 Well, thank you all for being here. I'm going to tell you a bit about Collabra and LibreOffice and how we work together, and a bit about the company. So we are a small subsidiary of a larger company. And I'd like to tell you a little bit about that. So the parent company, as you may know, is 150 people, something like that, based in Cambridge, UK, in Montreal, and it's probably the world's leading open source consultancy, I don't know, something like that. And we're a subsidiary then, a half owned by me, spun out of SUSE in 2013, around 48 people, 10 years old this month. So we're celebrating our 10th birthday. It's a little bit after the LibreOffice 10th birthday. And yeah, so, but there you go. We can't catch up at this point. And we're focused fully on Office and online. And we have a simple mission. It's more complicated, actually, when you write it out, but making open source rock is really the punchline. And that's the goal of our shareholders, and if we do that well, hopefully, we're all happy. And that means that when you give us money, typically we invest that in improving the floss software rather than Lamborghinis. And making LibreOffice better and building our partner ecosystem. And all our code is open source. Obviously, we are business, so we have to make money. Otherwise, it's very hard to pay the staff, but beyond that, it's all good. The parent company just quickly does a whole load of cool stuff. I mean, we're doing cool stuff too, but it's different spheres. Let me talk it through. So I guess the new areas there, I guess, virtual augmented reality, there's some great stuff there. OpenXR, so open sourcing and reverse engineering, proprietary graphics stacks and virtual reality stacks, and helping to standardize those with open standards. So the single Monado that's very popular there. Because it's possible that, you know, in two years' time, the meta universe will be here. And Jollywell needs to be open source. So it would be good if it was not Windows only. So there's some kind of strategic work there. It's quite popular, getting Linux and open source in cars and boats and ships. Hello, hello. Hello, hello, hello. Gabriella is tweaking a knob up there, fantastic. All that sort of thing. So everywhere, but really taking a strategic approach to getting open source in at the bottom of the stack. So if there's a medical device sitting by your bed that goes ping, it's very important that Linux is running on it at the bottom. And maybe there's proprietary bits on the top. But we want to help people build those stacks well. So the question is then, how do we push proprietary-ness out of the stack? And well, there are lots of ways. And we talked to lots of people. I'd like to give you one example here, which is the ARM Mali GPU. Now, maybe you've never heard of a Mali GPU. But initially, there was a real problem here. So it ran under Linux, but you'd have a binary blob, which was the driver. And for lovers of Nvidia, you'd be very familiar with the problem of binary drivers in Linux and elsewhere. And basically they work when you first ship the product, but after a little while, they break. And you can't do anything. It's just awful. And all embedded software people are hating this. So anyway, Calabra invested very heavily in reverse engineering the ARM Mali chipset and helped build this new now mainline driver, supports, I don't know how many generations of the Mali chipset. And we did it so well that eventually ARM embraced it as well. And we managed to open this huge binary blob in the core of so many embedded systems. And to give you just an idea, Mali is the world's most shipped GPU. Eight billion devices out there. That's more than obviously there are people alive. And we really wanna make sure that open source can come from the bottom up and advance up the stack from the bottom. So really driving openness there. Of course this is just one of our group's achievements, but I'm really pleased to see that happening. Now of course, Calabra productivity has a different role. We come from the top of the stack down. We are the ultimate leaf nodes in the dependency graph. We depend on everything in Linux. And so we really need to get going well on that. Let me start my timer, which is probably worth doing, isn't it? Yeah, so doing some cool stuff there, but starting from the other direction. And hopefully we meet in the middle. And the world is all free software. That's the goal. So what do we do? Well, obviously you have to make money before we can spend it. So we try and sell products that protect people's digital sovereignty, particularly Calabra online, your own private office in the cloud. And we build that of course on top of LibreOffice technology, which we support very heavily. And we sell SLAs and all sorts of compliance based things to make your deployment work beautiful. We also sell Calabra Office, which I guess is our branded supported version of LibreOffice. And we build online on top of that. And we also do consultancy, but we're really not a consultancy company. We do this just to help our product customers really get into all of the goodness. We sell through partners. And we really have a lot of them. There's 235 of them. And these are a few of the ones in our partner council, I guess, who come and we have meetings with very, rather regularly. And the problem is there are people out there who can just install our Docker image and claim they support it. But can they? Can they? Really, probably not. So it's really good to make sure you have a genuine partner to work with. And there's some really smart people there and it's just great to work with them. And that gives Calabra an amazing language base so we can talk to people in many languages, many countries. And so when people come to our website, they end up talking to a partner who understands their issues, we hope. And then we sell stuff for money like that, which is great. Primarily subscriptions, which is great. Another thing, and this is just an example of some great work getting that into the market is the Bundesministerium Interior in the Federal, the FBI of Germany has been doing some great work around the Open Desk project with open code, helping to fund and improve a Calabra online along with a bundle of other projects. You can see the logos below. I'm turning this into a product called Open Desk, or product, a project. And it's just great to have them to fund and work with all of these partners to improve the integration and make that much more beautiful together. So many of these features you'll see here have been funded in part by this project. And so we're just really grateful to have a level of funding that allows us to solve some of the longer-term problems. So back to the code. And the awesome LibreOffice technology underneath. So I'd love to give first a whole set of examples of things you've done to make LibreOffice better. And then I'll talk a little bit about some Calabra online things. But first, LibreOffice technology. So interoperability is a real issue because there are trillions, do the Google searches. There are literally trillions of documents out there in horrible formats. And we need to make that better. And so one of these then is, one of the things that stops this is the top writer issue. A vast, you know, numbers of duplicates, huge numbers of CCs, lots of broken documents. And it's just been great to be able to resource Miklosch to work on this for a long time and significantly improve this, particularly for government forms where for various reasons the UI these days in Word loves to float your table. As soon as you try and size it or do anything, it pops out. And turns into this very badly behaved thing. I mean, if you can see the text is wrapping around the bottom here. Ah, here I am. Well, I've got a laser pointer probably. I could kill someone with it. Second, ah-ha. Does it kill people? Ah-ha. Yes, can you see that? So you can see there's a nice table here, ah-ha. And then the stuff wrapping around the bottom of it here, which is the bit that didn't work before. Maybe that's visible. And it's surprisingly how impactful this is across so many documents, you know? Just to give a more pathological example, you can see sort of like corrupted mark all over the top of your table frame here that really should be, you know, in the next bit down, wrapping around it, as it were. So, and this is the Microsoft implementation there. So adding this feature is really very critical. Of course it's never one feature. It's a combination of a generation of different bits there. But really encouraging to see so many bugs getting closed there and fixed that have been open since open office long ago days. Another thing is compact pivot tables. So I heard about this a very long time ago, but Microsoft changed their default for pivot tables a long time ago. And we hadn't caught up, but now we have. So we have instead of these rather large and cumbersome ugly ones, we have, you know, smaller, prettier compact ones. And that really helps interoperability because pivot tables aren't perfect and people love to use formulae that copy bits of them out to do other arithmetic elsewhere. And that really then relies on the hard-coded layout. So that's really important. A document of theming, also very, very important. And yeah, just good to get, you know, our word or writer and Cookie here who's photographing avidly is responsible for all this good, you know, color theming and scheming and storing their arm in over here, doing wonderful multi-stop gradient stuff with allotropia for rendering better things to make your documents just theme nice. You know, of course we could flatten the colors, but we want to round trip them and be able to re-theme documents. And then just loads of other stuff. I mean, I don't know, what can I say? There should be names on each slide, but, you know, getting those huge documents to work nicely with Attila on 64-bit Zip or in numbering and frame anchoring positioning with just in multi-page TIF import, you know, lots of scanners generate these huge TIF files that have lots of pages in one file. It would be nice if you could actually see the pages. So Rashes just did this. Improving the fitting algorithm in press. Cropping videos are there. And improving the ability to make the sidebars look nice and lay out properly by simplifying the glade there. Mike Kikensky doing math sidebar stuff, making math easier to use. Sorry, I need a drink. A page number insertion. Uh-huh. Of course it was always possible to put a field in a header, but it's just nicer if you have a little wizard for the simple user that they expect. I'm making performance better pretty much everywhere. Lots of interest in this. And that's really, really cool. So, and of course it's really important for online as well, where we have lots of people using the same server. So that's one of the things I love about online. It gives us a really good reason to do lots of exciting performance work everywhere and make everything snappy and excellent for everyone. So all of those things are already in LibreOffice. They're already upstream. They're already shipping in many cases. But I just like to talk about some of those places where we use LibreOffice technology to get brilliant in-browser collaborative editing. And first off, we've got some just obvious stuff, like bringing nice features from LibreOffice technology straight to your browser. So to do that, it's not just a matter of showing all those dialogues. We need to do work for each one of them to make them asynchronous. So it's not just Shubham making this nice feature, but also Dashan doing the asynchronization bit so it doesn't block stuff. We've done a lot of work making bibliographies work really nicely with Zatero, thanks to Pranam. And so you can collaborate on citations and you can do great academia and legal stuff in government organizations and so on with Zatero. We've got lots of hyperlink improvements so you can have nice previews of links popping up as you edit the link. And I think video embedding too, which is pretty cool. And it's really nice to be able to embed that in the sheet and then stream it to the client and make sure your presentation goes smoothly. Font previews and dropdowns, so we've done some nice UX things there, thanks Shaman for that. And then lots of memory optimization. So one of the great things is that if we can use memory more efficiently, we can save bandwidth, we can send a bit more deltas, we can get more people on a server and more people editing happily. So lots of work there to improve memory. CPU optimizations as well. Wherever we look, there is a wealth of opportunity for improvement. So that's really good. Our forefathers have left us lots of good things to do. So no chance for boredom. And some great work there from Quailorn and others. We've moved almost our entire UI to using welded dialogues. So based on Quailorn's work at Red Hat there and Shaman's work to make that work with JavaScript nicely and a whole team of people moving all of those to the client side. To make them accessible, which is really important and themable, which is important for accessibility too, and just crisper and easier to Cypress test than our previous Pixel dialogues. On top of that we've been doing a whole load of work, Marco Cittetti in particular in Italy, doing screen reading and helping make the document content itself able to be used with a screen reader and navigated through the browser, through a remote web socket connection. So there's some quite exciting logic going in there that Marco's been working on. A dark mode, obviously that's been in the LibreOffice core and we now want to bring it and have brought it to online. Since you have to render differently anyway, you can do things like, hey, rendering pilcrows and spell checking and other different features in different views, so that becomes a per view setting. And we've also brought a very nice navigator to the client side here with all sorts of theme, themeing buttons and looking very good so you can navigate those huge documents easily. And online does scale beautifully, as LibreOffice does to huge documents. I routinely use a 300 page document just to test it, editing at the top and moving all of it down as and when I can. So yeah, all my one-to-one minutes have anonymized, ancient stuff in them for this reason, for my one-to-ones with staff. What else? Keybindings are really important. So we've got lots of keybinding work going on. Really focused there on the notebook bar and multi-character bindings as well. So some of these things, there's so many things in the toolbar that you can't do it with just one keybinding. You need like CD for some other thing, A, A, A, B and so on for some of these things, which is quite fun. So a nice pop-ups there to show that. And really a lot more. These are just some headlines and highlights of it, but it's just fantastic to be able to work with so many cool kids. This is the last 12 months, something like that. So thank you for all the people that contributed. It's great to work with you and just fantastic to have you as an extended part of the team. That's of course the people that don't work for Calabra. So external people, if such a thing exists, external, internal. And of course it's all built with Libroff's technology. So just that great technical foundation underneath there. What else can I say? Cool days. Well, yes, come and see us. So if you want to find out about Calabra line, get involved in technology, have fun with people, you can see some of the things that we've done here. I think this was skydiving last time. I can't, in Berlin actually, let's see, we're a time but one again. Punting, fooling around, lots of good people are there. There's some recovering people as well, but there you go. It should be great to be there. We don't quite know what we're doing this time, but book the date. Hey, wait a minute, where was the date? What was the date? It was the 8th to the 12th of April. And there's a QR code there if you like such things and you want to find the URL and register now. You should definitely do that. So last time there was some concern, I think, about some of the things I'd said or maybe the venue for saying them. And so I'm sorry that caused concern in my keynote. I think I probably mostly opened my mouth to change feet. However, I think it's probably good to build on that with some vicious personal attacks. So I thought I'd focus on myself. So here we go. So for a long time, my emails all came with this very helpful disclaimer in the bottom that said, pseudo engineer, itinerant idiot, which basically means having read my long opinion on something in an email, you should probably bear in mind that I could easily be completely wrong. And this is part of a discussion bringing me your feedback, as it were. And unfortunately, when I started to get customers with whom I interacted only briefly and often in my role as trying to sort out some problem, they didn't really understand and it became too difficult to explain them a long period. So instead I have a boring signature that says, that makes me sound like a posh person, which helps the customers, but perhaps not the community. So I'm sorry about that. I tried to have both for a long time but I just kept sending the wrong one to the right people and it was bad. So anyhow, this is rather a brief summary of my failings. So I thought I'd expand on that in some detail. So I think one of the problems I have is that when people criticize me, sometimes I'm like, yeah, that's completely fair. Like obviously. But other times, it's like, oh, that's a bit too close to the bone. I don't think I agree with that. And when I get on the defensive, when I'm attacked, it's very often that I can forget that I have actually lots of flaws. I mean, the other person may be completely wrong, but they may actually have some point to make. And so I think it's just difficult for me to listen to that. And I think if we look at our politics today, we see quite a lot of polarization in the wider world. Like it's this way too much of this not listening and I'm completely guilty of that. So sorry, if you're a victim of that, I'm sorry. Another thing that I'm particularly annoyed about and looking back retrospectively I think is that anyone who's dealt with bugzilla knows that there are flaming arses out there and they post horrifyingly rude things in bugzilla. And if you haven't experienced this, you're probably not a developer, you know? They question your ability, your parenthood, your, you know, how you can possibly write this and they point out you've lost them all of their software in all of their documents and how dare you live with yourself and so on. And our general approach I think as a project has been to laugh this off internally and try and get any good that we can out of the report. Is there some kind of pattern here? Is there some details? Can we calm these people down and work with them? And you know, and actually in some cases some of those people have even joined our community having been calmed and moderated and it's been a good strategy. Unfortunately, I think I have tolerated that inside our community as a way of discussing with each other for much too long. And I think a radical failing of mine was not setting boundaries there and saying, look, if you're gonna be a member of this community and part of it, you need to behave in a responsible and respectful way to other people. And otherwise, we end up with toxicity in the community that makes it very hard to communicate and understand each other. And I think I fail to take action on that. You know, it takes bravery to stand up and say this is not okay. You know, like this email is not acceptable, the tone is bad. You know, you can say what you need to say but you don't have to say it like that. And I think I fail to take action on that at TDF and I think lots of us have if we look at, you know, look at our discourse and probably we need to fix that. But getting back to me. Another thing I really struggle with is trying to get openness in there, not just for myself but also for my team. So it takes a constant work to encourage staff to discuss things in public, partly because customer names are involved and project names and they're confidential and so on and so on. So it's extra work to discuss in public just because of this. But more, it's extra work because other people might disagree with you and then you might have to work to, you know, work with them and build a bigger understanding and so on. So there's risk involved there. And often I don't think I've done enough to do that myself and I think, you know, I'm constantly encouraging my team. So if you're on my team, you know, do more in public, get it out there. And I think we had some drives to do that but they're not as successful as I'd like and we need to be more transparent and open. I think, I mean, I could go on for weeks here. Luckily I've only got a few minutes left because it's a compendium of my failings. I think one of the things that's particularly irritating to me is a time perspective mismatch. And some people can contribute to LibreOffice as volunteers and they have lots of time and they have a very relaxed approach and they want to discuss everything in great detail and that helps. That can help get a better quality result and that's cool. Other people are insanely busy and under cost and time pressure and this gives a very different time perspective and that can cause real communication issues and I've seen that across my career actually from before I was even involved in LibreOffice. A silly example is that I would love to be in lots of talks supporting individuals and my staff and so on at a conference but there are really at least three tracks. There's the hallway track, you know and then there's the conference tracks and I can't be in all of them and it's frustrating to me and I really love spending time with people. There are so many, I mean look at you, you're all wonderful and I love to go deeper and understand and get to know people. It's actually a luxury I enjoy but I completely fail very regularly to negotiate how much time's there. I also speak very quickly and so it's quite unfair to people who take time to process and understand what I'm saying and just having the patience to listen and trying to get the signal out of that is something I still don't know how to do this properly so if you have a brilliant idea, come and tell me but I don't know. Anyway, there are many more failings I have but thank you for your patience. Against that I'm really pleased that the Torsten and the Board here seem to be doing a really good job of working through some of these issues and getting help from outside and inside to improve TDF so I'm pretty happy with that. So 10 years, I mentioned we had 10 years. Open first, open source, this is our headquarters I guess for what it is. One of the great privileges of my life has been to work with so many talented people. In terms of code contributors in those 10 years we've had 75 Calabra coders, there are their names that have contributed to the code and done just the most amazing things and I've had the great privilege of interviewing perhaps another 150, 200 other people for many of these roles over the time so that's quite a lot of time talking to people and we've loved working alongside and I generated the number it's something like 1,400 individual code contributors and I made sure on code contributors because they're easy to count there are many contributions obviously but this is important and it's just like 1,400 this really encourages me that we're doing some really good things right at LibreOffice in that time and then of course it's much more than code you know we couldn't do what we do without the 40 plus people we've also worked with alongside for lots of these other vital roles you know when it comes to administration there's lots of things that need doing you know financing, I'm just sales telling people every day about LibreOffice technology and trying to get them to invest in it to tie their company to it, to buy it to deploy it, to get it out there and there's just a great set of people there doing really good things that perhaps many of you will never have heard of or seen we've done a lot of stuff now commits are not everything let me qualify what we're doing with that but anyway over 10 years you'll see that we've contributed really quite a lot and we're very blessed to have been able to do that alongside many others and the team there one of the things people often ask is how much of that was funded by TDF meaning to what degree is Calabra actually just the document foundation donors under a different disguise and that's a fair question the answer is it's about a 40th of our revenue and I've tried to put a little thing there so if revenue turns into commits eventually or code then it's something like that proportion so a real proportion but not overwhelming what else well our customers have filed and you know I'm always nervous with this customer filed tickets metric but anyway they filed something like 3,600 tickets as of two days ago and about 35 of them are currently open so we closed a lot of issues for our customers around 10 every day if we worked seven days a week for 10 years and we've paid all of our bills and our contractors and we've met and paid Sousa's unwritten obligations you know when Sousa came to me and said look we're shutting the team down and we're you know you're not going to work on Libra for this anymore that was kind of a wrench for me because two, three years into Libra office we'd made some pretty big commitments to the project at least individually and although there's no legal reason there's no written agreement to doing that I think it's really important Canabra has in those 10 years I think paid off Sousa's bill there and had great fun supporting their customers as well through that time another achievement I think is having an open culture internally so our salaries are private obviously well not obviously but yeah it is so but pretty much everything else that's possible is open to all of our staff our financials, our customers, our contract you know the whole thing we try and have a 100% open culture that makes it easy for people to get all of the information they need to help us make good decisions we have weekly management minutes are all published we have all hands calls, no questions, barred, etc we have something like 85 million docker image pools which is pretty cool when you consider that's a server that serves others yeah around 4 million paying users something like that 57,000 commits, 21,000 commits to collab online lots of partners and the partners are vital to our work we couldn't do anything without our partners we really prefer not to talk to customers but to talk to partners and in case you think this was easy because some people think it's easy my personal record for a working day is 36 hours it turns out you can do nearly a week of work in two days if you just don't sleep and you work back to back to back to meet the deadline and cover for the team and deliver finally what the customer needs in the time they need it to keep LibreOffice technology alive and to keep the project moving forward so that we don't lose a key account this makes my hair white and shortens my life and that's the commitment that we need we've also made lots of gorgeous products so the online and tablets as well as collaborative editing experience which you're perhaps aware with and the story is just beginning we're still hiring so if you want to join the team we have lots of people we want to expand our testing investment with a senior developer there for testing we have room for interns if you're good at marketing automation we have lots of optimization and our parent company has even more roles if you want to work on open source if you want to have fun working with other people just grab me afterwards so conclusion Collabra loves LibreOffice and we work pretty blinking hard on it and we do that to deliver our mission which is to make FOSS Rock we liberate people's documents by the hundreds of millions and it's all paid for by our customers and our partners we can't do anything without them so thank you if you're a customer or partner and you're watching you're wonderful to us and the whole project and it makes it it all happen and of course we can't do that without our staff so thank you for working hard alongside the community as open as possible and economics matters I have a talk on economics later I think on Saturday I'd love to see you there we should really use the strengths of our different kinds of organizations and our different locations to work together to deliver the best for software freedom it's a pleasure to sponsor the conference and thank you for your patience bless you