One Laptop Per Child Talks

Walter Bender of OLPC

In September 2006, Walter Bender, President, Software and Content, spoke at Ars Electronica Simplicity - the art of complexity about One Laptop Per Child.

The transcript of Walter Bender's session:

Announcer: Ars Electronica 2006. Simplicity, The Art of Complexity.

Walter Bender: OK, thank you very much, John, for inviting me to Ars Electronica. It’s a pleasure to be here. What I’m going to do with my few minutes is explain what it is we’re trying to accomplish with the laptop project, spend a fair amount of time actually talking about the design and the process we went through and some of the problems we’re trying to address. Then I want to talk a little bit about ways in which we expect the children to use the laptop. So that will be the gist of it.

I do want to make a couple of comments about simplicity and complexity, as part of the motivation, because this is a topic that John and I have talked about and argued about over the years. Last night we had a lovely dinner downtown in a little restaurant that was right in the middle of a wine festival. One of my favorite examples of the juxtaposition between simplicity and complexity is wine, because I never go looking for a simple wine. I like a wine that has some complexity to it. So, the goal, and certainly the goal with the laptop project is not to eliminate the complexity of the world, but rather, the goal is to build a device that is simple in its use but allows us to reach and enjoy and experience and leverage all the complexity in the world. So that’s sort of the fundamental relationship to the project and to the theme.

Our mission at one laptop per child is to get laptops out to kids so that they can use them as things to think with, as tools to learn and express and explore the world with. We’ve been working on technology and learning at MIT for almost 40 years and we really think through all our experiments and experience that computers are something pretty special and they represent an opportunity that we want children to have. Now of course, in year one it’s probably going to be the 100 euro laptop, not the 100 dollar laptop, but we’re working on it.

So this is who we’re trying to reach, we’re trying to reach the one billion school age children in the developing world. There’s a lot of need out there. There’s a lot of kids out there. They’re bright, they’re eager, and they’re lacking in opportunity. They’re all over the planet. The countries in green are probably the countries we are going to be doing our launch in Q1 next year. The countries in orange are probably the second wave of the program. Countries in yellow are countries that have also expressed interest but not at a high government level.

We’re going to very deliberately start in three continents, in South America and Africa and in Southeast Asia. You want to have a breadth of experience in the program. Again as I said, our premise is that children lack opportunity not capability. We want to give them that capability. The challenge to the program in some degree is manufacturing an inexpensive laptop but the real challenge is how do you take this and go to scale. It’s a real problem of scale. That’s the thing no one’s ever done before.

Now, how do you design a laptop for children? It’s not the same as designing a laptop for an office worker. What we did was take the opportunity to really rethink laptops from the bottom up, both from the aspect of hitting the price point but also from the perspective of trying to meet the needs of children.

So, a child’s laptop has got to be safe, it’s got to be robust, light, vibrant. Kids spend a lot of their time outdoors, and so the laptop should work outdoors as well as indoors. It turns out that I spend a lot of time outdoors too, and one of my great frustrations is, my laptop doesn’t work too well outdoors.

Communication and expression are fundamental, so the idea is you learn through expression and you learn through doing and sharing that expression and critiquing that expression. So we’re really trying to build an expression machine. And we’re trying to build an expression machine that allows for rich communication with teachers, with other children, with the world. We also want to build a laptop that lets them engage in an open ended discovery process. Of course, if we’re going to reach children in the developing world, it has to be inexpensive.

There are probably four things that are revolutionary in this laptop, and the first and foremost is Mary Lou Jepson’s dual mode display. She re-invented the LCD display technology, took a lot of parts out of the typical display just to make it less expensive, but at the same time built a display that’s much more versatile. So this display works with a backlight on. It’s color with a backlight off. It’s black and white so it works in sunlight. It’s got a failure mode so that if the backlight dies the display still works. It’s 200 dpi, so you have a very high quality display. It’s really the centerpiece of the whole project.

The first external design group we worked with is Design Continuum, and they came up with this notion of the brick. This is one of their early sketches. The wonderful thing about the brick is we married the motherboard and the display together into one unit. And that has a tremendous value in terms of robustness. The biggest single point of failure in laptop computers is actually the connector between the display and the motherboard, and we’ve eliminated that in this design. We’ve eliminated most of the internal connectors. We’ve eliminated the hard disk. We’ve eliminated most of the points of failure for laptops. It’s got to be robust. It’s also got to be robust in terms of the environment. It’s got to work in conditions of dust and moisture and big swings of temperature and the like. So we’ve designed a laptop that kids can carry home in the rain.

The next thing that was really transformative in the project was Quanta, who’s our manufacturer, insisted that we use this transformer hinge. We thought the transformer hinge was going to be too fragile and too expensive. They in fact said no, it’s less expensive and it’s more robust. That allows us to have a laptop that has multiple modalities. A laptop can also be an e-book. Giving these children access to books is certainly one of the big goals of the project. It’s actually quite a versatile design in terms of the laptop.

Finally, Michallas Bletsis, who’s one of my colleagues on the project who worries about connectivity, came up with an antenna design for the laptop for the WiFi. There’s these little bunny ears that pop up. They’re important for two reasons. One, having the antennas pop up above the display, we get about 2 dB better signal noise ratio. So this laptop’s going to have a better WiFi system than any commercial laptop in the marketplace today. It’s going to have in clean line of sight, about a half kilometer of reach with the signal. The other thing is that the bunny ears double as a protection for the USB ports and the microphone jack. So when the laptop’s closed, the bunny ears close over. They’re actually the latch that locks the laptop tight and keeps it closed.

Those basic elements were brought together by another design firm, Fuse Project. This is pretty close to what the final laptop is going to look like. Notice, for example, it’s got a nice handle. It’s nice to have a handle on your laptop, to carry it around. In this form, you see it in the e-book mode. It’s actually also used as a game machine. Kids are mobile; the laptop should be mobile and move around with them.

One of the most important innovations with the laptop is not with the laptop itself, but with the collection of laptops as a group. One of the things we are doing with One Laptop Per Child, is taking advantage of a saturation of laptops in a community. The laptops form the internet, the Ethernet structure automatically. They form a mesh network when they come out of the box. All the laptops talk to each other in a very high quality, high bandwidth fashion without any other additional infrastructure. Automatically, the kids take the laptops out of the box, and they’re all online talking to each other. The other thing that happens then, is that where there is access to the internet, all the laptops can share it. It’s a much more efficient way of building out the infrastructure for the laptops and for the laptop program. Its both robust and high bandwidth, and amazingly cost effective and economically efficient.

Another problem we wrestled with is power. Most of the places we are going are off the grid. Even those places that are on the grid, the grid is intermittent. I don’t think there’s anywhere in Nigeria for example, where there’s 24/7 power. Power goes out a few times a day even in the capital city. So we really had to think about how the children were going to power their laptops. We’ve got a variety of different solutions around. Hand power, and when Eric talks a little bit later you’ll hear a little bit more about different ways of thinking about powering systems, although he’s talking about maybe an order of magnitude less power than we need currently.

One of the other things we did is we really pushed hard on what are the power requirements. The typical laptop uses about 20 watts average power consumption. Our laptop’s going to be 2 watts average power consumption. Because if you’re going to have the kids cranking to generate power, you want that power to be used efficiently so you don’t have any waste in this machine at all.

And then the other critical design decision is more on the software front. We designed the laptop using open source software and free software. We were very deliberate about this choice. It’s not because we didn’t want to pay for software; this was a design decision that was driven by epistemology. We want to have the laptop be transparent in every respect so that the kids can own it and appropriate it in every respect. The idea that the laptop has a certain transparency to it that allows the children to reach in and understand everything about it, and change anything about it, is fundamental to how we think it’ll both be deployed and also where a lot of the learning will take place.

I want to talk a little bit about how we see the children using the laptop for learning. We’re capitalizing on three things, three things that are fundamental to being human. The first is that everyone’s a learner, and everyone’s a teacher. That is fundamental to being human. The second thing that is fundamental to being human is expression. We’re an expressive being. And then finally, we’re social. We’re taking these three attributes, these elements of humanity, and leveraging them. It’s really through this that we’re trying to build a learning process, and a learning machine.

The first is that we’re really interested in the children not just being consumers of knowledge and the consumers of expression, but also the creators of knowledge and the creators of expression. It’s all fine and well to have an iPod that you listen to music, but we’re as interested or more interested in the music that the children will create as the music they will download and listen to. They will be able to download music if they want to, but we’ve got a lot of tools of expression, particularly around music, actually, that will allow the children to compose music, to play the music across the mesh.

We’ve got a lot of tools that allow multiple laptops to work together to in a sense; each laptop’s a different instrument in a composition, things like that. The idea that you can build a very rich musical experience with the laptops is an example of the notion of turning the table on where does the expression derive from; it derives from the children themselves. This is an example on Barry Vercoe, who is the inventor of the programming language, cSounds, which is a language for music synthesis that’s at the heart of most computer music systems, has built actually a really lovely set of tools for musical expression across the mesh, laptop to laptop.

Second, is we really believe in debugging, that you want to get out into the world and experience it firsthand and wrestle with it and debug it. A lot of the learning we think is going to be in the world, not just in the classroom. It’s going to involve the kids building things, and building things in the real physical world.

One of the innovations that Seymore Pap is responsible for is the microphone jack on the laptop that doubles as an analog data input. So you can plug sensors into the laptop and have the kids be turning their laptop into an oscilloscope, for example. What you see on the left is actually a 30 cent conductor’s baton. I think that Sam designed the wooden spoon, and then we added a little photo diode on the end of the spoon, and plugged it into the microphone jack. You can use it to conduct the collection of laptops in the room automatically.

On the right is an example of a project from my colleague, David Caballo. It’s a project he did in Corachiban, Brazil. It’s not using a laptop, but we’ve enabled the laptop to do these kinds of things. A popular arcade game in Brazil is Dance, Dance Revolution. I don’t know what they call it here. It’s a game where the kids stand around the arcade machine and they do these dance steps. Instead of just playing the game, the kids built their own Dance, Dance Revolution machine. It’s that idea of letting them be constructors in a very physical way that we’re trying to enable with the laptop.

Another thing is we really want to maximize the exchange of ideas, and the exchange of best practice, the whole idea of being a storyteller, just the simple thing of writing and expressing to other people and getting feedback from other people is a fundamental tool for knowledge. Literacy is certainly one of the problems that we’re trying to address with the laptop. Again, the idea is that there’s no point in being literate if you don’t have other people to share that experience with, to receive materials from and send materials to.

Another thing that’s important to how we’re designing the laptop is the notion of presence. Again, because the laptop is on this mesh network, the learning doesn’t have to happen in isolation. Just as we’re revisiting laptop design from the bottom up in terms of the hardware, the software is being rethought from the bottom up as well. So, you’re not going to see a conventional desktop and file system on the laptop, even thought there is that underneath. There’s the Linux operating system sitting underneath. What you’ll see is an experience that really has two modalities. One is you’re always in the presence of your friends. You always see the other people around you and what they’re doing. And the other thing is that when you’re running an application on the laptop, it’s not just an application, it’s an activity. You have the opportunity to invite other people to participate, either in the role of a critic or in the role of a creator in everything you do. You’re one button away from being able to bring other people in to participate in all activities of learning in the laptop.

One little subtle piece of subversion in the way in which we’re building the laptop, I’d mentioned earlier that we’ve got this e-book mode. The e-book inside the laptop is going to be a wiki. The reason for that is not because we don’t want to receive knowledge. There’s a lot of great knowledge and a lot of great ideas that have grown up in the world. We don’t expect the kids to reinvent them; we want them to have access to the world’s knowledge. But at the same time, we want more than just access, we really do want appropriation. The idea behind the wiki model to the e-book is that you will have your article page, you’ll have a page about the content itself, but in a wiki, every page also has a commentary, a discussion. The idea is, whatever the children are reading, they’ll be able to write margin notes into that, and share those margin notes with other people; engage in discussion about the content.

Some content will be protected; most content won’t be protected, in the sense that the children will be able to add their own local experiences. Maybe it’s adding photographs and illustrations to the book. Maybe it’s adding additional material, or maybe it’s even taking issue with what they’re reading and saying, “No, that’s wrong, it actually works another way.” So the wiki really is a way of taking the knowledge that exists in the world and putting it in a form that makes it transformed and realizable by the children in a very different way.

Just one final thing that I want to talk about, I’ve been spending a lot of time actually over the last 24 hours while I’ve been here working on the keyboard. We’re doing a lot of work redesigning the keyboard, because we need to, first of all the keyboard’s going to be a little bit smaller, just because of the footprint. Also we want to make a keyboard that’s friendly to kids and smaller fingers. The other thing is, what are the keyboard mappings, because we don’t necessarily have to keep with the traditional keyboard mappings. I’ve been having a lot of fun thinking about what would be fun keys to add to the keyboard.

One of the keys that we’re going to add to the keyboard is one that was inspired by the web browser. One of the most interesting features of the original web browser from Mozilla was the view source menu item. Every page on the web, until we got Flash, you could actually look under the hood and see how that page was constructed. I’m convinced that the reason why that particular protocol, HTML, took off was because it was transparent in that way. Because the web browser let anybody see, “Oh, I can do that too.” So we’re building the laptop out of scripting languages.

It’s going to be sort of like an onion that you can peel away layers. And every time you hit the view source key on the keyboard, you’ll be able to, the children will be able to peel away another layer and see deeper and deeper into the machine, and be able to appropriate the machine at every level. Again, looking under the hood, transparency, appropriation, and sort of the social structure around the laptop are the hallmarks of how we are doing the software infrastructure.

That’s all I have to say about the laptop itself. We’re just entering what’s known as B Test mode right now, which means we’re going to have machines that include all the hardware and industrial design all packaged together. We’re getting those things out to kids to pound on over the next few months. Then we start the line at the end of Q1 next year when kids will be starting to have these laptops. So thank you very much.

[clapping]

Host: We have time for a question for Walter, and also now we have time for a question for Sam. I guess I don’t know how to say his last name. I’m sorry. Those who ask Walter a question will get a free t-shirt. So who’s going to ask Walter a question?

Man from audience: These laptops, these meshes of laptops seem like autonomous setups right now. Have you consciously not involved a project of information exchange with people over on this side of the fight?

Walter Bender: Well, the mesh itself is in order to define the local infrastructure. That local infrastructure doesn’t exist. The mesh is the most cost effective way of building it, because it gets built automatically by the laptops themselves. We’re working with the governments on ways of building low cost backhaul to the internet connectivity as well. So while a lot of the learning, a lot of the communication is going to be local, it’s quite natural that you’d have a lot to say to your friends locally, we have every expectation that there’ll be lots of international cross-pollination of ideas with children. In fact, one of the things that we’ve discussed with a lot of these governments is their desire to have their children communicate with children in the west in large part just to learn language skills, but in large part just to have exchange of culture.

Host: A question from Olga, in the back.

[Olga comes to the microphone]

Host: Olga is speaking in the afternoon, by the way.

Olga: I want to ask about the economic context of your work. Like, obviously you should explain some pressure from companies. For instance is it going to be 100 dollars in end cost, or what would be the cost of production, if you can talk about this issue.

Walter Bender: Sure. One Laptop Per Child is a non-profit association. Our mission is to get laptops to children. We sell laptops to the government at cost. So whatever the laptops cost, that’s what they cost. And we’re pushing industry pretty hard. Since our mission is to get laptops to children, there’s nothing in our mission statement that says it has to be our laptops. So if some other party comes along and can make a 50 dollar laptop that has attributes that kids need, that’s great. And it means we don’t need to be in the laptop business anymore, we can just focus on learning.

So we’re encouraging industry to take this marketplace seriously, and also rethink their practice, because what children need is different than what business executives need. We’ve done a lot of work to try to define this is what the children can actually leverage effectively. It’s possible to do this inexpensively; come on world, join us and do it. Now, some of these industries, it’s a little bit of a bitter pill to swallow in the short term because they’re geared up for something very different. So they’ve got to adapt. For example, I like to use the analogy of cathedrals and churches. The kids need churches not cathedrals, but industry has been geared up for many years to build cathedrals, so they’ve got to make that transition. We’ve already made that transition. I’m sure that industry will follow us and eventually get ahead of us.