 Well, thank you for the opportunity. What I'm going to present is some discussion that we have inside of Document Foundation about the way that we should handle all the documentation. LibreOffice is a complex application from the user point of view and we absolutely need to have a lot of documentation and all the offers that we have so far, documentation in several formats, a lot of overlap, a lot of work in terms of formatting and from the user point of view quite a lot of dispersion in terms of contents and documentation. So essentially what we expect to be the almost perfect documentation is a small pitch on a movie that tells us in the user point of view what is the ideal documentation. So basically, you remember this scene? So this is the ideal documentation. I really would like to do that the same with pilot tables, with styles, with paginating. This is the best thing that I could imagine and that's the dream and the vision that I want to bring here. So essentially if I could just press a set of commands and download all the information in my mind, that is the good one. So essentially from the user point of view, what is what we want? We need documentation. This is like I said, it's a complex application. It's something that people really like, that people like documentation. In terms of a community, documentation is perceived to be one of the easiest way to join a project, a open source project. People want to help. People like to be part of the team. This is something that I perceive in some cultures. People like to be part of a team and contribute, so it's a good opportunity for them. They like to feel responsible and documentation is a very good start because it's perceived to be quite easy to jump in. So the problem is that we have a hard work, which is the documentation costs times and resources. Whoever tried to write a chapter on a specific feature of LibreOffice, for example, is going to spend quite a lot of time. It's an extensive job because you have some rules to how to write. You need to be concise. You need to be precise. And it's not just describing a set of commands. It's a never-ending job because features get improved. Behavior of the software changes, so it's a never-ending. You need to update their documentation. And of course, you have national language requirements, which means that all this documentation, ideally, should be translated into as many languages as possible. And also, we have multiple targets. So far, what we have, it's too many media formats. We have XML formats. We have ODF formats. We have printed formats. We have online formats. And we have, let's say, wikis help question and answer, such as Stack Overflow or Xbox. So this is quite complex. And it's very hard to contribute. For example, people, we have a set of books that is inherited from the Open Office time. And what happens is that when you start to update the books, you face all the complexities of the structure of a document, which means you can't just type in information. You must be compliant with the styling, the formatting of the book, and the styling of the writing. So it's not just come in and start adding contents. You must really be part of the team and understand exactly what is styling. So whoever started to make a book is facing all the complexities of the very flexible styling resources that Open Office and LibreOffice have. On help, what we face today is very old and complex technology. It's an XML format. And this XML is bounding us in terms of the improvements that we can benefit with new technologies that are available in 2017 or 2018 like now. So we started to make a vision about that. And we would like to converge all these contents into at least one big repository that could, after that, generate all the formats that you want. This is a vision. I'm not saying that this is maybe a reality, like the other download of the piloting of the helicopter instructions, but it's a vision. So essentially what we have is we have the literature already available. We have the application help. We have the collaborative editing that we collect in the Wiki. And we have this collaborative support that we collect in Askbot, which is a software close to Stack Overflow. I, by the way, don't know whoever can really make some lines of code without looking into Stack Overflow today. It seems to be the holy grail for a lot of coders. And with that convergence of contents, we want to have not only the output in terms of online, but also we want to be able to print. And perhaps also to add into this point here, I just misread it, is the multimedia formats that we also would like to have. So in a sense, it's one unique and official content source. Why is it official? Because it is important that we be the source for many other kind of industries, like training industry, publishing industry, and get a unique and official source of information on LibreOffice. So the convergence is basically an open community, as open to community as possible. It has to be easy to join, like not only to contribute, but also to access. So websites, tablets, local desktops. Easy to join in terms of contribution. So not to have a steep entry barrier for the collaboration. Visible for user, searchable, accessible. If you use some search tools with LibreOffice, you will see a lot of information, but it's not very well indexed so far. It has to be better. And of course, if you search for, say, pivot tables in Google, probably we'll have 10 pages of Excel and one on Cork in your screen. So it has to be improved in terms of searchability. Good reference for users and developers, capable to develop and support the migration and training industry. It has to be also evolutive and versioned, which is also important to keep the same pace of the software and connected and synchronized with the software. Easy to translate as possible. It's also a very important issue. We have in LibreOffice and Apache OpenOffice, we have the help which is connected, loaded into a translation server, and we have as many as 50 language actively translated. So basically some achievements that we did so far. We have moved the help of LibreOffice into a browser, and also we had the help in your mobile. If you like to take a metro and raise the help, you can because it's available on your mobile. And also we are using a lot of resources that we use, that we have with modern browsers. So this is the current layout. We mimic the old help so far. We have not yet improved the contents because we still have some legacy situations that we must handle, but essentially this is the layout that we have. And also we did some work on transforming the chapters into an online reading. This is basically the HTML export of LibreOffice, and I had to do manually some adjustment just to make the layout better for online reading. So essentially this convergence means that we really would like to have only one repository of information capable of generating the several possible ways of documenting. If we come back here, we have on this side here, the contents of the current help that we have is also a user guide. So we have a user guide here and we have books that are user guides. So this is duplication of information and very often one jumps to the other. So the road ahead is to define a content plan for this convergence or a curriculum that could be able to preserve the legacy contents of the current help, preserve and address the translations, include multimedia and rich experience. Multimedia is far from trivial to generate. We have a lot of small tutorials in Google, in Google channels, but very often it's too amateur, it's not professional, it's boring. People want to explain their best, but they don't have the skills for a professional work. So this multimedia is well below the quality that is necessary. And also to unify the contents of the current documentation media, such as the help, the books, the wiki and so on. So that's it. I'd like to discuss, and I would discuss with perhaps Regina and also Svante, that we need how to address this issue of making a format that can not only be on an office document, but also on a web document to avoid duplication or to make the bridges between one and the other more easily, okay? And that's my talk here. I think I'm on time. I just want to perhaps show you what we have so far. Essentially we have the, this is my situation here. For example, we have the spreadsheet. If I open the dialogue for format of a cell and I press the help button, then I have my browser with the new help displayed directly from the application. So it will be a matter of time that we could drop all the old help and use the browser of your machine. And with the browser here, then we can insert specific styling or JavaScript to make it much more dynamic than a specific textual and extremely boring help to read, okay? That's it. Also, just to show you the last one is the transformation of the book into a reading online, which the export of XHML. I'm sorry for CSS. This is typically CSS issue, not to have it totally well aligned, but if I, let me see, okay. So if I shrink a little bit of this page, I could have it better displayed, but this is what we have and we can scroll the contents this way, okay? So thank you for your attention. I'm open for questions. Yes? The user documentation team, as last year they presented how they worked with documentation, how they documented your office in a quite deep extent. Who? The SUSE. SUSE. Yeah, no, I've not discussed it with him. Yeah, good pointer, yes. They have already summed up all the existing information. Okay, okay. Yes, thanks for the hint. Other questions? Okay, so time is up. Yes? Well, it's in the very beginning, so it's quite open for definitions. There is a lot of work to do, but essentially it's absolutely important to have the right definition at the beginning in order to really bring all the legacy contents and have it properly formatted and properly indexed into the convergence. But we don't have yet all the specifications. We have some ideas, we have some discussions, but it's not yet totally closed. But we don't want to be static. We initiated to do these movements toward more than help display and boot display in order to evaluate all the potentials that we need. Okay, thank you.