One Laptop Per Child Talks

At the MIT Museum, Walter Bender, President, Software and Content Development, One Laptop per Child, Senior Research Scientist (on leave), MIT Media Lab spoke about One Laptop per Child: Revolutionizing How the World's Children Engage in Learning.

In an informal conversation with an MIT Museum audience, Walter Bender describes the mission and progress of the One Laptop per Child (OLPC) venture. The brainchild of Nicholas Negroponte and the MIT Media Lab, this enterprise aims to put low-cost ($100 or less!) laptops into the hands of a billion plus children in the developing world. T

The full audio program transcript is below.

---

John Durant: Friends, welcome to the--ooh--welcome to the MIT Museum. I sound as if I’m coming from miles away! Great to be back for the first Soap Box that we’re doing in 2007, and to see you all here this evening, and to welcome all of the people who are taking part in this event online. We’re pleased to have those folks with us as well.

I just want to make a couple of announcements before we get into the subject of tonight’s Soap Box. The first one is, if you have a cell phone or PDA switched on, could you please switch it off, or at least turn the sound off? The second thing is that we’re trying to do a kind of ongoing survey of who’s coming to these Soap Box events, where you come from, what brought you here and so on. It’s part of some data we’re collecting, not least because of the partnership we have with WGBH-Boston over a collaboration we’ll be doing with them later on in the year, so you may find survey forms around. Please take a couple of minutes, if you would, and fill them in; we’d be grateful.

We’re using new webcasting tonight. The webcast should be a little faster, a little smoother, for the people who are online. I hope that’s true, and we hope very soon that when we get to the questioning and discussion stage of the evening that we’ll be in a position to take comments, e-mail comments from people online as well, so that we can extend the conversation that Soap Box is all about as widely as possible.

Now, we’ve organized this Soap Box, I have to confess, slightly at the last minute, and the reason we did was that we could get a hold of Walter Bender, on my right, who travels a lot, as does his colleague Nicholas Negroponte, not least, I think, because they’re busy now, implementing one of the most exciting projects that’s been associated with MIT for some time. And that project is sometimes known as “The $100 Laptop,” sometimes known as “One Laptop Per Child.” I’ll leave Walter to say what the official name of the project is. That’s the project that has taken you, I think, all over the place, because this is a truly global ambition that we’re going to hear more about shortly.

Walter was a founding member and former executive director of the Media Lab; he’s president of software development for this particular project. I’m going to let him say anything else he wants to say about himself. He’s going to introduce this project, then we’ll go into the brainstorming, where you get a chance to define the questions you’d like to pursue with him in discussion, and we’ll take it from there. But, Walter, welcome to Soap Box.

Walter Bender: Thank you!

[applause]

[2:58]

Walter: So, I’m going to try to keep a fairly brief introduction to the project, and have most of the time spent in discussion, hopefully getting some good feedback from this group and those who are out on the webcast. I started working at MIT back in the mid ’70s. There used to be a research group on Mass. Ave. called the Architecture Machine Group, and even then, it was pretty clear that what computers were for was interpersonal communication. They were things to think with; they were things to learn with. They were engaging... When I think back to when I did the real learning of my life, it wasn’t sitting in a classroom, it wasn’t reading a book: It was debugging a software program, it was trying to make something work, it was building something, it was trying to solve a problem. And it’s that sort of spirit of learning that we’re trying to capture in the Laptop Project and make available to the world’s children.

So, One Laptop Per Child is a... We’re a nonprofit company that spun out of MIT just about a year ago now, and our mission is to develop a low-cost laptop that we can get into the hands of children in the developing world. We think it’s going to give them an opportunity to learn. And I’ll go into, maybe, in the discussion, a little bit more about why we think that. If we could have the next slide... Somehow we missed a bunch of things in between... [directs projector to proper slide]

So this is who we’re developing the laptop for: There are one billion school-age children in the developing world. A lot of these kids’ school might be under a tree; they’re lucky if the teacher shows up. There’s just not much there, and so we are certainly supportive of efforts to build more schools, get more teachers into those schools, but we think that that’s sort of a “treading water” process, and we’re trying to do something that can do a little bit of end-run around the status quo and move a little bit quicker to reaching more children, and we think that the connected laptop is a part of that solution. Next...

So this is where we’re launching our project. The countries in green on the map are countries where we’re going to be actually starting next month doing some small-scale experiments with the next generation of the laptop. This is the B1 machine; B2 machines are going to Argentina, Uruguay, Brazil, Nigeria, Libya, Rwanda, which is not highlighted on the map yet, Pakistan, Thailand, and a couple of other places. The countries that are in red are countries where we’re probably going to be doing a larger-scale deployment sometime this calendar year. Countries in orange on the map are where we’ve been contacted by a minister of education or higher within the federal government asking us to do a laptop program. Countries in yellow we’ve been asked by somebody, not necessarily at the federal level. So, for example, 17 governors in the United States have asked us about doing laptop programs, so the United States is only yellow on the map. Gray is countries where we haven’t been asked to do a laptop program. So, Denmark is gray, and that’s why Greenland is also gray. [laughter] So there’s a lot of interest in getting this kind of technology in the hands of children. Next...

Hmm. How about the next one, then? Here we go. All right, so the laptop... Antonio Battro is somebody who we work with. He’s a brain scientist who studies the teaching brain and how teaching actually changes your brain. [He] did some work with Jonas Salk years ago, and Jonas Salk actually was interested in education, not just immunology, and made some interesting analogies between immunology and education. The first is that a vaccine is not a cure; a vaccine is an agency of change. And the laptop is not a cure; a laptop, in our mind, is an agency of change and an agency of opportunity. The other analogy is that a vaccine, unless it’s used broadly and to scale, is ineffective. And the laptop program, unless it’s done broadly and to scale, will also be ineffective.

[7:58]

Now, since the early 1960s at MIT, we’ve been experimenting with technology and children and learning. Seymour Papert’s work, his collaborations with Bolt, Beranek and Newman back then, were the beginnings of that sort of work at MIT. At the Media Lab we had a program with the Hennigan School in Jamaica Plain, a five-year program, which was a one-to-one computing program in the mid ’80s. Nicholas Negroponte and Seymour Papert did a computing program in Senegal in the early 1980s. There’s a long history of working with technology and children. The thing that’s different now is the opportunity to go to scale, and scale really does matter.

And the reason why scale matters is, you get people’s attention, and it becomes part of the culture. And by way of getting people’s attention, we have in the laptop a display which I’ll talk a little bit more about in a few minutes. We went to display manufacturers and said we wanted a display for a product, for a laptop, and we said we didn’t want a big display, we didn’t want a high-resolution display, we didn’t want a high-quality display. We said we wanted an inexpensive display that could have imperfections, and every single manufacturer we went to said, “That’s not what we do. We’re not interested.” And then when we said, “Oh, by the way, in Year One, we want 10 million of these,” all of a sudden, we got their attention.

So scale really does matter in terms of getting the educational industry actually to focus on a new set of problems. If it’s yet another pilot, a little program somewhere, they’re not going to pay attention. Just this afternoon, I had visitors from two different publishing houses, educational publishing houses, that came to us to say, “How can we get involved?” They wouldn’t have come unless it was to scale. Scale really does matter. But also, scale matters in terms of the culture around the laptop, what the children and teachers do with the laptop. Next, please...

So, just a little bit about how you would design a laptop for children: One thing you can do is you can try to make a cheap laptop. And certainly in our market, the laptop has to be inexpensive. But it has to be a little bit more than that. Next, please...

So for one thing, one of the things that a laptop has to do for children is it has to work where children are, and children are indoors or outdoors--sometimes school is outdoors. So the laptop display has to work outdoors. This is the first laptop that was designed to work in sunlight. So our display, you can see the difference between a typical laptop display and a laptop that was designed to work outdoors. [gesturing to slide] So that’s Scott Soong, who’s our display manufacturer, and Mary Lou Jepsen, MIT alumnus, who is our CTO and designed the display. Next slide...

Power. Power is a concern, as well. There’s a one-to-one laptop program in Dorchester at a brand-new middle school. They just built the school, and they have to schedule the classes around time for the kids to recharge their laptops, because even in a brand-new school in a wealthy country, they don’t have power to every desk in every room. I mean, that’s absurd. Absurd that they would, or absurd that they’d need it. So power is a real consideration. In places we’re going, and our focus is on LDCs, for the most part, there is--actually, I was in Abuja, which is the capital of Nigeria, the model city of Nigeria, visiting schools, and only one of six schools I went to had power at all, one outlet for the whole school. So there’s just no way that you can...

You’ve got to design a laptop that’s low-power. Kids aren’t going to have power at home, in many situations. So it’s got to be a long battery life, and in fact we’ve designed it so that you can hand-power the laptop. The power crank in the original model was on-board; it’s moved off-board onto the power strip so that the thing you plug into the wall also is either a crank or a salad spinner. And our goal--we think we can achieve this--is one minute of cranking for 10 minutes of reading. A typical laptop is anywhere from 20 to 40 watts; this laptop is two watts. So power is a real consideration. Next slide...

[12:50]

The third thing is connectivity, and connectivity not just to get to the Internet, but connectivity to make learning be a social process. And it turns out that learning is a social process: We learn from each other much more than we learn from books or from having a funnel attached to the side of our head. And we wanted to encourage that social process in the laptop design, so the laptop out of the box forms a mesh network, 802.11s is a new, emerging standard with the--and actually, a lot of the original work in 802.11s was done at MIT. We’re working with the Roofnet folks and others here at MIT on this project. The idea is that the laptops, without any additional infrastructure, all talk to each other. So every child in a village or in a school, they talk to each other, their teachers can all talk to each other.

They share a school server to get out onto the Internet. It turns out it makes the economics work in terms of the Internet access, even where Internet access is very expensive, because they can share it, and a lot of the traffic turns out to be local, anyway. So one of the things we had to do was, we had to change the way certain programs behaved. So for example, typically with chat, you might go to AOL or Google or Yahoo or wherever for your chat, so you go out to a server and you come back to connect to a kid next door. That’s inefficient; we can’t afford that kind of inefficiency. So there are service discovery protocols--in the Linux world, it’s a called Avahi; Rendezvous in Mac--and so a lot of the traffic actually stays local, a lot of sharing stays local. Next...

The laptop is all being designed open-source. In fact, I had a meeting with Richard Stallman this afternoon. He’s *almost* comfortable using this laptop. [laughter] There’s one little module that’s not free, and we’re working on it. It even has a Linux BIOS. But the reason why open source is important to us is not because we don’t want to pay money for software--although we certainly don’t want to have to pay money for software--but more important is the epistemological reason: We want the laptop to be transparent. In fact, on the laptop there’s a key that looks like a little gear--that’s a View Source key. So no matter what you’re doing, you can actually pop right into a debugger and see the underlying mechanisms by which the machine works. So that level of discovery, that engagement in the debugging process is again where we think a lot of learning happens, and that’s why we want to expose that in the laptop design. Next...

I want to talk just a little bit about how children are going to use the laptop. Next slide...

There are really three things that we’re working with, three things that are fundamental to be human, and it’s fundamental to being human whether you live in the First World or the Third World, whether you’re an adult or a child: We all learn, and we all teach. That’s fundamental to being human. We’re expressive; to be expressive is a human attribute. And we’re social. So everything we’re doing in the laptop exploits those three things. So it’s all about social expression, and it’s all about teaching and learning from each other. Next...

[16:35]

So, what we’ve done is, we’re putting into the laptop a number of tools for exploring, but we’re also putting into the laptop a number of tools for expressing. And the exploring tools are just sort of the usual gang of suspects: There’s a Web browser... There’s actually a Web server; eBook reader; chat and VOIP and e-mail; multimedia, so it’ll play movies, and there’ll be plenty of games. In fact, there’s a fun site with a couple of thousand games for Linux that we’ll probably get onto the laptop.

On the other side, there’s tools for expression, and probably the most radical thing that we’re doing with the laptop is, the eBook reader is a wiki. And the reason why the eBook reader is a wiki is because I want to raise a generation of kids who think that books are fungible, who think that every single page in every single book can support a discussion and commentary, and that commentary can get shared with other people. So we’re building that in as a fundamental attribute of Book in the laptop.

There are tools for graphical expression on the laptop, there’s a whole suite of very sophisticated musical tools in the laptop, and everything we’re designing we’re trying to design with a very low floor--and no ceiling. So our design philosophy is... So for example, Music, the most primitive musical toy we have in the laptop is a busy box, just instruments that you can click on and make sounds, and you can play the keyboard with that instrument and you can run it as a drum machine. You can record your voice and turn that into an instrument.

But from there, you go from the busy box to a sequencer, and you can begin to compose music, and it’s Garage Band, and each laptop on the mesh can be a different voice in that composition, and the kids can play music together and compose music together. But then from there you can go right into a synthesizer. And you can manipulate a synthesizer to create new sounds and new instruments, and fold those back into the busy box, fold those back into the sequencer. So we’ve got this whole gamut.

And then you hit the View Source key, and then you’re at the See Sounds, to see sounds in interpreted language for programming music, written by Barry Vercoe of MIT. So you’ve got this very simple door into exploring music, into as sophisticated a synthesizer as you can get anywhere. So that’s sort of the attitude we’re taking around the laptop.

It has an eBook mode, and... [gesturing to slide] now, this is just the difference between being a consumer of content and being a creator of content. We know there’s lots of tools for consumption; we’re interested in getting the kids to be creative, to be expressive, and that’s where the real learning happens. And that might be it. Oh, one more.

One of the ideas that Seymour Papert had for the laptop that we’ve incorporated--we’ve got five different programming environments programmed into the laptop, and one of them is a Logo environment. We’ve got a built-in mic, but you can also plug in a mic. The mic also doubles as an analog data port, so you can turn the laptop into an oscilloscope. On the left is a wooden spoon with a little photo diode on the end that Barry Vercoe used to conduct music with the laptop, just by plugging it into the--so he conducted a roomful of laptops using the analog port. On the right is a project, sort of in the spirit of the laptop, sort of predating the laptop, that David Cavallo, another colleague, did in Curitiba in Brazil.

I don’t know what they call it in Brazil, but up here they call the videogame Dance Dance Revolution. I don’t know how many of you have ever played Dance Dance Revolution. So what David did was not have the kids play Dance Dance Revolution but said, OK, you want Dance Dance Revolution? Build one! And so they did. So that’s a Dance Dance Revolution machine game that the kids built themselves. And that’s the kind of learning that we think is going to characterize--or at least the opportunity that the laptop will afford. Yeah, next... OK, now we’re done. So that’s my 20-minutes spiel. Thank you.

[applause]

[21:07]

John: OK, for those of you who’ve not been to Soap Box before, this is the point where we go over to you, so we’re going to ask you to spontaneously assemble into smallish discussion groups for about 10 minutes, maybe six, eight, 10 people in a circle, whatever feels comfortable to you where you happen to be. We want you to brainstorm what you’ve heard, and come up with preferably as small a number of maximally interesting issues for discussion that you can. There are, I think, tablet PCs... Oh, we’re back to cards, I’m sorry...

Walter: We have laptops...

John: We were in tablet PC mode for a week or two; tonight, we’re back to cards. Anyway, fill in these things, get them back to me or to Beryl or to John as soon as you can. You can use these laptops if Walter thinks you can, and we will then pop these up on the screen and reassemble in plenary for the main discussion of the evening. So, go for it. Thank you.

I should admit to something: Walter doesn’t know this, but we had one of the earlier generations--earliest, I think--of these green laptops on display at a grad student evening a few weeks ago, and I said to the person who came from the One Laptop Per Child project then that the museum, the MIT Museum, is extremely interested in eventually acquiring, you know, preferably, prototype number one of this exciting project for our collection. As you can see all around you, we’ve got some of these great breakthroughs and some of the significant steps of MIT research and innovation in this museum, and this looks like one we would really like, so that’s a sales pitch to Walter on behalf of the MIT Museum. [laughter]

But you’d expect me to do that really, wouldn’t you? This is a very exciting project, as I think is obvious, and listening to some of the conversations, it’s clear that it’s starting all kinds of questions and comments from you. John Bejeur[sp], my multitalented colleague sitting over there, had been pumping these questions into his laptop as fast as he could, and then he crashed. [laughter] So, he’s doing it all again, and some of these are now going up already. What we’re going to do is spend a couple of minutes reviewing some of the issues that have come up in your discussions, then we’ll see if we can group then together in similar clusters, and have as productive a discussion as we can.

So we’ve got questions like, how is connectivity to the Internet accomplished in an area with little power, with no Wi-Fi, presumably, and lots of other limitations, how does that work? What’s the end-of-life plan for 10 million-plus laptops, I mean, what are the issues to do with recycling and reuse when you go into what Walter describes to us is a very high-volume project like this?

One person whose question got lost when the computer crashed was a wonderful one from I think a 13-year-old who’s in the audience, or was: Why was the color green chosen, and are they any other colors, and could I choose a different one? [laughter] So there’s an interesting question. I’m interested myself in some of the design issues in this; it’s a very distinctive-looking and -feeling machine.

What about safety and security in urban marginal areas of poor countries? Who will ultimately pay for all these computers? How do you deal with corrupt governments? I did tell you to watch out--can I take one home with me? You see? As these computers are introduced to children living in poverty, have you considered that they might be considered of value to others, leading to theft and a black market? What happens when you $100 or $150 or whatever it is, your computer goes missing? Lots of issues here.

One person, I don’t know whether we’ve got that question, John, was asking other design-related issues. How did you, as it were, decide which facilities to put in? You said a lot about the philosophy, Walter, in your presentation, but was that philosophy something that you simply as a group colleague kind of came up with, or was it based on work with kids using computers? I mean, what’s the basis for these choices to go for this, rather than that, which you’ve obviously done? Can we start with some of those issues about how you homed in on the distinctive features that this computer has, and we could maybe, John, put some of those questions... because there are some of those... The poor old guy, he’s shuffling through cards as I speak. And then we can move on to some of these questions about how they’re going to be used in practice in the developing world.

Walter: Sure.

John: So, what was the research base for, the decisions you made about what to go for and what not?

Walter: Well as I said, technology and children is not a new thing, it’s not something we just woke up one morning and said, hey, let’s do this. It’s something we’ve been living and breathing for 40 years.

John: Yup.

Walter: And so, there’s a lot of experience, things that work, things that don’t work, that went into the design of this. I should also say that the design is not finished; this is a Beta 1 machine; we’re about to have Beta 2, which has a number of design improvements. The Beta 2 machines are going to be in the hands of kids, and we’re going to learn a lot from that. The design is going to change yet again; there’s going to be a Beta 3, and then there’s going to be the preproduction machine and the production machine, and we already have the Gen 1.5 machine lined up. So the design is an iterative process, and we’re expecting to learn a lot, both from the hardware perspective and the software perspective.

John: And when you talk about these Beta 1, 2, 3 machines, in what kinds of numbers do you imagine these early versions being produced?

Walter: Well we’re actually making a lot more beta machines than is typical in a manufacturing process, and it’s driving our manufacturer a little crazy, but we have a lot to learn, and we want to get as broad a distribution as we can, especially because we’re starting off in so many different countries and so many different cultures. Last summer, we built our A test machines, which were just the motherboards, and we built about 500 of those and got those to developers. Most of the developers were in the developed world for the A test machines. The B1 machines, these we built 875 of, and the bulk of those machines, except for the ones that the Quanta manufacturer kept for doing drop tests--in fact, actually, one of the things I want to do is, I want to do a green building drop with the video camera on. [laughter] We will do that at one point.

John: Not now. [laughs]

Walter: Not now. But these have been going out mostly--most of those 875 B1 machines went to Brazil, Argentina, Pakistan and the like, Thailand... The B2 machines, most of those... We’re building 2,500 B2 machines, and those are going to go mostly to kids in the developing world. B3, I’m not quite sure what the numbers are going to be yet...

John: Right.

Walter: ...or what that distributions going to be. So typically, we’re about an order of magnitude more on these...

John: On everything.

Walter: On everything.

John: Right. OK. And are they coming in different versions for different continents and different countries, or are they one-size-fits-all?

Walter: Well, one size, because kids are kids, but if you notice, actually, I brought the Nigerian machine and the Libyan machine with me tonight. So the only difference in these machines is the keyboard. So this keyboard for Nigeria has an augmented Latin character set because they use Latin characters in Nigeria, but they’ve got extra characters which they use for supporting the various major languages in Nigeria. For the Libya keyboard, I have Latin and Arabic characters, so you can switch back and forth to Arabic and Latin. The Pakistani keyboard, right now we’ve got an Urdu one; we’re going to need more than just Urdu in Pakistan. There’s an ñ on the keyboard for Argentina; there’s a ç for the keyboard for Brazil, you know...

John: I heard somebody come up and ask, and I don’t know if they put it into a written question, whether you’d considered making this the opportunity to leap beyond QWERTY, because QWERTY notoriously is an arrangement of keys which was devised in the days of the mechanical typewriter which has persisted right through the digital revolution, and given the volume, could you potentially have devised a more ergonomic arrangement of the letters?

Walter: Well, you know, there’s a little bit of urban myth to what you’re saying...

John: Oh, probably, go on.

Walter: ...in that a lot of people think that QWERTY was designed to slow people down; it wasn’t. QWERTY was designed to prevent typewriters from jamming.

John: Yes.

Walter: There wasn’t necessarily a correlation between speed and jamming. It more had to do with the arrangement of the letters to prevent certain combinations from hitting each other at a particular time.

John: Right.

Walter: Now, to my knowledge--and I might be wrong on this, but I’ve looked into it a fair amount--the difference between, and there are a lot of keyboard schemes out there, the differences between them from the perspective of typing speed are pretty marginal. Now on the other hand, there are some ergonomic things I wish we could have done [that] our form factor wasn’t allowing us to do, maybe things we could do later on. There’s some interesting split keyboards I think that are better for you in terms of repetitive stress, things like that.

John: Right.

Walter: But I think the key arrangement, in terms of QWERTY vs. Dvorak or whatever...

John: ...wasn’t a big deal.

Walter: ...is not a significant enough thing to bother with. On the other hand, you know, this is running Linux, X Windows; if you can to, you can go to the user/share/x11/xkb/sims and use your own key map. Change the Xorg.config file to whatever key map you want, and you can do that, so...

John: OK.

Walter: ...we’re not excluding that, by any means.

John: Right.

[31:39]

John: OK, does anyone want to pick up on some of these basic issues of design and design philosophy of the computer before we move on to other things? Yes, please. Use a mic when you speak, please, otherwise other people can’t hear you.

Man 1: Thank you. What about the keyboard for teachers? It’s empowering teachers that you can reach...

Walter: Yeah, our project should be called “One Laptop Per Child and Teacher,” but that’s just a little bit too much...

Man 1: What about the ergonomic design?

Walter: Yeah, for one thing, there are three USB ports on every laptop, so if a teacher wants to plug a bigger, larger keyboard or a different keyboard in, they can do that. And in fact, a lot of people do do that for the developers. But again, we made the keyboard about as big as we can with this particular industrial design, and I think the next generation, the keyboard, if I can, will be a little bit bigger. But you know, the other thing is that, boy, you look at what people type on for their SMSing and things like that... People can type on pretty much everything. I want to make it as best we can, but we also have a few constraints in terms of size and cost.

Woman 1: Kids’ hands are smaller.

Walter: Well, some kids’ hands are smaller, and some kids’ hands are bigger, but I’ve been using this quite a bit now as my main computer: I do my e-mail on it, I use it for word processing and browsing the Web... You get used to it. You know, it’s not great, but you get used to it.

Woman 2: The keys could be a little bit higher, I think.

Walter: The other thing is that there are vast improvements in terms of the feel of the keyboard between the B1 keyboard and the B2. The rubber membrane we put in because we want to be robust--dustproof in the Libyan desert; you want to be able to carry the laptop home in a monsoon. But it takes a little bit of tuning to get a rubber membrane keyboard just right, so...

John: OK, so you haven’t given up on that design criterion; you’ve just improved the design.

Walter: Right.

John: OK. [pointing into crowd] Please, one there.

Woman 3: Yeah, I was wondering about kids with disabilities, land mine victims or children who have been involved in bombings or war or other victims of, you know, farming accidents... Is there some sort of voice prompt or activation for that?

Walter: Well, in Linux, there is an accessibility suite that we’ll be supporting in the laptop, but there’s actually a lot of work to do in that space, just in general with computing. So anybody who’s got good ideas, actually tomorrow I’ve got a meeting with somebody who works in that space who’s blind, for example, but we need a lot of work. The industry needs a lot more work in that space. Yeah.

John: [pointing] Please, back there and then down here.

Man 2: Is there a target age of child you’re trying to reach initially? Is there a ceiling or a floor?

Walter: We’re sort of looking at 6-16, and that’s sort of a typical school-age child in the developing world. A lot of kids don’t get past fourth or fifth grade, but we’re hoping to sort of reverse that trend, to get more kids to stay in school longer.

John: There’s one here, and then one over there. Please. Could you just wait for the mic? I know you’re close, but you’re not so close to other people.

Man 3: This seems very empowering. I wonder who might be threatened by it, and who would be paying for it?

John: Good questions. Who might be threatened by this, and who would be paying for it? The question of who might be threatened by this is related to another, slightly cheekier question that was asked, but I don’t know if John’s got it up somewhere, which was, is this at risk of being overtaken by the iPhone, which you will have seen some discussion of over the last week, I know. But that’s a little bit of a similar question: What else might be out there that might conceivably fill some of these needs?

Walter: Well, let me back up a little bit and try to give a little more context to the question. I work for a nonprofit called “One Laptop Per Child”; our mission is “one laptop per child.” I don’t get a commission for selling laptops; I don’t have any stock options. Actually, I have the same stock options I had at MIT. [laughter]

John: As bad as that, huh?

Walter: Yeah, so my goal is to get laptops into the hands of kids. If some other company comes along and builds a better laptop at a better price, anything I can do to help that happen, I’m going to do. So I don’t think of kids as a market; I think of kids as a mission. Now there are certainly some ripples in the marketplace and within industry because of our setting a target where the price is a lot lower, but more importantly, our setting a target where...

See, the laptop industry hasn’t been obeying Moore’s Law, in the sense that if you look at the price of laptops, it should have gone down exponentially, but it hasn’t. It’s sort of plateaued, and the reason for that--until actually about a year and a half ago when we announced One Laptop Per Child--the reason is because rather than lowering the price, they add features. And what we did, and this is actually the answer to another one of the questions, is we went through and said, “What are the features that kids learning need?” Well, there’s a lot of things in laptops they don’t need.

In fact, there’s a lot of things in my laptop that I never use at all. We got rid of all that stuff, and we just focused on tools of expression. There are a couple of other things that we did, in terms of the design, so for example, with the exception of this hinge, there are no moving parts. There’s no hard disk to crash. If the backlight in the display dies, it works with ambient light, with some light. There are no internal connectors to break. The motherboard and the display are sandwiched together. So we’ve just eliminated three of the most common points of failure in laptops. So we’ve made a much more robust laptop. So we’re doing a lot of things, not because we’re trying to compete with other players in industry; we’re doing a lot of things where we’re trying to build the right device for the problem we’re trying to solve.

Now in terms of the laptop-vs.-telephone question which comes up quite a bit, telephones are great; cell phones have done a lot to expand communication and give people tools of communication--keep the momentum behind that. This does the same, but it does something different as well. I don’t know about you, but I... Form factor actually does matter and certain things have a certain form factor for a reason. So I think reading on a little display versus reading on a larger display is a different experience. And the creative process, the expressive process, I think, if you’re trying to do that through the handheld form factor, is very different than when you’ve got something like this. So we think that, again, because of the problem we’re trying to solve, it really has to be a laptop, not a phone.

[39:11]

John: Thank you. Henrietta.

Henrietta Davis: I’m Henrietta Davis. I’m a city councilor in Cambridge, and I’ve been working on the project to bridge the digital divide in Cambridge between the haves and the have-nots. And every meeting we go to, somebody says, “What about those $100 laptops they have down the street at MIT?” So my question to you, really, is, what role do you see for these laptops in bridging the digital divide in this country, where there is a lot of other competing technology? How would this help kids who don’t have a computer at home? Is this the thing we should be looking at, or we be looking at other solutions that do different kinds of things?

Walter: Well, here’s the thing. I think that this is on an interesting trajectory that would be useful to any kid learning anywhere. On the other hand, in the short term, I’m supply-constrained. I can only make so many laptops this year, and so then I have to ask a question: Where is the greatest need, and where is there opportunity or choice, and where is there no choice? So I don’t know what the... I’m just going to pull a number out of the air, but I think it’s roughly, that the average amount that’s spent per child--actually, you probably know the number for Cambridge--but annual spent for learning per child in the United States is on the order of $5-7,000 a year. More here, OK.

So if I take a $1,000 laptop or a $1,500 laptop and amortize it across five years--we’re designing this for a five-year lifetime--that’s $200, $300 a year. Two or three hundred dollars on top of that five, seven [thousand]-more figures... It’s money, but it’s not that much. In Guatemala, they spend $200 a year per kid, total. So where am I going to focus my efforts in terms of the limited supply of these things in the short term? It’s Guatemala, not here.

Henrietta: I was really asking, if these were available, would these be good enough to bridge the digital divide in this country?

John: OK.

Walter: I’m not quite sure I understand what the question is, but I think this laptop is good enough to do virtually anything a kid learning wants to do. As I was saying earlier, it’s got a low floor and no ceiling, and it’s a launch pad for any number of different things. So I don’t see any inadequacy in here, in the developed world or the developing world.

John: OK. Now you’ve already answered one of... I’m looking back at some of the other questions that you’ve been raising; I don’t want to ignore all these, because you put some work into this. Laptop life expectancy we’ve heard--you’re designing for five years.

Woman 4: Can I ask a related question?

John: To that particular issue? Please.

[pause while questioner obtains working mic]

Woman 4: I’d like to ask politically incorrect question of social responsibility.

Walter: OK.

Woman 4: As you create those laptops, and the children in the developing world are learning, they’re going to create more and more competition for our own children. That goes sort of against the human nature in evolution. Every single time in history when we went against human nature and evolution, it didn’t end well, an example being Communism.

Walter: Well...

John: Did everybody hear that? Yeah, OK. I think the... Sorry, did some people not hear that? OK, not. I think the gist of it was, are we going against the natural competitiveness of people by, as it were, subsidizing, and therefore making more effective, young people in other countries’... Will this all end badly, like some other experiments in social engineering? Is that a fair summary?

Walter: OK, so first of all, let me challenge a couple of assertions you made in your question. Subsidizing--I’m not subsidizing anything. I’m not buying laptops for other people. These laptops are getting bought by these governments for their children, OK, so subsidy is not an issue here. Now, I think it would be great if some of the haves help out some of the not-haves in that regard, but I think the world has got a lot of problems, and I think those problems spill over into this country, as well as remain in the places of origin.

One can just take a look at what’s going on in the Middle East right now, and the cost to this country of that involvement. Now, you can argue politically whether you think we should be in Iraq or not be in Iraq, or whether we should have gone into Iraq or not, but we are in Iraq, and it’s had tremendous costs by many different measures. Personally, I think that this sort of investment is going to pay dividends in terms of not having to deal with some of those kinds of problems down the road.

So I think that if you only measure through the narrow constraint or metric of competition, then maybe you have a point, but I’m trying to take a broader look at many more different measures in terms of what are we trying to accomplish on this planet, and what are we trying to accomplish for our children?

[45:10]

John: OK. I have a comment here, and then one over there. Please.

Man 4: I wonder if, in a relatively short time, the laptop may not be passé, and the keyboard will be replaced by something that uses voice, and the monitor is replaced by something that’s more like eyeglasses and very small. It’d be less expensive and more convenient, and more people could use [it], especially in these developing countries, [where] people really can’t read and write.

Walter: Well, I hope that one byproduct of this project is that more people learn to read and write. If there are more efficient technologies, if there are better ways of delivering the same set of tools of expression to children and their teachers, I’m all for it. This is just to get the ball rolling; this is not the be-all and end-all. And if this becomes obsolete, which it will, hopefully something else better will replace it. So, sure.

John: OK, please, over on my right. Here, yeah.

Man 5: You said something about... You addressed the issue of subsidies, and so for instance if the Carnegie Foundation wanted to buy everyone in Zimbabwe or Zambia a laptop, because of the economic implications of that, is it against your policy to allow charity foundations to buy laptops for others?

Walter: No no, not at all. I just said that we as an organization are not subsidizing laptops. We’re making laptops and selling them at cost. Now, how they’re paid for is another matter, and it’s actually something that right now what we’re doing in our launch process, because we needed to get up to scale. Because scale has an impact on the price, and some of the other issues I was raising earlier about scale, to get up to scale, we made a decision, rightly or wrongly, to go to ministries of education at the federal level in large countries distributed in three continents. That is not our steady state for this project. Our steady state for this project is to work with whatever good ideas are out there, and whatever organizations have good ideas about getting kids to have access to laptops for learning. And so we’re actually going to be issuing a request for proposals probably in the February-March timeframe of this year for this next phase of where this project goes, with the idea that that second phase will begin towards the end of the calendar year. So any good idea we’re interested in.

John: OK, and right next to you, if you’d like to pass the mic along.

Man 6: With regards to the efficiency of the distributed wireless networks: If I’m running on the Internet through a chain of three other laptops to a weak wireless connection, can I stream video and audio multimedia?

Walter: It really depends. Probably now is not the time to go into all the details on the mesh, but essentially what you have with the mesh... It’s actually a very high bandwidth... It’s essentially a g network in the mesh, laptop to laptop. Where you end up losing... Well, each hop you lose a little bit in terms of latency; you don’t really lose in terms of bandwidth, though. But what ends up happening is that if the mesh gets to be too distributed--too many hops---there’s a certain overhead associated with the addressing, OK, and you spend so much time and so much energy and so much bandwidth on the routing tables that that’s what causes the mesh to begin to degrade. So it actually has more to do with how many hops, not because of the bandwidth of a hop, but it’s got more to do with the management of the network.

John: One person had a question up behind me which is not currently being shown, but it was about the experience with these beta machines out there with kids, and I think the question said, “What are you finding in experience, or is it too early, that kids like most about these laptops, and what, if anything, do they like least?” Are you already getting feedback?

Walter: Well, nothing that’s really worth reporting yet, in the sense that there’s been lots of kids that have played with the laptops, but we haven’t had the laptops really in vivo in the real situations. That’s happening early February, that that’ll begin.

John: And do you have some feedback mechanisms in place to you, so that you can get useful data back from early-generation users?

Walter: Yeah, yeah. We’ll be having teams on the ground with all these deployments, and getting feedback.

John: Right. OK. There we are, yes. “Should the capability of the laptop be intentionally limited?” somebody, or one group, put up, and that was related to another thing I saw, which was, “Are you finding any governmental or administrative anxieties about what people can do with these laptops?” Are there constraints on you in some countries to limit the capabilities of this machine?

[50:26]

Walter: Well, I mean, there certainly is some concern in some countries about, you know, are the kids going to access pornography with the laptop, things like that, and that is a concern. There are lots of mechanisms in place that countries routinely already use to deal with that problem, with varying degrees of success. We’re not really in a position where we can do a whole lot about that. One of the things we can do, for example... You can imagine that an OLPC laptop can identify itself when it went to Google and automatically have the default be Safe Search. Now, how good is Google Safe Search? Well, it’s OK, but if somebody e-mails me a URL to something I shouldn’t see, well, I’m going to be able to see it.

John: Right.

Walter: So again, there are whitelists, there are blacklists, there are all kinds of mechanisms for doing these kinds of things that are fairly routine. That’s really not our decision as OLPC, and we’re sort of trying to keep at arm’s length from that decision.

John: A number of the questions that people have sent in or fed in are about what I think are sort of secondary implications of introducing laptops in volume into other countries. One point was about the possible creation of a so-called black market, with people, presumably kids, saying, “This is worth something--I’ll sell it, because the biggest thing that I could do to help myself right now is to get some money.” And again, I can imagine that this is not something that is easily controllable from where you and your colleagues are sitting at MIT, but you must discuss these things sometimes.

Walter: Oh no, we discuss these things a lot, and we’ve got some mechanisms to try to slow that down. So for example, one of the things that happens with the laptop is it gets a lease from the school...

John: Yeah. Oh, OK.

Walter: ...and it has to come back to the school periodically to renew that lease. And so if it is stolen... That’s one incentive for it not to be stolen. Another thing about the laptop is that laptops are going to get shipped in shipping containers, and if that shipping container somehow never gets to the school, the laptops don’t work. So there are a number of little things like that that we’re doing, but it all ultimately is going to boil down to having a culture around the laptop where the community begins to realize that it’s a resource for their kids.

There’s another thing. It’s a laptop on purpose. And the reason why it’s a laptop is so it will go out into the community and have an impact on the community, so the siblings can use it, the parents can use it, and it actually becomes a resource within the community. So it’s not just locked up in school when the kids aren’t in school.

John: But what you’re describing--and I’ll bring in somebody else from the audience next--but what you’re describing implies that... If we try to think what’s in our minds, if we imagine these things in potentially large numbers going into developing countries, it’s not, as it were, maybe, to take a caricature, vans driving into marketplaces and distributing these in exchange for some modest sum of money; it’s more likely, maybe--or is it?--schools working with governments to be provided with laptops which go to kids who may not have many other resources in that school. Is that a more realistic model for what’s going on?

Walter: Well that’s certainly how our launch strategy is; it’s exactly that model.

John: Right.

Walter: We’re distributing laptops through ministries of education, the same channels that distribute textbooks.

John: Right, OK.

Walter: And actually, one of the arguments for the laptops to ministers of education... You don’t have to argue with an IT minister, or a minister of science and technology, or a minister of commerce: They all get it. Ministers of education, they’re a little bit more nervous about this.

John: I understand, OK.

Walter: But this project, to some degree... Nicholas, when he was first talking about it, used the analogy of a Trojan horse. A laptop’s a Trojan horse. The horse, the thing that gets in the door, is the eBook. [folds laptop into eBook mode] Because now what this is, is just a textbook.

John: Yeah.

Walter: And ministers of education understand the economics of textbooks, they understand how you distribute textbooks, they understand that in most of these schools they don’t have enough textbooks; they have on the order of one or two textbooks per 50, 60, 80 kids. So this is a way of mitigating that need. But there’s some soldiers inside the horse.

John: Right.

Walter: The soldiers inside the horse are the kids with laptops, and that’s where the change is going to happen. And if you look at... Our project’s a learning project. We want to give kids opportunity for learning. And one way that learning’s going to happen is through schools. But that’s only one way. And in fact, a lot of the time, kids aren’t in school. Kids are in school a couple hours a day, and that time is already pretty well programmed. And so, we’re hoping that a lot of learning happens outside of school with the laptop. There’ll be some points of intervention in school with a laptop, which we hope are profound and of value. School itself will only change when the kids going to school come with different skills, different expectations, different needs, and the laptop will do that, but it’s going to take time.

[56:00]

John: OK. Thank you. Two people, and one over here. One, two, three.

Man 6: I guess this question is related to the previous question, or the previous issue raised about theft and security and so on, and that is that presumably there is a server that’s also associated with this program at each site. And I think in many ways the server is probably more at risk than the laptops for theft and reuse and appropriation, and I’m wondering how that issue is being dealt with, because that server is also going to be the link to the Internet, is that right?

Walter: Right. Right. We’re actually going through a little bit of back-and-forth about our server design right now. But you’re absolutely right: The thing that’s got the most street value of anything we’re doing in terms of our deployment right now is the hard drive inside the server. That’s cash on the street. Nothing else is necessarily directly cash on the street, but that certainly is cash on the street. And so protecting that--and we’ve got a number of different ways we’re thinking about it, but we haven’t completely resolved that issue--is a high priority.

Man 7: It’s wonderful to see this; I didn’t think I’d ever have a chance to see... I mean, I read an article in The New York Times, but it’s amazing. I have my feet in several places, but I’m involved in a program called the Ruwwad project, empowering Palestinian youth, and a project with Afghan women, empowering them, and as well as two days ago, I was on a Crown Point Navajo reservation, and I think that in any or all of these situations, something like this would be incredibly important, and wouldn’t involve... I’m kind of skeptical about governments, but these other things might be a way to get this wonderful instrument into the right people’s...

John: When you say “these other things,” what do you mean?

Man 7: I mean get these, the computers, to the right people.

John: Oh, OK. Do you want to comment on that, or...?

Walter: Well, there’s a certain skepticism I have about governments as well, but when push comes to shove, I’m a little bit of a libertarian, in that my personal philosophy about government is government should do the things that the private sector can’t or won’t do--and educating our children is one of those things. And I think that that’s true everywhere, and I think that there’s some need to engage government, even if you’re coming in through an NGO, you’ve got to have government involvement if you’re going to be impacting schools and impacting children’s learning. I don’t think there’s any way around it.

Man 7: USAID is involved in two of the projects, the one in Palestine and in the Afghan women’s, but I happen to be on the Navajo Indian reservation, and it just seems like they’d have a tremendous need for these things, and in a way they have the resources, but in a way, they don’t, so...

Walter: Yeah.

[59:24]

John: OK, there was a question over here, please, as well.

Woman 5: So my question is, who is going to teach the teachers how to use this machine, because it does have a certain learning curve that’s necessary? And what about those kids under the tree where the teacher doesn’t show up, or the kids who don’t have the $5 per year to go to school?

Walter: So there is training, and there’s preparation, and those are two different things. And there’s how to learn how to use a computer, and then there’s how to learn how to use a computer but have an impact on learning, and those are two different things. I don’t think that learning how to use a computer is something that needs a lot of training. I think that, you know, there’s things that are hard, and things that aren’t, yet discoverable in our interface, but I think that it is all learnable. An example is the Hole in the Wall project in India, where they put a computer literally in a hole in the wall. They didn’t put any manuals there, they didn’t have any training classes or Dahlsen’s[sp] or anything, they just put a computer there, and people learned how to use it, and if one person learned, they helped the next person learn, and there was a lot of... Again, they leveraged the social nature of that interaction.

I don’t know a kid ever who’s learned how to use a computer from a manual. There was a study done by Filmis[sp] in Argentina lately, and I think--I was startled by this number--I think something like 15% of the kids learned how to use a computer in school, as opposed to 85% just learned how to use a computer by using it. I can’t believe that 15% of the kids actually had to go to school to learn how to use a computer. Now, that said--and I think the same is true for teachers--but that said, I think there’s a lot to learn about how to use this as an effective tool, and there’s lots of way to use it ineffectively. When I was a kid, there used to be this technology called the mimeograph machine. We referred to it as the Purple Plague, and what it was, it was the technology du jour to generate worksheets.

[picks up laptop]

Well, this is the world’s greatest worksheet generator. And so, certainly one thing that could happen, the path of least resistance, is to use this as a way to keep kids busy doing nothing. And what we want to do is we want to have this be a way of having the kids explore and appropriate knowledge and be expressive. That’s the part that needs the preparation for the teachers, and exactly how we do that, well, we’re going to be running workshops and getting people involved and having examples of best practice, but it’s actually an open-ended question how to do that to scale. That’s probably the biggest challenge we have. Now, the technical challenges, well, you know what? The world knows how to scale up technology and how to build things. That really, in my mind, is one of the hardest problems we have to address.

John: OpenCourseWare was mentioned. You want to explain what that...

Walter: Yeah? You want me to explain what OpenCourseWare is?

John: Just in case anybody doesn’t know?

Walter: Well, OpenCourseWare is an MIT initiative to make--to begin with university content, and actually now they’ve got a program they’ve started on high school-appropriate content--available freely on the Internet for people to download and use. And it’s been wildly successful. It’s certainly one piece of the solution in terms of where content comes from. I don’t think that it offers a whole lot right now in terms of pedagogy. It could, but it doesn’t yet. So I think that... Part of what we’re trying to do is build a social network around learning. And it’s interesting to me: There’s a community out there that, in my mind, is better at collaboration and sharing of ideas and sharing of criticism and closed-loop development than any other community, and that’s the open-source software community. And what I think has to happen to really make this project take off is we’ve got to infect the education industry with the methodologies, the tools and the ethics of the open source community. And I think when that happens, it’ll all work. But it’s going to take time to make that happen.

John: We have a few minutes only left, I’m afraid, so I want to encourage those who’ve got issues that haven’t been raised yet in the course of this conversation to raise them now if you will. So I see a couple more hands, one there, and then there.

Man 8: I just want to comment on that statement:

John: Can you keep the mic close to you?

Man 8: I wonder if, instead of infecting the education industry, maybe laying something on top of the education industry, so that it has to compete.

Walter: Well, I think that actually, I would lay something below the education industry, and empower teachers, give teachers an ability to directly engage, directly spread the good ideas, interact with each other. That can exist; it hasn’t existed traditionally in the industry. But I can exist, so it can be a grassroots, bottom-up initiative. So then rather than laying something on top, building something from below.

John: Please, there’s a comment there, yeah.

Man 9: I guess the question is, say you drop one of these in the village. There’s no teacher there, and nobody in that village can read the native language. Is there a program that you put in these computers aimed at these communities where they learn, and then they advance up the thing to jump on the Internet and that. Because if they just poke around and find something that they like, will they stick to that? Do you have that type of program?

Walter: Well, we don’t exactly have that type of program. I’m not quite sure if that scenario exactly exists today. For example, we went to rural Cambodia. I mean, this is a place with no roads to get to it. And the very first word of English the kids all learn is Google, and the very first thing they did was look up on the Internet for all their favorite football stars. Not American football, but real football. [laughter]

[in response to question from crowd] No, in that particular case, we did not put it in their native language. Now, what we are doing with the laptop... It’s interesting, because the minister of education in one of the countries we’re going to did not want the laptop to be in any way in the native language at all. He wanted the laptop to only be in English. We sort of said, well, no, we don’t want to do it that way, but every single minister of education we’ve talked to, the one point of agreement that they all have is they all want their kids to learn English. Now, that’s not the only thing they want their kids to learn, but they all want their kids to learn English, because I think they think that’s going to make their kids be more competitive in the world. [gesturing to woman with concerns about Americans’ comparative advantage] Now back to this point.

We’re trying to come up with good tools for literacy, to get kids to become literate to begin with--that’s a basic goal for our project. Another basic goal is mathematical literacy and literacy in terms of science. We’re trying to make the laptop be usable without language initially, but have a collection of things that draw you into language and language use. That’s something that I think we’ve got a lot more work to do on. There are some beautiful examples of using computers for literacy, but we haven’t really integrated them in a very coherent way in the laptop, so that’s a challenge we need to face.

Man 9: So this would have to be almost an underground movement.

Walter: Yeah, to some degree it is going to be an underground movement, because we’re going to be giving all these kids an opportunity to have a voice and create things and say things and do things and share those with other people and critique their ideas.

John: Well look, there are other people who are keen to get in, but I apologize, our time is pretty much up, and what I want to do is just two things to close, actually three. One is a plea to those who’ve still got the two examples of the one laptop out there to hand them back, because I think we should allow our speaker to go back with the ones he brought.

Walter: There is one still out there somewhere.

John: There’s one there...

Walter: ...and I’ve got the other one.

John: OK. So if we could let Walter have that back, that would be in the category of “basic courtesy.” The second thing, let me just remind you, because Beryl’s waving it, the questionnaire which we’ve given you, which would be so helpful in organizing the events that are going to happen in the Cambridge Science Festival in April this year. I want to just give you a heads-up that our next Soap Box, which will be in February, is going to be on the subject of climate change in this community, believe or not, in Eastern Massachusetts. We’re going to be starting a quite exciting project with two scientists who are gathering evidence about climate change in this part of the United States, and that will be a citizens’ science project, which will allow people to start contributing to data collection in that very intriguing area. So watch this space for the next Soap Box.

But the most important thing I have to do is to say thank you to our speaker. Now, there are lots of different things that we’ve featured in Soap Box this far, but we’ve not to date featured any project from MIT which in my view quite so well captures the spirit of innovation that MIT I think prides itself on. Whatever you think about this project, it is certainly innovative, I think you’ll agree, and it’s also being bold and thinking very big about how to enable learning and development globally by very, very interesting adaptation and refinement of technologies that we’re all in one way or another very familiar with. It’s great to have a chance to learn about this from one of the key people in the project. Walter, thank you so much for taking the time to come.

Walter: Oh, you’re quite welcome.

[applause]