Watch videos with subtitles in your language, upload your videos, create your own subtitles! Click here to learn more on "how to Dotsub"

TEDxDubai - Joichi Ito - 10/10/09

0 (0 Likes / 0 Dislikes)
We're gonna go on to our next speaker, to the TEDx stage, and the question is to simply keep in your mind is: How can sharing change things? Next TEDx speaker: Joi Ito. Hey, thanks. Hello... I'm Joi Ito, and I'm gonna explain... I'm gonna talk from two different hats, originally it was going to be just about sharing. But I actually moved to Dubai in December of last year. I was supposed to be more moved, but my apartment was about a year late, and I'm gonna get it in November. That was the first word I learned: Inshallah! So, I think... So, Inshallah, I will be a Dubai citizen soon, or a Dubai resident soon, But as I got to Dubai, I realized that there were some really great things, and some things that needed some work. The reason I moved, one of the reasons, was that this was the place that I found the Internet and openness reminded me of when I was young, as some of the other speakers today. I'm 43 years old, which is one year past the average age of Japan. So, I'm starting to get this middle age crisis and decided that I have to get out of my comfort zone, so I moved to Dubai and, I am now out of my comfort zone. But I'll talk a little bit about my comfort zone. So I grew up in Japan and the US, going back and forth, and my venture capital had... I do very early stage investments all over the world, in Silicon Valley and the rest... Most recent lucky investment I did was Twitter, which I'm happy that everybody is using right now. And then, the other part of my job is to work on the open standards, that build the Internet. So, I'm from the Internet, and we're here to help. So, first of all, I want to explain a little bit about the basics of investing, because investing in venture capital, and investing in other things are slightly different, but they have a basic fundamentally similar premise. So, first of all, as everyone knows, you buy low sell high. Now, this may sound easy, but it's really important to understand that risk, everything has risk. Right? And the trick is, how do you figure out whether this risk and the price attached to this risk is cheap or high. Right? So, if you are good at something, like, good at understanding startups, good at understanding architecture, you may be able to, before everyone else, realize that the risk is underpriced, so you buy it, so you know the entrepreneur, you know the space, you're excited about it, and then you invest in the company. And then, as the company hits the newspapers, and becomes famous, and everybody wants to buy it and it goes public, that's when you sell. So, a lot of Japanese, when they asked the Japanese government when they were starting to buy stocks, they asked, "how are you going to pick stocks?" "Oh, we'll pick the safe ones..." The ones with no risk. Right? Well, that's the same Japanese who bought the Rockefeller Center, and then sold it for 1/100th of the price, or 1/1000th of the price. The thing is, the minute it's in the newspaper, in investing, we say "the information is in the price." That means that everybody wants to get in, and the price usually gets inflated. Well, guess what? There's somebody that's selling when everyone's buying when it goes into the newspaper. Well, it's the guys who bought it before it was in the newspaper. So, smart investors buy low and sell high. Now, what this means is, in startup space, as we heard from other people, there are lots and lots of startups who no one wants to invest in, because they don't understand. Well, that's when they're "underpriced", right? And the minute everybody wants to get in, like right now with Twitter, the price getting, maybe not "overpriced", but the right price. Right? So, I think the key to understanding buy low sell high. Focus on the areas that you understand. Take aggressive risk, as long as the costs of the risk is cheaper, Then, the real risk, over time you'll make money because then, eventually, the price will be at the right price to sell the risk. Let me describe a little bit of the differences, though, in a traditional company and a traditional business., you have like some kind of percentage revenue growth over time. So, as you're running a big company, you try to sell, you try to increase your revenue and it goes up at some level. But you have an infinite, unlimited, downside risk if you get a project and don't deliver on time, it could cost you more money than the revenue that you're getting. If something happens, a catastrophe, markets change, so downside risk is unlimited, and upside risk is limited to some growth in market share, or some change in your pie. And so, most people who are from traditional businesses are very focused on the cost. Right? Because the cost of doing everything is so high, and the upside is limited. Venture capital is very different. Venture capital, if I invest 5 million dollars, the most I'm gonna lose is 5 million dollars. The bottom is kept, especially if its a smart VC, where you're not hiring tons and tons of people. Basically the downside is kept, whereas the upside is almost unlimited. So, if you invest in a Google or invest in some of these Yahoo!, you know, Ebay, any of these guys, you're gonna get a hundred thousand extra money. Right? The trick is: how do you get into those deals that go off the charts, that knock it out of the park? It's by betting on a lot of the deals, making sure you're getting invited to the party. Now, this is kind of interesting, there are a couple of assholes in Silicon Valley, but most of the successful people are actually really helpful, really nice guys. Because, when you think about it, when we do an investment, if we are trying to figure out who to bring in to Twitter, if we're trying to figure who to bring in to a new deal, it's like who do you invite to the party. Because there's not enough room for everyone. It's like at TED. You have to find the best people, you have to get all the right people there. And so, what you do is, over time you start working with the other investors, and you say, "Okay, which of these investors that I work that really help the entrepreneurs?" "That provide a value?" So I'm the Japanese guy, the international guy, I provide early stage product guy who kind of understands, sort of, games and addiction. So when they need somebody like that, they invite me to the party. The trick is, for venture investing, is you focus on providing lots of value, being a nice guy. And the other part is this downside is kept, you can take the entrepreneur who's failing and push him against the wall and shake him all you want, but you're not gonna get that much money back. So, you shouldn't focus on trying to minimize your losses. It's about, if something's failing, let them move on, get out of it. Move on to the next thing. There are some big venture capitalists in America who, if they're not successful, they get off the board convert their preferred shares to normal shares, say, "Good luck, see you later." And they just move on to the next thing, because the problem is, a lot of companies, especially, I see in Japan, I see also in the Middle East, they're so focused on protecting this downside, they spend 90% of their time on the failing companies. And the good ones, "Well, they're doing it, okay fine." No, that's the wrong thing. Some companies doing well, you should be spending 90% on the 10% that are doing well, and let the other guys start something new, and let them go. And don't try to squeeze them for success. Now, it's different, before you make money, those guys that had spent 7 years developing the plan they have, that's fine, it's good to spend a lot of time in developing your plan and iterating and iterating, but as an investor, what you have to do is you have to focus on the success. And, when you bring an investor in, is when you, sort of, have that angle, the other thing that's really important, and we'll talk about this next is: the cost of failure is very low in good startups. Right? So, now, with programming languages, and with the fact that you can get cheap Chinese cup noodles, and Kool-Aid, you can go for a long time, without that much money. And if you look at open source, and if you look at all these different things, the cost of failure has gone down. Right? So, in Japan, trading company, when you do a deal with them, they have this thing called a "feasibility study corporation." They spend about 3 million dollars to figure out if they want to do something. Right? And then the average telco, to swing the bat once costs about a 100 million dollars. You know, maybe, for a big electronics company about 10 million. Well, my average investment is about 100 thousand. The average cost of a startup in open source is almost zero. Right? And if you think about, like, if you look at Linux, Linus Torvalds, when he wrote, when he started Linux, he wrote on Usenet, "Oh, I'm gonna do an operating system, I don't know how interesting it's gonna be, but, maybe you can join." Probably hundreds of people posted that day, opening a new open source project, you would never had guessed that that was the one that was gonna become the big winner. And, the thing is, that it didn't matter, 99-point-something percent, But 99-plus-percent of open source projects are complete failures with no one using it. But the thing is, since the cost of failure is so low, you can experiment with everything. The problem is if you're in a big company, and a big research lab, you have to spend so much money trying, that you can't take risks, so would you have funded Ebay? Or Amazon? Or any of these companies, if you were in a big company? No. Because if you can lower the cost of failure, you can try new things, and most of the really successful things in Internetspace, you would not have predicted had a very high likelihood of success. It's really also really important to notice that venture capitalists only talk about their successes. We don't really care about the failures, because everybody's allowed to fail, as long as you have a couple of successes along the way, then you're good to go. Which is, again, another problem that we have in Asia, I don't know so much about the Middle East, but, there's a huge, kind of, stigma to failure. But that shouldn't be the case, it should be that you should just keep trying, until you get one. And this is also the point about iterating. You know, a lot of companies can change their business model as they go along, this is also very important, but the lowering the cost of failures is very important, and part of this has to do with the standards. So the old way of doing standards, this is, now, I'm gonna use American paradigms, but, East Coast American, and then if you look globally, the telephone companies used to run the network. Right? And they had big research labs, and they would go to these big intergovernmental standards bodies, wearing business suits, and spending years and years and years coming up with these standards with PhDs and all these people. They would create these standards that are about this thick that would require thousands of people to develop, and then, they would outsource them to big contractors who would supply this. It was important because we didn't have agile rapid standards development. So this is how the telephone networks were built, this is how GSM was built, this is how most of the network and information technology was built, until the Internet came along. What happened with the Internet was, we got into this whole open standards, so, David Weinberger calls this "small pieces loosely joined". So, instead of big, huge, chunks of lots of smart people, with lots of funding, the Internet was about lots of little groups connected together, loosely around open standards, and the standards weren't decided by these places that were populated by big companies and government. It was 12 year olds, 13 year olds, who could go onto the mailing lists, anyone can participate, anyone can add to the standard, and we had the Internet Engineering Task Force had a credo called "rough consensus running code". So basically, instead of trying to anticipate every single problem, you write some code, and you talk about it, and you work on it, and you work on it, and sometimes we didn't really anticipate spam, and we didn't anticipate some of the security issues, but it was still much better and much more robust than you could've expected. This is what we call "The Stack". Now, for the technical people out there, you'll see that this isn't completely technically correct, but roughly, I'll explain, these are the open standards that make up the Internet. The top one you might not be familiar with, but if you look, they're very important, so, when Ethernet first came up, it's the thing that connects these computers, that cable that you see in the network protocol, and before Ethernet, you couldn't connect the Mac to a PC, I think, I guess you guys are all young here, so you don't remember, But I couldn't connect my Mac to my PC. I had AppleTalk and this Microsoft connector. Well, Ethernet wasn't the best, smartest protocol. It just didn't have intellectual property encumberment, it was an open standard, and we have one of the sponsors, I think, Cisco, and all these guys could create all these routers that connected everything together and we created an open standard. TCP/IP was a network layer on top of that that allowed the computers to talk to, like, address each other, so if you remember, I'm sorry, I keep forgetting, that the average age here is much younger. But, back when I was a kid, your Macintosh even if you connected them couldn't talk to your Windows machine. And I remember downloading free TCP/IP to both my Macintosh and my Windows machine, and suddenly, they can connect together, my computers could connect together, they could connect together with everywhere else on the Internet. And it became one big network. Now, neither Ethernet or TCP/IP was the smartest, most clever system, but they weren't protected by copyright, they were run by open standards bodies, that were non-governmental and ad-hoc, and for a while in Japan, it was called illegal to do Internet. And I started the first commercial ISP, as a CEO of the first commercial ISP in Japan. And they kept trying to shut me down, it was a political war, to win Internet. So Internet, beginning was a struggle, now everybody uses Internet. And you don't think of it as something that we had to fight for. But we actually had to fight for it, to bring Internet to the world. HTML was a little bit easier when it first came out, everybody, sort of, yawned and said, "Well, why do we need that? We can already log into the databases and get all the information, why do we have to make it easier?" Well, it turns out that making it easier created this explosion of innovation, the we couldn't have anticipated. That when the Web first came out, most people didn't really understand how much impact that would be. But again, very important, not patented, Tim Berners-Lee could've tried to patent and sell this thing, but the founder of the World Wide Web Consortium, that created HTML, he gave it away for free, created an open standard so that everybody can participate, and this created this whole layer of innovation. So, what you see common about these three layers is they're open standards, they're not copyrighted and it creates this environment of "small pieces loosely joined," because of these standards, you and I can create tools that automatically work together, it's called "interoperability," and we don't have to control the whole network, like the telephone companies used to do, we can just do our part, not know the whole of the network, but it all kind of works together in this kind of ecosystem. It reminds me a lot of an ecology or an ecosystem. So, what I'm saying right now is that Creative Commons, which is the new project that I'm working on, which I think is the next layer of standardization, is that now we have these networks together, what is the failure? Just like we used to have to have a software engineer and a network engineer to do everything, right now, we have to have lawyers, every time we want to do anything at the content layer. If you're a Croatian professor and you want to share your courseware with a Japanese professor, you have to go and hire a lawyer on each side to create a contract. If you add a third school, or a fifth school, then suddenly the cost of the lawyers exceeds the value of the transaction. We call this "failed sharing." So back before, when you used to go and fly to Cannes and drink champagne, I used to work for a television company, so we used to do this, the legal fees could be 10 thousand, 100 thousand dollars, that was fine, because there's a 10 million dollar, 50 million dollar deal. Now, the people who are collaborating on Youtube, they can't afford these legal fees. So what I'm trying to do is, I'm trying to create a layer of standardization of the contracts, for the lawyers, to allow people to share things with each other. And I think that that will cause an explosion of innovation, just like these other layers of openness have done. It also just, since Google, I think, is here... If you think about Google, and imagine... So Google use open source, they use, basically, mostly standard PC, and they just connected it to the network at Stanford. So it was google.stanford.edu. Before that, before the stack, we used to have what we call "X25," it was a network where you had to have, every user had to pay for traffic, you had to have a bilateral agreement with every country that you connected to. And if you imagine that PC being replaced by a big proprietary mainframe, probably Google would have cost hundreds of millions of dollars, taken ten years, run by the telephone company and it wouldn't work. Right? And the fact that Google was probably, the first server was built for thousands of dollars, and connected, at no cost, to the network, and suddenly the whole world can access it. It's the beauty of the stack. And this goes back to the earlier point, lowering the cost of failure, because the Google guys, I actually remember, I was running Infoseek Japan, they came and tried to sell us the algorithm. And they said, "Well, screw that, we're doing that on our own." Right? So they did it on their own, and they were successful. That's really the key to innovation, and that's why the West Coast, in the US, is beating the East Coast, in the US, because innovation is no longer in the big companies, it's in the little guys, and I think Google was doing a very good job of trying to keep that spirit going inside the company and trying to invent, reinvent, how big companies work. But this is very, very important, I think that people read about this in the news, but you kind of have to understand this as almost a belief system, before you can really monetize this and create this. And so, I'll talk a little bit about Creative Commons, this is my big mission right now, one of the reasons I moved to the Middle East was, the Middle East and Africa is the last area where we are trying to get adoption of the licenses, but we asked a bunch of creators, "What are the questions that you want to answer, or express when you mark your works?" So if you're a photographer, or a professor, or somebody, a writer, blogger, what are the things you want to be able to decide about how people use your work. So, we found that over 90% of people wanted attribution, so if somebody uses your work, you have to say it's from Joi Ito. Okay, fine. Some people said, if you create a derivative work, if you remix it, I want to have it shared back to me. So, a lot of musicians say, "Okay, fine to remix my music, but I want to be able to use it too." "Put it in my CD." So that's the share-alike. So, Wikipedia, for instance, is share-alike. You could use their work, but you have to share it back. Non-commercial, many professional musicians say, "Okay, you can share on the filesharing networks, but if you're going to use it for commercial use, on the radio, or in a CD, I need to get paid." "So, I want that reservation." The other one is no-derivatives. So, I think TED conference is, the videos are non-derivatives. When you try to tell a story, a lot of documentary people say, "I don't want you to remix this, this is a story," so you have non-derivatives. So we said that if you select these, answer these questions, you get six basic licenses, which encompasses 90% of things that most people care about. So obviously Hollywood and some of these people who make a business out of, they already have maximum demand, and just need to protect. They keep the All Rights Reserved, but many, many people in the world, like amateur photographers, or professors, or bloggers have, want what we call "Some Rights Reserved." And the key is that by marking your works with these icons, we create licenses in 51 different languages, so that people can use these works without asking for permission, so Obama's pictures were licensed under our Creative Commons licenses, and a broadcaster in Japan wanted to use one during the inauguration, well, they can use it without having to send a fax to the White House, and getting a fax back that says, "Okay." Because they already know that Obama says it's okay to use it, and Obama's like, "Okay, of course you can use it, because, you know, I'm happy for you to use this." So, in cases where you want to give permission, and you want to create distribution of your work, Creative Commons licenses are good. Just to give you a couple of examples, so, exactly at noon, when Obama was inaugurated, they updated the White House page, and they put Creative Commons on the White House page, so any content that a user uploads to the White House page gets a Creative Commons license that says that anybody else can use those comments as long as they provide attribution. Wikipedia, it took us about 4 years, but we finally converted Wikipedia to Creative Commons license. Now to explain a little bit about why this is important, is Wikipedia was created before Creative Commons, so they had a different, but very similar, license, that said you have to use it but share it back, but this reminds me of the early days of the Internet, it was a slightly different license, and then a Creative Commons license. So, the MIT professors who use Creative Commons licenses couldn't remix their content with the stuff that Wikipedia was using, so even though it was the same idea, because they were different licenses, you had them separated. And by converting Wikipedia to Creative Commons, we create one body of work. And this is why Creative Commons and me pushing, may sound a little bit imperialistic, but it's important, just like we don't want 5 different Internets, we don't want 5 different standards for this interoperability at the legal layer, it has to be a common language, so that all the lawyers are talking about the same thing, and that's why we had people in 80 countries giving us feedback, so that we can modify the licenses, we're just about to launch the first Arabic licenses in Jordan, next month, so that's a big deal for us. Thank you. Actually the first commercial broadcaster, professional broadcaster, to broadcast quality footage online was Aljazeera, which was great, and many, many broadcasters used their content, like RAI in Italy, and others, and they've told me because of they gave this stuff away, just like TED gives their videos away, they got so much recognition that they actually increased their sales, and developed commercial relationships with a lot of new distributors. Nine Inch Nails is a band, but they gave their Ghosts CD from last year away, so they said, you can, Creative Commons Attribution-NonCommercial-ShareAlike, basically means you can reuse it for non-commercial use, you can remix it but you have to share it back. But they sold a limited edition box set, a picture book, DVDs, and they made 1.6 million dollars on this website in the first week. Right? So, this is a big shock to the music industry, you're giving the music away, and you're making all this money on this other stuff. And this is a really important thing, is that you can actually, there are businesses built around sharing. I'm not gonna go deep into this, there are a lot of case studies. Publishers like Bloomsbury Academic in England, many big commercial entities are starting to use Creative Commons, Ridley Scott, the famous director, is going to do a prequel to Blade Runner, under Creative Commons license, so other people who will write other stories about the same characters, and reuse his work, because he wants to create a universe on the Internet, not just his own narrow thing. And so, that's, kind of, roughly what Creative Commons is doing. A really important other piece of this is the World Wide Web Consortium, so these are the guys who create HTML, and manage that process to create that standard. We created with them a standard to put the copyright information in the HTML itself. HTML is the metadata and the way you tag stuff on the Web. What that means is that service providers, like Yahoo!, and Google, and Microsoft, can now understand copyright information, when you copy-paste stuff from websites, it'll bring the copyright information. Actually if you go to Google or Yahoo!, you can go to advanced search, and say, "I want pictures of Burj Dubai that I can use for commercial use, as long as I provide attribution," they will look at the metadata of the Creative Commons license and find it for you. Eventually, this will be built into all of our software tools, and our network. But it's very important, again, to try to keep this technical layer together, because the technology, and all of the people who are using that, that, again, is trying to lower the friction, so that the software and the network understands copyright, so you don't have to hire lawyers to try to understand it, and I think that most people fail to follow copyright, not because they're criminals, not because they want to stick it to the man, although some of them do, it's usually because they're lazy. And I think once our tools are aware, it'll lower the friction and make it a lot easier. Which segways a little bit into this development model thing, this is a little bit technical, but I want to describe it, because these are some keywords that'll come up, but the Waterfall method is the old way of doing software, this is when things in the world didn't change very much. So what you do is write the requirements, design it, build it, implement it, verify it, and support it. And this is would take years to develop, and finally you'd get it, and you have this huge thing where the cost of changing anything was very, very high. So, cost of failure goes up, speed goes down, and you don't wanna change it. This is how most systems work, and it's, kind of, how when you build a big building, this is how you build it. In software, this doesn't really work, in software, we do what we call "agile development," so you look at Twitter, Facebook, all these guys are doing this, this is basically, we call them "sprints," so each rev is about one week, or maximum two weeks, sometimes less, and what you do is plan a little bit what you can do in a week, lots of what we call "stories," you write the requirements, and these developers get together, they write it, and they test it. And they write all these tests around it to make sure it does what it does, and they evaluate it, and they keep spinning out a new version every week, and then, suddenly, what happens is, change is built-in, and you create a system, so that the business people and the designers are all involved in this process. So that, suddenly, when it says, "Uh-oh! Wait! Twitter just came up, what are we gonna do about that?" Well, it only takes you, the maximum delay you have, before you respond is a week. Whereas before you would build something that's two years late, that's completely irrelevant, and unable to respond to anything that's happening to the real world. This is what most of the good software companies, and startups in Silicon Valley are doing, it's not that hard. I'll give you, again, a paradigm shift, from the way you used to do work, where you toss things into the system integrators. Well, this is not that hard to do. And this is something that I think we need to bring over. And then, one other part that's very important, I think is just a way of thinking, virality is just one way to explain it, but distribution and getting users. So everybody says, "Oh! So what's Twitter's business model?" Well, if you look at Google, any of these successful companies, have you ever seen a company that had a 100 million users that couldn't figure out their business model? Maybe Napster had their problems; there were a few. But there are very few companies that were able to get huge users but completely failed. Now, think of all those companies, that had a business model, had a product, had good entrepreneurs, but completely failed because they couldn't get any users. That is most companies. So the hardest thing on the Internet, even though it may sound, sort of, bubble and, sort of, fishy, it's the Internet is all about making it easy to find a business model. When I started on Infoseek, we used to try to charge each user for search. You know? And that was, kind of, stupid. We didn't realize it at the time. But, once you start to get going, and you get users, and traction, and Google was the same way, you can usually figure out a way to make money. And so, distribution, figuring out how do you get it, if you're Youtube, it was that they were attached to MySpace. Google, although they were really smart, I think that if they hadn't had the Yahoo! deal, which was giving all the search traffic to Google. And this is the investors saying, "Okay, well, you guys don't want to do search, let these guys do search." Distribution is really, really key. And this is, more than anything, just a hint to the entrepreneurs in the room, you really have to think about how am I gonna get the users in a viral way that doesn't cost me any money, that grows at over 30% a month, should be your target. And then, finally, this is my last slide, and I'm not going to go through and explain all of this, but it's an ecosystem, it's not one big institution, it's not one big university, it's not one method, there's a whole bunch of things that have to happen, right? And it takes a while, but you need all these pieces, you need good angel investors, you need good later-stage investors, you need founders who turn into investors, you need all this to work and it's very difficult to replicate Silicon Valley, but the key is to lower the friction, let all these different pieces of the ecosystem come in, and, again, it's going to be a different ecosystem here than in Silicon Valley, because the culture's different, and the people are different, but what you have to remember is that it's a whole, kind of, really is, like a rainforest. You have to have the trees, and all the little components in there are required to be vibrant. And, I think, one of the important things is it's not just one brandname, it's not just one person, it's a whole group of things that have to happen. And I see the seeds in the community here, of all the pieces of this ecosystem, but they all need to be encouraged together, and there's no silver bullet, and it takes a lot of hard work, and it takes a lot of failure. But, I think, at the end of the day, because Googles only happen once in every 10 years, but the point is as long as you get a Google every 10 years, all the other failures are worth it. And so, you really have to think about is not just investing time and money, you just have to invest failure, after failure, after failure, and those failures are teachings, so you have to let those failed people come back to build the Googles when you have a chance. Thank you.

Video Details

Duration: 28 minutes and 50 seconds
Language: English
Genre: None
Producer: Giorgio Ungania
Director: Giorgio Ungania
Views: 147
Posted by: giorgiotedx on Jan 9, 2010

In the spirit of ideas worth spreading, TEDx is a program of local, self-organized events that bring people together to share a TED-like experience. At a TEDx event, TEDTalks video and live speakers combine to spark deep discussion and connection in a small group. These local, self-organized events are branded TEDx, where x=independently organized TED event. The TED Conference provides general guidance for the TEDx program, but individual TEDx events are self-organized.* (*Subject to certain rules and regulations) Our event is call TEDxDubai, where x=independently organized TED event.

Caption and Translate

    Sign In/Register for Dotsub to translate this video.