 No, that was terrible. We're going to do this again. So I took German in secondary school in North Carolina and Frau Shin taught me to say, good morning, y'all. So, su du, good morning. Very good. Welcome to Dresden and welcome to Flock. That was like, you were supposed to applause there. That's terrible. Okay. Save that for the cameras. So I am here mostly to serve as sort of MC and do all of the fun logistical announcements. So let's start with fun logistical announcements. You are in Dresden and they speak German here, but you will find that your English works really, really well. So we encourage you to use that. If you have any challenges, don't hesitate to talk to the staff. You can find us at the registration desk right out here in front of the elevators next to the hotel's registration desk, who is also a great resource. And then people like me will be running around. On the back of your badge, please read all of these things. They're terribly important. Included there are phone numbers for Jennifer Madriaga and me, Brian Exelbeard. Do not hesitate to contact us if you have any challenges, questions, or other issues. The sessions are going to be recorded. We hope at better quality than in previous years, but we'll see. But yes, they are all being recorded. We have two requests. The first request is, don't touch any of the equipment, please. It has all been set up very carefully. This is triply true in the rooms that have wireless microphones because it could take us half an hour to get all the levels back correct and to pair the microphones right. And, like, it's all kinds of bad, so please don't do that to us. Instead, if there is a video or video challenge in any room, contact me. Contact Jennifer at the registration desk, and we will take care of it. Amate Hushowski is sitting here, is on lead for AV for us, and we are super excited that he is here doing this. So if you see him, buy him beers after the last session so he can still fix the equipment during the conference. But otherwise, please be very eternally grateful. Also Pavel Valena is running around as well and has been super, super helpful in AV land. Food and coffee. A couple of notes here. There are going to be coffee breaks set up outside. Today's coffee breaks are sponsored by ARM, so thank them. And so thank you, ARM. If you have... I'm sorry. Coffee gets lost. Coffee doesn't need... Coffee doesn't have to tell jokes. Coffee doesn't have to ship on time. If you have dietary concerns, food allergies, et cetera, please be sure to talk to the restaurant staff during lunch. Everybody who sent in a dietary concern, we have communicated those to the hotel. They expect to be able to accommodate most requests. How do I know if I get lunch? You have a badge. Because there was some question about that. Other than dinner on Friday night, you are on your own for food. Those of you who reserved in this hotel through the conference procedure do have breakfast included in your rooms. The rest of you will need to look at however you made your room reservations and figure that out. Let's start with a couple of other things. There are stickers available for you to put on your badge to express your communications preference. I encourage you to consider doing that and I encourage you to strongly consider respecting the communications preferences on badges that you see. So please take advantage of this. There is a quiet room that has been set up for your assistance. It is quiet, which means please don't go in there and talk. You are not to have a meeting in there. You are not to decide to do jazzercise on a really loud video. You are not supposed to go in there and engage. It is a place for people to get away from the conference noise and bustle and hustle and to take some time out and figure out what's going on, check their e-mail, relax, whatever. It is in the Loschnitz room, which looks like Lobnitz because it has a giant S set in it and it is straight down the hallway this way and there are signs that look like the no microphones rule sign all over that room. The hotel has accommodated us with an all gender bathroom. This is actually kind of fun having all these signs. The hotel has accommodated us with an all gender bathroom. The all gender bathroom can be found directly behind this wall on this corridor here. The other bathrooms are gendered so please respect the signs of the bathroom you choose to use. If there are any challenges along the code of conduct lines, you can find code of conduct information on the flyers that you were handed when you checked in and then you can report those kinds of issues to either Jennifer or me and our contact information is on the back of your badge. There is no Wi-Fi information on the back of your badge because there is only one Wi-Fi network that makes any sense and it does not have a password. Rumor is it is also not currently handing out addresses via DHCP but we are looking into that. I don't know, maybe we will implement DHCP V2. This laptop needs Wi-Fi so if it comes down to it and the presentations aren't working, everybody else has to turn their devices on. They did that once at an Apple Keynote. We are in the largest of the rooms that we have and you will notice the roundtables. These are not where we send the bad children at Thanksgiving. These are actually set up for use mostly during the roundtable sessions for groups to do informal meetings tomorrow. I think it is tomorrow, it might actually be Wednesday. Wait, today is Wednesday. It might be Friday. Check your schedule and you will see the words roundtables and that is when we put them there. You can sit there if you want like knock yourself out but the audio and video is set up for this side of the room so if you can't see or hear you are going to be asked to relocate as opposed to us fixing all of the video for your convenience. There really are supposed to be enough chairs in here if you guys would defrag and let the empty chairs be known. A couple of other activities. Tonight there is karaoke and game night and the candy swap is going to be held inside of the game night area. Game night will be here because we have roundtables if we need them. Amazing. Karaoke as I recall is next door in the Dresden room. There are two access points to Dresden. There is a door here in the back and then you can keep going straight down the corridor and honestly rumor is if you go down this corridor long enough you get to the Dresden room. It just don't know me while I was standing here. I have never actually walked that far in that corridor so let me know if that's not the case. But we are super happy the folks who have organized that for us and both of their names have escaped me so I apologize. They are listed in the program so please pretend that I read that. This morning at 6.30 in the morning there was yoga and my understanding is that we had people show up and that is totally awesome and I won't ask anyone to do a sun salutation right now but I encourage everyone to consider joining the yoga tomorrow morning at 6.30 as well. It is held in this room, mostly in this empty space right here so you do actually get to see like sun and stuff but you're on carpet and you have air conditioning which is about as good as exercise gets as far as I'm concerned so I encourage you to take advantage of it. We're super happy to have Suzanne doing that for us. We have lightning talks scheduled for the last day and we encourage you to sign up for those. Those sign up for that is at the rain registration desk. There are only plenary sessions today and tomorrow so you will only have the privilege of my announcements for those two days so I'm sorry. And I think, oh yes, two other notes. If you are interested in helping with audio video especially if you know you're going to be in one room a whole lot and want to act as a sort of room chair please see Mate who has left the room or Jennifer, now he's over here or Jennifer who is standing next to him and will coordinate getting your assistance. Lastly, you've all picked up t-shirts. We have given you the size that you said you wanted whether that size has any relationship to the size you actually wanted or not is something you have to search your conscience for. If we have extra shirts which we are expecting we will start doing swaps for different sizes on Thursday which is tomorrow but we want to allow everybody a chance to claim the shirt that they asked for before we give their shirt away. And I think that is everything that I have to say right now. I want to thank our two primary sponsors for the conference this year and that's Red Hat and Arm so if I can get a round of applause for them. And our first speaker is going to be Matthew Miller the Fedora Project Leader. That was Brian Exelbeard, the Fedora Community Action and Impact Coordinator or F-Cake. I don't think he introduced himself. So that's important too. This is like last releases or two releases ago's wallpaper but I really like it but I really am running the updated Fedora just in case anybody was wondering. Okay. Wait, water. Some logistics before we can talk. So yeah, I'm Matthew Miller, the Fedora Project Leader and this is the sixth flock which is kind of incredible to me. I had to count it like three times before I realized what it was. I like to do lots of graphs and things in my talks. This particular one is a personal one about my optimism level over the years at Flock. Last year things were kind of maybe a little lot on fire feeling. There was some carrying around of fires and I spent way too much time on this particular slide everybody. So but this year I'm really feeling good about things both the project as a whole and this flock I'm very excited for I think Hey, there we go. Things are unicorns and rainbows kind of feeling right now so I think this is going to be a really good conference and we're having a really good year in the project. So yeah, welcome everybody and let's have a great conference. I have a green dot on my badge here. I hope anybody who doesn't know me comes up and introduces themselves. I'm happy to talk. I'm on IRC and Twitter and email and basically everything I can manage except for Instagram where somebody had already taken my name. Please feel free to reach out to me at any time about anything. I'm really happy to talk and to listen. That's basically my job. So I want to start out with some messaging stuff because here we've got everybody in a room so I think it's a good time to talk about some of these things. Thank you to Langdon for this picture of the Fedora Museum at the Red Hat office in Boston. The question here is, what is Fedora? I think hopefully everybody at this conference has an idea but messaging is hard. There's a lot of things that we keep saying despite ourselves. One of them is, for example, bleeding edge. Fedora is not intended to be bleeding edge if there's blood on things due to using Fedora unless it was some sort of medical situation. Not right. Fedora is a leading edge distribution and we really try to not be bleeding edge and this is a matter of not just language insisting that we're better than we are or something but Fedora, we don't really try to be necessarily like arch. If something comes out, we have got it there the next day. We want to make sure software works before we get it to you. So, leading edge and we leave the bleeding to other people. Unless you run raw hide, then you're kind of on your own. That's what it's there for. But anyways, what is Fedora? The point of this here is actually that Fedora, these are a bunch of Fedora operating system DVDs and CDs, but that is not Fedora. Fedora, like solvent green, is made of people. This is what Fedora is. And so actually the serious point is, I would really like us when we talk about Fedora to mean Fedora the project. And if we are talking about something that Fedora the project makes, let's call it Fedora operating system or even better, Fedora Workstation, Fedora KDE, Fedora Python Lab, and Fedora Core OS, the specific thing that we're making, let's talk about it that way. Especially if we're talking amongst ourselves, whatever. But when we're talking outside the community externally, let's try and keep our communication clear that way. Because I think it helps people understand that we are more than just one particular operating system thing that we put together. This is like the wordiest slide in my whole talk. This is basically a summary of our mission. And this is the Fedora foundations here. I think everybody is familiar with these. And I think this is a thing that makes Fedora special is the way we have these values in this community. And it's something that I spent in my job, spent a lot of time at Red Hat fighting for Fedora and the thing that makes Fedora special. Fedora is kind of unique as an open source project. A lot of open source projects are kind of a corporate open source thing where there's a thing the company makes and they put the code out there. And you can participate in it as a community member, which means you can chat on the forums. If you can submit patches, they might be accepted, but you're not necessarily really invited to be part of the inner decision making. In Fedora, all of our decision making is in the public. Our Fedora council is open to anybody. We've got some hired positions that Red Hat funds, but we have general elections for full council member positions. And this is really kind of unique and special. And it's one of the things where Red Hat has a lot of open source projects and a lot of open source communities. And a lot of them, I'm not saying they're not genuine open source, but I think Fedora is the one that's big and successful and really has a community that's engaged in a way that is unique. And it's worth protecting and celebrating. So one of the things that I've worked on in Fedora, there's kind of a lot of different ways we could go with what Fedora means and what the Fedora brand is. And one way we could easily go is Fedora is kind of an underlying thing and that our brand is sort of a Powered by Fedora thing down in the corner and then we put, you know, other names up, you know, KDE, Powered by Fedora, or whatever, or CoroS, Powered by Fedora, or whatever server, those kind of things. But I've really fought for this idea that we want to have the Fedora brand to kind of celebrate a big umbrella brand where we allow a lot of things in here. And rather than saying, oh, that's not the way we've already done things, that can't be Fedora. Fedora has to fit this particular RPM format that has, you know, our packaging guidelines say that needs to be this and it needs to be this way on, you know, DVD to make it be real Fedora. I would like to have an expanded meaning for real Fedora. Things that fit these values are not going to give up free software or not going to give up open source and we definitely want to make sure that the community thing is there. We don't want things to be Fedora where there's something that, you know, Red Hat or some other company or whoever has exclusive control. Nothing that's based on who you are rather than, you know, your contributions. But things that fit into our values, I think we should be welcoming to calling those Fedora and putting those under the Fedora umbrella because that's a lot of what Fedora means to me. And I think that we have a hard time recreating this awesome community and things we have here under another name. It's often really tempting when you've got it something cool you want to start up to be like, okay, I'm going to make my own new community for this. It turns out that's really hard. And so we've got a lot, we've got a great community here. And one of the things that's kind of a challenge for this community that I want to offer is that we want to make sure that we're welcoming to new ideas and new ways of doing things in this community as the operating systems change in the future. Because the things that were important for Fedora 15 years ago are not necessarily what people in the world care about and need in an operating system today. So we need to make sure we're able to adjust to those kind of things. That previous slide was taken from the new Fedora dock site. So I wanted to put in a shout out here. This slide is to remind me to say, check out the new Fedora dock site if you haven't. It's very nice. This is also from the Fedora dock site. I don't have a good segue to the slide. I just wanted to put this here. At Flock and Prague, however many years ago that was, I stood up by a whiteboard and tried to draw a map of what Fedora looks like as a project and how all the groups are interconnected. And it was kind of a crazy web. We have over the last couple years tried to bring some structure to this. And this is what, especially with the Fedora Council in the middle, and this idea of having FESCO as an engineering oversight body and then the Fedora Mindshare Committee for the things that are not necessarily technical engineering decisions kind of bring some order to things. I don't think, I can't promise this is comprehensive and maybe some of these groups aren't active anymore. It's a big project and a lot of things move around, but this kind of gives an idea of the scope of what Fedora is. Okay, now we're moving on to charts. This is, I don't know, I've shown this before at Flock and at DevConf. It is a little confusing to read, but it makes perfect sense. This is using FedMessage or MessageBus, basically a measurement of Fedora account activity every week. And it measures things that are easy to attach to somebody doing an actual action. Like if you do an update to an RPM or do BODY feedback, that gets measured. Some other things like having a Fedora event doesn't necessarily get measured. Like being here at Flock doesn't get a little ding on the chart, right? So there's a lot of Fedora activity which isn't counted and even a lot of technical activity. So caveat is this doesn't represent everything. But the blue line is the number of, is basically every week the number of people who showed up that week and their account did something active. And again, it's things that are triggered by somebody doing something manually. Generally things like if you had a script running automatically, it wouldn't get counted. So it's actual activity. And that's somewhere around 400 or something every week. The solid bar here, the solid colors are people who have been active in the last year at least 13 weeks. So basically you've been active a quarter of the year, give or take, makes it solid. So the blue line is kind of everybody's involved kind of drive by kind of things, whereas the core is the solid numbers here. And the red represents people who have been, and it's not like danger, it's just the colors that happen to come out of the graph and they look nice and readable. But the red is people who have been involved at that week, they've been involved in Fedora for at least two years. Before that, the yellow is intermediate. People who had been at that time been involved, or their account, their first time it had been seen was less than two years ago, but more than one year and the green is new users. So I think that the main thing here is we're pretty solid, maybe a very slight upward trend but kind of flat. I would definitely like to see more growth but we're overall fairly healthy here. It's not like there's a drop-off. And I would definitely like to see the green and yellow be thicker lines. I'd like to see more people coming into the project. But it's fairly reasonable. We are getting new people in every year and that's good to see. But yeah, one of the things I really would like to see over the next year, the next vlog, is a lot more new faces and new people coming in and staying around. This is a new graph here and this is basically from the same data source for every release during the time of that release, the percentage of people who by which release they started in. So the dark blue at the bottom means they were four releases ago or older that they started. So for Fedora 27, the blue means they started at Fedora 23 time frame or earlier. And does that make sense? And so the green ones are people who are new in that release. And so again, this is, I did it, Brian wanted it not as percentages but it looks like just random matchsticks of different heights. So this is, I think, easier to read as percentages here. Basically, again, a lot of people, old school staying around, which obviously is good. If everybody was fleeing, that would be even worse. I'm not upset about the percentage of old school but definitely the new people coming on is kind of thin and it would be nice to see that a little better at onboarding there. But overall, I think this is pretty healthy and we've got something like, what, 20% relatively new people every release. So I think that's pretty good. This is an entire switch from that kind of measurement. This is just the Fedora magazine and I was looking at various things across the metrics that I could summarize. And I think this one is nice because this is basically page views per month for the magazine for the last couple of years, just taken directly from the WordPress stats. And I think the thing to note is, especially the black ones are the months where there was a Fedora release, which is obviously our most popular. And I think we're on track for some time in the next couple of releases to have page views a month view in the magazine, which is pretty amazing. If that were your personal blog, you could run it as a full-time job. So that's, I mean, if you wanted to put advertising on it, which we're not, we're not about that, but it's getting to be a really popular and useful resource. So thank you for everybody who works on that and has written for it. It's a really nice and useful thing with a lot of information for new and current Fedora community members and users. All right. Next comes the part of the slides where I have promised Stephen Smudgen, who's not here, unfortunately, to show dinosaurs before I show it, because this is basically Fedora mirror stats. It kind of shows us how popular each Fedora operating system release is. And this is done kind of observationally. We don't do any tracking, so we're just looking at our Fedora mirror stats. And unlike what SUSE does, we do not use a UUID, which I think SUSE is doing it right there, because this is really susceptible to network, the way the network is laid out. So if you've got NAT, so you've got basically one IP address for your entire organization, you get counted once in this. On the other hand, if you're moving your laptop around a lot during the day and going to 10 coffee shops, you will get counted 10 times, at least as long as you do RPM transactions at every coffee shop, and who doesn't. So yeah, so that's the caveat. And there's a lot of discrepancies in the number, but I think they're still useful to compare release to release. But I don't pay too much attention to the absolute number, because I don't really know what it shows. I would like, in a future version of this, to have a UUID, basically a completely anonymous, not connected to machine ID, random number that lets us just count the number of systems of each release out there, and I'll talk a little bit more about that's important. There's some graph discrepancies, which are obvious that that would help us with. But anyways, here is the release to release for the most current releases here. And you can see the Fedora 28 ascending with the bullet here. Hey, I have a graph in my speaker notes, which is not showing up. Oh well. The overall is about slightly over half are on currently supported releases. So like 30% right now of all Fedora IPs we see are 35%, I think, are on Fedora 28, and then 20 something that makes it add up to like 54 are on Fedora 27, and then everything else is previous to that. And you can see the peak there, Fedora what 24, very popular, and we haven't quite hit that with other releases. A lot of times that's simply because the new release has come out very quickly on the heels of the previous one. So basically every release goes up until a new release comes out. So we haven't explored, like if we just keep having, like when we had a release that was out for a year, that just kept going and going and going up. So I don't really know what the ceiling for any individual release is. We've never tried it. We keep coming out with new releases, which obviously we couldn't just have one release and let it coast, but just for a data experiment I'd be curious. So this is the stacked graph of all of the releases, and I kind of lumped them together mostly arbitrarily into these ages here, where basically the green is the Fedora next era. The orange, I decided we'd change colors since this is the first release with modularity enabled, and I thought, hey, that's kind of a watershed, so we'll change colors there. I've picked out a couple releases like Fedora 20 because that was such a popular release by itself. It's colored by itself. If Fedora 25 ends up flattening and staying high in the releases, I might make it be its own color. Anyways, that's really a digression into things I find interesting but are probably boring. Of the actual interesting things, we obviously have a nice upward trend overall, but I think there's some disturbing flattening at the end here, and one thing of particular note, you see a spike that goes down there. That is probably because when systems upgrade behind NAT, for one thing, you're usually counted twice that day as Fedora 27 and Fedora 28, and then also if you've got like three machines behind, or 20 or whatever behind NAT, and they were all Fedora 27 before, and then one day half of them are Fedora 27 and half are Fedora 28, suddenly the counting goes up, and then when they're all upgraded to 28, it goes back down again, so that's why we get the spike that goes back down and having UOIDs would help us untangle what exactly is going on there and get a better count of things. But yeah, I am a little bit concerned that it is kind of flattening off there at the top, so our growth is not the big upward thing we had been seeing, so I think that's room for some more evangelism and trying to get more Fedora into people's hands. There's definitely a lot of space in the market. What's that? Internet of things. Yeah, yeah, yeah. Internet of things, so there's two things. I was going to talk about that later. Peter Robinson reminds me two things. Internet of things is, for one thing, a big, big opportunity for a lot of growth where we could have, you know, or is a magnitude more devices running, which is cool. Also, a lot of Internet of Things machines are not updated. They're installed, and then they don't ever check for updates, so it is very possible that there are 10 million Fedora systems out there. You know, we see them sometimes crashed screens running Fedora 10. There's a lot of those out there that are not connected to the Internet that are updating that are actually there that are not counted as well. Is that what you wanted, Peter? So I mentioned modularity. This graph has a much... Wow, look, June 18 every time. That's awesome. That's because I didn't put the days on there. It's just the month. So you can see this is... Right. Good access labeling there. So yeah, this is over a much shorter timeframe, so the graph is flatter. This is basically the traditional repository and the... Yes, right. That is another thing. OS Tree is not counted in this at all currently, so that means Fedora Atomic Coast and Silver Blue and IoT stuff are not being counted. We need to come up with a way to count those things, especially as we go to more OS Tree-based operating systems. Anyways, the modular release... So the thing that's particularly interesting here, I think, is that we only enabled the modular repository for Fedora Server. So this kind of gives us a way to gauge what release, speaking of the UUID tracking, we don't actually have a way to see what addition people have installed or are running in the current system. So that's another little tiny bit of depersonalized data that I would like us to actually track this separately so we can kind of see which of our different efforts are resonating. So this, we actually created an experiment by only enabling that in Fedora Server. So even if you're installing a Fedora Minimal install that's not Fedora Server or Fedora Cloud and using it in a server case, that's not counted and we know a lot of people are, but this is like how many people are actually running and updating to Fedora Server Edition? So to cut to the chase, it's about 7% of systems seem to be checking into the modular repository, which, you know, that's probably a little disappointing to Steve, but it's, I think it generally fits pretty well with our earlier estimate that about 20% of Fedora installations are using some server context because obviously not everybody's running Fedora Server and not every Fedora Server system is updated right away because by nature, server users tend to be a little more conservative with updates. So yeah, I think it's probably a good guess. 20 to 30% of people using Fedora in server context, which is interesting because one of the things if you go and look on Reddit, people assume that Fedora is not an operating system to be used in a server context. They go to WNR CentOS or, you know, REL, which is obviously for many, many workloads an awesome choice, but Fedora is also very suitable for a lot of server use and it's an area where we've got a lot of room for evangelism and growth kind of based on what people are doing with it already. We could kind of tell that story better and I think get more people using Fedora in that way. Yeah, containers as well. Particularly good in containers, I think. This is a breakdown by system architecture. So we can see x86-64 very dominant and 32-bit very flat with some interesting drops. I don't know why it grew back up there. Again, there's dinosaurs in this data, so who knows what caused that to happen. But yeah, that big drop is particularly interesting. Like some large installation of Fedora decided no more 32-bit or no more Fedora at some point and that is interesting because that actually happened a couple months after we had a big discussion on the mailing list about is 32-bit Fedora viable and it was looking like no, and it looks like maybe somebody was following our discussion and decided that they needed to find a different 32-bit solution. And you can see ARM at the bottom here and I didn't graph Power or S390, not because they're not important, but because they would just be lines at the very bottom of the graph. They don't really hit. And this is the same thing by percentage here. I think probably ARM is underrepresented because of what I was saying about things not necessarily checking in for updates. But there's a lot of room for growth there. But ARM still hasn't replaced the 32-bit Intel architecture either, so make of that what you will. Okay, so I talked about Fedora as a big umbrella project and one of the things that I think is pretty important to Fedora that we don't always give enough attention to is Fedora Apple. Extra packages for Enterprise Linux, that's Fedora packages that are built to run on Red Enterprise Linux or CentOS or Amazon Linux or one of the other many rail derivatives out there. This is incredibly popular. Actually, I probably did these slides in the wrong order. Here, this is the Fedora operating system release and the red line is the Apple usage over time. And so it looks almost flat on this axis here, which is kind of depressing. I mean, if you look right there, you could Fedora operating system growth, but compared to the popularity explosion of Apple, it kind of pales. So even if you don't care about that red line, your impact as a Fedora project member on the world is probably largest in what happens in that red line. So it's worth thinking about and caring about a little bit, even if it's not sort of the immediate thing you're interested in. And I think it's something we should make sure we're... It's something we're getting a little bit of short shrift in the project right now. I think it's something we could invest a little more in. I'll talk about that a little more in a little bit. Yeah, back to this one. This is broken down by the Enterprise Linux release that they run on. And I think this is interesting for kind of two things. First of all, you know, REL7 has been out for, like, I don't know, eternity now. And it is just now almost taking over six. So Enterprise Linux users, very conservative. That's kind of why there's lots of room for Fedora to do what we do, because we move fast. Enterprise Linux doesn't. And there's a lot of demand for Fedora-like content on Enterprise Linux. So I think that's a really interesting thing. And the other thing I want to point out is that spike there. And this is actually a moving average. So that spike is actually flattened from what it is. That is Spectre Meltdown, one day. Yeah, it actually doubles. One day, for some reason, all the assistant men's in the world decided today is a day we're going to update our systems. And it causes very visible spike. And I think that's just amazing. Speaking of spikes, the other thing that I think is, back in this graph here, there's these dips here. That's Christmas every year. It's the holiday shutdown for Red Hat, and everybody's on vacation. Some stuff happens, but not nearly as much. And it's always fun to see the real world hitting your data, because stuff happens. Yeah. All right. All right, so enough with the charts. So the rest of the talk here, some things that are happening in the project and at Flock this year that have kind of caught my attention. It's not comprehensive. There are probably a lot of things that I care a lot about that I didn't put on the slide. So if you're not in the slide, don't be offended. In fact, everything at Flock here, I'm very excited. Every single talk, there's not one that I would not want to be at if I could be in multiple places. So there's a lot of great things going on. But here's some things I wanted to call out specifically. Yeah, first release with modularity enabled. This has been a big project over the years. So yeah, congratulations to everybody working on that. And Langdon's going to update us more on that. I think, but the reason I've got a slide here is this is something I think is important to us over the next year. We have this technology now. We've got something like a dozen different modules, which is not very many. This is something that enables us to do a lot of interesting things with different versions, different compile options, like alternate versions of software in a way that we haven't necessarily done before. So for example, if KDE needs a different version of a library, or a whole different like stack than different desktop offering needs, there's kind of a way to do that. And people can say, OK, I'm going to maintain this separately. Obviously it's nice if things can be coordinated, but software moves so fast. And there's all these different moving pieces. And out in the world, the basic problem that everybody who has an operating system faces is welcome. The too slow and too fast problem, where basically you want some parts of the operating system to move at a speed that you care about. And for some people, enterprise users, I want that to be very stable and never change for a lot of other people. I want the latest version of that always. And you want the other part of the operating system to either update itself or stay the same, but you don't care about it. But the problem is that's different for every user. Everybody's got a thing that they care about and a thing that they wish would just not bother. And it's pretty hard to make one operating system that appeals to all of those cases. Yet it's basically our charter to try and make an operating system that appeals to everybody. So how do we do that? Modularity and these kind of things enable us to make deliverables that have different use cases and follow different things. And we needed a technology which would let us package things up so we could solve different problems and not necessarily have every package be fit everybody's needs. So we're going to need a lot more modules. That's basically what this slide is about. So it was five years and two days ago when I gave this Fedora rings talk at Flock. And we still don't have rings in Fedora, although we're kind of getting that way. Basically the idea is, again, Fedora is a gigantic collection of packages and it is hard to treat them all equally and there's a big cost to trying making everything be one unified everything. There's some very nice things that come from having one gigantic repository with every package in it and I don't discount that. But it also comes with a pretty big cost for users and makes it hard to deliver things that fit exactly into use cases. So there's this proposal. Let's split things up by policy and have different rings. This is an updated version of that same thing that from a talk that I never gave, I think it's even more confusing than the first one, but Paul Fields is giving a talk later on kind of how a refined version of this and what we can do in Fedora and the advantages it'll bring us. I encourage if this is interesting and or scary to you to attend that talk. I talked about Apple and its importance. I think especially again with modularity this gives us some interesting abilities to make a better Apple. One of the things that modules have is this idea called stream expansion and that's a fancy way of saying you can make a module and that will just automatically build across multiple releases. This is cool because if right now you're maintaining basically the same thing on Fedora 27, 26, 25 or whatever and then on the Apple versions you might have basically one package that you're maintaining six different ways. This can let you say maybe I don't want to have six different versions that are kind of tied to what base they go on but I want to have them conceptually. I have one that is fast moving and one that's slow, one that conservative updates and one that updates to the latest version and basically you have two module streams like that and then you can build that same module stream for Fedora releases and for Enterprise Linux basically then letting the user choose independent of what base OS they want whether they want that package to move quickly or more conservatively and that then lets us also deliver solutions that use those packages in that way. So the point of that is we can have an Apple that is basically based on modularity at least once modularity tech gets into CentOS that's why we've got a CentOS thing over here and then with that basically it'll be very low cost as a Fedora package or to also have an Apple build of your package which again back to the red line there's a lot of demand for that and it's very high impact. So if you're doing this because you want to help users and you like to make sure that the software you work on gets out there and used by people this is an easy way to do that. So there's that. And the other reason CentOS is up here is because CentOS is an open source project in the Red Hat family we've got a lot of people who use both we've had a lot of people who work on both and we have not done a very good job of collaborating between the distros it's time to start doing more collaboration because we're leaving a lot of value on the floor I guess we're basically doing a lot of work twice for no reason and there are a lot of things we could take advantage of that we aren't by not collaborating so we should work together more and Jim Perrin from CentOS and I have a talk kind of about some very specific collaboration in DistGet that is coming up and I also encourage you to attend that although I think we're running against other interesting talks so you know if you care about collaboration another very exciting thing is Fedora CoreOS this is basically an operating system that's meant to be kind of an operating system as a service a very small operating system that will that's sort of specifically made or not sort of definitely specifically made for running your workloads in containers and having the operating system itself updated underneath that without disturbing things automatically updated and it is very similar to the ideas we were looking at with Fedora Atomic host and so Red Hat acquired CoreOS and the CoreOS project is being merged with Atomic host and we're going to have a Fedora CoreOS release with a lot of the technology from both and a lot of the ideas from both into an awesome best of the world container operating system which I am very excited for and I talked earlier about exploring new ideas there are a lot of new ideas that come in from that project that I want to make sure we are welcoming to in Fedora with automated QA, automated release and also automated updates in that system so I think that's very exciting and welcome to everybody from CoreOS to Fedora this is kind of a disturbing icon but that's the best so this is the MindShare HackFest badge MindShare is like I said, yeah right our new committee for basically the non-engineering sides of the project which are as you can see from the earlier graph if you remember a pretty significant part like we kind of tend to think of this as an engineering project but the things around community, around design around documentation and the ambassadors like are equally important to the engineering because if we don't have that the engineering doesn't go to anybody and isn't used by anybody and isn't useful for anybody so this is important we had a MindShare HackFest earlier this year I think it was this year, it feels a long time ago now and one of the things that came out of that is a plan for revitalizing the Fedora ambassadors basically are on the ground talk to people, sell Fedora group and I think that's very exciting there's some more MindShare HackFest stuff going on here at Flock and if you're interested in leadership in the non-technical aspects of Fedora if you're interested in helping spread Fedora helping those numbers go up I really encourage you to get involved in that kind of specifically I was actually talking to some of the product management people at Red Hat and one of the things they're very interested in is seeing Fedora grow into the university student space which we kind of have been less active in than I would have liked to be over the last I don't know time we've been the Fedora project leader I guess I used to work at a university but now I'm old so I'm kind of out of touch for what's going on at that level but this is something this is actually a project plan we came up with a while ago and didn't really implement on if you are a student or if you feel connected to you know academic world this is something that really is looking for leadership and work on come talk to me, talk to some of the MindShare committee let's see what we can do to help Fedora grow at universities and among students the next thing Fedora Silver Blue this is kind of a I talked earlier about OS Tree based technologies that's the sort of get for your operating system that Atomic Host is based on rather than having your operating system be a collection of RPMs that you manage directly with DNF or it's basically an image based operating system that you can update as a whole to one version and also go back to if there's a problem with the updates for a lot of people who are hackers and tinkerers this kind of thing might not necessarily be the easiest for us but this image based operating system is for deploying Fedora at big scale or deploying it to a lot of people who don't necessarily want to manage all the ins and outs of their operating system an image based operating system makes life a lot easier and it makes things a lot easier for QA because with RPM as it is right now every operating system every installation is an individual snowflake with the huge the combinatorial possibilities how you can put together Fedora as it is now are immense with this everybody can basically have a package set on their system and that makes it much easier to QA that also means that for actually running things that you care about like applications and so on we need new technologies for doing that containers are obviously one solution then that's one of the ways we want to have it's so that when you run and do your actual work you're working in a container even though your operating system is this kind of image based thing and also related to this is flat packs which is basically a container like technology for desktop applications snappy is another competing one so both of these things are things we're working on in Fedora and there's an effort to take existing RPMs that we've already made we do in Fedora a very good job of packaging up basically all of the open source graphical applications that are of high quality packaged up and maintained in Fedora in pretty good state which is good so thank you everybody and congratulations so one of the arguments for Flat Pack is that it'll kind of expand the ecosystem of applications I hope that's true but also we in Fedora do a really good job of providing these things to people and so I think that Fedora makes a good source of quality trusted applications so Owen Taylor is working on a project that automatically converts the RPMs we're already making into flat packs so I think that'll be kind of neat and it's a way also to increase Fedora's impact beyond the base operating system that we have so if someone else is working on some desktop Linux but their project is kind of focused on just the desktop Fedora has this wider focus and we package up all these applications so we become a good source of trusted high quality applications for other Linux distributions that's a pretty good way for Fedora to increase our impact and also gateway to the Fedora community for people who might have not start with the Fedora install CD so I think that's exciting here's another thing we've got Ben Cotton here our new Fedora program manager I just wanted to introduce him and we had this last release the first time we had a Fedora release that hit our target schedule which was awesome and Ben will be doing that for every Ben will now make sure that happens for every release from now on so then this I don't know this is a Penrose triangle it's impossible I guess and so there's a talk that Josh and Brendan who work on and on Fedora as well will be giving about the Fedora rel CentOS ecosystem and how it all works together this is kind of the evolution of the what Red Hat once talk that we've had people from Red Hat leadership give at Flock previously this is a little bit less of a what Red Hat once and a what Red Hat is interested in doing and is committed to doing and will be doing we've scheduled that as a plenary talk because I think that will be interesting to everybody that's coming up this afternoon and then immediately following this we're going to have some talks about the Fedora additions and our objectives the objectives are a Fedora council thing where we have something that spans like a 12 to 18 month timeframe with a measurable goal and we have a leadership position on the council to make sure that goal happens if you have something big you want to happen in Fedora coming to the council with an objective proposal is a good way to do that so we have some people who have done that who will be presenting on the state of their objectives and then we'll also have overviews from our primary Fedora additions and what's going on there also we'll have an awesome keynote about if you see something that you don't like how to make it better or if you have something you want to do how to make it better in general in open source and I saw a version of this talk at Tevconf and I thought this is something that we really could benefit from in Fedora community so wake up early tomorrow come to that keynote as well but up next the addition talks how are we doing for time here I wasn't paying attention we're fine okay so did you have them in order do you want to introduce you can introduce Matias Matias oh bad here you're a desktop person why doesn't this present there's a reason why it was best when I did present it doesn't go to the secondary screen do I just need to drag it over yeah quick right there we go alright I'll let you alright good morning everybody I'm Matias and I'll give a quick update on the workstation edition and what's going on there what has been happening in the last year and what we're looking to do in the next year and it's hard to go up after Matthew because I have no graphs at all but I'll try to make it interesting anyway so moving to the highlights one thing I wanted to mention here is battery life it's something that we have been interested in improving for a long time and in the last year Hans de Goethe has been doing some work in this area he's been changing a lot of things from hard disk power management to USB auto suspend if you want to know all the things he has done you can look at the change page that I linked up there where it's all described in detail and it's a very useful project I think not just because it lets us all get more work done on airplanes but it also gets us some quite nice headlines and visibility that I've listed to here so these things actually get noticed in the public and help us spread fit our work yes, keeping up with hardware this is an ongoing project of course here's one example that we've tackled in the last year Christian Kellner and the desktop team has been working on adding support for Thunderbolt devices to Fedora and that's a good example of a project that shows the scale of plumbing work that we often have to do just before we can put a new icon on the top panel or like expose something in the control center this fixing a new user space demon in this case it's Bolt that is the user space demon that manages the Thunderbolt security levels to all these things before we can actually put some UI on top and make it all look easy and just work but all of this is happening and landing in Fedora 29 I believe and there's again a change page that you can look at if you want to know more about this moving on to updates that's actually one thing I was going to mention for Thunderbolt as well that's another thing that we had to enable for Thunderbolt devices making sure that we can update their firmware and this AVFS is the project that Richard Hughes and my team is running for making sure that firmware updates solve problem on Linux and I think that's also a good project in terms of visibility a lot of the big names are present there now if you haven't had a look at the vendor list there's this website you should go and have a look there there's everybody from Dell, HP, Intel, Wacom they are all there and I'll support this and just recently I think some people were happy to notice that their ThinkPads actually get firmware updates now and yes moving on to more hardware support what do I have here LibInput is the low level component that manages input devices it's shared between Valiant and X it's a good thing and it has been around for quite a while so it's not really something that we started in the last year but I wanted to mention it here because Peter Hatara who works on this has now made a Cobra available with the very latest Git version of LibInput that you can enable and help us test and improve LibInput another thing I have here is the virtual box it's guest editions that we now install by default I think since Fedora 28 to make Fedora work nicely in virtual box and over here on the left side the screen chart shows some of the work we've done on improving the handling of multiple monitors and monitors in general in the desktop you can see like the high DPI scaling now supports not just once and twice but also intermediate levels of scaling things like that Pipefire is the next thing the screen chart here shows women and question testing Pipefire using a webcam I think there was actually testing like some effects or something and Pipefire is a new project that aims to do for video what parts audio did for audio I guess you can say it's basically managing video streams inside the kernel and efficiently and the short-term goals for Pipefire include replacing parts audio because if you can manage video streams then audio is not that hard and also supporting screen casting and desktop remoting under Wayland we'll use Pipefire for that as well and the longer-term goals include also supporting Jack applications and professional audio and finally having a unified base for media essentially and yeah that's very young we started that towards the end of last year and I think there's a hackfest this fall if you're really interested in contributing in this area you could look that up and work on that and if you're in general interested in more information about Pipefire you can also follow that link up there to the blog post that question wrote about it yes this is a quick run through some of the other things that we've done in the past year there's a lot of ongoing improvements in the desktop that just like come in from upstream some of them we are directly involved in some are just like stuff that happens in the community and I'm not going to read that whole list maybe just calling out one thing here you can actually run Firefox natively on Wayland now there's a separate package for that called Firefox-Wayland I believe maybe worth trying out and the screenshot here shows support for emoji that's of course everybody's favorite right you already saw my thunder here more or less Silverblue is a new initiative that we started this year at Defconf it's basically about making Fedora Atomic Workstation ready for prime time and expose it to a wider audience and I should take a step back and say that we had an atomic variant of the workstation for a few years now I think since Fedora 25 it was hosted inside Project Atomic and was a little under the radar I guess a little known few people tried it and liked it but we didn't really advertise it and Silverblue is about changing that and basically making it ready for prime time and using it to spread Fedora wider and yeah what does it mean to make it ready for prime time I've listed a few of the things that we are working on we need first-class support for Flatback and RBM OS 3 in Gloom Software to have a working solution for updating the software on Silverblue because Silverblue is using OS 3 as Matthew already said and that also means including support for package layering and support for automatic updates and it also means better support for containers in the desktop Debashir and my team is looking at writing something like a toolbox container that basically is the Fedora in a box that you can just use on an atomic system because you cannot just install an RPM if you miss something so you go into your container and you do it there we're hoping to land some preview of that in Fedora 29 and then we also look at integrating with the terminal so you can actually get some context information about whether you're inside a container or outside or things like that what else did I want to say here yes the name certainly has I've had a variety of reactions to the name it's not everybody's favorite but it's a new name and we also have a new logo and a new website this is just because we want to reach a wider audience here in Fedora as Matthew already said the graph look kind of flat at the end so all of these new things are means to that end names are hard so I think it's one of the things that sounds kind of funny at first but after you start using the profile it becomes very natural that's true with any kind of name for things right and another thing that we are trying to reach new audiences here is we are going away from using a mailing list and so if you're trying to use discourse like a web forum essentially I think Matthew is going to have a discussion about that is it today or look tomorrow I don't know okay so it's trying new things and trying to reach new audiences it's sometimes hard for old-timers like me but I think it's a good thing to do every now and then and this is my last slide a quick view of what's coming in the next year what we're currently working on there's things like a smooth boot Peter Jones maybe here somewhere is working on that and we're working on better support for the NVIDIA driver for X-Wayland and for hybrid graphics Vail and screen sharing and remoting I already mentioned that we are using pipe wire for that that's hopefully going to be supported in Fedora 29 game mode is about battery life again or basically the opposite of saving battery life it's making it so that you can run your games at full blast yeah Matthew already mentioned that Owen is working on building flat packs inside Fedora from RPMs I think Owen talking about that on Friday so you can go there if you want to hear more about that yes we're working on bringing automatic updates and full-up PMOS 3 support in Chrome software that should mostly land in Fedora 29 maybe some of it will go into Fedora 30 and yeah but I just mentioned we work on better support for containers and a toolbox container mainly for silver blue and with that we have a microphone to who's next in that case I'll just I made you a slide where did it go good morning so for those of you who don't know me I'm Steven Gallagher I have more or less been I don't like to say I lead the server edition but I think I more just make sure everybody shows up to the meetings because I really prefer that it be a community kind of a round table situation where everybody who shows up has an equal voice so I mostly do this because I'm the least squeamish about public speaking of the members so this last year has been a really interesting one in the of an interesting times sort of way for the Fedora server edition we have had in the same 12 month period a fairly some fairly large disasters followed by one of our I think what is probably server editions biggest success to date which has then led us into an identity crisis so let me see where to start when I stood up when I stood up and gave my state of the server union last year at Flock and I'm going to try not to cover too much of what I'm going to cover in that talk later today or tomorrow I was pitching that the server edition was going full in on modularity which Matthew has talked about a little bit and we did and we put Red Hat put a lot of resources towards this and we were building this really neat idea where everything in the system was a module you could move whichever piece of your system you wanted at any given moment and it turned out that that really was both not possible and really hard to get people to try out in which order sorry in which order it's not possible how could they try it well ok so the question was if it's not possible how could they try it you're catching on really there were a whole lot of bootstrapping problems so we were faced with a decision we got this was October or November of last year so a few months after Flock and we were either going to toss it and scrap the plan or we were going to try something new and several people basically had an international hack fest for about two weeks and came up with a way to reuse as much as possible of what we had already done and we came up with a new approach and we pitched it to Red Hat and they decided to keep moving forward with it and we got it into Fedora 28 and we delivered it as part of the default of Fedora 28 server edition and it actually worked for the most part we had a few bugs with package kit but we actually were able to prove that it is possible to build a modular system we just couldn't start we couldn't start from the ground up we had to take the existing system and start breaking off chunks until eventually we hope to get to that final place but we were approaching it from the wrong direction and why are we doing this Matthew also talked a little bit about that he gave most of my talk so this is a little tricky here it's about solving the too fast and too slow problem Matthew alluded to the Venn diagram of that I wanted to give you a visual and I was scrambling a little while ago to see if I could find one slide to put up here so I'm going to have to ask you to use your imaginations for this one think of any three or four year old child you've ever seen playing with an ink stamp that is what the Venn diagram of people I want this to be stable and I want this to be fast it looks like just punch a bunch of stamps on the page and that's what it looks like there is absolutely no two customers or users for whom those sets of packages are going to be the same so that's why we did we did this modularity thing now where did that lead us it led us to a place where we realized we've actually we've accomplished this there are a few bugs waiting to be worked out but once that is finished we don't actually know where to go next we've built something that once we can get it into Apple, once we can get it to work on an enterprise Linux system we now have to figure out what does server edition do because realistically we know that what people are going to do is they're going to prefer to take an Apple base and to take an enterprise Linux base and use modules in Apple there's still and there's still value to Red Hat at least for the server edition to make sure that that continues to be a platform where we do integration testing you know for things that will eventually become an enterprise Linux itself as well but we don't know we're pretty confident that that isn't going to gather a huge amount of community interest and we're not going to have people coming out of the woodwork to say oh I really want to work on that driver oh I really really want to want to get 100th of a percent of performance out of the out of the kernel today people are doing that and they're doing that outside of us the server edition needs a target and a goal needs to have a new purpose and that's going to spoil my entire talk for tomorrow but that's basically going to be the gist of it we need to find our new way forward and I encourage I beg in fact any of you who have any interest in this whatsoever to come to that talk it's going to be about 10 minutes of talk that will be very similar to what you just heard and then a discussion I really want to hear from anybody that has ideas so we are actually going to break a little early for the coffee break because of the way the speakers are organized so that should work out really well it is currently 10 18 so why don't we try to resume here at about 1045 so we'll give those speakers some time on the other end if we need it before lunch the coffee break today is sponsored by ARM and it is set up in the lobby and I encourage you to enjoy it there are regrettably no RPM transactions available with your coffee