 Hey, everyone. Good evening. Hey Yay, we've managed it with this is the last session Last talk say lightning talks. Um, we got them This many volunteers here. I'm very glad to report. Um, so more or less all slots are filled We're not probably not gonna run this until the very last second. Otherwise, we will get kicked out violently So I thought everybody got eight minutes and then we will 30 seconds before I will stand up And then at eight minutes, we will friendly cut the people off with a round of applause and we start with Olivier Thank you. It's awesome Good night, everybody. I'm going to present the Current status of the help editor. Yes Okay, so basically what we have is this editor is made for editing the XML and and We we I need my computer Well, okay. Let's let's let's go and basically what we have is is essentially This editor we can open and choose a file of Eight of help to open. I don't have any file in this local computer. So I can't pick one You can also save it. Okay, but you also can start documents from scratch and we have Set all these menus to make some XML blocks so that it makes it easier for us to edit The contents of a file or to create a new file this case here I will start a new HXP document. It asked me just because I'm going to clean The editor and then I will have to I will add some new things. So I say, okay, and it starts giving me this XML which is exactly the start the Initial content of an X of an help file. You have the meta. We have the the help document We have all the tags this editor here is code mirror Okay, and with this code mirror, we have implemented not only the edition but also several add-ons so I can display use for example The full screen editing we have a search and replace feature we have Auto complete tags everything is available for making easy to edit For example here, I start my My editing here in the body if I want to add something I just click here for example, and I want to make a title. So I'm going to type whatever thing here and then Just a moment then I select it here and I Format as a paragraph or for example heading such as each one it immediately Surrounds my my text by the proper tag and it adds a specific identification which is Unique inside the file. So something that is very cumbersome for whoever edits the Help files in in liberal office is to find a new string for the identification And when I'm done I can he come here with the tools and ask to render the page and Okay, so we have resolution issues here. Well, it's It should have rendered the page on the left Yeah, we have a resolution issue here it the result of this page is displayed here and You can see exactly what you are typing Unfortunately, it's not working here. I Don't Okay Okay, so here. Well. Yes. So here is the render page. Sorry It should have been on the on the on the right But then you have your title here and I have also implemented the possibility to make a choice You those who knows who edits help files knows that there is difference between the systems and also difference between the modules it's surrounded by this switching line or switch Tag that we use and here is so Here in the results. We have also other debug info. Okay, and also we have implemented some Some checkings for example, I want to check if the file is correct and it's It makes the validation. So first of all checks if the XML is correct So you have opened a tag and you have also closed the tag Check with the XML document type definition to be to see if the DTD is correct And also check if there is a duplicate ID in in the file Okay So before you submit your your file, you can run these checks and it will Guarantee that your file. It will be Correct for the help the help. Okay. We have implemented all the Other Nice t such as in the dock heating for example Just to come back here If I want to introduce a heading Sorry Where I am here Where is my mouse? Well, I lost the mouse Sure where I am Yes, it's on the other screen No Yes, it's closed. It's bounded here I can't I can't use it Okay, so here we are So we we we can for example introduce this chunk of Of xml which is the dock heating And it will introduce here lots of sorry lots of a section ID the bookmarks the heading At the paragraph with the a help everything is already there and with some Placeholders so that you can easily fill with the content. Okay Also, so base what we have is the bookmarks several bookmarks you can fill the sections Generic sections the section that is the related topics the how to get Uh Divisions that are for core basic code and python codes We have some Templates for tables If I click and add a table here For example, it it will generate all the table For me, okay With all the structure and then I fill the content of the table And we plan to One minute 30 seconds and we plan to implement Not only to generate a patch To be submitted to garret and also eventually to load the files directly from the master branch and some help pages the hxp reference That explain Okay Good have Hello, my name is kusha. I'm LibreOffice developer and I'm working in colobora I will talk about The Improvic LibreOffice Interoperability in LibreOffice When we want to open OO XML Based document in LibreOffice sometimes we can Have some problems To solve them first If we have a bug report first Reproduce to we need to report to reproduce the bug and second We simplify the bug document because We need to focus on the problematic part We should do it step by step And check it after every step the bug is still there If it's possible We can we should create the bug from scratch After We should try the bug In older LibreOffice versions if there is no bug another in older versions. It's a regression And We can use git bisect To to find the guilty commit The The documents ODT Docics pptx These all Documents it's zip compressed files When we open it unzip We can use read the XML files and We can pick some clues from that XML files The files we Are are readable, but XML tool is Correct the indentations If If it's possible to create a correct document We can compare A problem and problematic and non problematic files To solve the Bug the meld is a great tool for this compression It has a nice visualization and it allows us And compare the directories And after that When we pick up all the clues We should grab on the source code And debug and try to fix it I want to give an example That I solved before In this bug report We have a document when we save in LibreOffice And open again We missed Three digits after the comma In the chart label While I saw solving First I reproduced the bug I saw the bug I see I saw the bug is there The simple the document was enough simple And I tried it with all the LibreOffice bug bug Bug was there And it is not it was not a regression and for I examine The Chart label in Chart label properties in Microsoft Office and I saw the person to check box is check it I looked for the corresponding Feature in LibreOffice and I found it I debug it How to learn how it works After I unzip the file and I found a suspected attack Here the non-format Format code When I compare them Before and after The file Saving the file I saw the format code missing in buggy file and I read the related OO XML specs And learn What Detax Does I grabbed and Debug where the xlsx export Code and I found it And I have add the format code attribute In the non-format Attack And I saw this All about is Picking some true keywords To grab in LibreOffice The other part is your code reading skills and debug skills That's the patch That I sent And that's all. Thank you Hello test test Okay, I'm Svante Schubert. Please to meet you What is looking up? I'm going to give you a disruption warning now A 10 minute disruption warning I wasn't sure if it's three minutes or far or the five minutes So I only got 10 slides. I would have taken more So, um While he's trying to find Yeah, just found screw mode. It should be possible Windows, I believe Oh no, few. Sorry Great Hello So this could be LibreOffice and I hope it won't become LibreOffice So I give now, fair enough A very short disruption warning Disruptive in the sense of Just a second. Do I use this one? Yeah In the sense of Disruptive feature. I have in mind like Nokia One of the leaders. I love my Nokia phones Was very proud of my Nokia phones But then the smartphone came and they were gone, right? So I'm proud to be here with LibreOffice And I don't want that it happens again So What kind of disruption might happen there? So we have in the TDF A new toolkit, more or less new. It started with Sunnet In the beginning of 2000s And we put all the Java source code we have to manipulate ODF documents And it went from Apache now to DDF and was forked by By the OpenExchange for their backend And I merged it back and soon you see there's a snapshot One zero zero. This feature will be integrated And this is all Wait a minute. How can I do this? Okay, just a second Try it Oh, yeah, it is wonderful. All right. So that's very simple. So All we're doing now It's just a green box here Not a black box because open source So we drop in the text document and it becomes Totally equivalent sequence of user changes So we're entering now the realm of changes here And you can merge new changes back. Why are we doing this? Because We don't want to as a software developer zip our source code repositories and then send them to other developers If we share our work all we want to do we want to dispatch commits diffs changes So we are able to merge them back, but it's impossible We have zipped XML and these are like repositories and we need to come down to changes to really do collaboration Like an asynchronous offline collaboration where we can Where we can work together in a more comfortable way And this is becoming sooner tdf a feature And i'm going to document now. I'm just documenting. That's all the work you're doing now So in the end I would love to standardize even the changes So not only the full zip repository is being changed, but there's some odf change and you see It doesn't matter how we serialize it It's the user knowledge of these things the entities when I phone you about my change and it's going to call you Hey, by the way in the third paragraph the second letter, please delete it Okay, if you if you phone my mom she doesn't know anything on odf or so But but she's sitting in a phone libre office and you tell her there's the table There's the image then there's a paragraph and this paragraph go to number Now make the third third character the fifth please bold and that's the way we we go to the semantic level with stable And we are going to standardize these semantic entities and set them around So in the end we do a fine collaboration here. So Why is it cool? What can we do now? So we are first I standard We we no longer sending repositories. We're going to send changes, right? So we might be as github we can say there's some famous author now having his book outside and he doesn't want to Get copies because it's a very huge book But you can send a gift. Sorry a github pull request to say him. Oh, by the way, this 1,000 countable thing that's being a paragraph There's a type of please delete it and he can take just to take a look at this tiny thing a change and he can say Oh, cool. I merge it. There's no bad jokes in this Copy you might receive. So it's very easy and very comfortable and of course Version systems like it currently there are only line based and there's only binary file But if we go to a abstract semantic level then we we can say What are the difference between two versions of this Contract for instance and somebody say, okay, we just return these changes. We just defined and you say oh And the third there's a new third paragraph and the fifth one. We've deleted. Okay, so we can come to a new level of Collaboration and handling with documents, right? and I think one of the most powerful things is that these changes are already exist among multiple not only odf applications but In html there are paragraphs as well and there's both italic and tables and they do the same operations like insert a row into the column and You can even load an odf document a full featured document in a vi and text document emacs By simply saying every line is a paragraph and there's text and we just receive from all these changes these changes that are about paragraphs and tables and Reorder the positions to addition in the editor doing additions And then we merge these changes back and filling all these features that were known earlier back in So we are merging changes back and not like we always do with a flop it is load everything save everything so When you ever edit it with LibreOffice Word A very huge document and you make a small change and said oh, I better save. I don't want to lose it And waiting and everything is being saved doesn't matter It doesn't have to do because all you have to do now is to save only the change or just to merge it back So it's becoming much much more faster And if we all got copies of documents and you give me back your changes or a single change I don't have to look at all the documents and pass them I just take the changes each one and put them together. Yes compare these changes So the the merge is no longer dependent on the on the size of the document by the amount of changes, all right so And so we have a different level of interoperability and We will see how it's going to um, and why are we on danger because it's very hard to implement in the LibreOffice People are not Customers are not demanding it because it's they don't know where that's possible and it's very hard I heard from developers to implement it in a way that's being That you see something like I asked Michael style what he needs And he said I need a time machine because the design what was chosen the 80th Was totally different from floppy disk no internet to the current days where we're always the internet and we're collaborating with ourself all the time and so If there's no danger yet, right? There's only this Tiny backend tool not front and nobody is aware of this feature. Maybe you are but um No user knows it but if others taking this up and building let's say Like caka.5 which have changes in their model by design losing this to using this to load and edit Office documents with odf and embracing git then it's becoming very very powerful and then we are in a danger so What i'm suggesting here is That we are no, sorry. That was an old slide. We are we are we are really implementing it, right? We and if we are not doing it Then I think the document foundation if um should take care of and It's not for me because i'm not actually coding in lipo office, but um, I think it's important that we have this in our View, okay. Thank you for listening My talk is about bin bags Not really. Um, but the fostem Organizers came in before so sorry for a bit of disturbance during your talk But they said if we have any rubbish left, uh, we need to collect it together So if you've got anything you don't want to take with you tonight, we'll just put it all in the bin bag here Yeah, close the windows and because you know, I think they're all working voluntary and they're doing the best they can So we'll we'll leave that there now One of the biggest problems we have in an open source project is um crediting contributors and making them feel Um Showing that we're really grateful for everything they do. Obviously, we can invite them to conferences We can send them stickers and small merchandise and gifts like that as well But what are some other ways? How can we say thank you and make them feel really? Um appreciated and one way is with open badges that some of you may have heard of before So this is the super short description open badges verifiable portable digital badges with embedded metadata about skills and achievements. So these are PNG images with metadata inside to show what somebody has done why they are awesome as well The fedora project uses them. I'll show you in a second Good way to show appreciation to contributors And it's hosted. It's its own open source project. Here's how fedora does it So these are all people in the fedora project Active all their usernames and when they hit a certain milestone when they've done something they get a badge They have lots of automated systems set up for this as well And they they make it into a sort of friendly competition between people in the community. So if you've done 150 Git commits or wiki edits or 150 some things you get a certain badge as well So and then they show who's leading in the week who's leading in the month They gather all the information from all the different activity in the fedora project and then they give these badges Now you can decide you can define What these levels are what these thresholds are it can be for the first time somebody commits Actually gets their patch committed to the libre office source code tree. They get a batch as well, you know, cool We can decide all that as well So slightly boring technical stuff, but the badges are defined by json files as well. So just involves poking around in those three files the issuer which in this case is us libre office the document foundation The badge class that Defines the type of award we want to give so maybe for 100 wiki edits or for 10 code commits A thousand questions answered on ask libre office our user assistant website for instance And then this json file that's very specific for the person who gets the Uh gets the badge. So Here's an example again, not super exciting, but just so you can see This is who gets the badge. This is When it was issued, but this is very important as well this This is super important this verified thing because when we issue a badge And if somebody wants to show off that they have a really nice badge from the document foundation that they've written 20 code patches or answered 500 questions With this it can be verified that we issued the badge So it's not just somebody has opened a png image in a hex editor and poked all this data inside And this is this is good when people really want to show off their badges for future Work opportunities as well. So to say I've got a I've contributed a lot to a big project like libre office and I want to show that I want to put it on my linked in profile or Social media It's not just a little cutesy thing that that they're given but actually like this is real stuff that the libre office Project has shown they appreciate what I'm doing and I've contributed enough. So We have a couple of badge designs thanks to our community as well I quite like this one personally because I think it's quite quirky and fun But if again if people want to show them off on their on their linked in profiles And they really want to use it to to further their job opportunities then something like this may be better So we'll have a bit of a discussion in the design community Maybe use both let's see, but we have this working the tooling is set up as well so The plan is now let's go ahead and do it so After in the next few days after foster and we can start giving badges to people for their work We can thanks to Gillem and the infrastructure team as well We can get lots of statistics from different services and tools in the project So we can run commands to see who has answered 500 questions on ask libre office who has made 500 wiki commits whatever code commits as well I think even bug reports. There's there are a few other things we can do gather the data My plan is to initially give badges to people personally so to generate a badge and then email somebody and say with a personal thanks for for your contributions So to start off a bit smaller, but with a sort of customized message the way the fedora guys do it, which is cool Is it's all done automatically in this nice system? Maybe we can go to that as well. We can Aggregate lots of data and issue the badges automatically But I think at the start I want to send them to people and say thanks a lot, you know, we really appreciate it Here's something. It's not just a nice shiny png image, but this has data that you can then show to a potential employer Yep, so that's the plan Do it personally for now and then automate it later Then maybe some of you will be receiving badges from me soon as well So and then you can open them up in hex editor and check that they're real That's the plan and I think that's it. Yep. Was that eight minutes? Kenny Uh Okay, sure. Yeah, we're just using new design at the moment for testing purposes, but yeah That's a good point. I'll make a note for when we implement it live. Yeah, thanks Okay Oh Most probably it won't but we'll see So, um, let me just introduce myself um I have a fancy, um Up here and a fancy device Um, and I would like to show you what we did In the last couple of months. So my name is Nicholas. Um I work for ad finishes with um free and open source service provider And I will become a deputy board of directors for the tdf soon You can also follow me on twitter Where I used the um next lore Twitter name So what you can see here now is um LibreOffice more or less running on um natively on an ipad Um, I would like to explain a little bit what we did here. So this is not um online So I can actually uh disable wi-fi and everything will work as it should So the app is um using the same code base as um LibreOffice online does And it's also the same as we we are using on on and rates and kendi showed it um This afternoon as well, and I would like to show you a couple of Things that we can do with the app. So um If you open the app, what you actually see is the the native iOS file manager and I prepared a couple of files Just to show you a couple of easy things. So let's start with the text document There um You can see that it's more or less the same Use user interface as we have in in online, but we have some um Things that you can do um using the native functions that ios provides. So for example, I can use the emoji keyboard to add some um emojis or um, I can Copy and paste pictures from from the browser for example. So if I have This little animal here. Oh, okay. Yeah, so the browser is offline now as well, but um, Let's just connect To the wi-fi again. I can copy and paste pictures from other Applications to to the app as well. Um, we can use um Things with the finger so you cannot see that I touched the screen, but if I want to resize The table I can actually do that with with my finger. I can also just tap on one of the gray areas around the table To select things and make them bold for example Um, we also have quite big um handles around um the picture to just resize them with the finger You can also rotate them um You also have the sidebar for example, um, which you already know from from the desktop version Um, and there you can also see that we have quite big. Um Which it's to just use uh them with the finger as well um What I also especially like is um, if we use um the spreadsheet component Um, you can actually use the hardware keyboard. Um, that was not possible until I think ios 13 or something so now if you um Want to select the cell range you can do it with the handles as well But sometimes you just want to use the keyboard so you can just use the usual um shortcuts like shift and then um Just use the arrow keys on the keyboard as well Same also works for formula. So if you start to write the formula and then um, you can use the um The hardware keys to actually select the range as well and um Do things like that you can also just um Select the range of cells insert a diagram and um These things will also update. So if I just use um A much bigger number here you will see that Diagram will change and um This all works natively or or like you don't need an internet connection and we can also have a look at the impress part so um There's a slide deck that um I prepared and um, of course we can can edit things here um Change the text and then start the slide show Navigate to those slides have um charts here as well and um You can um edit everything as you would expect um There of course are still a number of things that don't work. Um, it's still work in progress. Um, but I think the Next release the 4.2 release um should bring quite a bit of um Knight nice nice new features one thing that I also would like to show is um The dialogues the native dialogues Um that you should know from from LibreOffice. They look a bit Um like iOS um native dialogues with those green switches and um Yeah, um things that are big enough to actually use with the finger So that's more or less. Um What I would like to show you can download it from the store from the apple store. I have it somewhere open. It's um right now, um You can use the colabora build And I would like like to send some thanks also the develop some colabora for their support for the heavy invest they did and um, it's fair to um, give them a Short applause or pay them a beer later on So if you use colabora off office the search term you should find the version right now It's 4.1 that you can find but I think you should get 4.2 Through test flight and also as a final build in a in a couple of weeks, I hope That's it. Thank you very much any questions No questions good Thank you Okay, so it's me So I'm cheating a little bit. So it's um what I'm talking about. It's not going to be um about um LibreOffice for a change Um, but it's about something that I care a lot about which is digital workflows um, so I was hacking a bit with boobly on something that Many of you might know Which is popular and ocular which is um, um one of the more um More popular um pdf viewers on on free operating systems. Um, so Since this is uh, not a native linux. I'm just going to play some Um screencasts. I was I was doing so the problem um that um Rises obviously these days is that you that you want to sign Documents digitally not not printed out and paper and then scan it again. Um, so that that's a feature that's broadly missing Uh, except for LibreOffice. So LibreOffice can do that. Um, but there might be documents that you're not generating in LibreOffice So what are you going to do? So well this document is generated now in LibreOffice. Um, but um, um, what I show you now in a moment While I'm trying to save that down Is that uh, so that there's two patches um to merge request one for a poplar. That's more or less Building on something that someone did four years ago Which started that and then um, I took quite a bit of code from LibreOffice. Um, that was done I think but with this crowdfunding from this, uh, uh, Willem Talks project that got this uh Partis um extension. So that's like time stamping authority and other nice nice things And put that into poplar, which is um the the let's say the low level pdf reader and renderer And then add it as a bit of UI on top. Um, which no that's not me That was boobly, but I was kind of just doing the finishing touches in the last A few weeks. Um, so While this is building, I wonder if I can just forward that it's getting a bit boring So yeah, let me let me just ramble on so, um Um, right. So, um, actually, I think I skipped too far. Um, so that's the, um The actual document that was signed already Unfortunately, I can't There's no keyframes there. So it always has to oh, yeah, there we go. So that's the the generated document with the signature stamp and the partis stamp of approval there in LibreOffice Well, I should probably scroll back a bit further So that you can actually see um That it's ocular doing that The workflow is essentially like like an um an an acrobat or other. So you just Yeah, you open some you draw some rectangle Okay, there we go. So that's the document You you you bump the sign button You draw a rectangle where you want to put the signature You select your your certificate. Um, you sign it Eventually I will do that there Right so That pulls the like like LibreOffice as well pulls that out of the nss Defaulting to fire folks but configurable keystore And then eventually Selects that yes, come on Okay, and I'm saves that and reloads that and It's all down Okay, so this is um right now. This is in review. The idea really is um to have um With pure open source to have um the complete set of tools that you need For a fully digital workflow But you don't need any proprietary software anymore You don't need to go and and have clumsy workarounds or worse printed to paper So and yes, there it is. Okay bought you enough Um, let's pause that for the moment any comments questions. You have a bit of time Great. Thanks All right, uh, you've seen a lot of very professional calls talks in this lightning talk session And even one with a linked in linked um In the early days of LibreOffice, it was a little bit more cowboy So who who knows what will happen if I press return here? right it takes some time, but uh After that my slow ssd should find that uh This hasn't been optimized from 1989 till LibreOffice 6.0 and there was a talk by Michael Stahl Are we optimized yet and since then we are optimized? So it was only 20 years. Um So trigger warning drugs What number do you expect to come out of that? It's not one line. So it's if you want to see a especially interesting one, uh, maybe look at that one Uh I'm not doing this to protect the guilty. Um Who knows what will we find with this? Yeah, that that was I think it was in when LibreOffice started and it's still in Who's german here? Okay, so let's let's quickly see this one for you guys And if you want to make a german very happy you can ask them to explain the joke to you And finally I also want to like this whole talk is essentially be careful get never forgets um The last one We also have storyteller The guilty are smiling then they know what's coming We also we also have storytellers in the project. Um I really love this source code comment And with that Thank you And uh, let's have a nice little round of uh Fun tonight