 We've been developing a set of web applications which whose idea is that they should look like desktop applications and function in the same manner. All the backend is Linux Apache MySQL PHP and the data transport is done with XML and the frontend is Flash which some people may find it unattractive. But again, it is, that's what's out there in the market. But as you can see, the whole concept is that the applications have basically the same functionality as every desktop application just give me a second. You can open full screen, you can do several windows at the same time and replicate many of the functions here. We are opening all our applications today. Later, they will be released under the AGPL, the Afaro general public license, I don't know if you've heard about it. It's basically the GPL with one class added which is usually, well what the GPL says is that if you don't distribute your code, you don't have to return the changes to the community. But with web-based stuff, if you serve them and if you provide a service over the web, basically you're distributing your applications. So the Afaro GPL license, what it does is just add some clauses doing that so that if you use it commercially or for any other purpose, and even if you don't distribute the code, you have to give it back to the community. So that's what the GPL does and that's the license that we're using to release this. And let me show you some more of the applications here. Oops, what we're releasing right now is all what you can see here, which is basically a file manager email calendar. Everything is in open formats, for example, the calendar uses iCol and the context directory is here. It is based on LDAP and vCards and all that. So the whole idea is that you are able to switch providers if that's what you need, for example. We believe that the monopolies that could be created with web-based applications, which seems to be the thing that's coming if you look at Google's strategy, for example, it looks like they're going to be heavily based, they are relying heavily on web-based applications. So what happens there is that as it is today with Windows, for example, your operating system is owned by somebody else, but as it would be in the future, even your data would be owned by somebody else. So we believe that in order to break this, web-based applications that provide the same functionality as desktop programs to users are needed in an open scheme to break this. So that's basically what I have to say and Carlos can talk about some of the technical details. Hey, well, I'm going to talk about how we use Linux to build this website. As you can see, we basically have all you can find on the internet, but just plug into the same page. We are using BossFix to email. We are using a storage based on GPS. And for example, the others book, Authentification, all is based on LDAP. We are handling about 100,000 users. All those users are stored in LDAP. We use NCCD because of the authentication must be as fast as possible. We build our own schemas for LDAP, for example, to handle all preferences. Our others book is based on LDAP as well. The calendar, as Oscar told, it's based on open standards, open protocols. So we use iCal, for example. We use big cars for exporting port data. So everything you have, for example, weft calls can be added to our calendar. We have a news aggregator, messaging, we use Java. You can add as many Gmail accounts as well. And well, we have very uses to this application thanks to Linux. We are giving this software not only as a web services, but also as a LSTP desktop, the Linux terminal server project. We are working with government, we've managed to handle, we've managed to build unattended Linux installations based on Debian. We build, we give the user a single CD and we package all those services, lamp, Java, we use Lame for MP3 decoding. We give the user a CD, the user just press enter and the unattended installation starts. You get a full server style it. Once you restart, you just configure networking and some default settings. And you start using SAP LISC. We added the LSTP project to launch about 15 machines without any hard drive just they put from the network. And we have this for about 20 schools in Mexico. So at the end is cheaper enough. And the whole idea here is that according to the open source development labs or it's a new incarnation, I can remember the name now, but 70% of all desktops require very basic applications only. So what we're trying to do is to put all these applications on the web with the same functionality as desktop applications so that this idea of web devices can really take off and it has worked for us, for example, there's a company in Mexico that has 12,000 desktops, it's a mining company and they started migrating to Linux recently and they are using some of these applications here because they, for example, had a survey amongst their users for switching them to web mail. They were using some windows and outlook and many of them didn't want any web mail. So they show them this and it was an acceptable compromise for many to have some web email that replicated some of the functionalities of a desktop email. As it is right now, we're opening it is, we call it in beta but web software is always in beta and we're opening it to the world to create, we aim to create a community around this and we aim to create a community that brings together both open source developers, free and open source developers and creative professionals. That's something which that's the culture of our company, that's how we work. Carmen here is a designer and she's the one who has created all this look and everything so we aim to bring more people into the free and open source community, people who usually wouldn't even think about participating in a free and open source community like creative professionals. That's something which we have, we participated in an event with web designers and all that recently and told them that we were opening and it was the enthusiasm that we could see was enormous and we already have 150,000 users, we have two commercial services, that's how we plan to survive as a company, not selling code, not selling software but selling a service based on that and we know that there will be competition, we hope that this competition is based on the same software and we want this to grow and be something else. So that's basically, we really don't want to steal more of your time so if you have any questions, but all right, shut up. Gosh, what do I start here, gosh. So you basically first said off that there's a trend that services owning a data. So I thought you were coming up with a solution that people can have their own data. Is that how I understand it? Like you give free software so that people can run their own desk on their own computer sort of thing. That's the whole thing. The whole issue is to promote services based on open standard, services based like this. What we want to do is to allow people to at the push of a button, get all their home and all their data here and if. But where's the data stored on your servers? On our servers, but the whole idea would be that if this architecture is replicated, if there are many providers using this same thing, it will be able to transfer your home somewhere else if that's what you wanted to do. So the plan is for you to hold the data for these services, but hold it in formats which are easily interoperable with other people providing competing services. Exactly, exactly. And allow people to move their data whatever they want to. I mean, this is something which you're probably not gonna be able to do with all the problems. Well, when I first thought of this, I thought this might be you're doing some sort of, you're latching onto a GNOME desktop, but I guess these are not GNOME applications. You're basically, oh, are they? I didn't quite get that. This isn't GNOME, is it? It's something else. I mean, this is not a GNOME. No, okay. This is not gonna go out there. Okay, so you basically had to re-implement every single application into the flash thing. But some of it's standard applications like Java. I don't get that. Some of them are standard applications like Java. Yeah, but I mean, re-implementing the whole freaking desktop stack is a lot of work. And... How do you want it to be GNOME? Huh? How do you want it to be GNOME? No, it's not GNOME, it's something... It's web-based, look, it's a browser there. Yeah, but... It's a URL. A web application isn't a flash application. That's, I think, quite fundamental to me. Yeah, you must have to rewrite everything. Well, but the whole issue is 70% of desktops use very, very basic stuff. 70% of all desktops are just email, calendars, contacts, and the basic office suites. What we're doing right now is providing open office through VNC so that people can do like editing of documents and stuff like that. So that's basically our concept. Create a very basic scheme for applications that replicate the functionality, the look and feel. It's a web application. But it's not a web application, it's a flash application. I can see this working with the... I can see this working. I mean, I like the idea of having open standards and a server backend. I mean, that's important. That's really important. That's why I don't think Google could actually succeed in the long run because they need to open up the server backend as in I own the data. So I thought that's a good idea, but I don't think this flash application is a good idea, really. You should be doing web applications, not flash applications. Well, the thing is that the only way you can create all this look and feel, all the functionality of the desktop today is basically with flash. We do it out of a pragmatic approach. Yes, you can definitely install the plugin in Linux. That's what you want. One of the things that they make very clear on the website for this product is that they would like an open equivalent of flash to migrate to. And that's the thing, Nash is there, but it's still in alpha and it has been in alpha for basically forever. So something has to happen there. I mean, what you have is the neutrality of the backend and the transport, which is XML. Basically, what you do with the interface, what we're doing now is using what's available out there, which is flash. If someone would come with something else, then I guess it would, everything would work. I'm coming from, I work on the Watt workgroup, web hypertext application technology workgroup, I work for the W3C. And basically using an API, which isn't really part of the web. So really got to make that clear to you. Flash isn't the web. And what a little thing I got carried away there. I don't think that there's no possibility of joining your top point of view. It's too complicated. Your is? My point is the one which I like. Google rights app, Google has quite a big market share. It's doing pretty well. Google are using web applications. Don't you see that Google is a success? Absolutely, absolutely, Google is a success. It's a strategy that works. It definitely, definitely works. What we want to do is to have people, to give people the possibility of installing this in their own servers, in their schools or anything else, and also provide their own services like this. It's like, look at the state of open source web apps like Squirrel, for example, Squirrel Mail. It's definitely back there in 1999. That's where we remain. Nothing else, nothing has improved, nothing has evolved in the open source with regard to web applications. This debate whether Flash is web or not, I believe you know how long it takes. If it's HTTP, if it uses port 80, it's web. Okay, I would just ask my question and I give back the mic to further discussion. Okay, I just want to know if you, in fact, I've got a small problem. I just want to know if you have a solution to provide people with sort of key in order to encrypt the data which you can place on the hard drive in order that this information keep our confidential only to these people here. In other words, when you come to, for example, a remote desktop, okay, you just plug in your, let's say, USB key with your cryptographic secret and it will be able to decode the content of the hard drive, but you need this key, only this key to decrypt the content of the hard drive so that even if you've got the data on your server, only the people who own the data really can only read it. It's a concern about privacy, in fact. I believe that those are the things that an open project would do. It would create all these derivatives based on people's own preference. I mean, if you have this in your servers, do you want to do this? Do you want to have encryption? Then that's something which we would certainly encourage you to do to create around this, you know, an encryption module. All this stuff, for example, as it is today, we have Java with the MSN transport implemented. That's what our users have requested. But the transports for Yahoo and for other messengers are out there, so the way we visualize it is that eventually it will include all this stuff and more. It basically depends on what you want to do. I mean, it's a small download. You just put it in your servers and run it and, you know, start doing whatever. Okay, great, thanks. Anyone want to ask a question anymore? Okay. Well, I mean, most people, that's what they want. Most people want windows. Most people want these functionality of a multi-window environment. Just to give my own point of view, I think that's ideologic things. And I'm not sure, maybe the best solution is to push Adopt to open his own source code. But I'll just point something. We have structure or development based on three layers. So we have backend, we have a control layer, and we have, well, what you see as flash. So our control layer just splits XML. So you can use XML translator and you can turn all these data and applications into a HTML page. So that is no problem. You just have to design your own frames or divs or whatever you use to deploy XML applications. And all the functionality and all the data will be the same. So, well, we are opening everything, not just the flash application. We are providing you a backend configuration. We are providing you with our control layer that you will get in minutes all this XML stuff. So, well, you can do whatever you want. You can use any Python, Parallel, or whatever script and language to process this XML. So, well, it is so independent. Now, well, I just forgot about backing something. I just forgot about backing something. Any more questions? Oh, I remembered.