 It could work, I don't know if the bandwidth for the network causes problems, it might be good to turn off your camera, but yeah, go ahead. Let's see how we do. I don't know if people need to see me, I think they probably know what I look like. Let me just set an alarm so I didn't go on for an hour. It was a risk. I want to take some questions. How long have they got? Half an hour? Something like that? Let's go quickly then. Here we go. 10 years of LibreOffice, or is it 20? Or is it 35? Who can say? I don't know. So, if you look at the Wikipedia picture here, which is very, very beautiful, obviously. We have 20 years, 45 years since Marko Boris releases the first version, or is it 35? 85 to 2020. It seems like a long time anyway. I think this was written in Pascal. But anyway, the key thing was Sun Buying Star Division and Marko open sourcing it in 2000. So, as part of that, of course, we dropped the email that used to be part of StarOffice and the desktop thing, there was a Windows 95-style desktop thing. This was really a collaboration, Nat Friedman, Miguel, both went to see Sun, and talked about the Linux desktop, and so we went with GNOME as a desktop evolution as an email client and OpenOffice as an office suite, and each company could then re-focus people on those bits, as I understand it. So, OpenOffice has launched, and in the launch party, I guess, or at least the announcing party, which I was at, which is San Jose, I think, Sun was funding some wonderful thing. Let me actually just turn off my telegram announcement, wait a second. I thought that was the problem with full screen sharing, is you get all your wonderful pop-ups. Yes, of course, I'm very kind of collaborating with it. They are paid by customers, but of course, but we have a larger end user base than just collaboration customers. For instance, as I said, improving the impress doesn't seem to be the objective of many paying customers of collaboration, and there might be a reason, because presentations are probably a niche application, and so there is a lower request. I think I might have been a little over-enthusiastic, closing windows there, so sorry for that. Let's try again. Are you still with me? Can anyone hear? So, at the initial launch party, we met at least three spots for the first time, and some positive and strong discussions. And Scampliqelevnet was going to be paid to create the most beautiful community in the world. And to be fair, this was a tough game, paying for by a company, building a community, in between them. But I don't think it was helped by us being a social scientist with no false experience at all. And, as I said, he told Star Division how thoughts worked, with lots of inflated rhetoric, and I think stepped quite a lot of them. And he was described by one interviewer as the Linus Torvalds of open office, which is extraordinary, because with no development experience, I don't know, it was just a little bit at all. And of course, the governance of open office was deeply gerrymandered on his set up, so there was no real... Yeah, but it's hard to describe how broken it was. And he also had a strategic focus. You know, he really focused on development or attracting developers from strategy up, so we could come up with a little... So, you know, I remember when we were shorted by the show, I know one of the community thinks it's so cool, but maybe I'm not releasing it to open office. K-Changes is anti-social. Well, he didn't speak for me. I thought that was deeply anti-social. Talk is cheap. Show me the care. I think that's a good way of looking at life. And so, you know, another great tip would be to get a developer to write and even test your building structure, so that actually the community project could actually build it. I'm here at some work that we did. David Ostrovsky from Siv and myself built the thing on the right, which tries to show you how LibreOffice works, you know, help people understand, but the thing on the left was built clearly to try and persuade people that they couldn't possibly understand it. And so much of the open office, you know, documentation was like just didn't bother. I don't know if that was strategic or accidental, but who knows. And the development process was crazy. People would fundamentally change APIs way down in the stack and email people to say, hey, you know, I just reordered the parameters and please update all of your code. Sometimes you could catch up with a compiler and sometimes you couldn't, like, you know, like this guy, hey, just copies the other way now. And they would mail out, please update all your code and move on to breaking their next, you know, the next day. And so go to windows. The build tree was in a permanent state of brokenness. You know, release engineering would, from time to time, branch the tree, try to make it a build, try to chase engineers to fix things, merge all that. And after about a week, they would produce a binary snapshot. Probably yes. They'd stay well gone or continue working with it. Because the code took a long time to build and all that. And actually, if you look at our credits, you'll see these committers are still on the commit count. They may be quite long. They're really very near the top, because they merged so many commits. So my experience with it was, come on, let's get it to the first build. Even though it's a little bit of an experience in software, actually. We have 10 years of history behind that one. I was just full of optimism. I think it's time to review the model. But if you've updated it, having got a build, you can get a better build. It took more weeks to get something working. So I had to create this build thing, which contained a public set of patches and fixes necessary to turn that into a natural release. And sometimes they may not have the distribution. They may not be able to actually adopt a viable distribution. Partly because 24 hours into a build, if it fails, you have a real problem. And I think sometimes you forget how it's going to be developed, and these are the world that make things really quick to build these better. It was pretty hellish there. So how did all this come about? Well, I just like to get you through a little bit of a timeline here. So Marko Bore is awesome, obviously. He's got a vision. He even stalked the other stuff. He left the building. He went over to York High League. He was a great guy. And eventually Google started him. And he went on to create Gmail, which you may have heard of. AppleBemba then was in charge of it, came out of the QA department and did, you know, see you as soon as possible from that. Around this time, I guess Zinnian was just chipping. We were actually contributing to open office at the time. And we did a whole load of pretty stuff, making the UX nicer, the icons in, alpha blending for your icons, so it's not single bit marks and individually named pixels, all sorts of good stuff. We didn't behave terribly well, partly because we weren't contributing all of it back. We were contributing some things, but not everything. Partly because at the same time we were in talks by some to buy us. They were very much in the form, hey, why don't you contribute everything back and then we'll buy you. And our talks were very much in the form, why don't you just buy us and then you'll have it all. These things happen, I guess. In the end, we were required by Nevelle, not Sun, and started to ship all that stuff upstream. Sometime later, the CISL license was dropped. So the Sun Industry Standard Source License let people take the code and not contribute back. It was like a BSD-style license as long as you didn't break the file format and interoperability. So it was an interesting approach. There was nothing particularly wrong with it, but it was a dual-license LGP-RV2 and CISL. And with the CLA, so Sun continued to own the rights for all of that. But essentially, IBM misbehaved. They didn't contribute their code back. They used the CISL to not do that, which was kind of bad. On the other hand, there was a CLA there, and it was pretty nasty having to get your code in for lots of reasons. In 2005, this is a quote from a mail that I wrote on time. I'm giving almost everything away. It's almost destroyed Sun's starvation's revenue stream. At this point, they had to have a whole load of budget cuts, and they thrown Nevelle the change in the license to the GPL. So we were shipping, of course, an open office in our distribution, by then, developed boards. That has many other people. And we were pretty against this change to the GPL because it would threaten all the plug-in ecosystem. It would put us to a disadvantage against Microsoft Office and so on and so on. And I guess they smelled money. Again, we said, look, it's bad for the product, but what can we do? We're not going to pay you for a proprietary license or some such oddness. We don't think it's a good idea. So it's not a vision. They cut a whole chunk of stuff. But it was the positive. So in 2005, Intel's open source group was in the Ascendancy. It sent us two people, which was great. I would train them here in my home. Still a friend of mine. And these guys, of course, were interested in the compliments then. They wanted their CPUs to run better with an open office at the time. Sorry, I keep saying Libra. They did a whole lot of performance work. They tried to improve IO and so on. They really struggled to get much of that in. But actually some of their work, avoiding atomic reference counting on single CPU systems and so on, had a big impact at the time for Intel. But yeah. Unfortunately, much of this predates GIT or SVN or anything that logged really who did what. And in 2006, actually one of those people that opened office or something was hired at Zahid-e-Bora. I actually went from some to Google and encouraged Google to hire people. So that was great. So we had Kai Backman was hired and he did some training with me, a great guy. And he started to do just extraordinary work. They recommended him to do. Just look at the build system. And get the build system, improve the build system. Because everyone got stuck in there and they were low-handed. They worked on PCS, they helped with their subversion migration and he could build. He could do a no-off build in a minute, which took half an hour for some build system. So he did some amazing things. Of course there's the Vell Micros to put the announcement, I guess. I can make some L filters that's still based on that. Work and some tough times there. Although Microsoft hired around 10 people effectively to work on open office. So there's some positive things there. The problem was there was an active process during that time. So we had our IT inks, there's a horrible Russian troll system. People made it extremely difficult to get things in. The release system was dire, feature-based, slipping. And lots of this stuff really, I mean it felt like treacle in any community discussion, doing anything. And a lot of the process was designed to exclude developers. People actually suffering from this stuff weren't involved in a quick teaching discussion around what should happen at the top. I don't know if it's a result of that, but Intel left in 2006. One year in, quite being there. And it's not that he didn't try. He moved to Hamburg with his family. He had face-to-face meetings with people who threw chairs and swore that he was lying. It couldn't be done and it was absolutely awful. And just the spaces there was amazing. He was urging me to fork and eventually he left to an open-side Google. It was too distressing to work at OpenOffice or with OpenOffice. But of course, many of us carried on anyway. So in 2007, IBM started to do this OpenOffice Foundation course. They tried to kick-start something there. We had a meeting actually in Boston. And Neville and myself were there helping all the people together and trying to skewer their plans to push Sun into a corner because Sun was doing a lot of the work and they were contributing a lot of the code. And it seemed entirely sensible that they should be leading the statement that IBM should be contributing. In particular, that IBM should contribute code back to OpenOffice, publish it in a meaningful way. Unfortunately, the net effect of all of this was that IBM got a proprietary IBM license from Sun and constrained them in all sorts of ways they couldn't share. And they also announced, seemingly in a connected way, Sun Solaris being shipped on IBM servers and IBM putting money into driving Sun Solaris at the time against Linux, including Neville, Cesar, and Red Hat. And so we saw our investments and our support in OpenOffice and Sun used effectively to compete against us in the marketplace. And the hopes that we had of an open foundation or project and a neutral solution where everyone could work together with Dash in a particularly irritating way. So at this point we were preparing for the inevitable. We were doing all sorts of trademark searches, branding bootstrap, fresh office, pure office, office unbound, maybe some of these cognitively similar. And we started producing Windows builds of Neville Office. And we started pointing at this contributor license agreement, this asymmetric license that Sun required and contributing back under the terms that Sun gave to other people. So things got a little more tense there. And yeah, it's a different form of IP. It brought to code a copyright ownership and using that to give some cell exceptions or something I think is a bit of a cool in a way that harms other contributors. But it was still pretty bad. Having signed this thing with IDM there was still no code. We didn't actually get anything out of it. Around this time, the OpenOffice brand which was opened by a team I think Team OpenOffice, EV. Which has some quite good people in the very opaque governance structure there. So Cornel, so I think was there. I would call him coming and telling me that they just gave him the brand to Sun. They were bounced into voting to transfer the OpenOffice brand to Sun. And at the time, I wrote yet, it looks as if it's a fairly benign nonprofit that owned the OpenOffice trademark which just transferred it to Sun. Apparently the intent is the club of people and not using effectively our version of OpenOffice. Which was pretty unfortunate. Of course, kind of doing it too in some ways. There's much that no Linux distribution would use. You know, the vanilla OpenOffice at the time it was simply unusable. So we then have the prospect of not just having the code ownership used against us but also the trademark used in the marketplace against us. And so to say, there was a lengthy trademark guideline discussion and so on and at the end of it all, actually Sun came up with something that sort of worked. And I forget if we rebrand it away from the OpenOffice development perhaps also. But it never used the brand effectively actually to build this business which is a shame because it could have done some really good things there. So IBM on the meantime, you might wonder why we didn't go to Apache with IBM. They were promoting this foundation in 2005. There was a press release in 2007. Here we go. IBM will be making initial contributions to developing blah blah blah, will be blah blah blah. Just a dedicated team of 35 programmers. It all sounded really good. We should definitely do something nice for IBM. A year later at the LibreOffice conference or the OpenOffice conference, they're saying, well, we've got a bad scorecard. You haven't done so well yet but we will be contributing. Two years later, we have this press release. IBM donates latest symphony source case in the Apache OpenOffice project. Amazingly, at this point there was still no code released and it was absolutely amazing. Only in 2012-05 did we actually get usable source code arriving at Apache 58 months after this had been promised in that press release. What do I take from this? You need to be really careful about people who say they're going to invest things. You should change everything and do stuff. Diversify your community by including IBM quickly because they're going to do all this good stuff without actually having a credible plan and seeing what's actually going to happen. Talk is cheap. And it's not like this is the result of a thousand men, supposedly, professional strategy team at IBM. They have a central strategy team. If we'd spent one of these guys doing some coding, we would have got in like five years more code than we got out of this. Having said that, it's worth noting that there are a whole load of other really unhelpful non-contributors to open office, NED Libre office that just take and don't give back. And that's a pattern that we live with even today. So anyway, here's the humorous bit before the conference. I went to UK conference with all these people talking about stuff, holding up these things. I had to take a picture of them. So in 1910, it was wonderful in theory having a QA member, a UX member or something like that or something else, a specification sitting down and talking before you changed anything. There's a picture of a spec actually on the previous slide that you should really write before you change the user interface. Now, of course, anyone actually using the user interface would realize that, well, a few paper cuts really help. Actually, well, it's still the same today, actually. Now, together with red tape, lots of fugs, not getting fixed at all. Of course, a contributor license agreement really stopped people getting involved which is really sucky. You can see when Fedora dropped it. Now, of course, TDF, when we launched it, we had this no vendor dominance. No one controlled more than 30% of the votes in any statutory body of TDF. That's of course a good thing. Although actually this is often around 10%, although the threshold is significantly lower than on the membership committees. It's around 10%. But it probably turns out to be more useful to focus on actual diversity and building an ecosystem that attracts many independent participants than overly obsessing about this particular detail it seems to be. It's also, in large part, marketing. When we created LibreOffice, SUSE was by far the largest contributor. We had about 15 people on it against two Red Hatters Well, luckily, lots of volunteers. But, you know, we really needed to emphasize that benefit neutrality at peace to make this not be seen as a Microsoft plot which was the problem at the time from a marketing perspective. I think the frustration thing is something that I think will be familiar to many people. You get a large organization and it becomes very static and there are many obvious problems and we think you can fix them but there's nothing that you can do. There's always someone saying, no, it won't be done, it's not possible and, you know, there you are. I think the frustration there is something that was epic and long-running. As you see, we lost Google and so on because of it, but yet I didn't want to give a perfectly negative picture. There were actually lots of really good people and I think, you know, guys that I'd love to sit down and have a beer with still pleasant and they saw the problems. I mean, it's not like these problems were unique to people outside the company. I think they were inside the company too. You could talk to frustrated people there but it was really a cultural disaster. So, well, we launched. How did we launch? We planned all these people. Susie bought everyone a very nice pill and walks and, you know, life was good. But launching was kind of tough. We got to the point where we got approval from everyone and, you know, there's some quite de-management chains here. All of which are very supportive, amazingly so. Quailon on the red hat side, myself, I guess, on the SUSE side and some of these people still involved. I mean, Quailon, myself, and he are actually still, you know, kind of involved in the free software world here. I think Yambol, the boss to the New Angeles. A red hat. Either way, and of course it's a tall press agency but everyone started saying, you know, but, of course, we were up against other people trying to play divide and conquer. So anyway, how can we unlock a Google quote? Well, I mailed Chris and Jeremy, old friend of mine, and, you know, he says, well, basically, you have to, these guys are outsourcing their due diligence. You're the press release team makers. So, we have all these people on board, if you do. So, you know, Double Rainbows and Penny and the good thing would be fantastic. And we waited. Well, and luckily, three days later, they came up Trump. So, good to be with you very quickly. Here's the proof quote. And this is essentially unlocked all of the other quotes from people and everyone else. With that, we've got SUSE, we've got red hat and everything else that goes behind it. They're perhaps going to get any credit or recognition of their role in making it possible to do a large launch here. And of course, you know, Jeremy's stuck around and done lots of good things in the air and people somewhere, Kate and so on. But often we hear this story that the community inverted commas and these days the word community is used to exclude the corporate ecosystem and its community against companies. And that the community by itself went and created the liberal first. That was not my experience. Volunteers play a vital role. Christoph, for example, creating a logo. Amazing, amazing contribution. But without the ecosystem and the companies behind it, you would just simply go on it. So, again, it's funny, man, I'm not sure I like it, but it's just a great page of great examples of logos. Lots of other great people, you know, like, it's here, I'm going to use the guys to do a different fantastic thing. And they released fantastic. And we've got, you know, Neficent, the Vel, the red hat, Mark Shuffleworth at the Ubuntu, science tips, supporting us obviously from the beginning and seeing various other people on it. There you are. And it doesn't look so different now as a batch, but that was our 3.0 release, hopefully it looks very familiar. Under the hood, of course, we developed one thing which was having phone calls to reduce conflict. Using email to resolve conflicts is a reliable way of making everything worse. Quote, anyone that's interested in conflict resolution, I always take about sitting around tables, it's very helpful as well, eating together, you know, understanding. Maybe large, it's fine, it's even made up. It's just terrible internally. At least from the technical perspective. The paper office is possibly one of the smoothest, the most pleasant and easiest to be involved with projects that I've seen. It's really, really beautiful. The engineering side. Now the marketing board, community, strategy, something, something else, stuff, isn't my dear total mess. But the engineering guys, they work together, which is great. So various things happened here. Of course, the time we had to get Oracle and some of all, some of you didn't. In traffic, you can see them, I guess, because of the dropout and the red chunk here, so the yellow to red is Oracle. Suzer spun out half of its Linux team. And, you know, can I say, they did that in an exemplary way. It was, I think, a run-to-business decision for Suzer, obviously. But it was also just a brilliant, brilliant way to go about that. And Ralph, and Gerald, and Nils, and the players to keep them popular, I think, so to be outstanding. And, of course, they continue to support a Libre office development, even today, although at a much, much reduced level. It was much appreciated. And I'm going to ask you for my best advice of all the good things that they've been funding to go. Ocon, Suzer. It was a really good, good, and, of course, CIB. And so another thing that you see here is the people. Lots of people go through this graph whilst companies come and get it, which is really quite common. And, quite twice, you could see, I guess, Oracle living there. But we survived the large decline, which you can see, I guess, before we started, which really lost as much activity. And there were lots of good things. And I'm going to skip through these, I don't have much time. So, what are the problems with the tech industry? It is really fashion-based. You've got to be seen to be there, even if you're not. You've got to have a finger in every theater. So, here we are, within a few months of branching, forking one of our open-office, we then had a browser-based version. Look at this. Fantastic. So, Red Hat invested in, I think, the GBK Broadway. It's just a fun, cat-weed type project, I think it's enough that Susan said we could have an in-browser office suite. That was demoed at our first Paris conference. Unfortunately, it took four years to actually find someone who was willing to fund turning that prototype into the LSD, in which time practically nothing happened. So, the code was there, the community was there, there was lots of good things that were going on, for sure. But in terms of this, there was just no investment. It's supposed to take an 80 days to finally get something that, you know, does that. Collaborative editing. We, again, needed to figure this in this pie. So, here we had a collaboration with Red Hat and Susan building a prototype of the telepathy tubes of cow shit that you can pick, because we thought it would be the easiest piece of the puzzle to attack. So, there you go. We are in 2012. This, of course, is still not fully formed. Svanter is still enthusiastically trying to sell the concept to investors left and right, and they're not getting any. Of course, there is collaborative editing built into Collaborative Online, although that happens in a different way. This was the multi-multiple synchronized collaborative editing view. Another thing that we did really was Android version. So, in Google's total code, and Susan filled out this in an input of flag on the mobile developers, and he did see what he pleased with each of us, as well as the work that I did here. Of course, it was difficult for Susan to be following Android development, because hey, there are Linux companies. And this was not really far away from anything that made business sense for them. So, we created this thing. It sort of worked, as you can see. And it was really seven to eight years of the investment to productize the proper Android version again, and so the companies came up with that. And actually, Android is quite interesting, because if you look at it, the investment is a bit jumpy. So, you've got a whole chunk of work in 2013 to fund two-hour rendering for their own proprietary by the way, as that. So, Susan, Canavera then funded the viewer, which is in shipping in 2014 based on Fennec. In 2015, TDA funded the space framework for basic editing, and we Canavera delivered on that. But then there was kind of a hope that the community would take it on and turn it into a beautiful application. And that didn't kind of work out. I mean, Christian did no really battle with the bitbox here and tried to make it build and fixed issues and things 40 years. There was more hopefulness. Canavera, you know, we tried our bit. We put a whole load of mentoring time and effort into Google Summer Code Initiative. It was stagnant. Then, of course, the TDA marketing community arrived. There was a video. There was evangelism. There were call for developers to come and help do this. Can we get people's conferences and conferences to talk about it? But the punchline was, oh, and even more, Google Summer Code Mentor Engine Collaborate. There's even a bit of work from a Turkish GBTAC, I think. But broadly, there was something that was so embarrassing at the end of it from the app store. And this four-year gap, you know, we tried pretty much everything to get an Android version there out of it. An absence, actual hard cash investment, which finally we got from Calabra and Anthonies, nothing happened. I think that's worth absorbing briefly. One of the questions about what is a foundation is interesting. So in 28th of September, we announced that we will make a foundation after 10 years of popular foundation. And this was great, but we haven't actually agreed what a foundation was or where, at that point. And so we spent the next 16 months discussing America, the UK, Germany. This may sound rather familiar in terms of looking for a jurisdiction. I'm actually from Luxembourg in there, probably. I imagine there are others for you, the Netherlands and so on. And the argument, I basically revolved around well, I thought the FSF, the Free Software Foundation, was a foundation and their 501C frees in the USA. Or an association. But were some people in our art steering body who said no, it's a schtiftung and it is in Germany or I quit. When I said foundation, I meant schtiftung and nothing else. You can't win all arguments. So there you go. With trepidation, it was then legally established in Germany and there were lots of consequences to that. We wrote our statutes in stone, set them in concrete at considerable speed. And people who hold them up and say this is the last word in everything, I think are computing themselves. They've done well for 10 years, but I think it is worth not treating them as 100% possible, believing that every consequence of them was well considered when they were written. Particularly that the world has changed since 2010 and so did we. So in 2010 the Linux desktop was the future. We expected 50-some article developers and 10 IBM developers. We expected things, the nice things in life to stay the same while we changed radically. And that actually didn't happen. In 2010 we expected that the 20 to 25 devs would be there building the future of the Linux desktop all the time and that's what Linux distros are putting in. We believed economics to someone else's problem. It's not something we had to deal with. We had lots of good advice for sales people and marketing people. But we never sold a marketer video. So in 2012 it became more clear that just the Linux desktop economics sucked. It was really hard to get to market. There was aggressive competition. There were subsidised drivers for the windows to buy. And there was the Ubuntu effect. Ubuntu came and marketed wonderful free everything. Everything is free, free and forever. It's all brilliant and just how to get it from us and from our brand. And the consequence of the Ubuntu effect was gutting the desktop teams. First Mandrake, Mandreba, Sousa and Red Hat all were significantly deleted by the fact that you couldn't now sell the desktop anyway because the price point had been set at 0 or negative by canonical. And that's economically problematic. I saw this effect happen to my team, my desktop team at Sousa and amongst other things. And of course the effect on not having sufficient investment in these things. In the end of 2012 Sousa spun out half of its 15 person team and moved the other half elsewhere. 2017 Canonical retired their one developer for 80% of the Linux desktop market for one developer to zero develop. Great shame, we lost fuel. In 2018 Red Hat moved several of its developers away. The world looks very different in terms of needing to have a clear cut economic understanding. Still, we had fun. Look at this, FOSDEM, everywhere, cakes, pet fest, our first ever conference. It was cool. Look at it, Lecontein in Paris, doing good things. The German Federal Ministry of Economics and Technology. Look at all these cool kids there thrilled to be working together to do something useful and meaningful. And you know, continuing to this day. Look at this, Italians, Albanians, Almeria wonderful. And of course Nuremberg, sadly virtual, but hopefully we're all still loving it. I'm not going to look at all the features. I had a thing to talk about features, but I think if we get all of the surroundings and the economics and the community and the fun around the project right, the features come by themselves. So you might just look at all the talks that we're going to be accrediting. The marketing is really important. How we deal with our trademark, our brands how we present ourselves is probably more important than the minor features that we have. And we can see that in the open office comparison. So if you come to my talk on economics and why it's important we get those right. So what happens in the future? Well, we still need smart people to invest to create artificial document self-creation. You click the button and everything works nicely. But my advice is pretty simple. We need to focus the community on the development. Everything we market, translate, document tell people about. All of it flows from a strong technical focus, strong technical leadership. And we have to invest in development mentoring and education that goes alongside that and revitalize the fun of contributing to the code. Focusing on our primary production, not our tertiary strategy unless that strategy drives more investment, more development and more good things. And it seems to me that the user experience despite huge work can still be improved easily. And it's just a great place to invest. There are lots of things we need doing in terms of bringing people into the project this has to be one of the easiest places to make an impact and see what you're doing and get good things done. So there we go. The future of the key is bright. We're going to fix these things. We're going to work together. It's all going to be awesome. But let's try and get some conclusions. So, we did what we had to do to make the code base survive. I'm still convinced that we did the right thing. It was to news, Oracle and some. And thank you so much to all the many people I didn't name. There were lots of extraordinary people doing things. At the beginning, but also through the lifetime of both open office and repurpose. Economics buys the really capable, incredible promises of future contribution that you use to inflict present harm to win a nebulous future. Because as it's been seen, it is easy for people to promise and not deliver. Even giant, credible companies. Beware of the Ubuntu effect. Everything is free and can we get it from us without paying anyone? Beware of free lunches. Please understand that investment shapes the future. Of course community contribution shapes it too. But if you look at the big gaps in our roadmap on how we look to our product distribution, it seems pretty clear that you need people to put serious money in. And ideally you need to come back again and put more money in afterwards. I think we're struggling with. But really be careful how you use these collectively created assets. Codes copyrights, trademarks, whatever it is, you have to be super careful or you create real promises to yourself. And even professional strategists perhaps you are a strategist. Even professional strategists that work in large teams and talk to talk can fall foul of development realities. Who is actually doing the work? Surely the character. And that concludes my talk. Thank you. Please do ask any questions. I think we have a bit of swag time. If there is some time for that I'd really have to answer. Michael, I wonder