One Laptop per Child's breakthrough software replaces the standard PC look with a design for the networked age
The so-called $100 laptop that's being designed for school children in developing nations is known for its bright green and white plastic shell, its power-generating hand crank, and for Nicholas Negroponte, the technology futurist who dreamed it up and who tirelessly promotes it everywhere from Bangkok to Brasilia. What has not received much attention is the graphical user interface—the software that will be the face of the machine for the millions of children who will own it. In fact, the user interface, called Sugar, may turn out to be one of the more innovative aspects of a project that has already made breakthroughs in mesh networking and battery charging since Negroponte unveiled the concept two years ago.
Sugar offers a brand new approach to computing. Ever since the first Apple Macintosh was launched in 1984, the user interfaces of personal computers have been designed based on the same visual metaphor: the desktop. Sugar tosses out all of that like so much tattered baggage. Instead, an icon representing the individual occupies the center of the screen; "zoom" out like a telephoto lens and you see the user in relation to friends, and finally to all of the people in the village who are also on the network.
It's the first complete rethinking of the computer user interface in more than 30 years. "We're building something that's right for the audience," says Chris Blizzard, the engineering project leader for Sugar. "We don't just take what's already there and say it's good enough. You can do better."
The audience he and his colleagues have in mind is the hundreds of millions of poor kids all over the world. Negroponte came up with the nonprofit "one laptop per child" idea when he was chairman of the MIT Media Lab and observed the failure of standard attempts to use computers in education to improve the lives of underprivileged children. Typically, a handful of computers, designed for business applications, are installed in schools; students only use them in special computer classes and are forced to share. Negroponte's idea was to give a laptop to each student that he or she could take to every class and bring home at the end of the day. "OLPC is child-centric, designed to be a seamless part of their lives at home, at school, and in play," he says.
Nearly a dozen countries, including Brazil and Thailand, have committed to buying the computer, now officially called XO. The UN Development Program will administer the program locally. About 2,500 beta test machines ran off assembly lines in Taiwan in February and are now being shipped to participating countries so they can kick the tires on the technology. The final version is supposed to be ready by August.
"You Just Do It Right"
While XO has been greeted warmly by many, some technologists criticize Negroponte and his colleagues for not testing out their new ideas on underprivileged school children earlier in the process. And that goes for the user interface as well. Jakob Nielsen, a user interface designer and principal in the consulting firm Nielsen Norman Group, falls into the critical group. While familiar with the design of Sugar, Nielsen’s criticisms focus on the process. It’s only in the coming weeks that they’ll begin to get feedback from kids. “It’s always dangerous to release any product without the safeguard of user testing,” says Nielsen. “But it’s outright reckless in a case like this.”
But XO developers defend their approach, which grew out of a core philosophy of the MIT Media Lab known as "demo or die." Researchers are encouraged to build new things, critique them, and then make improvements—rather than doing a lot of concept-testing up front. They're backed up by John Maeda, a user-interface design guru from the Media Lab who has been watching the XO development process from its beginnings. "They're using the Steve Jobs method," he says, referring to Apple's famous chief executive and design whiz. "You don't use focus groups. You just do it right."
When BusinessWeek visited the OLPC offices in Cambridge, Mass., in mid-February, one of the XO designers had just achieved something of a milestone. He had loaded a game modeled on Tetris on a test machine and was trying it out. This scene took place in a large, brightly lit room where a handful of XO computers were scattered on tabletops, many with their miniature circuitry exposed—a reminder that Sugar is still very much a work in progress. "You're the first to see Tetris running on our computer," said Walter Bender, OLPC's president of software and content.
The game, called Block Party, is being used as a sample of how developers should create applications for XO. "We're showing them how to 'Sugarize' their applications," Bender explained. That means conceiving applications from the start as activities that take place on the network and are shared by groups of youngsters and their teachers.
"Sugarizing" also has a technical side: The software is built on top of Red Hat Linux, and is an open-source project itself, meaning that any interested software programmer could write software to run on the machine. But the programs must be small—the XO has no hard drive—so existing PC software must first be rewritten.
Sugar has a look and feel all its own. When you start up the machine, you see the image of the so-called "XO Man," an O on top of an X, placed in the middle of a circle. A darkened border frames the display, lined with icons representing activities such as e-mail, a simple word processor, a photography program (XO has a built-in camera), a Web browser, instant messenger, and an electronic book reader.
There are also icons representing the three different modes—home, friends, and neighborhood—that are integral to the "zoom" metaphor. In home mode, a user sees the XO Man, and, when she clicks on the icon to launch an activity, the icon for that activity pops into a gray ring encircling the XO Man. In friends mode, she sees icons representing her circle of friends, each identified by nickname and chosen color scheme. Next to the friends are icons depicting the activities in which they're engaged.
If several friends are sharing an activity—say, working on a school report together—they are pictured clustered around the appropriate icon. Our user can ask to be invited into a group activity or can start one of her own and invite others to join. The neighborhood mode gives a broader view of all of the individuals and clusters of friends on the network at the moment and the activities they're involved in.
Wi-Fi For the Village
One of the key technologies behind the XO computer is its so-called mesh network. Created by Mikhail Bletsas, the OLTP's chief connectivity officer, the XO mesh connects all of the XO computers in a village via a Wi-Fi network. If any one of the computers is connected to the Internet, they all get Net access. And the computer's antenna is always left on so the network remains active—though networking draws less than half a watt of power from the computer's battery. The children are expected to keep their computers powered by occasionally turning a hand crank or operating a yo-yo type device that keeps the battery juiced up.
From the start of the Sugar project last summer, Bender urged his small team of programmers to keep the interface simple and to organize things so children could learn by doing. Even now, the Sugar development team is made up of just eight full- and part-time contributors. Several of them, including Blizzard, work for Red Hat Software, the leading distributor of the Linux open-source PC operating system. One of them, lead designer Marco Gritti, an Italian, gave sugar its name.
Then there is the open-source community, which the organizers are just now engaging. Any programmer who is interested is free to view the core software code on an OLPC Web site and suggest improvements. And a handful of OS efforts have formed to create applications for the computer.
The project leaders hired Pentagram to help out with the visual design of the interface. The Pentagram designers have kept the icons spare and universal so kids can understand them instantly no matter where they are growing up. Text labels are kept very short. The "zoom" feature is being designed so kids see the transition from individual, to group, to village as if they're in a helicopter lifting off from the earth. "We're trying to use as many references as we can to the physical world so it will be easy for kids who haven't used a computer before to use this foreign thing," says Lisa Strausfeld, the Pentagram partner whose team is working on Sugar.
Kids Can Tweak the Code
The interface is being designed to encourage the users themselves to explore—and improve on—its inner workings. Bender chose a software programming language called Python that's simple enough for more-tech-minded children to learn. If a student is playing a game on the computer, he can actually look at the game's code and modify it—say, changing the colors on the screen. If the student makes a mistake, he can restore the program to its original form with the click of a button. "The machine is a tool, but it's also an experience. It's a way to be creative," says Blizzard.
To that end, Sugar offers a simple technique for moving objects—a document, say, or an image—from one application to another. A student can pluck a photo off of a Web site by clicking on it and dragging it to the left side of the frame. Then, after she launches another activity on the display screen, she can click on the icon for the photo and drag it onto the screen. The drop-off spot on the frame is conceived as a "pocket" that the kids can use to carry around things they want to use later.
The Journal, another nifty feature, allows youngsters to create a record of what they did with the software and what they thought about it that can later be shared with others. Once a Journal reaches a certain size, earlier entries are automatically shifted to a more powerful computer on the network. Documents and photographs can be stored in the same way.
While the Sugar team hasn't run any formal usability tests yet, Bender et al. have received unscientific feedback from a number of children who have tried the machine. Typically, says Bender, the kids get totally absorbed and disappear for hours. Now he’s glad to be able to test Sugar in real-world situations. The first 2500 Beta 2 machines are being distributed to schools in participating countries, and OLPC will finally receive more substantial user feedback. "I’m sure we have the basics down, but we can still make changes," he says. "We want to get this right."