 Okay, so everybody last slot of the day lightning talks We got actually four people signed up here. First one would be oh rules are six minutes and I will do would be hard and forced So ten minutes ten seconds to the end of the talk. We're just we're gonna stand up and applaud So it's very nice and run some way to cut you off Okay, first off it's one two What is this? Okay, I call it distributed ODF applications Where is it? Is it going this way? Okay, so from a bird perspective right ODF applications adjusting can we Toss and toss and I've got a screen problem. Can we switch it that it's fit? Should should it's should switch the screen. It's a it's over here. So I'm Yeah, I'm just a pedantic. Sorry. Sorry. Sorry. I'm waiting six minutes for a brilliant thing I just leave it this way that's it's good enough for me full screen that's much better, but it's it's should be wait a minute No, then it's Why is it working? Okay But but give off the page you know of page feel it Fit page fit page is better fit page is good. Let me switch. That's much better. Okay. Sorry. So I've got one minute left Okay from a bird perspective, okay So every every every we have all the ODF applications Microsoft Office or they have toolkit or so on It's always starting with your load the ODF file into yourself And then you there the state machine the user macros they adding deleting or modifying Document right and at any given time you flush it out. You save it export it to another valid ODF So this is from a bird perspective everything that Libraff is doing right, okay Very abstract right because this is this very important, okay And the next thing is when we have multiple of these state machines working together like here Then and we want to collaborate in a perfect sense like we do with software then we need to You can think of even as not even as three Libraff's versions or Microsoft Libraff So but it might be git repositories and look this about this guy here is adding something is inserting your second paragraph and The others are pulling these changes and this one is just modifying the third paragraph And this one is just doing nothing just receive the changes, right? This is our goal. We want to go that direction so currently we are not able to Receive any of these changes these high-level changes We cannot receive some XML snippets because in the internal model. We don't and know anything about parts we can only load everything and Drop everything so this is very very important if you want to do collaboration that Libraff is learns to Accept changes so these changes are just some high-level Fused changes so what I'm going to do is I'm currently working on I did it tomorrow There this morning that this ODF XML grammar is a nightmare It's a very huge text file with a lot of XML notes. So what I do to answer questions like What is the late? What is the? What is the minimal document? I loaded in a graph database because this is exactly what they can do What is the lowest smallest ODT possible our paragraphs nested at all and all these things you have to look into this long text file And this is horrible. So what I did is you see a lot of using a graph database and Making visualize it and this is only the table table element with all the child elements and The red things are the attributes, right? You can even see it from here. So I I'm going now to improve this graph because this is just an internal model and There are few things that we have to to to skip this To to to view better, but we have to extend this grammar. We have to have a better tooling Otherwise, I mean one otherwise we cannot cope this complexity, right? And this is I think this is the brilliant way to go using graph databases annotate more annotate these changes because we think XML all this now we specified everything but in our lines and all the users minds the user changes of all the same You can in every ODF application you can add a paragraph add to lead a table and so on or like I said Modified format a paragraph ground and we have to define these things and we have to find it in an automatic way and specify this Three so we have something like ODF git repositories like we have real coloration where I can oh here's my pull request, right? You have your ODF book outside and as I read it and said oh, there's a typo I've got a pull request to you and you don't have any ideas you can internally have ideas But you we should be able to have a coloration as powerful or even more powerful than the kid. That's it. Thank you So I'm having a quick talk about this state of labor office at the city of Munich First of all who am I I work for the Linux team since 2016 and I had of development for LibreOffice of the city of Munich and Presenting the city of Munich in the OBS PA and also in the advisory board of the TDF And I'm also working on GIMP dark table and a lot of other stuff, but mainly doing packaging but You may ask wait, didn't you decide to switch to Microsoft Office? not really the government only decided to switch to Windows and The city council will decide in November if they are going to switch or if they want to switch or not so but even if we switch to to Microsoft Office, we need LibreOffice for the next four or five years and at least two or three or maybe four releases and so we are going to continue the development as usual and Also continue our community efforts and the advisory board membership What have we done in 2017? We started development of a new relief for office release based on LibreOffice 5.2 We finally got a new build server which improved building time a lot building for releases at the same time We also started development for one looks 18 Which will be released on one looks net and Jmux did a lot of work upstream on scheduling and bug fixes and We also sponsored a lot of work at CIB and KDA be mostly CIB re-rotating images and Also the border frames you talked about signature lines as I talked about today and improvements in hyperlink sending KDE 5 plasma back ends which is based on GTK3 a lot of work by KDA be in cooperation with CIB and More than 40 improvements in OO XML and also many many more bug fixes and performance improvements and What is to be done in 2018? We want to improve web-duff support Polish with KDE 5 support For our next buzzed line 6 no release which is based on or which does use KDE 5 We also want to roll out LibreOffice 5.2.8 All patches for this version are upstream and back ports are also available on our private branch We are getting rid of more than 100 internal bugs fulfilling a lot of feature requests by by our users and finally getting rid of LibreOffice 4.1 plus 300 or something like that back ports and patches Which is diverging from 4.1, which is upstream And we also are going to start development for the next release which is based on LibreOffice 6.1 We want to make change tracking rate again, but it doesn't crash and But we are going to pay for it ourselves so More improvements of our OO XML support and also at bug fixing and some user requests And hopefully we can host a bug-scratching party Munich last year. We didn't have a chance, but maybe next year. Oh, we'll see you Okay, that's all. Thanks Okay, while Thorsten is preparing the A Website hopefully my website I sent you a link to a site stops in clock Okay We're talking about GSOC as a pro organization we applied for being accepted and Application has been submitted. I have the dates here Announcement if we are accepted will be on the 12th of February and Student application follows between March 12 and 27 so we are Hopefully looking for a lot of mentors and of course students, but since here are more mentors and students I am happy to Be able to announce. Yeah, we have some Success stories from last year that could be you the next year So that's the successful projects from from the last GSOC Season and Yeah, so um, so we've been doing GSOC in one way or the other since I think 2008 back in the day with with OO build and before that I was doing that I think two years with with Sun at open office So roughly, I don't know. Let's make it 10 years of of GSOC Between I think between always between four and twelve students So it's a great opportunity both for For mentoring for getting into mentoring for students. So if you know students in your community or Friends and family encourage them to try GSOC. It's a great thing. You get paid for open source work It looks great on your resume And you get a t-shirt and That's yours Is it really no the left side is for you? Left side is mine. Oh, cool. I got the left side and have two minutes left I'm really happy to have chance to advertise some ideas from the design team We just We have a really large list of topics that Could be improved for instance change tracking. No question Something what I would like to highlight is the addition thing addition means to integrate our Extensions more closely into the program integrating means at the place where you use function you get the chance to add things from a certain place our extension site or if you are in a in a company you could Restricted to the company extension for instance if you are in the templates manager You can click on the link to add a template. You don't need to go over the side Should be a quite an easy thing just a framework it downloads A file from a link kind of configuration should be behind it and the dialogue Sounds quite easy to me Blurry shadows, that's something we really Desperately need we have a really hard shadow a shape has a shape behind and that looks really terrible blurry shadows makes the draw even better if it's possible at all in this sense we could also improve the The lines if we have some kind of brushes so that lines get Looks like scribbled lines. It makes tool a really great mock-up creation tool Font handling I talked this afternoon about font handling we have some topics around font handling where Mentors are welcome. I think students should be really smart if they do work of that Next one. Welcome scream is two minutes left It's not really a welcome scream. We have a problem with user Configuration we are still placing things in in the user directory under slash four and everything what the user Configured themselves in the past is not overwritten So if you update you don't get the updated information and that's not that good The idea here behind us that you can Show a screen for the updated when the new releases Reheable what of the previous Configuration should be overwritten so that user can say yes, I accept it and this dialogue could be used as well for Welcome and tip of the day right so this was all the user experience High-level great ideas That you can really make an impact with and a big splash and the right side is a few sometimes a bit more low-level project ideas just a random Subset of the ideas page page maker this document liberation project for import Domain specific language for UI testing. This is to make it easier and quicker and nicer to write UI tests Smart art that got recently the last year had the import finally done. So now will be great if you could also edit that and Charts and or course or because I say special calc filter import filter Python support for into a liberal officer clips and VML import for writer VML is a part of the OXML standard Okay, and we're running out of time So Yeah, just a summary and two links to the resources Please do go out and advertise Encourage people encourage mentors or mentor yourself or you student yourself. Thanks everyone Okay, and we have I think Olivia, okay. Yes. Let's see if we can see so Here we are I want to show you a very little. Oh my god Sorry. Yes I would like to show you everybody knows in the in the work office that the help files are quite Hard to to write so we had in the past a small extension that allows you to use a writer file and generate a XHP file, but We had some issues with this extension we have some problems, so I decided to and One of the issues that you have to review the XML that is generated and sometimes fix each of the points of the tags, so I Decided to look into Kate Kate is the kitty editor and Kate has a very nice Feature which is some widgets that you can have here and When you click on the widgets you you can generate all the XML and just Insert the contents that you want so I'm gonna just show you how it works. It's very simple. For example, I have this Content which is purely text and if I want to start generating a specific Help file for example, I come here with this. Oh, there is With the here new XHP file so double click here and then he it generates It generates The the XML I can let me So I can here in page topic Write anything I want. Okay. The page title also I can type anything I want and The fun name is automatically generated. So I want now this hidden heading one I want to generate the XML for this heading one and I get here a specific widget Okay, so H1 And it generates. Oh, I'm sorry. So I select I Select and I Double-click on each one and then it generates Paragraph role equals feeding ID This this ID is the most cumbersome Activity that we have to do because each ID has to be unique level X equal one XML language In us and your content here. So if you have For example here, I want Emphasize This Put emphasize into this String I just double-click here and it inserts the tags Okay, so and then I want to trust on that into a paragraph and Double-click and you have all the XML things that you need to do. So quite a fast way to quickly get some contents and generate an HP file and Most of the very boring things that you need to do With the IDs it's automatically generated it's a set of JavaScript that runs in the text and You can have most of the X HP Tags implemented here. So this is what I wanted to show and You can There is a page in the wiki where I describe how it works. So no more excuse for Like missing help pages Hey, thank you Okay, thank you very much that concludes the very last session of the day Thank you all of you for being here. Thank you all the speakers for For great information for great talks. Thank you for them for hosting us and Yeah, I have a good evening and a great Sunday. I don't know. I think