 App Engine is another example of this infrastructure like Amazon EC2 where developers, instead of having to, again, not to belabor the point of building your own infrastructure, but if you have a very popular software application and you suddenly get a whole uprising of traffic, the typical response of most applications is to go down due to lack of resources. So if you're the developer of that application, what you need to do, can you hear me in the back? So where was I? We're talking about App Engine. So instead of having to actually make a scramble, purchase more machines, get these machines into a data center, and also be purchased more bandwidth. Worry about all these infrastructure things. Developers can write their software, push it into our infrastructure, and we will scale that and run more instances, handle more data, provide more bandwidth as necessary. And as Simon pointed out, multiple companies are doing this sort of thing. Open source code is obviously a big part of the internet. Open protocols such as TCPIP, which is what the internet is built on. Various open source systems and software are sort of what got things kickstarted, whether it's Sendmail or other software such as that. We've open sourced over 100 different projects so far in the last four years that have provided our developers and our consumers a better experience and better ability to innovate and use those. To touch briefly on one is Android, our open source mobile operating system. Android, the goal of it, we have a number of different goals with it, but one is it allows our hardware manufacturers to have a common platform focus more on hardware and innovating in their hardware. It allows developers to focus on developing applications that can run on a common platform across multiple phones. And lastly, consumers get a better internet experience on mobile phones, which is very important to us. But what we do think open is important. Open is integral to the internet. It's integral to the continued success of the web. But I want to take a different angle on that. I don't think that open is the end goal of all of this. The end goal of this really is two things, and that's choice and trust. These are two big things that made us really successful. We always say that our competitors are only a click away. If someone wants to try another search engine, they just type a different URL into their browser. It's just that simple. But let's think again, go back to the 90s and talk about software and how you acquired new software. Now, the first thing you would do is you'd go to the store. You'd purchase a box. Inside of that box was a diskette, and sometimes if you were lucky, a manual. You'd come home, you'd put the diskette in your computer, install a software, reboot seven or eight times, and then you'd have this wonderful piece of software you can use. If you decided you didn't like it and you wanted to switch to another piece of software, you'd go back to the store, you'd get another box, another diskette. You install that in your machine, and you try that out. There's a really high barrier to entry, to getting started, to just trying something out. Now, you don't have to do any of that. It's merely a matter of typing a few characters into your web browser to try another application through another website. So that's your choice. The barrier to entry, the ability for you to try new things out is extremely low. So you can give something, get a taste of something. If you don't like it, you can move on very quickly. And that leads to trust. I personally think especially that it's important that now more than ever that consumers trust the companies that are providing their services to them and posting their data. Because it's so easy to switch. It's so easy to change, move around. Again, I'm speaking specifically to consumers here, much less on the enterprise level. The other example I have on this is if you run a flat, let's say you run a flat on this side of town, and it's a little expensive, but it's been good for you for a couple years. And you find a better deal across town with a view of the river. It's a lot larger. You decide you're going to move out. So you go tell your landlord, well, I'm going to move to this other flat, so I'd like to give my 30 days notice that I'm leaving. And they said that's fine. You're welcome to move out, but you're going to need to leave all your wedding pictures, your children's shoes, your journals, your diary, your furniture, and your clothing. But other than that, you know, have a safe trip. Ah, please. I have to sabotage this. Next, I'll speak with a glass of water balance in my hand. Thank you. Okay, we're back. Is this better or is this better? This. Yes, okay. So this is, so if a user, if you have this landlord and you try to move out of this apartment and you have to leave all your possessions, many of which are very valuable to you, you're probably going to wind up staying for considerably longer. It may get to a point where the cost of living there goes so high or it's so inconvenient for you that you will eventually leave all your possessions behind you. But I guarantee you will never do business with this person again because you've been betrayed by them. You haven't had this trust with them. If consumers give trust to a company by putting their data in them and using their services, there needs to be a way to revoke that. Okay, there needs to be a way to say, I no longer trust you. I'm going to take my toys and go home. So this leads up to the team that I started a couple of years ago. It's an internal team within the company called the Data Liberation Front. We couldn't really agree on a name when we decided to start it. There's an internal team and we took a cue for money, Python's life, for Brian, and named it up to the Newtonian People's Front in this case. But we saw ourselves as sort of a subversive group of engineers, although our idea within the company was anything really but subversive. We actually decided, instead of trying to convince all our products to make it easier for people to take their data out of our products, that we would actually help them to do that. Because when we talk about...