On March 10, 2011, I was in Cambridge at the MIT Media Lab meeting with faculty, students and staff, and we were trying to figure out whether I should be the next director.
That night, at midnight, a magnitude 9 earthquake hit off of the Pacific coast of Japan. My wife and family were in Japan, and as the news started to come in, I was panicking. I was looking at the news streams and listening to the press conferences of the government officials and the Tokyo Power Company, and hearing about this explosion at the nuclear reactors and this cloud of fallout that was headed towards our house which was only about 200 kilometers away. And the people on TV weren't telling us anything that we wanted to hear. I wanted to know what was going on with the reactor, what was going on with the radiation, whether my family was in danger.
So I did what instinctively felt like the right thing, which was to go onto the Internet and try to figure out if I could take matters into my own hands. On the Net, I found there were a lot of other people like me trying to figure out what was going on, and together we sort of loosely formed a group and we called it Safecast, and we decided we were going to try to measure the radiation and get the data out to everybody else, because it was clear that the government wasn't going to be doing this for us.
Three years later, we have 16 million data points, we have designed our own Geiger counters that you can download the designs and plug it into the network. We have an app that shows you most of the radiation in Japan and other parts of the world. We are arguably one of the most successful citizen science projects in the world, and we have created the largest open dataset of radiation measurements.
And the interesting thing here is how did — (Applause) — Thank you. How did a bunch of amateurs who really didn't know what we were doing somehow come together and do what NGOs and the government were completely incapable of doing? And I would suggest that this has something to do with the Internet. It's not a fluke. It wasn't luck, and it wasn't because it was us. It helped that it was an event that pulled everybody together, but it was a new way of doing things that was enabled by the Internet and a lot of the other things that were going on, and I want to talk a little bit about what those new principles are.
So remember before the Internet? (Laughter) I call this B.I. Okay? So, in B.I., life was simple. Things were Euclidian, Newtonian, somewhat predictable. People actually tried to predict the future, even the economists. And then the Internet happened, and the world became extremely complex, extremely low-cost, extremely fast, and those Newtonian laws that we so dearly cherished turned out to be just local ordinances, and what we found was that in this completely unpredictable world that most of the people who were surviving were working with sort of a different set of principles, and I want to talk a little bit about that.
Before the Internet, if you remember, when we tried to create services, what you would do is you'd create the hardware layer and the network layer and the software and it would cost millions of dollars to do anything that was substantial. So when it costs millions of dollars to do something substantial, what you would do is you'd get an MBA who would write a plan and get the money from V.C.s or big companies, and then you'd hire the designers and the engineers, and they'd build the thing. This is the Before Internet, B.I., innovation model. What happened after the Internet was the cost of innovation went down so much because the cost of collaboration, the cost of distribution, the cost of communication, and Moore's Law made it so that the cost of trying a new thing became nearly zero, and so you would have Google, Facebook, Yahoo, students that didn't have permission — permissionless innovation — didn't have permission, didn't have PowerPoints, they just built the thing, then they raised the money, and then they sort of figured out a business plan and maybe later on they hired some MBAs. So the Internet caused innovation, at least in software and services, to go from an MBA-driven innovation model to a designer-engineer-driven innovation model, and it pushed innovation to the edges, to the dorm rooms, to the startups, away from the large institutions, the stodgy old institutions that had the power and the money and the authority. And we all know this. We all know this happened on the Internet. It turns out it's happening in other things, too. Let me give you some examples.
So at the Media Lab, we don't just do hardware. We do all kinds of things. We do biology, we do hardware, and Nicholas Negroponte famously said, "Demo or die," as opposed to "Publish or perish," which was the traditional academic way of thinking. And he often said, the demo only has to work once, because the primary mode of us impacting the world was through large companies being inspired by us and creating products like the Kindle or Lego Mindstorms. But today, with the ability to deploy things into the real world at such low cost, I'm changing the motto now, and this is the official public statement. I'm officially saying, "Deploy or die." You have to get the stuff into the real world for it to really count, and sometimes it will be large companies, and Nicholas can talk about satellites. (Applause) Thank you. But we should be getting out there ourselves and not depending on large institutions to do it for us.
So last year, we sent a bunch of students to Shenzhen, and they sat on the factory floors with the innovators in Shenzhen, and it was amazing. What was happening there was you would have these manufacturing devices, and they weren't making prototypes or PowerPoints. They were fiddling with the manufacturing equipment and innovating right on the manufacturing equipment. The factory was in the designer, and the designer was literally in the factory. And so what you would do is, you'd go down to the stalls and you would see these cell phones. So instead of starting little websites like the kids in Palo Alto do, the kids in Shenzhen make new cell phones. They make new cell phones like kids in Palo Alto make websites, and so there's a rainforest of innovation going on in the cell phone. What they do is, they make a cell phone, go down to the stall, they sell some, they look at the other kids' stuff, go up, make a couple thousand more, go down. Doesn't this sound like a software thing? It sounds like agile software development, A/B testing and iteration, and what we thought you could only do with software kids in Shenzhen are doing this in hardware. My next fellow, I hope, is going to be one of these innovators from Shenzhen.
And so what you see is that is pushing innovation to the edges. We talk about 3D printers and stuff like that, and that's great, but this is Limor. She is one of our favorite graduates, and she is standing in front of a Samsung Techwin Pick and Place Machine. This thing can put 23,000 components per hour onto an electronics board. This is a factory in a box. So what used to take a factory full of workers working by hand in this little box in New York, she's able to have effectively — She doesn't actually have to go to Shenzhen to do this manufacturing. She can buy this box and she can manufacture it. So manufacturing, the cost of innovation, the cost of prototyping, distribution, manufacturing, hardware, is getting so low that innovation is being pushed to the edges and students and startups are being able to build it. This is a recent thing, but this will happen and this will change just like it did with software.
Sorona is a DuPont process that uses a genetically engineered microbe to turn corn sugar into polyester. It's 30 percent more efficient than the fossil fuel method, and it's much better for the environment. Genetic engineering and bioengineering are creating a whole bunch of great new opportunities for chemistry, for computation, for memory. We will probably be doing a lot, obviously doing health things, but we will probably be growing chairs and buildings soon. The problem is, Sorona costs about 400 million dollars and took seven years to build. It kind of reminds you of the old mainframe days. The thing is, the cost of innovation in bioengineering is also going down. This is desktop gene sequencer. It used to cost millions and millions of dollars to sequence genes. Now you can do it on a desktop like this, and kids can do this in dorm rooms. This is Gen9 gene assembler, and so right now when you try to print a gene, what you do is somebody in a factory with pipettes puts the thing together by hand, you have one error per 100 base pairs, and it takes a long time and costs a lot of money. This new device assembles genes on a chip, and instead of one error per 100 base pairs, it's one error per 10,000 base pairs. In this lab, we will have the world's capacity of gene printing within a year, 200 million base pairs a year. This is kind of like when we went from transistor radios wrapped by hand to the Pentium. This is going to become the Pentium of bioengineering, pushing bioengineering into the hands of dorm rooms and startup companies.
So it's happening in software and in hardware and bioengineering, and so this is a fundamental new way of thinking about innovation. It's a bottom-up innovation, it's democratic, it's chaotic, it's hard to control. It's not bad, but it's very different, and I think that the traditional rules that we have for institutions don't work anymore, and most of us here operate with a different set of principles. One of my favorite principles is the power of pull, which is the idea of pulling resources from the network as you need them rather than stocking them in the center and controlling everything.
So in the case of the Safecast story, I didn't know anything when the earthquake happened, but I was able to find Sean who was the hackerspace community organizer, and Peter, the analog hardware hacker who made our first Geiger counter, and Dan, who built the Three Mile Island monitoring system after the Three Mile Island meltdown. And these people I wouldn't have been able to find beforehand and probably were better that I found them just in time from the network.
I'm a three-time college dropout, so learning over education is very near and dear to my heart, but to me, education is what people do to you and learning is what you do to yourself.
And it feels like, and I'm biased, it feels like they're trying to make you memorize the whole encyclopedia before they let you go out and play, and to me, I've got Wikipedia on my cell phone, and it feels like they assume you're going to be on top of some mountain all by yourself with a number 2 pencil trying to figure out what to do when in fact you're always going to be connected, you're always going to have friends, and you can pull Wikipedia up whenever you need it, and what you need to learn is how to learn. In the case of Safecast, a bunch of amateurs when we started three years ago, I would argue that we probably as a group know more than any other organization about how to collect data and publish data and do citizen science.
Compass over maps. So this one, the idea is that the cost of writing a plan or mapping something is getting so expensive and it's not very accurate or useful. So in the Safecast story, we knew we needed to collect data, we knew we wanted to publish the data, and instead of trying to come up with the exact plan, we first said, oh, let's get Geiger counters. Oh, they've run out. Let's build them. There aren't enough sensors. Okay, then we can make a mobile Geiger counter. We can drive around. We can get volunteers. We don't have enough money. Let's Kickstarter it. We could not have planned this whole thing, but by having a very strong compass, we eventually got to where we were going, and to me it's very similar to agile software development, but this idea of compasses is very important.
So I think the good news is that even though the world is extremely complex, what you need to do is very simple. I think it's about stopping this notion that you need to plan everything, you need to stock everything, and you need to be so prepared, and focus on being connected, always learning, fully aware, and super present.
So I don't like the word "futurist." I think we should be now-ists, like we are right now.
"Remember before the internet?" asks Joi Ito. "Remember when people used to try to predict the future?" In this engaging talk, the head of the MIT Media Lab skips the future predictions and instead shares a new approach to creating in the moment: building quickly and improving constantly, without waiting for permission or for proof that you have the right idea. This kind of bottom-up innovation is seen in the most fascinating, futuristic projects emerging today, and it starts, he says, with being open and alert to what's going on around you right now. Don't be a futurist, he suggests: be a now-ist.
Joi Ito is the director of the MIT Media Lab.