what has killed other ambitious computing education projects in the third world (for instance, Egypt).

The OLPC doesn’t really address any of these problems, Negroponte being rather dismissive of infrastructure problems (though I think some of the other team realise that these are critical now). The handcrank is of course a joke (you only have to do the calculations on the back on an envelope to see that), which is probably why Trevor Bayliss wasn’t very interested in getting involved, so you have the problem of paying for electricity still (which in much of Africa will kill its use at home, and possibly in schools), access to the internet will be limited by the infrastructure. Now if the project had addressed these problems first, that might have led to some interesting solutions.

“Adults may have access via costly internet cafes in cities, but I believe you are blowing smoke if you are asserting that most children have such access.”

Most children don’t have such access (though given what most children actually use the internet for, I can’t say that this seems terribly important – is the ability to plagarise wikipedia really that educationally useful?), however given the speed and cost of internet access in Africa, I can’t imagine the OLPC doing much to change that.

I’m really not at all worried about that—developing countries will get computer systems and internet access just as they’ve gotten other modern electronics (radios, TVs, DVD players, cell phones)—which is to say, through low cost versions of standard technology imported and sold by local entrepreneurs. Look at the forest of satellite dishes sprouting everywhere in developing countries and be reassured.

bemused: So much ignorance, so little time. First, as to #9:”computers and net access will be low cost versions of the devices we all use rather than custom MIT-designed technology for poor people.” The XO-1 is a low cost device (it costs about $180 US at the moment). It is the most environmentally sound….

Yes, I’m aware of the technology. To the extent that the battery, screen, or mesh web technologies are truly revolutionary, they will be incorporated into devices for all of us and sold to the developing world in that form. For example, are the sunlight-readable-as-black-and-white displays really a better display technology? Is that what people really want? There are many millions of cell phones and digital cameras with display screens that are often used in daylight conditions—would people like them better with XO screen technology? We’ll see (but I doubt it).

you don’t need to do research to know that the money spent on a computer would be better spent on things like maize porridge

are hard to argue with, of course. It’s like the Lomborg argument against the Kyoto treaty. In fact it works quite well against most any initiative with opportunity costs.

Which is not to say opportunity costs are always irrelevant. But in this case I suspect they are.

The third world is littered with failed top down technology projects, of which the OLPC is an example (which was the point I was making). The initial requirements for the device came from Nicholas Negroponte’s assumptions about what was needed for a third world computer. Wherever those assumptions came from, they didn’t arise from any familiarity with the third world, or technology development there. The “need” for such a device came from Seymour Papert’s (lovely) ideas about education. Unfortunately, Papert has never had much success in proving those ideas, and they didn’t have a huge affect on education either. I have no objection to further research in this area, nor would I have to Papert testing them in the third world. I do see it as highly problematic if you’re going to try to sell them to impoverished third world countries as the solution to their problems.

The production and demand for the product was largely driven by Negroponte lobbying the UN and government heads. Is there demand on the ground for such devices? Nobody knows and nobody much seems to have bothered to find out. Was there serious testing of the prototype? Any exploration of other designs, other approaches? Apparently not. It has lots of innovative ideas, like the wireless networking [nice idea], but has it been tested in realistic conditions? Do they even know what realistic conditions are (and how much variety there is likely to be between an African village, an urban slum, etc, etc)? Lots of ideas like that which seem like a good idea in the west, turn out not to work out so well in the field. Good to know before the things have been sold.

And I still don’t understand why nobody thinks that very poor families won’t sell the things. $50 is a fortune in many of the places that they’re talking about placing these things. Hell, it could pay for a child’s school fees for a few years – or food.

OLPC has worked closeley with education faculties and technology groups in Latin America and elsewhere in developing the technology.

Well no. They’ve worked closely on developing some of the educational software. This is very different. Has anyone run trials in villages, urban slums? Asked teachers what they think? Carried out projects with teachers? These are all standard practice, and yet…

This is another project that is in the line of SUCCESSFUL technology projects, because it is small, distributed, and maintainable in country.

These are hardly primary criteria – its supposed to be an educational device. How do any of these make it more effective educationally? They’re infrastructure. Its like saying a school will be successful, because the building has a roof. I’d agree that such things are important, but…
As for the criteria. Well maintainable I agree with. The other two may be true, but they’re hardly iron cast rules. A small computer is very easy to steal, for example. Distributed tends not to work very well when you have small communities separated by considerable distance.

In addition it bears the hallmarks of successful projects done in the US, namely by the association with the education ministries and universities in-country to provide integration with the local curriculum standards and practices.

As well as the hallmarks of failed projects carried out in the US. All projects of this scale will involved these things for what should be obvious reasons.

As for the “little evidence of effectiveness” perhaps that is because it has not been possible to try on a large scale. The only large school laptop experiment in the US of which I am aware is in Maine, and it was carried out solely in middle schools.

Little evidence that widespread use of computers in US schools has had any affect on education. Evidence on laptops is mixed, but broadly similar. The Maine study could only find evidence of an affect on writing (though they give the usual bullshit excuses for why no other affects were found on any of the other tests), due to the ease of revising (this putting it broadly in line with other work looking at word processing). However, given that there was no control group its impossible to say how much of that was due to laptops (the fact that not all groups used laptops is indicative, but there could be other reasons for why they didn’t use laptops, which would independently affect their performance on the writing tests – and of course there may always be a group that doesn’t want to use laptops, thus limiting their effectiveness). In addition, obviously this study only shows the affect on particular types of teaching (those practiced in Maine), and there may be non-technology based forms of teaching that, while less trendy, achieve equal or better effects. The US school system is fairly unimpressive globally, so I wouldn’t expect to find best practices there and interestingly countries with better educational systems often use far less technology.
The other link is largely a puff piece, with most of the benefits being about student and parent attitudes. These may, or may not, have an affect on actual education. These may, or may not, last. They really don’t tell us anything about what will happen elsewhere in the world, though.

the magnitude of the change they introduced is dwarfed by the impact of OLPC in places that have no access to the Internet or computers at all.

Most places have access to the internet and computers – even in Africa. Experiences in America don’t really tell us much about what will happen in places that are culturally and linguistically very different.