 So it's 430 and Tell me who you're here to see Who? All right, is that what's on the schedule? So the who should we get? What you would think right? The universe is a mystery place And Los Angeles is where we make adventure movies. So what did we bring you? We like to think of it as Johnno plus it's Johnno with an adventure story In which Johnno may or may not be here because if we knew Johnno was gonna be here, there would be no drama If we knew Johnno could not be here There would be no drama But as of literally 30 seconds ago We don't know if we're gonna have Johnno Well, that's a good question. Now you think that's kind of relevant, isn't it? See that's what I wanted and he's telling me he left his GoPro at home But if he goes back to get it then that whole thing Well, then we get a fine wrap but next time but if we did it next time then that would require planning and there wouldn't be drama There was a plane issue is what it boils down to when you let me know the plane was just taking off He was keeping me up to date every minute while the plane was in the air the plane landed I got word a few minutes ago. He said his uber driver picked him up and the uber driver I in LA Well, here's the best part. Here's the best part His his uber driver when he picked him up said, oh I think we can do that in about 20 minutes And Johnno not being local said Hmm, I think he said 20 minutes and texted me my uber driver said 20 minutes Which we know to correct as three hours, but Yes Not driving but yes Burbank So the bottom line is this we had a good time with Q&A before we had a little break. We're back We can continue that conversation Johnno may join us Johnno may not join us Johnno's uber driver might join us instead. We don't know I'm getting updates by the minute. We'll keep you informed Take it away Okay, so I was sure why not could happen. Yeah Well, of course Johnno's talk was lessons learned from a boom to and I think I can sum it up if he so if he can Get here he can elaborate, but I think what we really learned from a boom to Was how to love So there's that Um, but yeah, he may get here fingers crossed. We'll see until then it is Q&A time And so it's such a good time. You have more people here, which is great because um Not that we're here for Q&A, but we have a great time in Q&A because you can ask any a boom to related question you want You can ask any Question you can sort of try and spin as a boom to related And we will give you an answer. It may not be the right answer To your exact question, but we will definitely give you an answer Um, and this is just kind of a fun time to uh, you know ask, you know, whatever is going on um, and sometimes we have audience members who Know the answer when we don't and we welcome that too because you've been to is all about community and working together And so that's just this one So I think it's time to um open up to the audience. Um, you're burning Urgent boom to questions and also, um, are you going to mic around or are you going to So so rude to so rude to get your phone The uh skill avi hates when i'm holding two mics Let me point them at each other For the best mic win big question When does 18 Maybe not you do have to eat the mic, but the uh also So when does 1804 come out? You've been to 1804 will be released. Um on avril 26 to 2018 and on once April 26 2018 UTC midnight rolls over Every time somebody asks that they delay they're released by 15 minutes So when does 1804? Yeah um, if you're running a boom to uh, 1710 you'll have an update. Um Prompt basically right away and if you're running 6004 you won't get prompted until six at 1804.1 comes out in july mid july so You can force it, but otherwise no big first day bug surprises when on your LTS upgrade so uh And then there's a question back there This is a good deal for me. He's the one on the bike all the time and i'm the one that needs to exercise him He's running excellent. Why is maria db in the uh universe repository not in main? That's a fantastic question. Um That is a good question actually I'm not sure i'm not sure So maria db should be a complete drop in replacement for um My sequel i don't I don't know why um Who here's from canonical this weekend? I can harass about that now. I want to know because Why is uh maria db maria db in universe and not main? maria is pretty awesome. Yeah, that's a very good question because I run next cloud and I would like to use the better less evil database so Um, of course when I use databases, they're all evil when i'm trying having to support them so and Better opinion. Yes Mine's related more towards uh networking. Um, the fact that uh, I up up down being deprecated for I forgive what the name is for the new convention Why is that being deprecated? That's a good question. Apparently, um, it's been decided upstream that the old ways were Horrible and wrong and outdated and we can do better and um Every so often with free software with technology they say well, let's let's do things better and yeah the networking in the last year So has been one where like You know what was wrong with networking all these years is having um short identifiable readable names for all interfaces what if they were Automatically generated based on the mac address and unpronounceable and 64 characters long um If you're running cloud stuff um to be able to do that to have automatically generated names that don't change because um If you swap PCI cards, which is not a thing anymore Suddenly your interfaces change. Uh, so uniquely identify defiable uniquely identifiable interface name It's sometimes a good thing And so speaking of networking And transportation You're gonna want to get the mic back I know Yeah, you know, I told him the lesson we learned from linux was how to love So do you want us to stall for another five minutes while you boot up? You want to stall for another five minutes while you boot up? Or are you? Okay, are you running gen two? Uh, there was a question over here. I think yes How do you manage conflict in the community? I almost feel like I want to give this to johno But basically since I am a member of the local community council um It's tricky. Uh, what you have to do is you have to I mean the winter has a code of conduct, um that Governs how we treat each other and so it means think we Be respectful um, this is one of the biggest things and what we want to do is um We try to talk to people to each other with an open mind and that doesn't happen often. There's a governing body There's a um, you're working on a subsystem and there's someone kind of in charge Or you're in a local team and you can't um the local team members Can't decide on on something there's a conflict. Um, they can escalate to the local community council We can uh, there's a community council on top of that the local community can't local community groups Locals are short because we're crazy, but I've been to our groups of um of boon to enthusiasts and advocates throughout different countries um in the The u.s. It's per state. So ubuntu california is here today manning the booth and running everything and um the president of ubuntu california and Liz is a member and John is a member and richards member so if uh if uh People in california work an event and had disagreements they could come to to me and and myself and everyone we'd say well We tried this tried that If that didn't work they could go to the local community council If and I would recuse myself it was a california thing and then There's a community council for for for community community things for technical things. There's a technical board In the ubuntu project You know mark shudderworth's title is self appointed benevolent dictator for life And so he can also kind of snap in and and soothe things In the end we always want to assume good intentions and just try and hear each other out and see how that goes So All set Fantastic. Well, thank you for your patience and to great excitement and no further ado. I'd like to introduce john o'bacon Thank you everybody my apologies for being late. I hate being late Anywhere somewhat hilariously I had a flight that was delayed So I rebooked on to the earlier flight And then that was delayed to the same time as the delayed flight So this has not been a good day for traveling. Um, all right, so um First of all, thank you everybody for coming to the ubu con. Um, It's it's great to see so many friendly faces here as well and so many new faces so Back in 2006 I joined canonical, uh, where I I joined as the ubuntu community manager and This kind of launched. I think one of the most interesting One of the most interesting phases of my career for sure Back then I you know, I'd spent a couple of years Working and thank you richard and experimented in an open source Um, what I didn't realize was the journey that was about to to happen So I'm actually going to play you a short video I know that we haven't rigged any audio up to this So I'm going to put the microphone in front of it But this is actually a little video that one of my colleagues robbie williamson canonical produced and I think it really kind of encapsulated The sense of what ubuntu was all about. So this gives you a sense of the journey that preceded anyone remember any of the pieces in that video It's like a thousand memories Crammed into that video it for me was you know, the eight years that I was a canonical was thank you ma'am We're just an incredible time And what I wanted to go through here was to just kind of share some Some some of the lessons that I learned over that time that have been particularly relevant since I've become a consultant And worked with a number of other different companies. So when I was a canonical It's by no means was was me. Um, we had just me we had a a team of people And it became a bit of a common theme that we took these horrible boy band shots Like this I'm not going to deny that I was absolutely driving this. Um, this was actually This was actually a shot of the team And this was taken on my last day at canonical in malta just before I left and this was A pretty nerve-wracking time for me because I basically grew up kind of in this position You know when I was When I started out there it was I was very new at this and I thought I had a sense of what I was doing, but I really didn't And uh, here we can see some members of the team as well as some members of the community who Were as close and it part of the team as the guys who worked for me who were paid by canonical You know, it was a real sense of team spirit. We took we we looked after each other When people got stressed we wrapped around each other like a glove. It was a pretty remarkable time So first of all, I want to get into some of these lessons There's hundreds of lessons that I took uh while I was there While I was participating in a bun two and a canonical So this is just a snapshot of some of them that I think have resonated Most with me and my experience the first thing is you miss every shot you don't take And this for me started out right at the beginning um in 2005 I was at uh Guadeck in Stuttgart And I was hanging out with the GNOME community And this guy walks in and uh First thing I thought was look at that haircut, right? I still think the same thing it turns out. Um, I'm just envious honestly So he walked in and I had a meeting with him and I was at the time a really enthusiastic bun two user This was very early um in in in in of course in the Evolution of the bun two project because a good friend of mine is called scott james remnant Um, and jeff and this other guy jeff war were very involved in the first stages of canonical And they'd kind of clue me into some of this stuff that was happening but um It's one thing to be excited about A project. It's another thing to go to a multimillionaire and ask him for a job and uh When I heard about this a bun two community manager position, which had just kind of sprung up I'd never seen the term community manager ever like ever. I've never seen it anywhere. And um And the first thing I thought was like I know I can do this job, but I'm probably not going to get it. There's already some members of the bun two community I've never worked for a distributor before like why should he give why should he give it to me? I had the most awkward job in interviewing his kitchen. Um It was I that's the story for in the bar. Um, but you know The thing I think a bun two is taught us is That showed us with open source is if you if you don't have a go Then you'll never succeed. And I think the bun two story has been littered with Store like examples where At the beginning somebody said you're crazy You know like 12.7 million dollars for a crowdfunding campaign is ridiculous, right to try and raise 32 million dollars We had to go at it. We got most of the way there. So So have a go at it and that's something that's definitely followed into my career is that if you don't have a go you you will miss every single shot Okay, so um And I think this was not just this didn't just resonate for me in my career But it's also resonated. I think throughout open source as well The second thing here is that diversity of participation is important I remember when I joined one of the things that I really wanted to do my main mission with a bun two was to diversify the types of participation Back in 2006 when I joined Linux was a place where you know Generally to be successful. You had to be a a man and be a programmer And when I started at canonical, I really wanted to make sure that the bun two project was a place where so many different people From different backgrounds were able to participate Now i'm not going to claim but for any Any second that I am a diversity guru, right? I was as equally Unaware of the importance of diversity particularly in terms of gender diversity back then As I think a lot of people were and that's something of of course has come later in in in Technologies evolution But at the time one thing that was very important to me was diversity of participation not just programming but documentation and advocacy and testing and all these different ways that people could participate what drove me To pursue this was that I'm a basically a bit of a crap programmer You know, I have a go at it And I can write a bit of python code, but my code is pretty terrible So this was completely self-serving that I wanted to make sure that more people like me could participate because there's lots of us out there um and a good example of this actually was a project which I kicked off in sort of 2012 sort of time, I think it was 2013 called the bun two accomplishments Which was a gamification platform for for a bun two. It was kind of just an idea that My best friend and I started kicking around and playing with and what I loved about this is that I wrote a chunk of the code About two other people wrote a chunk of the code But you know people did artwork and people did translations and testing and You know, I remember this client. I put it into launch pad and 24 hours later It was translated into 12 languages just this one tiny example A lot of open source projects. I don't think look at it from this angle I think there's such a focus on developers particularly in a lot of modern open source projects that we sometimes forget that Translations and testing and all kinds of other things are critical. So not just diversity of gender sexuality um Religion all that kind of stuff, but also diversity of participation and collaboration Now the third thing is one of the things that uh You know, I think when we look When we're in the future when we look back in the history books at at at a bun two One of the things I'm I'm most proud of is that I think The notion of a release cadence came out And the person I credit to this credit this to is mark shoalworth You know, he was hell bent from day one that we are going to release every six months regularist clockwork Now, of course, that wasn't a new concept. Debian where it had a certain level of of release cadence Some other distributions were playing with it as well. Um, some of the projects have done this But in a bun two, we said every release comes out and you know Every six months and then part of that six month release cadence means that at the beginning of the release you're planning Um, you know that there's an a bun two developer summit coming up You're putting your blueprints into launch pad. You know that there's going to be various string freezes and qa freezes and uh, you know all these different regular points in the timeline and The reason I think this is powerful for two reasons One is that it gets people into the same mental mindset of where we are in the life cycle of a piece of software This has become particularly relevant since I've become a consultant where I work with lots of different companies around how they implement community strategy And one of the big challenges with community strategies You can do a million different things you can do blogging and Meetups and social media. You can build on ramps for your for your community. You can do all kinds of stuff How do you pick the right things and make sure you ship stuff? and Boiling that down into a release cycle means that You're able to optimize it that the release cycle itself becomes as much of a project as the project itself So in a bun two what we would end up doing is Is you know, we'd have a release cycle We're executing it some stuff will go well some stuff will go less well And then we would optimize it. Okay. Well, maybe we need to adjust the these different deadline dates Maybe we need to provide a little bit more time at UDS There's all these different optimizations that we made to how we work together And everybody who was part of that process could feed into that as well So the way in which we work together to me is as much of a product as The actual thing you're building and that to me was a really insightful thing at the time Which I had no idea about to me. It was just normal. Yeah, that's how we do it It wasn't until I came out of canonical that I really realized how Unnormal this this was As a small example You know one of the benefits of this is that when you start planning a bunch of work into a release cycle That means you can start generating things like burn down charts Which is where you can basically say we've got all this stuff to do Are we going to make regular progress on these different things to accomplish our you know to get to our deadline? and What I loved about this was that people in the community Would feed into this just as much as my team would in fact There was more work going on in the community than than my team was doing in many cases And this was broken out across many different teams And you know the thing about workflow is that it's imperfect. There's never going to be a workflow that's perfect for everybody So it's there's always opportunities for refinement seeing data Experimenting and then improving and I think that was a really really powerful lesson That's really shaped the rest of my career and I think can shape a lot of open source projects now Another thing here is that that I really learned a lot about was The importance of on ramps and what I would call enablement Um, and I don't mean hardware enablement in this context. What I mean is um Every one of you sat here at one point was new And some of you will be new right now and you'll think I'd like to participate in the buntu or another project Whatever it might be In my mind the way I tend to think of it is that to be successful in the community that person needs to go from being new And then getting to creating something of value for either them that's both valuable for themselves and for the project It's not just important to generate value for a buntu. It's important that you can You can derive a sense of personal value as well, whether it's satisfaction, whether it's helping your career Whether it's increasing your your skills, whatever it might be To go from not knowing how to do something to know how to do something is what I would call an on ramp and The on ramps vary generally Depending on different types of participation So for example, I've talked about this in some of my previous talks So some of you will have heard me blather on about this previously But for example for for developers typically I tend to think of it as the first bit is you need to discover How to participate secondly you need to set up a tool chain Thirdly, you need to be able to learn the skills to participate such as using documentation And training then you need to be able to find something to do then you need to be able to get help And then you need to be able to Have your your contribution reviewed And then you need to feel a sense of validation that the work that you did if it was good was appreciated Steps one and seven on that on ramp are always the same. It's always discover and validate Because that's how we build a sense of belonging in communities, which to me is the most important thing we should be focusing on so When I was a canonical and when we were working on Ubuntu This was always at the forefront of my mind. I Consistently beat myself up that it was way too hard to contribute to Ubuntu. It would drive me nuts and I'd You know rant and rave at my team about You know people will come up to me and say, you know what? I I tried to contribute to this particular project But it took two hours to get the tool chain set up or The worst thing for me was yeah emerge, you know merge review has been has been sat my merge proposal has been sat there for a week It's like this is crazy. So Looking at how we optimize that Is I think essential and so many open source projects don't do this, right? Don't look at how to do this Like to me what you do is you put your brain into incognito mode and you look at the world from the perspective of a Of a of a new contributor and try to throw away all assumption and see what you can do to to optimize those different pieces It reminds me a little bit of a really close friend of mine and I I I'm really into smoking barbecue And one of the things they say in barbecue is always write down What you did in a cook because then when you do something that works You can go and see what you did and you can learn from it and I thought I'm not going to use a pen It's ridiculous. First of my handwriting is terrible. So I built a web app And the reason why I got into smoking was because it was time away from a computer It was you know fire like an analog medium and then I ended up spending more time writing barbecue pad than actually smoking Which is the way it tends to work So him and I kind of you know We've had so many of these little projects before and we we came up with an idea with how it is actually quite a complicated Workflow because we wanted to make this ultimately flexible And we thought it was it was genius and it would work really well And then we did some user testing. We actually asked people to do a cook and and use it and it was a complete disaster It turns out that when we we were trying to put ourselves in the minds of our users And we had no idea what was going on. So this is I think really really really really important um This is actually where a chunk of my team's focus got got got sank into developer dot ubuntu.com. It was How do we provide an environment that gets people up and running? You know elegantly quickly and easily this actually This actually relates to another lesson that I learned is as well is there is a very different relationship that you have I think between And people who contribute to the core of an open source project and people who build apps on top of it, right? So people who Contribute to ubuntu Care about how it's governed The licensing they care about the the culture. They care about the politics, right and so much of the So much of the discussion but also so much the noise Was wrapped up in our community was wrapped up in in in the operational dynamics of how this project was running One of the things I've learned is that people who write apps for platforms don't care about any of that stuff They just want documentation that works an sdk that you can build software with and the ability to get your software out to people So when we started working on the ubuntu for devices apps developer strategy Uh, I actually explicitly didn't want to focus this on the existing community because I saw it was almost two different demographics So that's one of the reasons why when we when we build this We're actually built this from the perspective that you don't have to care about, you know You don't have to care about the latest flame walker on a mailing list You just have to care about getting software having great documentation and building great apps All right, I'm going to try and get through these as quickly as I can because I think questions are more important than me ranting on about stuff Expectations are important and often poorly communicated. This is again something that is One of the things I've learned as I as I get older is there are There are so many things There are so many future insights In your life. They're already in your brain But you need to put yourself in a different scenario to unlock them. It's kind of like a horrible video game and one of the things I've learned from working as a consultant is that um Is that the expectations of your stakeholders you have to manage them effectively and we always kind of knew this when we were doing this in Ubuntu and we obviously had Very very explicit stakeholders, right? I mean an obvious one is mark shuttle. Right. He's an opinionated guy He knows what he wants and he will tell you what he wants So his communication his expectations you'd think Weren't poorly communicated They were loudly communicated But they were sometimes poorly communicated and that's not his fault Because a lot of people when they whatever they say they want one of the things I've learned is It often isn't what they actually want So for example, some of my clients that I've worked with I'll talk to them and I'll say, you know, what do you want and all all of them will say I want growth. I want a big community wrapped around my product to my service or whatever the whatever we're building the community around But when I delve into it often what they actually want is validation in the market more customers Brand recognition any of those types of things often isn't just growth so One of the things I've learned is that particularly for those of you who are working on an open source project that's got a company attached to it or if you work for a founder Is going through the process of understanding expectations is really important Because it gives you two things. First of all, it actually gives you a sense of what the goal should be and what the deliverable should be But secondly is that if you do it well and you write it down and you agree it and you go through a process of kind of like understanding documenting reviewing Then what happens is when that person changes their mind a thousand times you can say, okay, we can change your mind But if we can't do this if you want to do this Then we can't do this So you're gonna have to make a decision we can't do both And that was where I think that is where I think so many companies get stuck is that they Is that there are stakeholders who want to do everything and you can't do everything and it kind of gets embroiled in this kind of lack of communication so Going through an explicit expectation setting expectations and you know requirements gathering I think is is really important Number six scale is tough. I don't mean the conference. The conference is very easy. Okay This is literally the only place you can make a joke about that like that Um This was something that um You know again like when you without turning this into an autobiography like when I Look back on my time at canonical The thing that I think I internally wrestled with more than anything was how to deal with scale Um, you know the early stages of the of the community And the company started looking like this This was in mountain view Then it started growing. This was in Spain I think it was Uh started growing more. This was in Budapest This is in Orlando more and more people. This was in Copenhagen. I'm not in that picture because my son was being born um More and more people were joining, you know, we had over a million people on the forum. We had hundreds of developers um You know a constant Microscope on the work that everybody in the project was doing Um, and of course that hasn't stopped to this day. Ubuntu is a powerhouse in technology and it can be difficult to know how to manage that the thing that I wrestled with repeatedly was How do we build scale but keep the personality of the project? Because that's what got me excited about Ubuntu when I started out was it wasn't just an open source project I didn't just get a website With an on-ramp where I can you know know the steps to go through and nowhere to submit the pull request and Know how it gets reviewed and you know, you get your little icon in the corner of the website I didn't want Ubuntu to become that I want you to Ubuntu to retain the character like this The thing I love about this is that when all of us are in the bar tonight or in a restaurant tonight Or just hanging out with our friends tonight We'll talk about certain people the certain characters in the Ubuntu world that Many of us will know You know like Corey Berger, Jeff Spalletta, Benjamin Mako Hill Richard Solman he's a fan So, uh, but those characters I think, um They shape the community they give it life they give it character and uh And as you grow it can be increasingly difficult to kind of maintain that character And one of the things I've learned since uh, I left canonical Was has been the only way to accomplish that is to be explicit in how you how you react to scale Not the conference is that is that you have to build For all intents and purposes mentoring programs where Where people can guide people to be successful, but you you have to have Um places for the community to be social with each other Um, one of the things that I think was a great, um Financial decision, but a poor community decision was when we moved away from in-person UDSs This was a this was a uh A victim of scale the early days when I my first Ubuntu developer something in Mountain View there was like 80 people there Um, you know we're good all fit in a restaurant The last Ubuntu developer something that I went to there was about 800 people there So it got too big it be they became too expensive and I think canonical made the right Decision as a startup to try to move that to an online setting, but it had a pretty detrimental impact on the on the That character of the community because so much of that was just meeting people in person and hugging each other and all the rest of it So scale is tough, but I think it you can do a lot by by Being intentional in the way you in the way you address it seven I think it's 12 so we're nearly there Most people need direction, but they don't need necessarily need options. There's a there's kind of a misnomer. I think in the open source world that You know, it's all about choice. You know, this is the reason why we've got 600 linux desktops um I don't think people actually necessarily want choice and choice is a burden right you need to know your options Um, if you if you if you want to anyone who's built a pc knows what i'm talking about Where you have to pick which ram you want which processor you want all the rest of it So, um There's a the the offshoot of this. I think is that what happens is you say to people Here's all of these ways in which you can participate. Here's all these ways in which you can join our project and be successful um And then what happens is they don't actually make a decision They don't participate because they don't know where to start and they feel a bit unusual about volunteering The thing that i've learned and you've got to be a little bit kind of forward about this is if generally If you go up to someone and say hey, what do you mind doing this? Nine times out of 10. They'll say yes unless it's you know, like money laundering People don't usually like that sort of thing So when you start a community and I think when you continue in a community Making decisions about okay. Well, we think we want to do this want to move in this particular direction and then saying all right Jane will you do this? All right, bob will you do this? All right in do will you do this Is You'll actually get people to say yes And then if you get if you ask those folks to commit To those actions in a public setting then they'll usually Get you know end up getting done You know we used to do this a lot again at the bunch of developer summits where we'd have all of these sessions and one of the things I used to You know go on and on about in my intros at the bunch of developer summit was Get actions written down. Don't just go and sit down and have a nice chat, right? Let's do something like this is a limited opportunity. We're face to face. Let's make something happen and And a lot of those sessions that I was in at least and and I know other people in the community The end of it was someone saying okay, this sounds great. Will you do this? Okay, will you do this? Okay, will you do this? Okay? And the burn down chart gets filled. So yeah be intentional Eight well focus work can deliver great results so When we started the Ubuntu devices strategy, I'll always remember flying to budapest I just moved to california And I realized how far california is from budapest I spent two days traveling and I got there You know and my beautiful new wife is at home. So I don't particularly want to be away from away You know from home, but I'm looking forward to seeing everybody I get in on sunday evening run in kind of like how I did Matt just now Mark shellworth comes up to me and he says We're gonna Which was we had a bunch of people do some designs the design team approved them at canonical And then we went out to people and we asked them if they wanted to write these apps and have early access to this sdk I'd never seen this model done anywhere else And what blew me away is that it worked You know people started building apps. This is actually deco. So this was kind of out of it, but Like we got this incredible group of people would fly them out to sprints and and and they did a ton of work And all of that was based upon trust You're like no one else had access to this sdk. This could have been a terrible idea a terrible waste of their time But but I think when you when you intentionally say we want to accomplish this and go after it It's amazing what people will do If I think there is a really good sense of kind of equity in the community that people feel part of that My goal with this and I asked David Panera who when I quit canonical he ended up replacing me I asked him to run this project and he did a great job and the thing I said to him was like We have to make all of these folks feel like There is there is valued and has participated in as any canonical employee because this was before all of this had Kind of gone live and public with uh with you know with the broader community Nine change in habits take time Uh, the general scientific thinking is that it takes 66 days to develop a habit Which is actually a really useful metric if you want to get anyone to kind of build a habit in anything Whether it's not smoking or whether it's going to a forum every day um But change is tough And one of the things uh, we went through of course with with a bun two was unity um Bit controversial at the time um And the idea here I thought was a wise one was so it all started off with this thing. Notify us day Anyone remember this? Yeah Michael whole definitely remembers it at the back so uh You know the idea was like let's just make something and see if see if there's an interesting kind of like Bringing it into the community and people participate in all the rest of it um, and there was the usual level of kind of like, you know Is this some way of canonical kind of stealing all of our our project and this that and the other and I think we tried to um Tempor that as best as we could and tried to be as collaborative as possible. This was based upon like a kde um spec this was all built on we tried to work with with kd and go around this um, the next thing that came in remember this Messaging menu of course, um the early version. This was awful. Uh, it got much better In in later in later years and and the indicators I'll always remember of course back then if you clicked on an indicator and you moved your mouse It wouldn't move to the next indicator Whereas with this you could like you could do this and your indicators I remember sitting there just watching it go over and over again. My wife was like, what is wrong with you? You're crazy But you know it was like these things that we the bugged us forever We started we started focusing and then of course the desktop piece started coming And then we of course we have the the the dash and all these different components And I at the time genuinely thought the way in which we were going about this was the best that we could could do with With the goal to differentiate, right? The reason why unity happened was so a bunta could differentiate from other distros, right? Because In the earlier days of Ubuntu the only differentiation we really had was a hardware worked With a lot of other distros. It just didn't work and b you got the cool stuff earlier But the nature of a competitive landscape is that your competitors are going to get better, right? So I think in the early stages of Ubuntu and this is no disrespect to the other distros But Ubuntu was just categorically scientifically better than fedora. It was better than debby and it was better than suzer It was just better But as Ubuntu kind of got older these other distros really got their game going right? They were shipping the the the hot new stuff earlier and earlier and earlier as well And for a company behind it you have to kind of differentiate in a way unity was kind of an effort to do that But to still collaborate and participate with with the broader community at the time I don't think again I'd ever I don't think I'd ever seen this kind of model applied and I think there were certain things we did That worked well and there were certain things that we did that really didn't work well And the thing I'm happy about is that we've actually we've got like a some case law The things you should and you shouldn't do but at the heart of this a lot of this was about change and and and I think some people in our community were had Very valid concerns about the implementation and the approach and stuff like that But I think there were certain people like change is bad. I don't want this because it's different And that's a hard thing with communities. It's hard to pivot Because you cannot let those people win You cannot let people Tie you to the past and never change and never improve Because what will happen is you will stagnate and we've seen many open source projects suffer from this So I'm proud of the effort But there's definitely things we could have I think could have done better, but Again, I think being intentional about change is really important You know like sometimes not changing things is actually a good idea, right if it isn't broken don't fix it So but with you know with something as contemporary in this wide region has been to we really had to kind of evolve it and keep moving it forward 10 Code is an artifact of community There is a massive misconception in the open source world that it's all about source code And I don't think it is I think source code is The tangible evidence that a community is working Um Or that the collaborative piece of a community is working There are many communities the right code that actually have pretty broken communities. I think but You know as we've seen various cases examples of you know, you have to It's important to um Well put it this way You can write code and have a crap community that no one wants to participate in But if you build a really solid community You will by definition get a good product at the end of it because it's well structured So to me a lot of this is about again being intentional in in strategy now Obviously, I'm biased. This has become my life for my career Like every element of how we build communities is fascinating to me Um, and I think we have to be intentional But a lot of communities just kind of rock up and say well, let's have a go at this Let's just let's just install a discourse instance and we'll register a github account and off we go And if you do that unless you are lucky um In many cases that will that will fail. This is one of the reasons why You know the majority of repos in github and githlab are empty Right, it's the same with every platform. You'll always have a small number a small group of um Of of users that will do great stuff and the broader The broader range will do less great stuff or no stuff at all So it's important to be intentional for us to get that quality up 11 The biggest ideas can be carefully, uh managed what I mean by this is sometimes you have A mark shuffle where say we're going to hire 150 people and build a devices strategy and the first thing you think is Yeah Right, there's kind of a lot to take on But I think even the biggest goals the biggest ideas and we see this more and more I think you can say let's take this let's break it down into smaller pieces Let's understand those pieces and be intentional in how we tackle them right and and This is I think really important particularly if you work professionally in In open source or in communities where it's only a matter of time before you have a CEO Or somebody else is going to come and say we're going to do this giant big thing And the first thing you think is I don't want to do that because it's too big But actually we can break these things down into smaller pieces and manage them I'm I'm a huge iron maiden fan And I made and have these enormous arena shows all over the world They hire a jumbo jet They take half the seats out They put all the gear in the back and the lead singer of iMade and flies the plane, right? It's nuts And their manager of 35 years rod smallwood was once interviewed and they said how do you pull off this incredible elaborate You know this elaborate tour with with all of the logistics and the visas and the band and all the rest of it And he was like make a plan And stick to it That was his response And the make and the plan bit is easy just sticking to it is the hard thing and it's as simple as that in my mind Everything can be broken down into smaller pieces, you know, whether it's something like this the devices thing or whether it can be something like The bun to edge this was that was so much fun the bun to edge I think I've got a prototype in my kitchen drawer still actually It doesn't work. It's just a brick Um And then number 12 You can think you know it all and you don't and this is uh, this is again a somewhat personal one. Um I uh when I Was kind of getting to the point of leaving canonical because I was just ready for something new I didn't think I knew it all but I felt like I had a reasonably good handle on how you build communities You know, I'd written a book organized a conference Do a bunch of talks all over the place. I felt comfortable in my expertise. Let me put it that way, but still a lot to learn um Having been out of canonical now for Four years I think it is. Yeah four years. Um When I think about what I knew when I left because you know, it's like a milestone in my life It is just mind blowing to me how little I knew back then compared to what I've learned in the last four years From working to get her of an x prize and consulting and Meeting people at conferences and all the rest of it I think it can be so easy to suggest you think you've got to handle on something whether it's community Whether it's software engineering, whether it's something else And it's amazing to think just to to realize just how much more learning there is to do and for me It's the that's the biggest thrill this is without sounding like a cheesy self-help thing One of the things I love is when you realize you've been doing it wrong When when you say like why was I doing it that way? There's a way more efficient way in which I could have accomplished the same results You know, you know to give you a tiny example from my business I used to retain clients on a daily basis So I'd be retained with a client for like two days a week three days a week Monday's Wednesdays fridays for it for example And I took on a bunch of clients and I was struggling a little bit because I was like, okay Well, I'm retained for these clients But what if I what if someone wants to have coffee on a Wednesday and I'm retained I'm retained for a client all day on Wednesday I can't do it. So I was every lunch break. I was meeting people And you know, I was doing calls in the evenings. So I didn't have this conflict And then I was actually in a pub one night with a friend and he just said Well, why aren't you on like a normal retainer? So what do you mean? He said, well, they book you for a certain number of hours during the month And those hours can be utilized whenever during the month and then they can have access to Access to you whenever they need it and then it completely frees up your calendar Seems like a good idea Made the change infinitely happier And I love moments like that because it's like I think too many people dwell on oh, well, I screwed up, you know, I I made a poor choice But there's actually there's so much opportunity for us to Make improvements to change and to grow And I think when you can be in a position where you're doing this every day where you're always learning new ways to look at the world And that's one of the things I love about open source Is there's always a new way to look at something there's always a lesson that can be learned like I talked earlier on about unity Did we make some mistakes? Of course we did But at least we can look on those mistakes and learn from them. And I think that's that's a pretty good situation So that's everything I have. I think we have a few minutes left any questions This is a shifty looking individual at the front I know him by the way. I'm not just being mean to a random guy So I will repeat the question and most of you probably aren't gonna under this is a bit of a ninja So the question is uh, how many I'll translate you said tomatoes tomatoes How many tomatoes does somebody have to hear at a conference? Whenever we did in a bunch of developers summit one of the standard slides I always used to have was something that said, you know, um, don't drink too much Eat healthily get some exercise like take care of yourself this week because at the end of the week everybody got something called ubu flu Which was, you know, someone from some far-flung corner of the world brought something to the event and everyone got sick and um So this was kind of like it became a bit of a running joke about, you know, eat a vegetable and uh, um The irony the irony of this is in the 12 emergency summits that we did I didn't give one of those slides when I wasn't hung over complete hypocrite So, yeah, it is the way it is So yeah, any other questions And by the way, I'm unoffendable. So if you want to ask something hard Go for it Michael Hall puts his hand up immediately Favorite moment at canonical God, I don't know so many favorite moments Clinical was the best job I've ever had by miles. It was so much fun Um, and as you know, Mike might used to work there as well It was the people that made it I think um I actually don't know there's so many. Um Yeah, I'm genuinely struggling to think of I mean, there's loads of it like the one thing I'll never forget was um in budapest We had a friday night jam session. We did this in the last night of every bunty developer summit And everybody being crammed into this little this little club and everybody just dancing and having fun and watching like You know people were translating the bunty into bulgarian dancing with the CEO of canonical and Mark shuttle worth awkwardly singing come on baby. Like my fire on stage Uh You know things like that There's like real social moments like that that I think were that I think were really cool that the One thing I loved as well was just walking into places like airports and seeing broken airport monitors running on bunty, you know that always kind of give me a Give me a You know kind of a boost one thing that was kind of funny as well I'll always remember was I was actually consulting with Deutsche bank on the side And I got bumped up to first class never flown first class before So I was you know like taking pictures and sending it to my wife and uh You know, there's some guy in the on the flight walks on has got like the first class jersey on He's very proud to be flying first class. I was thinking oh, what a pathetic human being Opens up his laptop a bunty Misjudged Yes Yeah Yeah, I have to say the the political communities I've not really spent a huge amount of time with I mean, I've seen great work that other people are doing But I think there's I think there's two comments in my mind on this Um, there's somewhat abstracted responses I think a lot of the consistent things that we do in successful communities can be applied in a multitude of different areas Which is what I think is so neat about this Is you know, make a plan stick to it Make it easier for people to participate Um build, you know, social connections between your your members You know have I think giving people a mission I think is the most important thing where I think it would tie with the political realm Is I think one of the reasons why a bunty was so successful and continues to be so successful Is because it really just rallied this kind of like energy in so many different people That's the reason why I got into it and it reminded me when I first saw a bunty It what it reminded me of firefox when that first kind of came out and people were like making crop circles and You know things like that with you know to promote it So I think having a mission is so important and that's where projects like a bunty and Certain political communities have a benefit is Is that it's easy to rally people around something because it's it's often an emotive thing Like a lot of people in this room who are interested in the bunty will do it because of Software freedom and ethics like it's the right thing to do One of the things I've discovered. I mean obviously you can see I'm running mac One of the things I learned was that for me, it's actually the community collaboration piece is actually more important to me than the software Which is the reason why I don't mind running a mac But the the collaboration between people and doing it in an open and transparent way is is non-negotiable Now if you're a company and you're trying to make a community of people who are enthusiastic about why a wool products It's going to be a little bit more difficult. So I think I think political communities can almost certainly look at many of the similar patterns in open source communities and vice versa The other thing as well, which I would just make one comment on is One of the things I think is a problem in communities and we're seeing this in our current political landscape is The growing inability for people to have differing views and have a respectful conversation And one of the things we add our you know, dirty laundry in the bunty community But by and large people are always very respectful with each other people who you'd radically disagree with Could sit down and have a civil conversation Um Not necessarily like each other, but just have a civil conversation. I worry about the political discourse that we're seeing um, you know where people Can't seem to actually debate Seemingly complex issues in a simple way. I guess is what I'd say Yep, what is canonical's biggest challenge moving forward? That's a great question. Um It's not profitability because canonical's doing well I think the tricky thing that canonical's got moving forward is The main area where I would say canonical are playing these days is in the cloud and the infrastructure world I mean, there's still a lot of work going on for sure in Like snappy and desktop snaps and all that kind of stuff. I mean, there's there's you know, there's teams of people working at canonical and that stuff but I think when you look at a lot of the work that's happening It's around the infrastructure, you know the keynotes the a lot of the messaging is around the infrastructure side of things I think one of the tricky things that canonical's got to wrestle with a little bit Is that I think a lot of people associate canonical as an operating system company The company that builds a bunty even though canonical builds a bunch of other stuff as well And I think operating systems have become a less central role in infrastructure these days with containerization like an operating system is a A thing that lives inside of a container whereas You know when when when we started out with a bunty an operating system was one of the most fundamental technology decisions that you make And I don't see that as a technology problem for canonical. I see it as a marketing problem and every company has this at some point where The hot new stuff becomes less hot because there's other things that that they're working on You know, I don't Have any concerns really about canonical in terms of execution and delivery Um, you know, there are things that canonical is working on that I agree with and some things that I disagree with But canonical continues to hire really good people and I think mark has still got like a really exciting and inventive vision but um I think the race is on to make sure that canonical continues to position itself as a very relevant organization Which it is there's no ambiguity around that, you know, so Yep Yeah, that's a really great question. So what Just to repeat it for everybody. Um, there is a bit of a stigma the linux desktop You know, it used to be a bit broken and maybe it's still broken, but it's actually a very, you know, comprehensive and Compelling technology. How do how do we get over that stigma? Um, I think it depends somewhat on your goal um You know, again comparing a little bit to when I was at canonical The reason why I was personally like many other people Working ludicrous days on Ubuntu was because I thought this is our shot to To take on microsoft and apple um To be like as valid as kubernetes is today And linux is today on the on the infrastructure world on the desktop And I think if if that commercialization goal is the goal It's it's it's going to be hard because the You know the hard bit is not the technology like linux desktops are great Right, I think we're I think the linux desktop landscape today is better than it's ever been You know when I think of the work that is going on in buntu with fedora the folks at endless with endless os I mean, it's amazing. What's that? It's amazing how elegant and how sophisticated desktop linux is today But the only the hard bit is is the distribution, right? It's it's the it's getting it out to people in products that they can use it's crossing that chasm and With Relative exceptions such as endless. I think I'm not convinced that that's going to happen anytime soon I you know I feel like we had our shot now that I might be wrong. I'm historically wrong about these sorts of things So I don't want to demote anybody at demote to demotivate anybody however if the goal that you're looking at is An awesome operating system that people like us who go to scale every year care about and love and participate in and have fun with like I'm more interested in that personally I actually not I don't particularly care about linux dominating the commercial world on the desktop anymore I care more that my buds can use it and that Kids in indonesia can use endless os and you know People in developing countries can can have access to information that they never could have previously and stuff like that So kind of honestly, it sounds like a kind of a bit of an arrogant answer But I kind of think people who think that stigma idiots, you know, I don't really care what they think So the people who I care about are already understand Desktop linux and I think and I think it'll continue to grow. I actually like the fact where I prefer where we are now I think we're in a good space I'm not sure how much time I've got. Oh, I'm over All right. I think we're done. Thank you everybody. Have a great scale So I want to thank you all again for coming to ubicon at scale It was a lot of fun a lot of challenging and Um, although it wasn't planned. It looks like everything from planning to execution was last minute today So thank you for your patience. You're a good humor um, you're uh camaraderie camaraderie and We hope that we see you not only having fun out on the scale expo floor this weekend, but also, uh, if you drop by the Ubuntu booth and say hi, we'd love that too. So, uh, I Think the expo floor is open for another 25 minutes. You'll have fun. Have a great dinner I think there's a happy hour thing. Check your schedule And uh, have and get some sleep so you can enjoy the expo floor tomorrow. Thanks for everything And for those of you who like to travel there are ubicons around the world. So drop by ubicon.org And as you have suggestions for next year here Drop us a note there and you can tell us what will happen here next year. We'll see you then