Innovation in Open Networks

November 10, 2011

Joichi Ito

Michael Zielenziger
, McKinsey Global Institute Senior Editor

Joichi Ito, Director of the MIT Media Lab, spoke at Japan Society about the power of open networks to encourage innovation.

In his late teens and early twenties, Mr. Ito said, he considered a career as a journalist or a TV producer. But he remembers thinking that under the social norms of the day in Japan, he would have to wait until he was 45 to achieve anything important.

For "a very impatient guy," the early days of the Internet were a revelation. "I realized that this was a place where you didn’t have to have permission. Everyone could connect, and actually being young was better than being older, because the less you knew the more creative you could be."

The big rival to the Internet in Japan was a protocol called X.25, which was developed by an intergovernmental body through an elaborate, formal process. The Internet, in contrast, was "built by small pieces, small groups loosely joined by standards created in ad hoc committees"; "any 12-year-old could participate." The aim was to create "rough consensus running code, which was let's just kind of figure out what we want to do and then build it and keep it simple, and then let's let it evolve."

The Internet won out over X.25. Open source and free software have brought the costs of collaborating on new software products and distributing them to "almost zero" today.

With lower development costs, the nature of innovation changes, Mr. Ito said. If you're a venture fund and someone says to you, "'I’m going to create an encyclopedia that’s better than Encyclopedia Britannica, and the way we’re going to do it is let anybody edit the page,' you would never fund this project. Right? It sounds like a stupid idea. And, in fact, just about every successful Internet company when it first starts out kind of sounds like a stupid idea."

A Japanese telephone company would spend $100 million to kick the tires and do a feasibility study on a Wikipedia or a YouTube. The trading companies, which are smaller and more flexible, would still spend $10 million. Today, a venture capitalist can test an idea for $10,000. "You don’t sit around and worry about whether this is a good idea or a bad idea. You just let them do it."

The idea of the agile company goes back to Toyota and the notion of just-in-time manufacturing. "The Toyotas of the world only had a very, very short inventory, so that when they found something wrong the only thing they had to throw away was that short amount of inventory," Mr. Ito pointed out. "This is exactly the same way that Internet startups in Silicon Valley work. They don’t stock a lot of code and they’re always pivoting and they’re always changing."

"It’s not planned. You have what I would say a compass, so you have a general trajectory—we want to be massively scaled and we want to be in video. But how exactly you get there, you figure out as you go along."

When the Tohoku earthquake happened, Mr. Ito was in Boston and his family was in Tokyo, he recounted. He and a friend who is a hardware designer looked for information online about radiation levels, but couldn't find any. They wanted to help supply radiation detectors to the region, but couldn't find a source to buy them ready made. So they decided to put together a team to do it themselves.

Within a week, they had people on board who were experts on Geiger counters, sensor networks and data management. They found a group of young people at Tokyo Hackerspace who could solder metals and make tools. They put up a plea for donations on the web, and got $45,000. They recruited volunteers to bring the Geiger counters to Fukushima and make measurements. Tesla and Aston Martin contributed cars to the effort.

They got the Knight Foundation to provide $250,000 for work on visualization, so that data could be open for all to analyze. "And we now have over a million readings. We have more complete data than any other entity in the world on the radiation in Japan." A design for making the Geiger counters in China is online, downloadable for free to encourage further development.

At the MIT Media Lab, where he became director in April 2011, Mr. Ito will encourage students and researchers to put more of their work online and to create a more open network, he said. "We'll still have scarcity of the degree, but I want everyone to be able to learn from us."

Reflecting his own background as a self-educated technology entrepreneur, he said that he wants the Lab to "allow a little bit more learning through creating, a little bit more learning through doing.... We focus a lot more on rendering things into physical or software stuff that you can see."

This "helps a lot with interdisciplinary stuff as well, because it’s very difficult to get a sociologist and a physicist to agree theoretically on something. But if they’re forced to work on building something together, suddenly their ideas kind of mesh."


Michael Zielenziger of McKinsey, who moderated the discussion, asked the first questions:

How did institutions in Japan such as the health ministry react to the Geiger counter project?

"Initially no one wanted to talk to us," and the group was seen as "troublemakers," though over time this has changed, Mr. Ito said.

"At the beginning, we really didn’t know what we were doing, but because we were open and we were sharing everything on the Internet and everywhere else, a lot of people started to contribute their information and make comments."

The people of Fukushima were much happier with Mr. Ito's group, who sat down and told them the details on what their Geiger counters measured, than they were with the government officials who came, took measurements, and left without disclosing the results. The group was able to give detailed advice: Move your kids from the third floor to the first floor, because cesium is sticking to the roof and the readings are higher on the top floor.

The project has inspired local entrepreneurs. "Little companies in Fukushima want to start building the Geiger counters there, and we're helping them with the design."

How do you deal with the question of trust and the Internet? People worry about hackers, fraud and identity theft, yet there's enormous potential for economic growth.

"Every decision that we make in the United States or Japan to create tools that prevent people from having anonymity, or Facebook and Google implementing real-name policies, these affect the ability of the Egyptians, the Syrians and the Tunisians to have a dissenting voice," Mr. Ito said. The arguments in favor of censoring Internet content for copyright or trademark reasons come from Hollywood and from big brands—it's "very financially motivated."

Members of the audience joined the Q&A:

Growing up, what helped you overcome challenges that seemed daunting at the time?

"Basically it’s just failing," Mr. Ito said. He started his first company at age 18, which failed, and kept on starting companies and failing for 10 years. Eventually "I started to kind of figure out how to do it....For me when somebody says don’t go there, I have to go there and fall before I actually learn it. So, I think it depends on your personality."

Living near Detroit until junior high school, Mr. Ito said, he encountered devastating racism. When he went back to Japan and went to Nishimachi International School, "suddenly the kid who could speak English and Japanese...actually was the best kind of cultural category to be in," and that gave him the self-confidence to take risks.

Could you comment on the potential for collaboration technologies—in particular, reputation systems where people vote on who's making good contributions?

"What I find with information right now is because the systems don’t work very well, it’s good old-fashioned human networking" that has the greatest potential. "My best content comes through email and tweets from people I trust," Mr. Ito said.

Part of the work of sorting out what's worthwhile and what's mere noise "can be automated, but it should be automated after you really understand how to do it by hand," he added. "This is why blogging was so successful. Blog software is written by people who wanted to write blog posts who also knew how to write code."

"If a few people who really knew stuff about information, about journalism, about getting rid of noise... actually open the black box and build the tool themselves," then "we're going to get some really interesting stuff."

—Katherine Hyde

Topics:  Business, Policy

Calendar of Events

September 2020

S M T W Th F S
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      
All content © 2020, Japan Society, unless otherwise noted. |
333 East 47th Street New York, NY 10017 Phone: 212.832.1155 |
Credits | Press | Contact Us | Privacy Policy