 Hi. How are you? I'm Chris. I'm going to skip all this stuff. So I run open source and public sectors engineering for Google, which means I have a small portfolio of government focused projects, as well as my open source work for the company. The agenda is very simple. I'm going to yammer for a little bit. And then you can ask some questions. This is a famous criminal in the United States. And then we'll go on from there. So first of all, before I started talking about Google at all, I wanted to talk about open source itself. We did a study of how much code was out there and available to you as a nation, individually, within your companies, to use freely. In so many definitions of the word free, it's kind of scary. But here's what's out there for you. We have somewhere around 31 million unique files containing program code that you can use across approximately two billion lines of code. And so what it comes down to, and the reason why I show this slide is that this is the equivalent of 10,000 engineers working for six years straight. This is a staggering amount of productivity available to all of us. And it's one of the very real reasons why Google was able to innovate and come out of the gate so very productively, so very early on. Only by 2009 did we have 10,000 technology people at Google. Google's about half technology people engineering and operations and the rest and another half everyone else. So it was fairly staggering to basically be able to double the productivity of the company from the very first day it was founded, for some definitions of double. The real question is why is all this code out there? And a lot of people have postulated and written stories and written books and the rest about this, but we decided to actually ask people about 10 years ago. We asked 10,000 developers on source origin. We said hey, why are you actually releasing source code? And it was actually a pretty amazing thing. They weren't doing it because they hated Microsoft or because they were long haired liberals from San Francisco, although I am from Mountain View which is about 40 miles south. It was because they were interested in computer science. It was because they wanted to become better using their computers and ministering their systems. It was because they wanted to sort of identify with other people who wanted to stand on the shoulders of programming giants as it were. And only 11% of them didn't like Microsoft or didn't like some company or some proprietary way of doing things. And this was something I identified with at the time and still because I never saw computer science as being combative between large companies or between countries or between people who like this language or that language, whether they're spoken or programmed. I always saw computer science as the way that we could do things, you know, better and do things in an interesting way and process information and what I felt was a very compelling way of life. So this was a great thing and this is why I've identified with open source and been involved since 1995. It's not. Oh, there it goes. I'm just going to skip those because we really don't have a lot of time. I don't know how far I'm going to cut into the break. But let me continue. This is Google. You probably recognize it. This is Google about 11 years ago. They basically walked up and down the hallways of their lab there in Stanford and they found every computer that wasn't plugged in. They took it, they brought it to their lab and they plugged it in. And if people got upset, they gave the computer back. So this is the last time Google ran a heterogeneous network of multiple operating systems. You can see here that we have IBM's, Sun's, but mostly x86 machines running Linux. We got a little better in that we decided to put everything in a trash can and mix it up to make computers. No, not really. For those of you who are computer architects, system administrators, this will look horrifying to you and I apologize. There will be drinks later. No. So what you're looking at here is how much computer you could fit into a rack in 1999. It was hot and anytime you break one computer you'd be breaking four of them. They shared power supplies. There was a fabric of fans behind it. And whenever you would pull a computer out to work on it, you would usually snap something off on those cables on the side. So that was a bad design. But it worked really well for us. You can go see these at the Computer History Museum if you're in the Silicon Valley. The hard drives you see there are resting on top of plexiglass on top of the memory sticks. So, but we were able to do a lot of computation very, very quickly, very unreliably. And that unreliability was programmed around using open source software. These machines were all running Linux all the time. And from then on, we were basically running Linux on everything. And it would be Linux with software that was built using open source tools like GCC and Python and the rest, combining open source libraries and our programming together to create Google. This is a couple of iterative pictures. You can see our advanced cooling system there on the bottom right-hand side of the screen there. That fan has been at every data center I've ever been in. And I was very happy when I got to Google to see that we had one too. At Sourceforge, this fan lit on fire and we had to bring the site down for a little while, but we won't go there. You can also see the advanced power distribution network that it's cooling down, using the power strips hanging from the side there. But then we got pretty good at things. We would fill up a data center this size in about two hours. We would unwrap machines and set them up and use up all the available power in that data center. So we would often fill up a quart of the data center and use all the available power in AC. And that didn't work. So we went on to the more modern lines where we have our own data centers, our own machines, our own everything. So what's funny about all this is that to a great degree, the commodification of computer components for desktop machines and to a much lesser degree server machines is what fueled Google. We always used desktop components. We always used desktop A6 and all the rest to drive our machinery. And the reason for that is because we could get the best prices for them and thus get the most power out of them. And we were very much dependent upon the raw computing power and the raw storage that the internet and scanning it and indexing and searching it requires. And then we would run Linux on top and derive all of our sort of reliability in the software layer. So that's Google. Google is, you know, our own racks, running PC class on other boards, and, you know, really big flippin' hard drives, you know. And then we wrote our software on top of Linux and it's been working really well for us going on 10 plus years now. So I talked about that kind of. Oh, why? Why? So I'm going to tell you a story about another company to tell you a story about Google. Back in 2004, France Telecom wanted to bring out push to talk capability on their cell phone network. Push to talk is when you hit the button and it goes, it's really loud. It's often used in construction sites. So it acts more as a walkie-talkie than it does as a cell phone. I'm sure you're all familiar with it. And if not, I've just described it. They went to deploy this. They had a two-year plan. They spent $2 billion upgrading their carry infrastructure, very expensive hardware, and an enormous racks machinery to make it work. And they expect out a cell phone to take advantage of this new feature. It was running Windows Mobile. And this is actually not an anti-Windows talk. I'm not that kind of person. It's a what happens when someone else controls your destiny talk. And so which I think when a country decides it wants to be serious about computer technology, this resonates. So they had a problem where every fifth call would die. It wouldn't go through. It would fail. And often the phone would need a reboot. And they couldn't get past it. And they finally tracked down the error to a shared library in the operating system. They contacted their customer representative and they said, hey, we found this problem. They're like, hey, we found it too. They're like, great. We'd like to fix it. They said, we have fixed it. They're like, awesome. This is a fantastic, you know, this is terrific. And they go, can we have the fix? No, you can't have the fix. And they're like, well, why not? They're like, it's in the next version of the operating system. Okay, we'll take the next version of the operating system. Okay, it'll be ready in 18 months. And they're like, we ship in three months. And they're like, we're sorry, we can't help you. Now, this was not unreasonable for Microsoft to do. They did a very specific kind of math that you do in computer companies where you say this, you say, how much money do I make from this? And is it worth the extra cost of having a special version of the operating system? And if the scales bounce one way or another, you have your answer. In the case of France Telecom, they were two and a half percent of Windows Mobile revenue and the cost to have a special cut of the operating system just for France Telecom for that time period was determined to be too high to give them access to the code. Now, if it had been Linux, they could have done what we had been doing all along. Just change it and push it out. And you know what? Maybe the Linux kernel team will take your patches and maybe they won't, but you'll be able to control your destiny. And that was a very, very important thing to Google. And it's something that our employees and our engineers and our founders were used to doing. Here's open source software. We can do with it what we wish and make Google better. And you know what? That's a funny thing. I could have stopped my talk there and maybe for the sake of time I should, but I still have a little bit of time left, so I'm going to keep going. You can stop right there and say, you know what? I am a very good of open source software and free software. I know how to use it to make my life more productive. My computer scientists happier. My engineers more useful. And that's a pretty terrific thing. 10,000 engineers in your pocket. Literally in the case of Android. But we wanted to go a little bit further since we recognized how important open source was to us, but more importantly to the future of the internet, the future of a fair and competitive internet, which you cannot have without free and open source software implementing the open standards that Mitchell and Joey talked about. We decided that we want to do some other things. So we took three broad tax supporting open source infrastructure. We host over 300,000 projects on code.google.com now. We also help fund the open source labs, which hosts groups like Mozilla and others. We release open source code ourselves, which I've always thought is the real coin of the realm when it comes to open source software developers. We've released a little over 900 projects to date. Those are smaller projects, usually between 20 and 100,000 lines of code for a total of about 18 million lines of code into the world of open source software. It is not a lie or hyperbole to engage in to say that every single one of you is running some code that we have released from Google. Even if you're running an Apple iPhone or a Macintosh or a Windows machine or Windows mobile, you don't have to be running Android or any specific Google technology to have been taking advantage of the work we've done. And that's something I'm actually quite proud about. We've patched into literally every project you can imagine. So we lost track. We decided actually to stop tracking patches directly. It was just everywhere. We did a survey about two years, about a year ago, and we found over 87,000 patches had been done by Google that we could find. And that was just the ones with the at Google address. And then we've released huge and major projects like Chrome and being the open source part of that and Android. And Android is now shipping a little over 250,000 units a day. So the thing I want to wrap up with before I go to questions, assuming you have any, you know, because I'm sure I've covered everything, is WebM and WebP. So WebM, we purchased a company by the name of On Technologies and they have a very very innovative and effective and modern video codec that they put together. We released it under a patent granting version of a BSD license, where basically we have the BSD license and then a patent grant on the side that says here take it. If we happen to have any patents covering this, you're cool. You don't have to pay us for that. Because we want people to adopt this very very quickly. We feel the state of video codecs is such that the costs incurred by video sites, encoders, and software is too high. And so we released this. We think this is probably one of the most important things we've released in the last five years. And WebP is the single picture version of that. They're efficient and truly free codecs, which is something we haven't had ever. You don't know this, but I'm sure many, raise your hands if you have a phone or a HD camera in your possession today. We'll make it easy, it's not even necessary. So every single one of those is licensed by MPEG LA for non-commercial use only. Non-professional use too. So if you consider yourself a commercial or professional person, you probably don't have the right camera for it. Maybe that guy in back at the tripod. He had to pay probably three times as much for his camera though. And it's restrictions like that that you don't even think about. But you're breaking that license every time you post it online. And that's something we wanted to see the end of. So yeah, I think that's basically, oh, I went to, I took off a whole bunch of slides about VBA. So I figured you didn't want to see signals to noise graphs and the rest. I'll wrap up in case there's any students in the room. We have a yearly program called the Summer of Code. This year we had a thousand, just a little over a thousand, so I usually pick a prime number over a thousand. I know that's really nerdy, but. And they worked for over 150 organizations across open source software. There were over 2,000 mentors looking after them. And they're paid $5,000 if they succeed. And we have an 81% pass rate across 94 countries. Generates about 3 million lines of code a year now. We've regenerated about 13 million lines of code through this program. And this is separate from the other code that we've released. This is our way of bringing new developers into the world of open source. And these developers have gone on to both work on projects like Mozilla and the rest, but also to become active members of open source communities all over the world. So there you go. And now I'm happy to take any questions you might have. There you go. Okay, now you'll need a microphone. There's microphone people I hear. They're over there and there's people over here. He's running up. It's very exciting. I assure you. So thank you very much for that. Anytime. You know, one thing I think Google, I think they've been revolutionizing the way, you know, they're neutralizing the playing field across the internet. I mean, openness not just into technology, but even information. I mean, seeing every single day you have Google labs, something new comes out, experiment, tinker with it. I mean, those are your things. But I mean, have you ever been thought about like opening labs up even more to people around the world? I mean, even though that started by Google's using your amazing Google 20 principle? So we have. We've had some very weird ideas that didn't pan out as well as we wanted them to. So the way we approach that, the way we take our data systems and our machinery and our programs, we try to open them up via APIs. And sometimes we're successful, like the Google Maps API is very successful. And sometimes we're not. I think people aren't really clear as to how the whole Doxon Spreadsheets API can be used and is effective for them. So this is in a lot of ways an evolving place for Google. You know, one of the things that we looked on with just like glorious envy and admiration was like the Netflix contest. I don't know if many of you are familiar with this Netflix release a lot of data and they had a contest with with an amazing monetary prize. If you could improve their algorithm for matching and recommending new movies. This is something that we considered doing. We've always hit this on privacy because it's extremely difficult to really search data without releasing lots of very personal information as AOL found out back in 2006. And as Netflix found out because they had to cancel their second one because there was too much personal data in just the way that people looked for movies. It was astonishing a research group was able to match up the movies search with the human being almost immediately. So, so yeah, no, we think about this all the time and we express it through APIs more often than not. And if you have any ideas for new APIs that you want to see, I'd love to hear about them. So, find me at the break, I guess. Yeah, we have someone up here. I think there's someone back there too. Yeah, it's like the lights work against you, you know, but he'll be next. Yeah. Yeah. First, I was attending the Java day conference in and Google was one of the sponsors. But actually, I am I want more more investment from Google in the region. And there is no right. Yeah, there is no development center for Google in any Arabian country. Okay, I am I'm lucky like to have one of these development centers in Arabian country. The nearest one is I think in Turkish, there's solution center for development for Google. But and also there is, I have friends working in Google in Arabic Arabization, but they are working from Switzerland. Yeah, we have a huge engineering office in Zurich. Why, why Google didn't open any development center in Arabian country? We're pretty clueless sometimes. You know, it's hard to figure out where to open up new engineering centers is a long one of the things we're doing is we're having Google developer days in Egypt and Jordan in November. Sorry, in early December. And I see that as sort of like a Google getting its head together as to where these developers are. Zurich is easy for us because we could pull a lot of the European nation developers to Zurich and to London and to a lesser degree to Ireland, although we have a huge footprint in Ireland. Honestly, we're just not not that sharp about the Middle East sometimes either. You know, we have a lot to learn about the region and we have we have an office in now in Dubai as well as in Saudi Arabia. I think it's Saudi Arabia, Saudi Arabia somewhere else. But yeah, no, we and those are more than just sales offices. They're also sort of like cultural touch points so we can learn more about the region and come to understand where the developers are. You know, so Doha can make a huge step forward by just getting those kids from CMU and and VCU and the rest and and get them hacking. Well, I'm the CMU graduate too. So. But yeah, you know, it's like we you know, the other side of this is there are developers everywhere and you always do that math. Can you bring them to you or do you have to go to them? What keeps people where they are? You know, it's how we ended up in Australia actually. We so wanted Lars and the only way he would stick come to Google is if we were there and so we had to open a place that the same same was true with Craig Neville Manning in New York City. So but we haven't opened up something in New York in New Zealand for Craig Neville Manning. So you have to do the math all the time. But yeah, we think about it all the time. Hey, Chris. Hi. How you doing? It's well. Thanks to you for being here. I'm going to run a little bit here. If you can hear me. Yeah, I can't. Okay, great. Great. I couldn't see you. That's why I'm doing it. Okay. Basically, okay, it's great to have you here. And we've all been encouraged and we've all been talking about openness and how it makes sense on a conceptual, factual and more philosophical level and based on his production values and develop it keeping this in mind. I would like to talk about Google's open Android platform. Okay, bring it on. Okay. First of all, you are league. Google is legally called legally named in Android to be an open mobile operating system because it gives out the source code. However, I think it fails into get, you know, dealing with, you know, the more moral values of openness, which is more of a contribution of effort and development effort from consumers. Yeah. So in that case, it's, for example, it's not being updated to users on new features. What was that one say that one more time? Basically, all the new versions of Android is not always released because of the contribution from users. Right. That's one thing. And I'm not saying this personally. I'm saying it's from, you know, people who use them as well as, yeah, it makes sense. Okay. I've got a second, second point in the question to make, which is scarios. It seems carriers are always hurting the openness of Android, even if there is an openness in Android. Right. Right. So I want you to comment on that. And also thirdly and lastly, comment on other true openings or open operating systems, such as the Symbian Foundation. So Symbian is working on opening up their code as fast as they can. And they actually open up a huge repository of software. So that was great. It was under the EPL. I believe they chose. Which makes it even harder for carriers to adopt it. It's funny because the things that you want from the openness side of things can be very, very hard when it comes to getting carriers to actually stop the operating system and ship it. So we try to do this balance where we have carriers, handset manufacturers, chipset developers, software developers themselves running applications on top of the platform and trying to get them all the information they need to ship phones. And it's a balance. And we haven't pleased everyone during this time. I've been happy in that we've released the code for every version to the outside world and people have been able to make devices from that. And they've been able to make them using Bing or using our competitors. So that shows you in a lot of ways what Mitchell said about forking and about having people do with your operating system maybe what you're not thrilled about. And that's when you know it's really time to get started with open source. But a lot of people want more than that. They want us to take more user input. And that's historically lagged. JbQ, who's on our open source team in the Android group has done a fantastic job of bringing in user patches. But it's just it's one of those things where over time they might converge a little better. It's a lot better now than it was two years ago when we first released the software code. I don't see carriers as the enemy though. A lot of people like to say are the carriers stink for doing X, Y, and Z. The carriers have a huge investment in infrastructure and they're the ones selling the phones and having to support them. So they'll often do things that maybe I don't agree with philosophically. But the overall trend is towards more openness over time and I'm patient. So I would say what is Qtel doing? Can we get them to use Android? And of course you know the answer to that is we have to have better Arabic language support and that's something we're working on. And I think I answered most of your questions. But I'm bad at remembering multiple questions at a time. And then do you have anything going on here? Do I have to shut up? Yeah. I have to shut up now so you can have a break or something. Brian is going to come up here and tell you what you're going to do. Thank you.