Fun Fact: Either the school you are currently attending or the last school you attended
Ask Zivthan Dubrovsky what it takes to turn kids into lifelong coders and he’ll mention a spark.
Dubrovsky leads a bio-inspired robotics team at Harvard’s Wyss Institute, and knows a lot about spark. As he explains it, it’s an event that happens early in a person’s experience with programming or computers—a formative experience that inspires a desire to learn that comes from within. He’s pretty sure he’s figured out a way of creating that spark, too: Root, an educational robot designed to teach kids—and adults—how to code.
“We’re confident that we have the right product and the right interface,” he says. “Now our next step is to share this with the world.”
Root is a small, squat hexagon, a cheerful and friendly ‘bot reminiscent of the robot vacuum-cleaning Roomba. Scanners dot its bottom, bumpers line its sides, and there’s a touch interface on top. Root knows when it’s on a magnetic surface. A little holder for a dry-erase marker sits dead center on the bot, which can draw on a whiteboard and erase it, too. Using that whiteboard, an iPad, and a language that’s super-easy to pick up, kids learn to code by teaching Root how to interact with the world.
Root joins a long line of tools, from simpler programming languages to children’s books that teach coding concepts, that the code literacy movement has created in recent years. Even the idea of an educational robot isn’t new. Two years ago, Harvard researchers developed the AERobot, a $10 programmable robot meant to get kids excited about science, technology, engineering, and math.
But Root differs from those efforts. The AERobot, for one, was primarily an academic endeavor. Root was designed with consumers in mind. Dubrovsky’s team enjoys a unique position at Harvard. Though immersed in academia, it has a mandate to consider how it can bring products to market—and team members have deep industry expertise. (Dubrovsky himself did stints at Sonos and iRobot, which makes the Roomba.) Now, after three years of heads-down research and stealthy product development, Dubrovsky says Root is ready for the real world.
“We’ve met with teachers, we’ve gone into classrooms,” Dubrovsky says. It’s time now, he says, to prepare Root for the world beyond Harvard.
Drawing kids into coding isn’t an abstract mission. Computing jobs are growing at twice the national rate of other types of employment. By 2020, the Bureau of Labor Statistics says, the US will have 1 million more computer science-related jobs than graduates qualified to fill them. In December, President Obama announced the Computer Science for All Initiative, pledging $4 billion in funding for computer science education in the nation’s schools.
Yet all kinds of dysfunction keeps the country from closing the deficit in computer science talent, according to a survey by Google and Gallup. Yes, school budgets are a problem, and teachers have a limited time to devote to additional classes. But schools still don’t recognize the scale of demand for computer science skills.
At the same time, another study by Google found that high school students with an opportunity to take an advanced-placement computer science exam were 46 percent more likely to show interest in majoring in computer science. In other words, exposing kids to coding before college is crucial to closing the gap.
To be fair, there is no shortage of kid-friendly robots, and some of them are pretty cool—Ozobot, Dot and Dash, and Sphero come to mind. But their ability to inspire kids to code has boundaries. “Kids find the limits of things really quickly,” says Dubrovsky. His team’s research included playing with every toy robot it found. And the team discovered that kids are very good at playing with a robot, figuring out exactly what it can do, then losing interest and moving on.
The challenge, then, was to create a bot that keeps kids engaged. Dubrovsky and his team decided to stuff Root with sensors, components, and features that would allow it to accomplish a wide variety of tasks. “What we had in mind was having something have really high limits. You can continue to unpeel the layers of what you can do with the robot,” Dubrovsky says.
Start Their Engines
Out of the box, Root works with an iPad and a whiteboard. The software language created for Root is a simple, graphics-based interface that follows an “if-this-then-that” framework. Often, Dubrovsky says, the first thing kids try to do is stick Root on a whiteboard and create a simple program on the iPad to control its direction—tilt the tablet left to make Root go left, tilt it right to make it go right, that sort of thing. It’s also possible to program Root to light up and play simple notes. It can detect nearby Wi-Fi devices, and has light-sensitive sensors. An expansion port allows adding simple hardware add-ons.
But the really neat part is the stacking of commands. Draw a racetrack on the whiteboard, for instance, and Root robots can be programmed to race each other. You can add other behaviors like getting Root to bounce off a “barrier” at the edge of the track. You can even program Root’s motors to accelerate when it drives over the color green or stop when it drives over red.
“Kids can create this very rich experience, with a completely free-form space like a whiteboard, and code with each other,” Dubrovsky says. “They think through the sequencing, the function, the variables and the priorities.”
Root in Real Life
Shay Pokress, a computer science curriculum developer and lead writer on the National Framework for K-12 Computer Science, got an early preview of Root and was impressed by its functionality. At the same time, she sympathizes with the pressures teachers face as they try to do so much each schoolday. “[Computer science] can’t be an add-on,” she says. “We can’t tell teachers, ‘Hey, make sure students are ready for the common core math test, and also add in two hours of computer science.’”
But the beauty of Root—and educational robots as a whole, according to Pokress—is teachers can integrate it into existing lessons. She suggests some wacky, but feasible, ideas, like using Root robots to recreate historical events. More obviously, Root has a place in math and science classes. If teachers are willing to bring Root into their classrooms, Pokress says we could very well see the little bot gain traction. “It comes down to great curriculum material, and great training for teachers,” Pokress says. “And if teachers see success from early adopters, well, maybe Root can be successful.”
Of course, Root’s sophistication and ease of use doesn’t mean much if it isn’t affordable. Dubrovsky and his team drove down costs to around $200, about the cost of a college-level textbook. That gives it a shot of rolling into classrooms and homes. Now it’s up to someone else to help make that happen. “Root is production-ready,” says Dubovsky. “This is our press release to the world to say, ‘Hey world, do you think this is important?’ If the world says yes, we’re hoping the right investors come to us and help us launch.” If someone steps up, Root could reach consumers by the middle of next year.