Complexity Now!

By: Jason Moriber
Mod Robotics makes toys for children to help them thrive in our complex world

The vision for this series is to surface and share insights of thought-leaders and trailblazers who live at the cutting edge of technology. While the opinions featured may not necessarily represent those of Verizon and its employees, we still believe that we can each learn from experiences and opinions of others, which is why we’ve chosen to feature them here. This dialogue is how we take the first steps towards making innovations that matter.

Though there always seems to be a cultural momentum towards simplicity, Eric Schweikardt knows the world is growing ever more complex. From his point of view this means rather than run from complexity and strive for simplicity, we humans need to adapt and learn how to better thrive in an ever increasingly complex world, and to realize the world was never simple, nor will it ever be. For Eric, the best thing to do is prepare our children in ways that are fun and engaging. The solution … toy robots that look like building blocks. But these blocks are “smart.”

Eric and I sat outside for our conversation, under an overhang of a former postal building in an industrial section of Boulder, Colo. This building is both the HQ and the manufacturing plant for the toys. They are all made right there in Boulder.

Eric grabs two handfuls of the robot blocks and places them on the picnic table we share. The sun streams down between the roof angles. It’s much warmer in the sun than in the shade.

“What do you call these?” I ask.

Modular components are the core

“Cubelets,” Eric replies with a giddy smile. He has a slight Peter Pan glint in his eye, but with a Ph.D. on his resume. He begins to introduce the different types of cubelets. “Everybody loves blocks like Legos and those construction kits. I think lots of creative people played with construction kits and this is what happens when you put robot stuff inside of a construction kit. So battery Cubelet… flashlight Cubelet… and distance sensors… so that’s the distance-sensor Cubelet. ”

“So each block has its own function?”

“Each little Cubelet has a tiny computer inside of it. And categories, all of the clear ones are action blocks; they do things. All the black ones are sensor blocks; they sense things like our senses do. So just with sensor blocks and objects are neighbors. I’m not sensing anything. I’m sensing something far away and then I’m sensing something close. So action blocks have those types of data values, they light up. Especially in bright sun. Flashlights are a warning, wheely block, and a distance sensor, battery block….

Eric starts clicking blocks together then unclicking them into different configurations. Based on how the blocks are put together, the robots behave in a different way.

…Soon as the distance sensor detects something it starts driving. Flipping around the other side as soon as the distance sensor detects something and scoots around the other way. Simple 1-to-1 reactive robots with flashlight on the front. It does its thing… but it can get complicated. We can start to do things that most people would consider programming. What if I want a robot that drives and then stops before it hits something? I’ll take off the flashlight and put the distance sensor on the side and now I have a robot that when it drives, it flips the behaviors and, I’ll take a pink inverse block, put it in between the sensor and the actuator, now I have a robot that drives until it sees something and then it stops. So to little kids, it looks like I just built it and made it to do what I wanted it to do, like I programmed; in fact, whenever kids or adults build robots, that would follow the same paradigm. You build platforms with your metal robot body and you shift your attention to a computer and you nerd out and you write programming.”


Mixing and matching blocks creates new robot brains

“We just got a robot to do what we wanted. To switch its behavior we just built a robot by snapping some physical blocks together. By snapping the physical blocks together, you’re obviously building the body of the robot, but you’re also building the code. And it’s a different model of building robotics for every other robot out there. But it’s a model of complexity; it’s a model of systems in the world where lots of little modules we understand are doing their own thing and then we zoom back and clear our eyes a little bit and go, ‘Whoa  we have a tough environment, we got a bad economy, we have corruption and strife in the Middle East.’ This emergent behavior happens and we have a hard time understanding complex systems and emergent behaviors. And we have a hard time understanding how like millions of different variables contribute to making an environment that’s not doing so well. Or how millions of individual actors are acting and doing smart things, but then we zoom back on a bad economy? It’s really hard to think about complex systems…like a flock of birds, everyone thinks the lead bird is in charge. That’s the boss bird. It’s actually not like that.  It’s about how aerodynamic they are all, looking to get together to solve problems in those big systems. We need to get smarter about them and we can get smarter about them by getting Ph.D.’s in complexity science and going all Greek letters and going about that, or we can get smarter by going to little kids, giving toys to little kids that shape intuitions.”

I’m suddenly reminded of conversations I had about Punk Rock back in 90’s. About how Punk wasn’t a style as much as a collection of diverse thinkers seeking to do things differently, on a new set of terms that matched closer to the reality of the time versus an ideal version of a past age, and disrupting the status quo to make positive impact. I realize Eric is the most Punk person I’ve met in a while.

And kids intuitively understand this

Eric continues, “So they can begin to gain intuition and build complex systems so that when they’re older, it won’t come as a shock that there are rarely simple answers in the world and that it’s rarely black and white, good or bad, red or blue. And that the world is actually more complicated.”

“Have you given up on adults?” I ask half-joking.

“I’ve given up on adults,” Eric replies straight faced, then cracks a smile, “Not seriously, but kids are malleable, adults are stuck in their ways. If someone has made up their minds, like if they don’t believe in evolution, it’s impossible to change that. You can have a really great scientific talk, but… I had a neighbor, Cathy, who would smoke cigarettes and talk to me over a chain-link fence when I was living in Pittsburgh and her gripe with evolution was, “Well look around, look at all this magic, these Audis and love and poetry, computers and politics, that’s too perfect and unique to happen by itself, it needs to be designed.” So how do you talk to someone like Cathy about that? You can give her scientific explanations, you can give her primordial soups and large numbers about how this stuff can come out of nothing and kudos if you want to listen to that. With kids you don’t have to do that. You can just give kids experience and intuitions about how behavior of a robot drives until it falls off the edge of the table … how you can build something like that with some of the agents you can understand has some of the complex behavior that we see in the real world. That’s a better way to affect change in the world.”

“So when you say kids, what age group?”

Real life is not robotic like that, real life is a little more messy, a little distributed, and a lot more complicated.

“Starting as young as four. There are a couple of ways to play for little kids, four-year-olds. Snapping and building and making stuff light up and drive around the table, that’s cool. Older kids can be a little more self-directed. Older kids, middle school-super motivated or high school and college, you can go in and program any or all of the Cubelets. You can go back and make something brand new, which is cool! Because typically when you build a robot, you build your one robot, plastic or metal body. And you write your one program on what the robot should do. Drive forward, turn 90 degrees, drive forward, one meter, and open the fridge, whatever top down very God-like program. You can reprogram one Cubelet, you can reprogram all the Cubelets. You can reprogram one Cubelet and see how it has a ripple effect that changes the whole robot. You can revert that Cubelet to stop, you can program other Cubelets and see if that can affect the same change in the world as that other Cubelet that you reprogrammed. It’s an experiment where it’s not as simple as turn right 90 degrees. Real life is not robotic like that, real life is a little more messy, a little distributed, and a lot more complicated.”

“I asked about the middle school kids because that’s with a lot of programs what I’m finding, that they’re focusing on the middle school kids than the younger kids.”

“Middle school is at a time where kids are trying on lots of different personalities. Kids would circle through personalities. And by the time kids get to like high school, at least for like gender and STEM, by the time kids get to high school if girls for example decided that they’re not going to be scientists, it’s kind of too late. A lot of our work at least in getting to kids early and preparing them for more technical careers, the more innovative and more based careers and to get them to middle school.”

“So in the bigger picture…”

The future requires complex thinking

“What’s next for Cubelets and what’s next for mobile in general and tech and IoT is figuring out how to manage complexity. So we’ve gotten really good at making individual devices. And we got them all over the place, individual purpose devices and we’re starting to see backlashes with them now like smart homes; everyone’s frustrated with smart home. Nest is reaching into your house and disabling your fridge and like all this crap. Nothing works together; you have to be a programmer to get things to work. And typically people respond to that kind of system and desire to make things simple or we’re going to simplify your home automation or we’re going to simplify this for you, beginning with mobile phone, we’re going to simplify this for you we’ll give you a wallet card and etcetera, but that doesn’t really work. When you simplify a complex system, you end up with like stupid ham-fisted solutions and problems that really require detail and analysis and nuance and a smart system policy and answer and a really good solution. That’s my definition of good design. A good designer is someone that can create a really nuanced complex solution. There are other solutions in which we develop other devices that speak a bunch of different languages and embrace that complexity and embrace the notion that there are lots and lots of different stuff that you need to deal with and that’s what we’re trying to do with our products. To give kids exposure to that and let them figure out how things work and expose them to, “Hey I got this robot I want them to change this behavior by changing this one block and seeing how it has a ripple effect throughout the whole construction.”

“And what advice would you give to an entrepreneur reading this story, someone who is excited by what your saying, how do they get started?”

My first piece of advice on getting anything done is make something. Just make something.

How can we get started making our own complex solutions into reality?

“My first piece of advice on getting anything done is make something. Just make something. Ideas… and this might sound jaded or strange, but ideas are worthless. As far as that’s concerned everyone’s got ideas. I have a million ideas a day. I know you do. Everybody on my team has a million ideas a day sifting through those like every great idea; somebody’s had a million times before. All these great ideas and an idea don’t do anything. I’ll never sign an NDA with somebody, “Hey I want to tell you about my thing where you got to sign an NDA.” Tell me your thing. So have an idea, that’s great. The next thing would be to make something; prototype of your idea. If you’re inventing a new type of electric car, you don’t have to build the electric car from scratch to get conversation started, but you can build prototypes of the one innovation that you think will make your electric car much cooler than the other. You got an idea for an app; you don’t have to build the website. You’re not a coder that’s still fine. But you can build some story boards and you can build some screens and you can talk to some people. In terms of getting started, and getting your foot in the door and making something, and be able to talk about that thing that you just made, is an ideal conversation starter. And that comes from kids too, it comes from what we’re doing and it comes from the legacy of constructivism and constructionism, and education like learning by making stuff. That’s how we all learn…by failing early, by building something, by testing, and by trying it out. And for some reason, especially entrepreneurial communities, you hear about a lot of people who it never occurred to them to make something. They’re talking about getting off the ground. They don’t have any traction because everybody’s trying to figure out these ideals and they’re all over the place. But as soon as you make something, you can talk about it with somebody and it’s extremely powerful.”

“And your ideal state?”

“I guess, I mean our mission is to make the world a better place with millions of tiny robots. That’s what we’re here for.”

About the author(s): 

Jason Moriber is a creative communicator with a background in social and digital for CSR, tech and start-ups. He’s working within the Communications team at Verizon, charged with developing a new model for corporate and brand communications. Connect with him on Twitter @jasonmoriber or on Instagram @jasonmoriber.