INNOVATION ON THE EDGES

INNOVATION ON THE EDGES

Joichi Ito [6.21.12]

Today, what you want is to have resilience and agility, and you want to be able to participate in, and interact with the disruptive things. Everybody loves the word "disruptive innovation." Well, how and where does disruptive innovation happen? It doesn't happen in the big planned R&D labs; it happens on the edges of the network. Most important ideas, especially in the consumer Internet space, but more and more now in other things like hardware and biotech, you're finding it happening around the edges.

JOI ITO is MIT Media Lab Director and a leading thinker and writer on innovation, global technology policy, and the role of the Internet in transforming society in substantial and positive ways. A vocal advocate of emergent democracy, privacy, and Internet freedom, Ito is board chair (and former CEO) of Creative Commons, and sits on the Boards of The New York Times Company, the John D. and Catherine T. MacArthur Foundation, the Knight Foundation, Mozilla Foundation, WITNESS, and Global Voices. 

Joichi Ito's Edge Bio Page


 


INNOVATION ON THE EDGES

[JOI ITO:] I grew up in Japan part of my life, and we were surrounded by Buddhists. If you read some of the interesting books from the Dalai Lama talking about happiness, there's definitely a difference in the way that Buddhists think about happiness, the world and how it works, versus the West. I think that a lot of science and technology has this somewhat Western view, which is how do you control nature, how do you triumph over nature? Even if you look at the gardens in Europe, a lot of it is about look at what we made this hedge do.

What's really interesting and important to think about is, as we start to realize that the world is complex, and as the science that we use starts to become complex and, Timothy Leary used this quote, "Newton's laws work well when things are normal sized, when they're moving at a normal speed." You can predict the motion of objects using Newton's laws in most circumstances, but when things start to get really fast, really big, and really complex, you find out that Newton's laws are actually local ordinances, and there's a bunch of other stuff that comes into play.

One of the things that we haven't done very well is we've been looking at science and technology as trying to make things more efficient, more effective on a local scale, without looking at the system around it. We were looking at objects rather than the system, or looking at the nodes rather than the network. When we talk about big data, when we talk about networks, we understand this.

I'm an Internet guy, and I divide the world into my life before the Internet and after the Internet. I helped build one of the first commercial Internet service providers in Japan, and when we were building that, there was a tremendous amount of resistance. There were lawyers who wrote these big articles about how the Internet was illegal because there was no one in charge. There was a competing standard back then called X.25, which was being built by the telephone companies and the government. It was centrally-planned, huge specifications; it was very much under control.

The Internet was completely distributed. David Weinberger would use the term 'small pieces loosely joined.' But it was really a decentralized innovation that was somewhat of a kind of working anarchy. As we all know, the Internet won. What the Internet winning was, was the triumph of distributed innovation over centralized innovation. It was a triumph of chaos over control. There were a bunch of different reasons. Moore's law, lowering the cost of innovation—it was this kind of complexity that was going on, the fact that you could change things later, that made this kind of distributed innovation work. What happened when the Internet happened is that the Internet combined with Moore's law, kept on driving the cost of innovation lower and lower and lower and lower. When you think about the Googles or the Yahoos or the Facebooks of the world, those products, those services were created not in big, huge R&D labs with hundreds of millions of dollars of funding; they were created by kids in dorm rooms.

In the old days, you'd have to have an idea and then you'd write a proposal for a grant or a VC, and then you'd raise the money, you'd plan the thing, you would hire the people and build it. Today, what you do is you build the thing, you raise the money and then you figure out the plan and then you figure out the business model. It's completely the opposite, you don't have to ask permission to innovate anymore. What's really important is, imagine if somebody came up to you and said, "I'm going to build the most popular encyclopedia in the world, and the trick is anyone can edit it." You wouldn't have given the guy a desk, you wouldn't have given the guy five bucks. But the fact that he can just try that, and in retrospect it works, it's fine, what we're realizing is that a lot of the greatest innovations that we see today are things that wouldn't have gotten approval, right?

The Internet, the DNA and the philosophy of the Internet is all about freedom to connect, freedom to hack, and freedom to innovate. It's really lowering the cost of distribution and innovation. What's really important about that is that when you started thinking about how we used to innovate was we used to raise money and we would make plans. Well, it's an interesting coincidence because the world is now so complex, so fast, so unpredictable, that you can't. Your plans don't really work that well. Every single major thing that's happened, both good and bad, was probably unpredicted, and most of our plans failed.

Today, what you want is you want to have resilience and agility, and you want to be able to participate in, and interact with the disruptive things. Everybody loves the word 'disruptive innovation.' Well, how does, and where does disruptive innovation happen? It doesn't happen in the big planned R&D labs; it happens on the edges of the network. Many important ideas, especially in the consumer Internet space, but more and more now in other things like hardware and biotech, you're finding it happening around the edges.

What does it mean, innovation on the edges? If you sit there and you write a grant proposal, basically what you're doing is you're saying, okay, I'm going to build this, so give me money. By definition it's incremental because first of all, you've got to be able to explain what it is you're going to make, and you've got to say it in a way that's dumbed-down enough that the person who's giving you money can understand it. By definition, incremental research isn't going to be very disruptive. Scholarship is somewhat incremental. The fact that if you have a peer review journal, it means five other people have to believe that what you're doing is an interesting thing. Some of the most interesting innovations that happen, happen when the person doing it doesn't even know what's going on. True discovery, I think, happens in a very undirected way, when you figure it out as you go along.

Look at YouTube. First version of YouTube, if you saw 2005, it's a dating site with video. It obviously didn't work. The default was I am male, looking for anyone between 18 and 35, upload video. That didn't work. They pivot it, it became Flicker for video. That didn't work. Then eventually they latched onto Myspace and it took off like crazy. But they figured it out as they went along. This sort of discovery as you go along is a really, really important mode of innovation. The problem is, whether you're talking about departments in academia or you're talking about traditional sort of R&D, anything under control is not going to exhibit that behavior.

If you apply that to what I'm trying to do at the Media Lab, the key thing about the Media Lab is that we have undirected funds. So if a kid wants to try something, he doesn't have to write me a proposal. He doesn't have to explain to me what he wants to do. He can just go, or she can just go, and do whatever they want, and that's really important, this undirected research.

The other part that's really important, as you start to look for opportunities is what I would call pattern recognition or peripheral vision. There's a really interesting study, if you put a dot on a screen and you put images like colors around it. If you tell the person to look at the dot, they'll see the stuff on the first reading, but the minute you give somebody a financial incentive to watch it, I'll give you ten bucks to watch the dot, those peripheral images disappear. If you've ever gone mushroom hunting, it's a very similar phenomenon. If you are trying to find mushrooms in a forest, the whole thing is you have to stop looking, and then suddenly your pattern recognition kicks in and the mushrooms pop out. Hunters do this same thing, archers looking for animals.

When you focus on something, what you're actually doing is only seeing really one percent of your field of vision. Your brain is filling everything else in with what you think is there, but it's actually usually wrong, right? So what's really important when you're trying to discover those disruptive things that are happening in your periphery. If you are a newspaper and you're trying to figure out what is the world like without printing presses, well, if you're staring at your printing press, you're not looking at the stuff around you. So what's really important is how do you start to look around you?

Well, one thing is you've got to be anti-disciplinary because if you're in a discipline and you're worried about peer review and you're knowing more and more about less and less, that's by definition an incremental thing. You've got to be anti-disciplinary in the way that you've got to have the freedom to connect things together that aren't traditionally connected. You get that ironically by not being rigorous, and by building. It's practice before theory.

We're right near a very famous university. A lot of really famous university people I know will, if they run an experiment based on a theory, like economists do this all the time, and the data doesn't match the theory, they start questioning reality. I think it's got to be the opposite. You don't need to have theory. If a thing works, you sort of figure it out later. It's kind of like innovation by the seat of your pants. I don't think everybody should be using it. You can't only realize peripheral vision. You don't want the mushroom hunter to not be looking at the road when they're driving. So it's okay if most of society has plans.                 

What's really important for the future of science and technology and the future of just the society, is that people need to have the freedom to be able to have the peripheral vision, look at the innovation on the edges. Companies like Google have 20 percent time, and things like that, Scotch and all these other people who have been doing innovation, give a little bit of extra time for the people on the edges.

There is a skill to pattern recognition. You can't just give somebody a bunch of free time and then expect him or her to come up with brilliant ideas. There's a little bit of method to the madness. One of the things that I'm trying to understand and trying to turn into practice is like the mushroom hunter, how do you create a practice where you can unfocus and you can start picking out those really important patterns? Because I do think that more and more, as the focus with big data and everything else becomes pulling out patterns and important causalities out of massive amounts of information, it really requires a certain kind of intuition about patterns that right now goes against the grain of the somewhat planned and incremental nature of scholarship and research.

I have to be a little bit careful about what I say, now that I'm a university administrator at MIT, but I think that there are roles for everyone. If you're really trying to get to the bottom of something, you do need to drill down. We have lots and lots of departments at MIT who are really good at scholarship, at drilling deep and figuring out theoretically how things work, and then rendering those into practice. Intellectual property is really about turning ideas into practical methods. But ideas, by the way, aren't patentable. Ideas are really, really important, because ideas are often what inspire people to create intellectual property, people to do research, people to build the supercolliders to figure out whether the thing is actually good. There are different roles for different pieces of the institution.   

The whole idea of scholarship and rigor is really important, but I think it's way overemphasized because what's interesting is, I would call it, the obsession with assessments and the ability to measure. People are afraid. They're afraid of risk, they're afraid of not being able to measure things.

I'll talk about a related topic: I was with a bunch of foundation people recently. I was with John Palfrey, who ran the Harvard Library for a while, and now running Andover. We were talking about the future of libraries, and how libraries are really, really important. But many foundations give libraries money, and they use a metric, which is circulation. What does that mean? We are rewarding libraries for how good they are at warehousing books, and is that the right thing to be reinforcing, when we're trying to figure out the future of libraries. Shouldn't we be giving all the librarians money and say: we don't really know what we're looking for, but try something, and it might be that libraries become the center for getting people excited about learning. Maybe they're the center for peer learning. Maybe they're the center for all kinds of different things, but they're probably not going to be the center for physical book circulation.        

The problem is that you can't figure out what to measure for the success of a library until you know what you want the library to be. But how do you fund the design of a future of a library when you don't know what it is, right? This gets back to venture capital, which has the same thing, which is if you can measure it, it's probably not that interesting, because it means you already know what it is, right? And you can't really have real progress without venturing into those spaces where you don't understand it. Venture capitalists get this, right? Venture capitalists, if you look at the history of venture capital, all of the home runs typically are contested. They're the ones that almost didn't make it through investment committee. They're usually the ones that are a little bit outside of your investment thesis. In really good funds, the investors don't sit and try to get the companies to provide detailed reports of every single company. They're looking for the home runs. So the funds that had Facebook in the portfolio, none of the other companies matter, because Facebook knocked it out of the park. Do you ask how do you measure Facebook's success? No, of course not.

If you put in $100,000 and it turned into a billion dollars, of course it's a success, right? So what's really interesting is there's an incrementalism built into the whole notion of assessments. If you think about the way we do research and the way we do academia, counting patents and figuring out how many citations you have, and the whole scholarship of it all, it's important, but it's sort of incremental in definition, because it does require these assessments. When you try to build large institutions, and build patent portfolios or build jets, you do need to have order and assessments and promotions and things like that, but what's really important is, well, who thought about the idea of trying to fly a plane in the first place? Those people weren't trying to solve some sort of challenge that was given to them by some sort of RFP, right? What's important is that you separate those two things, and I think that they're just different roles.

What I am very interested in is how do you create a group of people who are given the complete freedom to come up with these crazy ideas. Most of them, just like a venture fund, are stupid ideas and aren't worth making. But when we find those few ideas, and we can demonstrate the importance of those ideas, then what happens is those scholars and those engineers and those companies then turn those things into intellectual property and products and things like that. In the Internet space, it turns out, since you can just write the software, and in three weeks you can build just about anything that minimum viable product, it's already happening in the venture space.

But when you start looking at things like biotech and hardware, where you need a little bit more infrastructure, you need a little bit more math and science. I think that an academic institution like the Media Lab is a good place for that kind of freethinking to happen, where people can just go down into the basement and build just about anything they want. I will add that we have a healthy respect for all the other institutions that turn our ideas into stuff. But I think that people underestimate the importance of the freethinking.

The Media Lab was created around 28 years ago. If you remember 28 years ago, we didn't have the Internet, we didn't have personal computers. Nicholas Negroponte was in the School of Architecture, and that was when they had those CAD computers, it was some of the first graphical computers came out of this whole sort of architecture machine group. Nicholas was a real visionary, and they were saying we're going to have these multimedia workstations on everybody's desk, and everybody said, no way that's not going to happen. If you look at the Internet and personal computers, a lot of the ideas that were floating around the Media Lab back then made their way into personal computers and social media and the Internet.   

The Media Lab had this tradition of, if the bandwagon starts to fill up, they get off the bandwagon and they move onto the next thing. So in a funny way they said, "oh well we helped invent the Internet, but let's go on to the next thing because we're done with the Internet." It turns out the Internet kept doing cool stuff, so they did deemphasize the Internet for a while. But another interesting thing is that the Internet DNA that I love so much, which is David Clark's term, 'rough consensus and running code,' which is very much just build it, and figure it out as you go along. That DNA turns out to be the same DNA that the Media Lab had. The Media Lab got it from a slightly different place because at the Media Lab, Nicholas was an architect, and in architecture, what you do is you build models and you look at the models. The models really help, because if you have a model of a building, the civil engineer and the artist and the architect can look at a model and have a rigorous discussion about it, even though they could never write an academic paper together. The idea of building stuff first and then talking about it is actually, there was the old term 'demo or die,' rather than 'publish or perish'. But this sort of de-emphasis of scholarship and the emphasis on building, that DNA is actually similar to the Internet.

The biggest difference is, when Nicholas and all of the faculty at the Media Lab started out (back when Nicholas had just written a book, Being Digital, and personal electronics were coming out), it was really about empowering the individual, it was about man and machine, it was about user interface, it was about TVs and consumer electronics and computers. And it really was about how do you take technology and augment the ability of an individual to do lots of stuff? It was really important because what happened is that's enabled this explosion of innovation because the cost of production, the cost of interaction with computers, bringing designers into the mix, that all happened around the Media Lab.

What's different now that we have the Internet is that the emphasis is no longer on man/machine; it's on the community, it's on the network. The pivot that we need to make at the Media Lab, and which is sort of happening at the student level already, is stop focusing on things and start focusing on the network. Stop focusing on individuals and start focusing on communities. Stop focusing on top-down and focus more on bottom-up. Stop focusing on single experts and start focusing on the Cloud. There's a bunch of fundamental changes that the Internet has brought to the nature of innovation, the nature of society.

It's interesting because I love Nicholas. I have dinner with him a couple of times a month, and he and I are really, really different, but we're really similar in certain ways. That "just do it" aggressive innovative and antiauthoritarian nature, I think makes us very similar. But we're very different in some ways, because Nicholas is somewhat, how would I put it? I guess somewhat top-down. He's very design-oriented. He grew up building beautiful objects. My life has been very bottom-up, very community-oriented, very Internet networking stuff. It's interesting when he and I talk because there are certain things that we completely disagree on, and certain things that we totally agree on. The big shift that we're doing at the Media Lab right now is to pivot from this man/machine individual thing to the network thing.        

It's going quite well. I think that Nicholas totally gets what I'm talking about, and I think that the faculty and the students know that the world is going that way. It's happening at multiple levels. So in the work, the Internet and the networks and things like that, have already happened. Most of the big data work that we're doing is about networks and things like that. The layer that hadn't really been Internet-enabled is now; I'm opening up all of our content, the meetings where we show our research used to be closed; I made them open. I'm inviting outsiders in. We started a blog, we started posting the videos on YouTube, and we've been opening up the lab using Creative Commons licenses.

I think it makes some of the sponsors a little bit nervous because many are focused on the intellectual property, but I'm trying to deemphasize intellectual and emphasize more the network that we're creating between the companies. I changed the name; I don't call them sponsors anymore. I call them 'members' because I want the members to feel like it's a club. One of the things I'm thinking about is firing members, and only having interesting members where they want to hang out with each other, and making it feel more like a community of people, or a network, rather than this hub and spoke style of "we give you money, so that you create intellectual property that we can take back to the lab and turn into a thing," I want it to be much more of: how do we design an ecosystem together.

Yesterday I gave a graduation talk to our MIT students that are graduating. It's been about a year. All my life, I've felt like a misfit. I was doing all kinds of things, and the metaphor that I used in my remarks was, you know the X-Men? Where you have all these kids who have these super powers, but they're all misfits in their little communities that they were in, and then suddenly they all show up at a school and they realize that there's a whole bunch of other people like them, and they suddenly bond. It turns out I'm head of the school now. The Media Lab was originally all the misfits of MIT and it was kind of a department of none of the above. It's still the same way, and all the kids are amazing. But they're all really different.

My background has been sort of like that, and I wasn't really a venture capitalist, I wanted to do music, I loved Creative Commons. But I had my finger in all kinds of pies, and I did nonprofit, for-profit, Japan, Middle East. It was tearing me apart. I felt like a global citizen, but I also felt like I wasn't anything. What I found that was really interesting for me personally, and that may or may not be interesting to you, but about two weeks ago I got an email from some student, and I try to reply to all emails. I went to sit down and write a long reply to this kid, and I said, wait a second. If I'm going to spend 30 minutes writing a reply to this kid from this other university, I've got to help some kid at the Media Lab. I suddenly got this shock. This is the first time in my life that I felt loyalty to one institution, like they were my family.

It made me feel creepy for a minute, because I am a really open person and I feel like I want to save the whole world, and I don't like patriotism and nationalism and loyalty and stuff like that. But then I started thinking, well, if this is my family, I take care of my family before I can help save the world. It was the first time in my life that I felt this sense of loyalty. It was interesting because it's the Media Lab, first of all, the Media Lab is interested in everything, so there isn't a single thing that I'm interested in that I can't do at the Media Lab. There's no excuse not to do whatever I want to do at the Media Lab. I'm responsible for an institution for the first time in my life. It's been a really interesting shift. I can bring everything in. I mean, Media Lab connects with Creative Commons; Media Lab connects with venture capital. I'm just helping all the students with their startups. There isn't a single thing that I've been doing all my life that doesn't have somebody at the Media Lab who's more interested than me, and is better educated about it than me. What the Media Lab has done that's been really amazing is I've taken all of these 45 years of just crazy networking, and it's solidified it into one place. That's been really, really interesting.

At this commencement, another really interesting thing that I noticed was the diversity of the parents. I walked through Harvard Square, and I have nothing against Harvard, but I noticed that most of the parents are kind of similar, they know each other and it's kind of junior and junior and junior. A lot of the kids are there because the parents went there. The Media Lab parents are completely random, and to me that's really interesting, because it shows the diversity of our students, but also shows the way that we select our students, which really had nothing to do with where they come from, but really about what they are.     

If you look at the students individually, I basically went through and have been having meals, lunches and dinners with every single student, either in small groups or individually, just to try to get to know them. It's pretty impressive: this is not statistically accurate, but I'd say half of the students have probably played musical environments semiprofessionally. I would say most of them have some design skills. Most of them have some sort of technical skill, they can write code. I think all of them know how to use basic machine equipment. All of them can build stuff, all of them have some sense of design, and lots of talent around art.                 

They're neurobiologists who play violin, or they're people who want to do opera, but have an obsession with sensors. One of the key things that we say for both faculty and students is, if you could do what you wanted to do in any other place, you shouldn't be at the Media Lab. We're doing a faculty search right now, we're trying to hire two new arts faculty. What's amazing is that all of the faculty that we are interested in don't have offers from anywhere else, or offers they're interested in. Because by definition, what we want them to do at the Media Lab, or what they'd want to do at the Media Lab, they couldn't do anywhere else. It really literally, not jokingly, but literally are the people who are misfits, who can't fit in anywhere else.

What's interesting is, when you put this collection of misfits together, you get this really interesting capability. The other point about this is it's not interdisciplinary. Interdisciplinary is you've got an engineer and a physicist and a designer and they work on a project together. That's an interdisciplinary project. An anti-disciplinary project is when you can't tell who the designer is and who the engineer is, and the engineer knows designing, and the person who's dancing is going to be the one also doing the molecular biology. That's what is important and unique about the Media Lab. It's really, really hard for a Media Lab person to get a normal job because we have a degree called 'media arts and science.' Now, what the hell is that? You don't have job listings for people who have media arts and science. The other thing is, every one of our graduates is so different. Just because they have that degree and they're from the Media Lab, you know what kind of DNA they have and roughly what kind of personality they have, but you have no idea what they're really good at. You know that they have certain skills, but you don't know what they're obsessed about. The Media Lab prepares kids for a certain type of role, like Chief Innovation Officer or Designer, but they end up in all kinds of places.

A lot of the companies that are members of the lab recruit our students, but a lot of the students go on to academics, they've gone on to become entrepreneurs, or social entrepreneurs. But we have a pretty good diversity of companies. We've got 75 companies that range from Northrup Grumman to Mars to ESPN to Hallmark to Hasbro to GSK, so it's incredibly diverse. Having said that, they're mostly corporate. One of the things that I'm working on at the Media Lab is to try to bring more philanthropists and foundations. The Knight Foundation, which I'm on the board of, funds the Center for Civic Media. I'm trying to bring different types of stakeholders in so that the students have access to opportunities in different places, but they're pretty good at finding those.

Media Lab students, by and large, don't have that much trouble finding jobs because the Media Lab has a brand. It turns out the world has a lot of misfits. I think we probably only accept like two percent of the students who apply, and sometimes even less. Some groups have 300 applicants for one position, and we benefit a lot from the MIT brand. MIT gives the parents the comfort that their student is going to MIT, but then they end up with the opportunity where they can do anything they want, and so we get a really interesting group of students.

We have our academic program is called Media Arts and Science. It's a Master's program, which is two years, and a PhD program, which is five years. We're weird in that we don't let people take our PhD without taking our Master's, because we're so weird that you kind of have to go through our indoctrination before you can do your PhD. We do have undergraduate students come in through the Undergraduate Research Program. We've had like 150 undergraduates that interact with the lab, and then we have 150 graduate students who are our main research team. Then we have 26 faculty members. Our whole ecosystem is about 400 scientists and researchers and staff, and 75 companies that work at the Media Lab.

We don't really do classes that much. One of the things that MIT allowed the Media Lab to do, which I think Nicholas and Jerome Wiesner pulled off, was that at MIT, normally labs have the fundraising capability and the money, and academic programs have the degrees. And it's like church and state, they're separated. The Media Lab is the only place where they're together, where we raise our own money and we have academic programs. One of the decisions that we made was that we were going to really cut down on classes. We don't have many classes at all, and most of the work is based on research, so its interest and research-driven learning. We do have some classes on how to use equipment and things like that. A lot of the students will take classes off-campus, and there are some classes in the Media Lab, like how to use the shop and stuff, that kids from other departments will take. There's a lot of collaboration at the sort of research project level. But the Media Lab is kind of a uniquely unstructured place.

Just to give you two specific examples, Cesar Hidalgo is one of my junior faculty. He's a physicist, but he's now applying his physics brain on the economy and has figured out a way (it's a big data problem) he took all of the import/export data of all the countries in the world, worked with the Kennedy School, with Ricardo Hausmann, but he did the math on this. He crunched it up and created this map of the product space of what countries make what product and came up with what's called a 'complexity index,' which is sort of how complex of products can a country make. It turns out this is a much better predictor of the competitiveness in the growth of countries than anything IMF uses, and better even than the World Economic Forum Competitiveness index. He used it just by applying math and physics to the economy. Now all of these banks love him because he's able to predict the future of countries and their growth. Now they've got faster computers that can do this at Cities. 

There's another group working with Sandy Pentland's group where they did, there's this thing called 'social investing.' It's like a social network and you can go on and you can buy gold or buy foreign currencies, and you can follow other people. So if I think you're good at making bets, I can follow you and just put money on you. And what it is, is like a social networking site where you bet your money based on whether you think somebody's a good investor or not. It turns out that if you don't follow anybody, you don't have as much information and your returns are low. Then your returns increase as you start participating in the social network, but after some point, everybody's following each other and it becomes a bubble and the returns go down. 

There's this sweet spot of where a social network should look like for the highest returns in the social investing thing. So one of the researchers, what they did was they figured out a way to tune the social network so that people are connected to not too many people, but enough people so that you hit the sweet spot. Every asset has a different sweet spot. It turns out that you can improve the returns by two or three percent just by tuning the way that the people are connected to each other, by giving bonuses and coupons and things like that. So these are two examples of how we applied math and science to problems in ways that economists would never be able to do to increase returns by a couple of percent. 

In the early days, when I invested in Six Apart and I had my blog and I blogged several times a day, I remember it was really old media versus new media. Before, it was blog versus the mainstream media, we would always cheer mainstream medias losses. But as I got a little bit older and as I saw the damage that we were causing, I started to feel really bad because I was very involved in trying to overthrow the Japanese government. I would march on the streets, and I was an activist because I thought the Japanese democracy was broken, and it was.

I realized how important independent media was in keeping democracy in check. In Japan, it was a failure, and I don't think, and I didn't think, that social media is going to replace mainstream journalism. It may eventually get there, but there's this really, really important social role for mainstream media. As part of that, that's why I joined the board of the Knight Foundation, because they were giving money to give to teach journalists about technology, to upgrade the mainstream media so that it could integrate more with social media and make it better. To me, that was really important. But I really realized that we need to somehow save the old media because without The New York Times and some of these mainstream media outlets and professional journalism, I think that even now the government is out of control, but it would get even worse. 

When The New York Times asked me to join the board, I think my first meeting with them was really interesting because they clearly get that they need to move to the next step. They have a lot of obvious legacy stuff. But the reason I did join was a couple of things. When I met them, and the board, and particularly Arthur, it was clear that they had the resolve to really try to change the newspaper and take it to the next step and embrace the online stuff. They're doing a CEO search right now, which I think will be critical for their future. I felt two things: I felt like an obligation to society to try my best to say what I think is one of the most important institutions.

The other thing is, (it's a tangent, but it's important) I was in Tunisia hanging out after the revolution. I had been there before, but hanging out after the revolution, and we're sitting around and there's like Berkman Fellows hanging out, they're all looking at America. And America's not working, right? And Jim Fishkin was telling me he did this thing, a deliberative democracy session in China. He was in some village in China and he asked a Chinese guy, "how have you found democracy?" He goes, "well, I don't think voting works, really." Jim says, "oh, of course it works." Then the Chinese guy says, "well, are you in a swing state?" Jim was like, "uh, no," and he said, "Isn't your district Gerrymandered?"  And Jim was like, "well, yes." And the Chinese guy said, "Do you really think your vote counts?"

What's really amazing is whether you're in a village in China, or you're in this activist meeting room that I was in, in Tunisia, they know everything that's going on in America, all of its problems, they know about the super PACs. Unless we get America working, the rest of the world, they're not going to follow America. To me, one of the key components of getting the American system working properly is a healthy, functioning, strong independent media. The New York Times is, I think, the key thing. Part of it is the challenge of really helping and thinking about the future of The New York Times from the perspective of venture capital and technology. But part of it, it almost feels like social obligation on my part, to try in the capacity that I can, to make America a role model for the rest of the world because otherwise, then, we're really screwed. 

I can't remember the year, but I was in Davos, and I was shouting at the top of my voice about how broken Japanese democracy was. I was with some friends, and this is when the economy was crashing, so everyone wanted to hear my revolutionary ideas about Japan. At dinner I sat next to Sadako Ogata, who was the high commissioner for refugees for the UN from Japan, she is amazing. He said, "Joichi that was a nice talk, but you know, it's really parochial of you to talk about Japanese democracy. The whole world's democracy is broken; why don't you look at the rest of the world?" She shocked me into realizing that I thought it was something big, but I was thinking small. Then I looked around and I said, okay, well, I kind of understand the U.S., I kind of understand Japan, there's lots of people working on U.S./Japan stuff. How could I butt myself out of my comfort zone and attack a real problem? Or do something really interesting.

So I visited Bahrain's first conference, and I looked around in that region and I said, the Middle East really freaks me out and scares me; I don't understand it. I'm going to move there. So I bought a place in Dubai, I went to Syria and Jordan, and Ramallah and Riyadh, and all the different places in that region. After four months, I was starting to get a little bit of intuitive sense of how that works. It's a tremendously important region, but it helped me a lot, because I think best when I'm not in my comfort zone. I think most people do, but they don't realize it. For me, moving to the Middle East was really about knocking myself out of my comfort zone. Now I see a lot of opportunity from a demographics perspective, the theory of strong ties, and weak ties, the Granovetter paper, and the strength of weak ties. The fact that there are so many things in the Middle East that aren't connected to the places that they need to be connected. It's so easy to add value in the networks like that as well. So it's a connector type message that's also really fascinating.

I still have my place in Dubai, and it's not the right time to sell anyway, unfortunately. I have my place in Japan; I just bought a place here in Cambridge, so it's not on purpose but I have three houses right now. I'm trying to move everything to Cambridge. I'll probably still keep some presence in the Middle East, because long-term, I still think that the Middle East is hugely important, and I'm trying to bring Media Lab stuff to the Middle East as well.