 All right. Hi, guys. So the whole point of this lab is it's going to give you a nice overview of raw security technologies. What is a requirement to participate is a laptop with SSH client and a web browser, of course. Without that, you're literally just going to be just reading the laptops. So I'm going to talk in an introduced environment like for five, 10 minutes, and then you guys can jump around. And I'll go through that. You guys can jump around, do whatever you want in terms of the lab extras. There's about 11 security technologies that you're going to see, 10 or 11. I can't remember if it was 10 or 11. And you guys can jump around and do whatever lab technology that you would like to do. So first of all, a quick introduction. Do you guys want to go first? I'm working on SLM's technology at RedHead and maintaining SLM's policies for RedHead to president and president. Hello. My name is Daniel Subecek. I work in a special project scheme. I work here on the SB Guard, 8 and NBDM on security. So my name is Lucy Kerner. I'm working for security across the RedHead portfolio from a corporate standpoint in terms of, OK, what is our internal messaging for security? What do we want to say about security across the entire RedHead portfolio in terms of as a differentiator for RedHead? OK, then you can say, well, you're not a security company. You don't sell security products. What does that even mean? And the idea is that we have security built into our technologies from RedHead to PriceLinux and also all the way up the stack to help developers build security into their application, management portfolio to automate security and compliance, things like that. So that's the idea. So I create the messaging. Also, technical content for labs like this working with engineering and highlighting all the things that our engineers create and showing the security benefits of our technologies. That's at a high level what I do. I spent three years actually in the field as a cloud solutions architect for US public sector. So I helped our public sector customers in the US use our cloud products, OpenShift, Ansible, Satellite, et cetera. In a very, of course, US public sector, they are very sensitive to security in terms of the very strict security requirements. And I helped create proof of concepts for those kinds of use cases. So that's at a high level what I do. So I'm going to jump to this section. So the lab doc is here. Everyone must do lab zero. That's setup steps. After that, you can do whatever lab you want to do. Everybody has their own unique lab environment. Every lab exercise is its own VM, which means that they will not override each other. So the idea here is do lab zero. After that, if you do whatever other lab exercises you want, this is self-paced. The lab environment is going to be on. I have it automatically shut down and die at 6 PM tonight. If you want to take this, not here, but go somewhere else and do it, I don't care. It's all cloud hosted. And that's it. So after that, the lab instructions that you'll see is very, very detailed. So just go at your own pace to whatever lab exercise you want to do, and then we're here to answer your questions. Thank you. Go to this link, go to lab zero. And then after, everybody do lab zero. Then after that, you do whatever other lab you want to do. And then if you have questions, of course, raise your hand. It's right there. But then it says that they have to request lab. Oh, OK. Yeah, so that was just an example. You just use that exact thing. Yep. It's up to you. So it's like whatever interests you. You don't have to go in any order. They're all independent from each other. To do the whole thing, I don't know, maybe three hours. It depends on the person too. Some people are low linux much better than others, right? So it depends. The goal was not necessarily for you to finish all right now. The goal is kind of just introduce you to well-secured technologies as a whole. Someone who needed a laptop is not here. Let me. Oh, you did already. OK, OK. Who has the longest arm? You, right? You have the longest arm. Sorry, that might be a stupid question. It's OK. But who is paying us to press reset to work stage? Oh, don't do that. Don't do that unless you want a new GUID. Only if you wanted to get. So you don't need to do that right now. Yeah, that's only if you wanted to get another GUID. You have to restart to reset it. But you don't need to do that for this. You just need one GUID. That's it. Yes, sorry, yeah. OK. No, this one, middle one. Get the audience with us. Already done for you? This is just FYI. Already done for you. FYI. So you just start here. Right here. Yeah, you do. Yeah, see? Lab user. You see GUID here? The GUID you got from lab zero. Insert that here. OK. No, I put password up there. Yep. OK, uh-huh. Which one do you want to do? OK. All right. Instead of this, put in your GUID. OK. Can I see the GUID? This is your borrowed laptop or your laptop? It says SSH. Oh, you have it twice. That's why. It's OK. Try it again. That's Alex's laptop? Must be some sort of fat fingering. Fat finger, yeah. That's the secret way of spelling red hat. Oh, you even photo bombed our picture. That's funny. Yeah? Do you have a picture somewhere? Oh, no. You got red. No, this is not the issue. Oh, photo bombed. It was good. I like that. OK, I tweeted it. You could check. You could check. I tweeted it. That's a good photo bombed. These are, yeah, I started at 75. This means that it's being used by someone. Like, this one is not. This one is. Yeah. It's OK. There's a shitty connection. Oh, no. The face is, like, shutting down several times. Great. You have three times. Yeah, and he's in disconnection. It's this Wi-Fi or dependent on it. It's very slow. But look, he got an OK to ask at work bench. Yeah, he's also got it, but it's very slow. Oh, it's slow? Yeah. We don't have control over the internet. Yeah, the internet is just dependent on the internet here. Yeah. Yeah, comprehensive, right? Yes. Did you like the fact that it was comprehensive, or what did you like? Just some things you didn't know? Prepared. Oh, I see. Thank you. 6 PM is when I automatically shut it down. Yeah. It's just because it's cloud hosted, so I have to pay every hour. Right, so. But you don't even have to be here. No problem. What? What infrastructure? It's actually behind the scenes. It's cloud hosted. So this environment changes from Amazon Azure. It depends when I provision it. In this environment that I have it in, it changes depending on where the workload is less. But it's all cloud hosted. That's why you need internet. Yeah. But you don't have to be here. You just need internet. Yeah. After 6 PM, today, or? If you really want it, I can. I can. But I mean, that's only if you will really use it. Because otherwise, we're, you know. But I will if you want. I would like to really use it. I would like to recreate the environment. OK. OK. I can't leave it on indefinitely. But I can, if you want it for an extra, I don't know. What time were you thinking? Like, make. Oh, no, I can't do that. Because we pay hourly for this, so. Like, I mean, for the GitHub? Oh, no, no. GitHub is public. Yeah, of course. GitHub is public. I'm talking about the lab environment itself. Yeah. It's going to shut off at 6. Yeah. The GitHub is always there. Yeah. Tuesday. Tuesday from Prague at, like, 2 PM. Yeah, I'll come in first thing. Oh, oh, good me. Good. So, like, I don't know how to, the, the, into the surreal security, I put this, I don't know what time, and I don't know who I should invite. That's my problem. And I asked Peter, but I think he's busy. So I don't know. OK, so. So here, see, to discuss, well, here we, oh, shit. We want to showcase that summit, so the lab, and discuss the Royal Security Deck update, OK? So what I mean by that is, I have a, I have a slide deck that are, that, that, that accompanies this. Sorry, what? OK, like, I have a slide deck. Royal Security Technologies. Yeah, OK, but it's kind of, it's some talk, or? Yeah, I have a talk version of this. So I was just trying to get some feedback. But the more important thing is the lab, right now. That's the more short term. These are the things you'll be hearing from inside. No. But what I do is, like, I can show you from here. So, like, our, our, our sales team and our partners, they can access security content from this product page, OK? So this is the page I created. And this was for all of our products. So for example, I have one for Royal Security Technologies. OK? And I have this one that I created. And then I also have, oh, here it is. But I want to update this. Like, and I created this diagram. So this is like, goes through all the basic, like, like, in slide, like, goes through all of these things at a high level, right? So it goes through it at a high level. And then I also, but I want to update this, huh? So it's very, very basic. See? It's very basic. But I need, I want to update it. And I want to, I created this diagram recently. It's part of the overall security deck. But this later, let's show me the rotation list. OK, where did my calendar go again? OK, I got to find where the, do you remember where I put it? Oh, I see. Yeah, it was to here. OK, it's OK. See? Let me just show you this with the Royal One. This is the all, like, entire Red Hat portfolio. This is the one I'm creating right now. But for example, for the Royal One, I try to do this defense in depth picture. Here's, instead of these box, like you saw earlier, this is my next idea. OK, for polish as an awareness, you can use this. For physical security, this perimeter security, internal network, host security, application, right? Maybe applications should also be in host, I don't know. But it could also be application, data, right? I was just kind of trying to get some feedback on things like this. That's the point, OK? Can you show me the imitation? These are all your kind. OK, so now you know the point of what I'm trying to do in the meeting, OK? I don't know who else to see work on. Oh, I see Linux product owner, yes. You added? I'm going to talk to the guy who just walked in. I think you just walked in. So basically, oh, you already got it. You understand? OK, OK, OK, thank you. He's the one from yesterday. The one right next. I said, who are you? Yeah. He's also a sealer next. OK. What other product do you think that if we say RHEL security, that's too much probably? It's too much. Too much, right? This one? That's everybody, right? That's not good, yeah. I don't know who else. Like who's the groups? Who are the team leads? Who are the team leads in RHEL security? Who are the team leads? Niko's? Who else? OK. Do you need Simo? Simo? But if he's there, I don't know if we also need Simo. Do you? You guys have this mojo page with all the RHEL security? I don't know. It is to do all the same, to do all the same special projects. OK. So add them. I don't know who they are other than Marik. OK. So then Marik Kaichman is another one. He's already there, OK? Then me, it's the link already there. Make sure. Let me make sure on this mojo page to find it. No, there it is. OK, here we go. OK, so let's see. Martin? OK, so this one is Marik. OK, so let's see. Daniek, OK. You mean him? No. Who's this? No, I know this, but I mean, who should be the? Oh, you and Zenek. OK. This one, Daniel's already there. Paul Morris, this is now, should he be there? So we've got everyone. What about the subsystem team? What about these guys? I should add them too. Let me ask, do I need him too? I don't think so. I don't know, but I don't think so. He wanted me to add him. Last time he sent an email, he said he wanted to collaborate, so I'll add him. Mark's leaving, so let's see. I don't, probably not him right now. OK. What room? What room should I do? Sorry. Which room? OK, so 14, and which room should I do? Let's sum up some in graph four. You're asking for the room, but that's for the people. That's more and below three. So we will have extra chairs. OK. And also, for example, we'll see. Oh, there's, we didn't get. This is four, not two, and we'll do this. OK, OK, OK. So I believe Peter won't attack Peter. OK, OK. The Demetri sometimes does, we'll see. He's busy. Oh, no, you didn't put the room there. Yeah, good. You didn't fill it. It's already here. Oh, OK. Yeah, let me make sure. OK, hold on. Let me put it in here. Let me copy here. This is in TPC. People will know. They will understand. You don't need to recognize it. Oh, OK. Because it's room four. Time do you think I should do? I don't know. I think Nicos has a team meeting or something, right? Let's see. What is this, crypto meeting? Well, no, it's busy. What do you think is better? Should I do it later in the afternoon? This looks OK. OK. This is much better because it's certainly putting the video here again. So people will be trying maybe tomorrow late. Late, OK. So three to five is OK? Yes. OK. All right, then I'll send. You're going to go? You want me to go, too, or are you going to go? Right, too. I can. I guess you are up. Is the internet problem? The machine is there. There is an internal error for HTTP. But for another, yeah, yeah, maybe. And that worst case, you could, if it's active, buddy? Oh, OK. I was just going to say you could get another good, because I have plenty, but yeah. Yeah. I mean, we have a lot. The directions you can recreate it. But yeah, the lab environment, I'm shutting it down at six. Yeah. But if you really wanted to let me leave it on, I can, but not for forever, right? Like an hour, maybe extra three hours, whatever. Yeah. What's the context? Yeah. We can see where we are. We can see where we are. Yes, but it's on the other side. It's disabled. It's disabled. There is a moving. It's disabled. And the context is disabled. Now, we found a match with this match. Thanks. But what do you mean by that? But this states that you use that search to HTTP. These are your search committee members. Yes. So we have limited tools. Yeah. You chose the NSE switch. The main thing is the NSE switch. The main capacity. You know, attributes. Inside the attribute, the domain. And this domain is a lot of stuff. It's better. There's no way I could fit there, though. The machine, for my young wife, it tells you to change the password. It tells you to change the password. To IDM1. OK, I'll keep a note of that. Thank you. This is all the slide I have. It's just the lab dock. Did I say that in the lab dock? I don't remember off the top of my head. Oh, you haven't. I don't know off the top of my head. Let me find out. It's so hot in here. Unbearable, huh? There's like a hole there. So the birds might fly in, I think. I think there's a chance that the birds will fly in because there's a hole there. Are you hot? Are you guys hot? This may be the first time, then. Is it better for you guys with this open? So far it's OK, yeah. I took it off because it was too hot. OK, so if the bird flies in, then it's not my fault. But that's why they would come inside because it's cold outside and no bird came in. Not that it was reported in the organization telecom. Did you ask? Is it this one who asked the question about the lab environment? OK, so I can't leave it on indefinitely, but I can leave it on. How long do you want to... It turns off automatically at six. Oh. Mm-hmm. Mm-hmm. The lab dog, of course, doesn't... It's on GitHub, so any time, yeah. Yeah. Oh, so that course is... You're talking about the one for Red Hat Training, right? So this is not a Red Hat Training course. This was actually delivered by us at Summit. It was a Red Hat Summit, and Red Hat Summit is not Red Hat Training courses. Yeah, I know, yeah. I know about the one. You mean the new one, right? Yeah. The new one? Yeah, I know. They came out like the Linux and Cloud, that one. What they did is they took some of the content here. So you may see they took some of this, and then they added to it. Yeah. But that one is like we sell that class. This is what... This is mainly for like events like this, or like it was delivered at Summit, Red Hat Summit, things like that, yeah. Cool. You might see some similarities because they took some of this content to build that class. Yeah. Mm-hmm. Mm-hmm. Were you hot? Mm-hmm. You was hot, yeah. I'm just scared that the birds might come in, but maybe they're not outside. I don't know. Huh? No, because they'll shit everywhere. If the sparrows come in, I'll take your picture with them, okay? Okay. Then you can put it on your Twitter. I was looking for the class. I think it's something to do with object class. I haven't played around with that one, but I'm looking. The doc. That's fine. Mm-hmm. Mm-hmm. Which... What doesn't work? What specifically? Are you sure that you... Mm-hmm. You left off at 8.5 or 8.4? Where did you finish? I finished the 8.4. And you're right now trying to do 8.4? I'm trying to do 8.5. But it doesn't work. Yeah. Let me look. Log in what? Fail? You can't do this part. Yeah. Okay. Let me try. And then it worked? Okay. Okay. Yeah, because I didn't think you had to do anything special from what I'm looking at. Yeah. How long? Morning. You will work on it? If people always say it, and then they're like, ah, I'm just going to go to the pub. I don't want to do it. Okay. Yeah, sure. So tell me what your goo it is. The code. What's your code? So what time do you want it till? Noon? Noon. No, no, no. Tomorrow. Tomorrow, okay? Yeah. Okay. Just tell me what your code is. Yeah, yeah. 15C? Yeah. D or B? D. D, 15C. Okay. Let me add. Okay. I'll turn it off at noon tomorrow. Okay. Hi. It was interesting. Was it helpful? Yes. Mm-hmm. I made a small request for the age. Oh, okay. Documentation. Awesome. Thank you. I have it. Yeah. I'm going to get it. All right. Thank you. Thank you so much. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you so much. Mr. President. Mr President. Thank you. Thank you. Thank you. Thank you. Thank you very much. Thank you. Thank you. Thank you very much. So… Thank you, sir. Good. Three hours. OK, I'm going to, I don't want to do the math to figure out. Which time is math? It's like, OK, I don't want to do the math. I'm just going to keep it on for one day, OK? Wonderful. OK, so tomorrow 1.30, it's going to turn off, OK? Yeah. All right, this is yours, right? Yeah. Thank you very much. Thank you. Nice. Nice meeting you. Thank you. You too. Get up? Yeah, I'm not going to. That's public. That's public. The lab will turn off at 6. The environment will, yeah. The reason it's like this is because it's, it's, we don't, we don't have physical machines at Summit. This was delivered at Summit last year. So there's, there's laptops, but the Vagrant is like, if you want to have it all, there's not, there's no way like, it's not, it's not as repeatable, right? Because with this way, how, how we have it set up, we don't even need to provide laptops, right? Repeat, yeah, exactly. Because if you want to repeat it. Because if you want to repeat it, you're right, you're saying if we had Vagrant VMs and you can just, you can just, yeah. Yeah. It's, it's a matter of time, right? It's a matter of like, we create the environment here. This environment doesn't let us export into that format easily. So that's why it's not, yeah. So, so the machines have been customized a lot? Like the, the things you see there, like, these steps have been done for you. You saw that in the directions. That's pretty much what we did to it. Otherwise it's raw 7.5 VMs. And for some of them, like that's the Linux lab, that, the scripts that he wrote in, in a directory. I don't know if you saw that in the exercise. Yep. And actually it's on GitHub. It's already on, yeah. It's a link on GitHub, yes. Yeah. For Summit, I'm thinking of making, like, you know, this is from Summit 2018. For 2019, I'm thinking of designing a whole new one. But I don't, I don't know, I'm not artist either. I have to think about it. Maybe I'll put the phishing. Where's this guy in there? I don't know. I don't know what to make. I'll just give it to you. So, I'll ask Peter, I guess, tomorrow. Is he gonna be in the office? Maybe, maybe not. About who from his team is able to go to Summit. Because I'm pretty sure that it's gonna, that it's like, I'm like 99% sure that it's gonna get accepted. I'm like, very positive. But I mean, I could be wrong, but I mean, I'm pretty sure based on the conversations I heard at Raleigh two weeks ago. I don't make the final decision though. The REL team does. Not like, not Mark Thacker, but there's the REL team, yeah. Remember, it doesn't matter because I turned it off at 6. You can do this in the hallway. Oh, I see. Yes. Actually 10 minutes, right? Yeah. Okay. Yeah, 12 minutes. Thank you. Oh, I see. When do we have to leave the room? Yeah. Think of like other cool swag because we have a budget for Summit that we can use to make stuff, me. So like, I'm trying to think like, what else should I make? Like a T-shirt. Last time I did this sticker. I don't know who I gotta think about it. Oh, thank you. Thank you. He was asking me about this, about IDM. Like, I don't remember off the top of my head what this glass field is. So like, I don't know if you know, I don't remember off the top of my head the ad user. Like, there's the, where's the ad user? Oh, an IDM? Yeah, basically like, I created all of these screenshots took me forever. It's the basic. No, I didn't. I did all this. Okay. Yeah. Basically I redid all the entire documentation because I needed to make it look consistent. Oh, here it is. This? Yeah. I don't know what that, I don't know off the top of my head what that is. I've never used the class. Do you know off the top of your head? I couldn't find it. I've never used it. I've never used it either. No, I left it blank. Yeah. Usually we leave it blank. So I don't know off the top. Let's see if you can find it. Seven five. We're all seven five. It's gonna be updated to roll eight for Summit 2019. And also different exercises for roll eight. I don't know that it's actually, I don't know, I don't think we can discuss all those kinds of details. Yeah. Because it hasn't GA yet. So. It's still not in the kernel. Yeah. The roll eight is already in beta. It's in beta. Yeah. It probably would not be included. There are stuff like that. You do clock work in the kernel. All right. I can't commit to that because I'm not in the roll eight product management or product like engineering team. Like they, they make the final decisions on that. It's going to be blank. Excuse me. We have the answer to that. Yeah. So those are all the classes. So it's whatever your object class, it will be automatically filled in whenever you create a user. The only reason you would ever put something in there is if you wanted to put something that user is automatically inserted into. For instance, when you create an object, it's automatically inserted into whatever classes where you did the object. Like in that org, like organizational person, like everything is in top, right? That's what it is. So the only reason you would ever put anything in that box is if you put something, you wanted to put the new user into a class, it's not an insignificant hierarchy. Uh-huh. Uh-huh. Uh-huh. So it's basically a black class. Yeah. It's necessary. It's an educated thing. Yeah. So we'll get kicked out in five minutes, but like I said, you can... Anywhere you have internet, you can finish it. It shuts off at six. Okay, so, um... See you tomorrow, right? In the office? All day. Yeah. I leave Tuesday. Okay. Is this in the office? Yes. What is this? Pochikon? I don't know. What is this? Pochikon? What is this? Yeah, this is the address. That's the street. Name of the street. In the...in the Bernouro office? Yes. Ah. So I have to go to Bernouro office if I want to go to this? That sucks. But it's weird. I believe he made some mistake there. He's there. Okay. Because I thought when you put an address here, I thought it was up here, but it is in Bernouro. Because I got some, some cases in the past, where they wanted to be able to... they didn't even try to sympathize with what you're trying to do. It's... Thank you. Thank you. Yes. I have to check with the maintenance. Are you ready? I don't know who it is. Because... Hi. Thank you for the work. Yeah, I hope it was useful. Yeah, it was useful. Yeah, it was useful. So some more things? Mm-hmm. So some more things? Yeah. Yeah. Yeah. Okay. You've done it in the future. Right. Hopefully you do. Yeah. Thank you. Thank you. Do you have any devices? I don't know. So you can get this. Okay. Okay. Okay. I understand. Okay. Thank you. Please come on the screen. Okay. So I am the one who manages it. Uh-huh. I have one. And one of you, I just take one. We haven't seen him in forever. Will you share it? Uh-huh. Okay. But I will do it this way. Oh, dear. Yeah, okay. I'll have to... I'll have to do it this way. Okay. Okay? Yeah, I didn't know that part. But then frankly I took a note of the... I'm saying I understand here. So, thank you very much. Thank you, hope it was helpful. Oh, which one? So the slide, it's just one slide. So, but you are calling it a zero? You are calling it a zero? Because they can't do the lab without the lab environment, so I wasn't sure. What should I do? It's going to cause confusion. Yeah, I mean, there still will be a video. Oh, yeah. But it won't work, right? I mean, it's a lab for a particular day, right? Because I'm already shutting it off. But the github will be there. In the video, they can see the github and they can see at least the lab. But they will never know. Because it's like a picture. No, it's inside an Amazon Azure environment. Yeah, which you launch specifically for workshops on it. Right. But my email address is there on that first slide, so they could technically email me. I actually wanted to participate in the workshop, but I've got lots of discussions. Oh, no problem. So that's kind of interesting. Yeah, no problem. The Fedora boot over in E has the QR code and URL for it. I think I can wrap this up. Okay. I'll just go through. But then you can pick as many as you want. I hope so. I hope so. All right. This one's for you. This one's for you. Okay. And you should use it somehow because it's okay. Are you going to use it? No. It's for half an hour. Yeah, like you should use it. And it's recording to the camera. And come on over here. So once you attach your laptop with a gem I need to put like press notebook. And so it will turn off. And all the else time it should be a portrait. So this slide is shown. I'm going to show you what I'm going to do. I'm going to go to sleep. Okay. I think I might have a couple for the clock. Is it okay to sit in the first row? Yeah. Yeah. We'll wait till Matthew gets here. Okay. I will just come home. It's a place. Okay. Till Matthew has to answer all the questions. Thank you. Are you going to marry? No. Can we go here? I think they're trying to talk. That's the price he pays for being the big boss. What are my friends? He has a professional task to enable SEDs on 25,000. So I'm going to show you. There's some Linux engineers here. Thank you. This is the guy that was here. Let's go outside. They're waiting for you. I know. He's trying to attend the session. All of them. You're welcome. It's a wonderful thing. It's always good to have a minute. I mean, you can't mute it. So this is a new... Yeah, yeah. This is for you. Thank you. Boss? Oh, yeah. Last step, please. Boss and Sam are working now. Okay. Yes, so Boss and Sam are working. Okay. But there is no audience room. It's only for recording the mics. Okay. So whenever somebody will ask me a question, please remember to repeat it on the mic. So... You don't need to unmute it. I'll mute it again. Just mute it for the whole session. We will manage the audience. I'll probably bring it from both of you. I don't see who we have here. But I would say... I mean, when I was... I... Are we recording? Second one. You can't make a wrong place if this goes by. Which way? We just threw up there. It's like Middle Eastern. So should I set the first row or something? What? Should I set the first row? Should I all stand up front? Should I all stand up front? Which direction should we be facing? This way. This way. This way. This way. This way. This way. This way. This way. This way. You're better. If not the stage is kids. I would like to let the last question to the floor. I might have had a question. Number one. Number two. Number three. Number four. Number five. Number six. Number seven. Number eight. Number nine. Number nine. Number eight. Hello, we thought you were ditching us. Yeah, no, I just got hijacked on the way over here. They have two of the law points for us to cross around. There are only four recording, not the other way around. Thank you. Okay. Do you want to keep one of the mics on you and we'll pass it around the other? Or do you want to pass it around the other? I can keep one on you. I don't know. I don't really know the plan for this. I was just going to, yeah, yeah. I'm going to show some charts, then we're going to talk. That's the plan. I thought it was because we were showing it. Naturally, more so the audience. So, Obie? That's great. There are two microns. If somebody asks something, I will please ask you to repeat it on the microphone so it will be recorded. There will be ten minutes left. I will be showing the ten minutes, then the five minutes, and then out of time, which means, yes. If you need anything, let me know. More water, whatever. Something awesome works. If it's too hot, we can try to open the window. Additionally, if you could, copy the slides on the USB so we can keep it in your mind. We'll do it afterwards. That would be okay. That's how we're going to do it. The predictor should work, but we have to change probably there to the other one. If not, the backup plan would be to copy the slides on the USB. How do I drag this window? That's all I want to do. You have the mirror anyway here, so you can... I haven't seen it. I haven't seen it in the building. I've got a keyboard. Keep it between them and the same. Is it time to start? What's that? Are we starting now? Go for it. We are starting now. Welcome, everybody. This is the Fedora Council BOF, Birds of a Feather, People Who Care about Things. I am Matthew Miller. I'm the Fedora Project Leader. We have other Fedora Council members up here who I will let introduce themselves in a little bit. First, I'm going to show you some charts because that's what I like to do. Mostly we are going to... I'll show some charts. We'll introduce ourselves and say who we are and what we do, and then we will do questions. That's basically the session here. I didn't have a keynote in which to present charts, so I'm just going to present them raw here without a lot of analysis or talking. The first thing to show is the dinosaur slide, which I show because Smooch has asked me to. The charts I'm showing are based on some observations of IP addresses and not very much... We don't have any deep tracking and don't want any deep tracking, so we're kind of... It's like we have gone into the jungle and are looking for dinosaurs through binoculars, and sometimes disasters occur and the data is rough. That's the caveat. Here is... This is the chart. Basically, unique IP addresses seen per day for each Fedora release for all time. Normally, I glossed over this chart because it is a little bit hard to read because of all the different things here, but this one is super interesting because here's Fedora 29 and wow, that's a gigantic jump. This is a very popular release, at least by our metrics here. Honestly, I can't explain why. It's a great release. We've been doing a good job and everything is awesome, but I don't know that it's 40% more awesome than the previous release. Maybe it is. We do have a new program manager. This also could be a network topography change because we count unique IP addresses, so if something is behind NAT at a big institution and they're using IPv4, all of those will show as one IP address. Maybe there are 20,000 machines at a large company that are using IPv4 and then shows up as a dozen IP addresses because they're all filtered through that funnel. If that company switched to IPv6 and then all of those machines are now showing up with unique IP addresses, maybe suddenly they're all being counted differently, so maybe it's something like that. I like to think that there's probably some increase in popularity here as well, but again, there's a gigantic dinosaur there. We'll see what happens. If we look at this, that last one was a seven-day moving average. This is just kind of a zoomed-in view of the last of the per day versions you can see these go up and down with fewer check-ins on the weekends and more check-ins on the weekdays. That's pretty much the typical pattern there, but again, it's pretty high numbers there. There's not some weird spike or something. It's the normal kind of growth we see in a release just steeper, so I don't know, maybe we just are very successful and everybody should be proud of themselves. This is the same thing in a stacked graph and we can see, again, the overall numbers here. A few years ago, we were looking at the total number of deployed systems checked in as somewhere around 200,000. That doesn't mean there's 200,000 machines out there. Again, the IP address thing, and then not every machine checks in every day, as you can see with the up and down graph. I don't know. I think this probably represents 2 million Fedora machines out there in the wild somewhere, which is pretty good. Again, a lot of growth there, it's absolute growth. It's not just that the older releases are being cannibalized to the newer release, but we also see that a lot of people are upgrading to the newest releases. There aren't people running Fedora 8 here, this light blue thing. There's still a sizable number of them there, but it's finally decreasing so the absolute percentage of it is reasonable. Are KD3 users, I guess? KD3 users, yeah. We see the same thing with Fedora 15, doesn't have its separate thing, where people didn't want to go to Genome 3. If that's back here in this graph here, where this release right here had SystemD and Genome 3 in it. Again, there's a lot of possibilities, but I think that was just a lot of change all at once right there, and that took a little while to recover from. We want to manage change carefully in the future when you have big changes like that. This is the same thing, but for Apple, so this is 5, 6, and 7. This is again a 7-day moving average. I think there's several interesting things here. We have the REL8 beta out now, and just in time for that, the Apple 7 has finally become more popular than 6. You can see that the Enterprise Linux audience is very conservative, which I think should surprise nobody, but there's some really fun things also here. REL5 went EOL like that day. A lot of machines dropped off the network, but then some of these continue. I think that this might reflect a config difference between CentOS and Enterprise Linux. Some of them kept checking into the mirrors even though the release was EOL and some of them didn't, but I'm actually not quite sure on that. Anyways, that dropped, and then this spike right here is super fun. Again, this is a 7-day moving average, so actually this number goes up way off the charts here. On this one day, a whole bunch of people decided, wow, I should apply security updates today. It's kind of amazing. When I looked at that, I was like, what happened? Oh yeah, I know what happened that day, so I think that's pretty awesome. This is the same thing stacked here. Again, see there's a lot of growth. Oh, this was going to be the one with Fedora and Apple both on them, but I guess I reduplicated slides. Yeah, so I just have to explain. The Fedora numbers we can see are going up pretty well, and these Apple numbers are also going up. The Apple ones are in order of magnitude over Fedora release, which means that people who are working on Apple have a huge amount of impact in terms of the actual places their software is deployed. That doesn't mean that the Fedora release doesn't have a lot of impact as well, because that Apple release and the Enterprise Linux stuff that it's used wouldn't be possible without the faster moving innovative side, so we kind of expect on the innovation curve for there to be a smaller number of users on the faster moving thing, so that doesn't mean the impact. If you're only interested in working on the Fedora leading edge stuff, it doesn't mean it's not important, and I don't want to downplay that part, but it's also important to realize how much impact we have with Apple as well. We wouldn't turn down being double Apple. Right, yeah, right. Yeah, exactly. It would be nice to have all those numbers up really high. Okay, this switches to my other chart here, a totally different thing where I, instead of dinosaurs having to do with IP addresses, these are things that we see in FedMessage. FedMessage is a Fedora message bus where a lot of things where people make wiki page edits or check-ins to get or QA feedback all get sent over the message bus with your username attached, so I counted the unique usernames that show up. Over the course of a year, this ends up being about 4,000 different people who are seen somewhere between two and 4,000, depending on how you draw some lines, and I've tried to filter out bots and automated things from there, so mostly these are humans. The blue lines are the ones that basically are mostly drive-through contributions, so every, the graph here is for each week, so basically somewhere like 350 people, every unique names are seen doing some activity in Fedora every week. Of those, the solid line ones here, these are people whose names keep showing up for basically at least one quarter of the last year if they've shown up for that amount of the year, so these are basically the people who are solid all-always-around contributors, and then the colors show how long that person was first seen, so the red ones have been seen for at that point more than two years, and then the green are new users, and then the yellow is in between there. So this chart isn't great. I mean, it shows that we've got a nice consistent core base of contributors and we're not dropping things off. I think there's kind of a slight downward trend over the last three years here, which I think is concerning. It's hard to really draw a line through the top, but it kind of feels like there's a peak every year, and then it drops off. I don't know why that would be, like people get excited in January and then their New Year's resolutions don't pan out. Yeah, right, yeah, something like that. But yeah, I think that... What's that? Oh, yeah, yeah, right. The drop-down every year, that is Christmas, which I also find a fascinating thing, because it's nice to see the real world in your data. So your red hat has a shutdown and a lot of people just aren't around doing things. But you can also see that during that shutdown there's still people show up and do work, so it's not like we're off in those times. But yeah, what I really would like to see is that green line get bigger and the yellow line be a little more healthy than it is. I don't want the red line to go away. That's maybe unfortunate coloring. It's not like a problem. It's actually great that we have so many people who stick around. What's that? Yeah, exactly. So I would like to figure out how to do that. And I talked about this last year and you can see we haven't changed that picture very much in the year here. So hopefully we can do better this year. One of the things we've talked about is kind of in the Fedora Mindshare Initiative, which is kind of our outreach to new users and outreach to the community. It's our side of Fedora. We want to have a lot more small events and we've actually asked for it to be $250 events every week by that group. So it's kind of the challenge we put out to that group. We've put funding in place for that. Yes, Brian? I just have a question about... Yeah, the question is do we... Right, and yeah, so one of the things that questions is do we see this trend other places or maybe is it just a downturn in the things that we're measuring? Yeah, and actually one of the things that should be noted here is one of the things that feeds into this is Wiki page edits and we've done two things to the Wiki. First of all, you have to be in another Fedora group in order to edit the Wiki because we're having a huge spam problem. So that makes the bar for Wiki editing higher and we've also tried to push things to using docs.fodoroproject.org instead for a lot of things that are documentation and those things aren't measured here because of... it's hard to count those in Pagger properly. Yeah, the question was have we considered allowing the spammers to make the graph go up again? Well, one thing that did Brian's point that I think would be interesting to compare to is Fedora Magazine, right? Which I think, as I recall, it's still on a pretty good upward tick, right? As brilliant as I'm sure it is, I bet it's got a limited user base or reader base. It's not going to be general tech readers all that much, right? It's going to be at least Linux people specifically. And also Brian keeps track of social media metrics for us and we know that the Fedora Twitter account keeps getting more and more followers and is... Of all the upstream projects that I track between OSAS and... Yeah, so of all the... of all the projects that Brian tracks I'm repeating for the recording, all the projects that Brian tracks, the upstream projects that OSAS, Red Hat, Open Source Group tracks, Fedora has the steepest growth curve for both Facebook and Twitter followers. So we're definitely seeing growth there that fits more with the IP address stats as well. So yeah, I think, yeah, there's definitely some other things we could do here. So I don't want to get too freaked out about this, but I still think that that growth should be contributing to more green and yellow in this number. So I'm not super freaked out about the blue line going down, but I'm seeing the other things go up if things are working. Okay. Yeah, Tomasha asks if we get package automated will this graph collapse even more? Yes, I think it should, but I'll need to find another way to measure people's involvement because we're not trying to get rid of people with that. We're trying to put the people into doing more interesting and useful things. So I think that'll collapse my graph that won't collapse either. Just want to I think this is a very, very important piece of data and I really, really think we need to do a better job of capturing it if we're not sure how right it is because this is all bread and butter right here, right? This is us showing revenue growth, right? This is really, really important and I really think we need to stop having these presentations without this being good data or having a pattern for getting it. I recognize it's a lot of work, I'm just saying. Max has definitely volunteered right. And just like personally for me this is something we should find Fedora budget for and go hire an intern like we need to do this better if we don't have the time to do it as the regular contributors to the project then we should go pay somebody to do it. One of the challenges is a lot of this stuff needs to be in order to measure it easily this way because it's going to be linked to your Fedora FAS account and a lot of the activities especially as we expand into other ways of working, you know, telegram it's very popular, like that's not your FAS account that's not stuff that we can easily track in this way so I need to figure out ways of doing that so yeah an intern who's interested in this that'd be awesome. Actually, I actually could probably generate this as a research project for BU Okay, sold. Can you generate this as a research project for BU? Yeah. I've got a question about your height because if you have something about your height because maybe they might also show us something like if we are doing, for example, that the job testing, gating, whatever your height becomes more attractive because there is probably some Yeah. You can see the difference between it and the bottom of the graph. Yeah, exactly. Yeah, no, it's interesting. It's actually Yeah, right, this is raw height back in ten years ago a higher portion of people were running raw height as a percentage, a gigantically huge portion but also the absolute number was higher back then. So, yeah. They also might be right. And also, some people do call it branch now but they don't call it raw height but they don't call it branch or they don't call it bait. Yeah, it may be reflective of people wanting, like they're doing their work differently right to Dan's point. I'm not sure, do we want the raw height number to go up? Or do we just make it so that there's a perpetual beta that's kind of like raw height to Adam's point a lot of people are using the branched version, the beta versions before we get to raw height and that's actually, you can see that on the graph here that's actually fairly popular, that's this here. Yeah, right, this is yeah. Yeah, exactly. So, yeah, that's pretty good too. Should we let some other people talk here? Let's let people introduce themselves and say what you do I'm looking at Dennis because you're standing on this side here. Yeah, except for where did I put the mic? It's in my pocket. Trouble's going to happen. Stop talking. Otherwise all we get on the recording is you saying let me take off the mic. I'm Brian Exelbeard. I'm the Fedora Community Action and Impact Coordinator and I work around mindshare and other community focused events and other components of the project. I'm Langdon White, I've been here a couple of times. I'm the Modularity Objective Lead. Hi, I'm Tomas and I'm one of the two elected representatives from the community. Hi, I'm Dominique Rapide and I'm the CI Objective Lead. Hi, I'm Yonah and I represent the Diversity and Inclusion team. I'm Ben Cotton, I'm the Fedora Program Manager. I'm Peter Shabbatan and I'm the Currently Serving and Generating Representative. And I'm Dennis Gilmour. I am the other elected member of the Board of the Council. It used to be the Board long ago. So actually, I have a question for Dominique. We talked about gated or better raw hide. Can you talk about better raw hide for, you know, what are we doing there? Sure, so without having this written up completely, we're going to have to have a lot of work to do to get the data center around. What we really want to do is gait raw hide. We want more things to work and being a place where we can develop new things and try out new things without being blocked but others are also working on new things. I think that's, as a community, that's what we want to have. So we want to have, if we want to have bleeding edge, we can go to upstream, right? But if we want to have, we can line that with automated packaging where you can pull things automatically from upstream. Thomas Tomaszek will work on that. And yeah, I think we have some good plans on how to get something that started, hopefully by flock. So what is gated raw hide, as a package or what does gated raw hide mean? What is the gait and what do I have to do to get through it? Excellent question, what gating actually means. We have seen some things, we started testing packages in Fedora a while back. For various reasons, that was not quite as successful, there were usability issues, and the question is how that actually worked out and was limited to atomic packages. So what we want is, first of all, a package level gating. That means when you look at a package, what does that mean? It's not just the code, but if we look at the distribution, the Fedora, we should also define how to work in the distribution. So I have tests that cover how those packages work. So when I submit something, submit a new build for example, we should run those tests, but not only those tests, but also other tests from other packages. So that when I say, I have the dream that when I pull raw hide, it can actually boot. It's far-fetched, but things like that, just have some stories where some basic things that should be able to compile other things or maybe even see a desktop. When I have those stories, those things should be tested and if I submit something that breaks those things, I should be gated. So, for example, we should say that update is not going to go in right now. Imagine a testing system, so I'm not sure about the implementation, but we'll want feedback, but imagine a testing system that says my karma is minus one on this because your update will not let me boot my Fedora. So the package will be blocked from getting in and you have time to fix the issue before you would get it in and people would send you a lot of emails and saying, thank you for breaking my installation. When will you fix that? It might feel like a slowdown a bit because tests will be run when you submit something else, it's not going to go automatically in, but in the end, I think it will be a more comfortable development place because you don't break other people's workflows and you have the chance that when we test against raw hide, when you really try it out, things are not as broken and you can try out new features. So that was a lot of talking. Yeah, so from a practical experience say I have an update to Sudu and I try to put it into I'm a package maintainer for Sudu, so watch out everybody. Which is currently broken. Yeah, so let's say suddenly that doesn't let you log in, which is fairly... You can no longer get root even though you're in the wheel group and so we've got some tests for that it worked on my system because I happened to have on my local system where I'm building it I've got some other local configuration which is non-standard, which happens to let it work so I didn't notice it. It worked fine for me. I build it in raw hide. What happens next? We'll see there's this new build and we have tests then that are attached to your package for example and they're really good tests or say we have other packages that depend on Sudu and say hey, when Sudu changes run our tests too because Matthew keeps breaking our stuff, right? So we'll run those tests for you all the tests are in a standardized format there's their wiki pages and also documentation pages on that and thankfully we don't have to run those tests on your system because we have this cool thing called CI, right CentOS CI, so we talked to Brian Stinson about that. We run those tests and if those tests fail we'll say well that does not get into raw hide we're sending out a notification to you and say you failed these tests here you can look up where those tests are these are the logs here's the person to contact let's figure it out. So my package is built, goes through Koji but then it doesn't get into the compose and then I get email about it or how do I know that I mess things up? I think I think for Sudu you might get more people but yeah I think there are different ways we can do that right now I think we have the notification system that we can use but I think the email will be the default for that. The current status because I didn't know why. Yeah the question is a very good question we've been talking about this for eight years what's the current status of it? Eight years is an exaggeration it's really only like six. Yeah so the current status is that we do have tests that are run for packages but they don't have an effect because there are various usability issues so gating was disabled for a while in Fedora because people are saying this breaks our workflows we can't really use it properly so we have worked out a new plan we don't know the details yet ironed out with Fedora and for for example we have an idea of how to do this properly using tags if you want to know the details so that essentially your workflow doesn't change except for the fact that there's a slight delay between when you've built something and it goes the rest of the way through Bode so we do we do run the test for a subset of packages right now but not all of them so that will change and whereas right now the test results don't have an impact then moving forward they will prevent packages from going into Rahide with the asterisk at this point that we believe as a human we should have responsibility or as a packager so if you say for example I don't believe in that test result there's a way to wave those things so it's always like if you're a certain I've tested this if Matthew says I don't believe that test everyone should set up their machine like I set up mine just push it out to everyone then he can do that Matthew, I'm going to pass over to Lisito yes Will this process be any different? So the question was whether this is also going to apply to stable branches I think it should but this is scope to Rahide for now because we believe that this is where the development happens I think this is where we feel a lot of the pain also doing the process where we can catch things early and unblock others so I think the testing system should scale there's no reason why you can't run those tests on all the other updates but let's not try to solve everything at the same time Adam you can come get a mic if you've got a lot to say come we don't want that so I just wanted to add to what Dominic was saying there were there's two sort of angles on this when we tried turning on gating we did it for stable releases via Bode I showed up a bunch of inadequacies in the process which people have been working on since and that hasn't been very visible work but there's been a lot of work done in Greenway there's been a lot of work done in ResultsDB we put bits of the puzzle together to make it less of an exciting experience than it was last time and the other thing that Dominic was kind of talking about but it wasn't clear if you didn't know what he was talking about is this is different in Rohide because we don't have Bode so the big piece that was missing for doing Rohide package gating is there was no mechanism by which you could say these three package builds have to be tested together otherwise all the tests for all of them are going to fail and we kind of had to come up with a plan for what Dominic was talking about about tags the idea is the fed package is going to make you side tags basically but we had to go through a few different design changes to come up with that idea and that's what's getting done now that's basically the idea the question was does this mean that side tags generation is automatic and the answer was it's on request and automatic at the same time from two different people and apparently apparently Mike McGrath is not involved in physically doing this we could probably talk about this for another two hours but let's see if we can move on to some other topics are there other things people want to talk about in the audience or should I start prompting council members here Mike McGrath the youths what are we doing to attract the youths do you want to take this youths young people that's what is going on because so as also Matthew mentioned a bit before especially mind share is working on this so like for example release parties especially this time like it's even easier to organize a release party and to have your budget approved so like people that are not only ambassadors so we are trying also to lower with the barrier not only ambassadors can organize a release party they can help release party a small event let's say at their country their city and in this way they can go to universities for example they can have a talk there but especially release parties know that it's like especially user friendly for new people so like just talking about the community what it is and also like trying to install on their computers or things like that so this is one of the things that we do and we think that especially the release party is a nice way especially for new contributors to join but not only but this is one of the things and especially lowering the barrier not only ambassadors we will have also other people that let's say they will be kind of process so they can start organizing but it doesn't mean that you need to be you just need to know about the community how it is how to join and so on what was the term for that we made up a new word we've been using advocate but we're not titling it out okay get out of here with your labels Matthew the reason for labels is there was kind of a thing where people felt like being a Fedora ambassador should require a certain amount of commitment and involvement with the project to get that title and to get that sort of a badge of the ambassador should be a certain level but we also wanted people to be able to just like show up and get involved and previously that you've got to become an ambassador before you can do things meant that people were like never mind and that wasn't working so well what's that yeah David says it got a little weird Ben suggests that we should go to Jen Madriaga next Jen is our events coordinator and has been working very hard on getting things set up for flock for this next year no I'll just stand here it's okay come here on video Jen there's video isn't there video yeah I assume there's video so so Matthew has asked me to make the official announcement for where flock is going to be this year and I think this is the earliest that we have ever made the announcement because it's still January so I'd like a round of applause for that so we've been working on sourcing location we're still confirming where the hotel is going to be but we are going to be having it in Europe again this year and we are going to be located in Budapest hungry hey and so I hope all of you can come this year it'd be great to see all of you there and I'm sure there's questions about why we're in Europe a second year dates okay so August August 8th through 11th yeah plus or minus one day August all the months just keep coming and one of those months will be right so it's going to be in August like it always is August 8th to 11th followed closely on the heels of another conference that will announce I think did we already announce def comp us followed by def comp us which I will also be at def comp India is the week before flock and then it will be def comp us in Boston so if you had any plans for August to say someone's going to be doing a lot of traveling so like I said we're still negotiating with the hotel and we want to just make sure everything's squared away with some of the details there like internet because that's important hopefully we will be able to have that announced in the next couple weeks or so but we are going to be doing a launch of registration and I don't know when we're going to launch to Eventbrite back do you want to give us a within the next 14 days I have to go to Fozdom so it depends on how bad it is okay so after Fozdom we will have the registration site up and then we're also going to update the flock to Fedora website so you can now mark your calendars and make travel plans it's definitely going to be in Budapest we just can't tell you which hotel it is yet but we should have those details we're going to be giving you from Amazon and Facebook and so on we will have a sponsor perspective as well yes so please sponsor I think that's it, does anyone have any questions I'm sure the main question is why it's in Europe again and maybe someone from the council can talk to that yeah it's Mike's fault yeah basically the reason that we're in Europe again is Mike's fault no no seriousness we like a lot of other conferences have to recognize that we exist in a universe where lots of people have lots of other commitments and so one of the things that we had found out was it was very hard for certain people to get travel funding from a certain company in the years that we were in Europe because it was time drawn for that company that company is Red Hat just for the record so when Red Hat Summit is on the west coast it's very expensive to go to also Europe that year that's exactly what I was about to say and so for the record when Red Hat Summit is on the west coast it was hurting some travel budgets and so it was making it harder to send folks to Europe so we were asked to change the TikTok and well I like Europe so we stayed in Europe no it was actually slightly more democratic than that but Europe is actually really good for us and we felt like having a second conference here to reset the TikTok was going to be a good move forward and if you're interested in the further conversation there is a council ticket on this you can go and look in our and there's a closed council ticket that explains all the rationales that we went through Brian can you just make it clear for the recording that we will be continuing the TikTok we will be back in North America in 2020 and Jen and I hopefully will be able to make an announcement in 2019 this time for where 2020 will be so our goal is to be able to put up a banner stand at Flock with a city name on it that's even right I'll promise you a banner stand with a city name let me put it that way and apparently a Lord of the Rings costume but that's a different speech every day of Flock the name the name will change too expensive any one of our contributors in Alaska has been lobbying for an Alaska Flock but the other thing regardless of what continent it's on we really wouldn't want to have the people come face to face who make Fedora work and who can benefit from the connections we make when we have Flock and get everybody talking in the same place and so like previous years we're trying to decouple the funding from whether you're a speaker or not we want to make sure that we get key community members to Flock wherever it is in the world we don't have an infinite amount of funding but that's why we're asking for sponsorships that's what the money goes to we want to make sure that everybody can participate because it's a really important thing for the community I would just add to be very explicit about it we are not geo-locking the funding so tell your friends that are important contributors who don't happen to live in Europe that they are not geo-blocked out of funding we look at contributors worldwide and for those of you with companies that do have travel budgets where speaking is an important part of that travel process let us know because we can work with you to get you letters that explain what's going on and how we're making our decisions and why you're so valuable to us to try and help you make those business justifications so we can release funding for other contributors Adam That's a new question Are we delaying Fedora 31? Yeah, that's it The question was are we delaying Fedora 31 and you will begin answering it I'm going to turn that over to our program manager who's scheduled Can't get the mic all fast So barring any like complete freakouts about that we will not be delaying Fedora 31 the schedule is in draft form it's not published to the community yet that will be very soon but for the time being we are looking at continuing with the twice a year from May and October into November hopefully not release cycle Put that on the mic Do we have anything to share What about the new logo Okay, I don't know if everybody know that we were working on a refresh for the Fedora logo Does anybody not know this? Yeah So the design team has we've had this logo for a while it's not since the beginning of the project but a lot of us are very attached to it it has a number of challenges it's kind of weirdly off center it doesn't work in two colors we had a big back and forth with the awesome people at font awesome who wanted to make a black and white version of the logo which looks hideous and made the design team spasms so we asked them to not do that but we actually want our logo to be available and easy to use to work on places where we need to do screen printing or things like that so that's a problem it's kind of weirdly off center which is problematic and we have Facebook people here again the F is easily confused by most people in the world with the Facebook icon logo and I went and looked back we're being recorded I'm not a lawyer but I went back and looked at trademark dates we were first with that F but just by a little bit and it's been a long time now and I feel like they may have beat us on brand recognition so we're looking at doing some adjustments to do that so Mo Duffy from the Fedora design team who is really awesome with user experience and also with the way things work in a design sense has kind of looked at and worked with the rest of the design team and with community members to go through some refresh ideas for the logo she has a really good blog post we link to it on I think on the Fedora magazine it's on her blog on the Twitter somewhere I really encourage you to look for that post because it really kind of goes through the background of the logo and why we're changing it some of the candidate ideas for what a new logo will look like we don't have a finalized new design yet and we're going to make sure we go through some more iterations of community feedback make sure it's a really open process you forgot the best one Fedora Fedora, oh yeah another thing is if you look at the word Fedora from a distance the font that's used the A kind of looks like an O and to the fact when I look at searches coming into our websites that are typos Fedoro is one of the high up ones so I feel like there may be some real confusion in the world over that one what is Fedoro maybe it's a typo but I think it really does at small sizes look like that so we also the font is not open source in the original design so we're replacing with an open source font which is seems something we definitely should do I can try and find that blog post now but I can do it while you're talking and then you're going to put it on the screen from your phone Langdon how's that going to send it to somebody who can then project it why don't we why don't we keep talking while that happens somebody else talk about something else so the question is what how's the release cycle Fedora coro s going to be different from Fedora and that's actually a great question because I don't work with the Fedora coro s stuff Matthew do you happen to have that answer while you're trying to furiously Google that's my understanding as well so the answer is it's going to be similar to atomic but with three streams every two weeks so the question is what does that mean for older hardware that doesn't have RAM that atomic won't run use something else I guess I think we're Fedora as our main sort of standard distributions still run on fairly low powered hardware Brian has a better answer it looks like oh sorry I wanted both microphones so now I own the entire Fedora council so my understanding and since Matthew's distracted he cannot correct me when I'm wrong is that even if we choose to move forward with using a default of desktop that is built on atomic or related technologies that does not necessarily mean the other builds will go away and it will not be an addition anymore but if there's an interested community that wants to keep them around there will certainly be something that we can continue to build Fedora has in general always tried to run on a large set of generations of hardware and this is something we're going to have to deal with and think about and I don't think anybody on the council is going to vote out of quickly vote let me say it that way to say alright we're just going to kill off people who don't have this much RAM we like you and so I think we will find a solution there you have seen some similar things around certain architectures where we may not have those architectures supported fully in an addition but there's a group of people who want to bring that about and we're seeing builds for it and things like that if I'm not mistaken I would actually say three numbers but I think I have the wrong three numbers in my head they are was it 386 sweet so it's 386 or 686 I don't know 429 is that a thing sorry I don't do PC architecture when we have the blog post now because you're not on thing you should read the whole thing I'm going to skim through here but she's the next girl with no eye yeah and you don't have a thing like Ryan oh yeah microphone this kind of goes through the background and history and shows things that are problematic and yeah there's the Facebook okay through the process here and some of the design things and now we're to the comments so there were a couple different things suggested this candidate number two actually turns out to look very much like an existing logo somebody else so we're kind of focusing around this candidate number one and there's some refinements to this that are being worked on here still but this is the idea so it kind of keeps it keeps the three elements of the voice bubble the infinity and the F which actually stands for freedom not for Fedora who knew but yeah so it kind of keeps those but moves the F away from the Facebook like F changes the font around a little bit so it's not like a gigantic we're throwing everything away it won't be recognized it won't look like the Fedora logo and I think the exact shape is still subject to some tweaking and there's going to be more feedback on that but that's the direction this is going yeah it eats its own tail is that bad Mike I'm excited well your feedback is welcome one of the other things is that we're also looking at guidelines that allow it to be used without the bubble in cases where that makes more sense where you need something a little more simple because our current one again is a complicated thing and it's the actual guidelines require it to be used in a complicated way I think Langdon's shirt is probably out of spec right should I try for yeah so there's the logo other questions logo related or not there must be some Mike you're the longest serving Fedora leader how you doing I'm doing alright thanks I think so a lot of it is we set up this I am the longest serving Fedora leader it's almost five years now which is something like three times the normal thank you I'm glad people are clapping for that rather than booing because that could have been the other way that went I think a lot of it is that we set up this structure like this and I have Brian to help me and Ben is a very active program manager and the rest of the council both to like help with a lot of the different things that are required to run a project this big you know like I said 4,000 people a year is crazy but also like so that I don't have to have all the good ideas and I don't have to like we don't depend on me constantly innovating as the leader with the top down kind of thing that would be impossible and I'm not that smart so we have this council where we have you know people coming in with objectives and all these different areas which kind of help spread out the leadership to the community and we think that structure has worked really well yeah yeah Jim asked what my thoughts are on the Fedora desktop I think it's important spin more elaborate so we have an initiative at Red Hat to put more resources into making Fedora desktop the thing for Red Hat sort of desktop ecosystem in a way that it hasn't been before so I'm very excited about that and we are going to be starting with we have interns in this is also bringing in the youth for Mike we have intern programs at Red Hat here in Brno and at the Boston office where we have working with Boston University and we are going to be working on focusing on making sure that we have a desktop offering that really appeals to those students and from that we're trying to build out relationships with some hardware vendors so I hope that we'll be able to actually ship Fedora on hardware like the XPS 13 for example not necessarily the Dell ones we were talking to other vendors as well not necessarily that model but that kind of thing so that we can get Fedora desktop into the hands of more people Kevin provocative question from Kevin ready for it to do an awesome desktop experience for the students do you think this is achievable with No? I don't think much any other Kevin who works on KDE is questioning whether it is achievable to have an awesome desktop experience with GNOME and the honest answer is yes I think it absolutely is but it would also be possible with KDE it would be possible with a lot of the modern Linux desktops are amazing and can have a good experience and I know people have different preferences and I know a lot of people love KDE a lot of people love GNOME and we have the resources to put it into GNOME and that's why this is based around GNOME but I also would like having GNOME working well on these laptops should make it the hardware enablement on these relationships will make it very easy for the KDE experience to also be really good on it and I would like it to be easy for those people who prefer KDE to switch to it is that okay? I got a little not Dan Walsh asked what we're going to do about codecs you know MP3s can play fine now it's a waiting game can we get IBM to buy all the patents can we get IBM to buy all the patents yes we have this plan where Red Hat is going to get acquired by a large company to whom this is not an issue that's the are we thinking about making the desktop very developer specific for example as a developer why should I choose Fedora over Ubuntu yes so the question is are we looking at making the desktop very developer specific so that was kind of the idea with Fedora workstation we wanted something that would be developer specific and we never really had the kind of the resources to make it that way and there's just also there's a lot of like basic like do the codecs work does the machine work that we didn't really have time to make developer specific things however with the work we're working on right now and again you know it's kind of focused on our developer interns as a use case and like a focus group for it I think that we will it won't necessarily be developer specific but we want to have that developer experience be very very for the for the addition the developer experience will be at the forefront and we really want to make it a smooth experience for that and red head is interested in making sure that you know there's a smooth experience for getting to open shift and kind of other development tools in the open source red head ecosystem Langdon just going to say as probably a likely lead of that because my affiliation with Boston University and my heavy bias towards wanting that I'm looking forward to trying to subvert it into much more strongly developer focused workstation that's my hope Kevin, I have a question about the developer orientation what kind of developers are we talking about now because the web developer needs very different tools from C++ developer or from a C developer or even from a go developer say so far I think the focus has been mostly on tools for web developers or scripted language developers so I've actually been forgetting C++ and C developers so the question is basically when you say developer that is not a one size fits all answer right and I'm very aware of this problem I think with the university focus it is likely that there is going to be a lot of focus on like web development but that said a lot of the interns that will be working directly with head engineers and the ones that are primarily based in the Boston office are a bunch of C and C++ programmers so I think we will actually see some of that so if you kind of knock it down a little bit I would say like web developers especially kind of related to like open shift and that kind of deployment style I think we will have a bias towards it because that's what we are going to have of people right and then the other side of that will probably be like C and C++ mostly because a lot of those guys are like graphics drivers people and so it will be a little bit more focused on that. I think we will hopefully be I'm pretty cognizant of this problem so I'll hopefully be trying to steer some of it or have some impact on trying to make it so that it's either generic or we say hey this is what we are doing for this part or we are going to try to do it this way so that we can have a model where we can actually try to grow it for different types of developers but yeah it's a huge problem there's not really much we can do about it except say yeah we are going to tackle some aspects and hopefully it will be good for everybody or it will be good for like the big thing we want to do with the university students though is we want to go where they are rather than try to get them to come to us so if they are doing web development I'm going to try to make it a good web development experience for them Dan again is it going to be rolling updates or is it going to be release based a lot of this comes down to what we can work out with hardware vendors to make them happy and this also goes to kind of an LTS thing although I try to avoid saying LTS because we don't really mean long like rel we mean like four years at most and we don't necessarily mean that we are expecting like our the time between having to click the upgrade button which doesn't necessarily mean no new kernels doesn't necessarily mean that it's all backwards compatible it just means that you don't have to do a reinstall we need to and the hardware vendors want to make sure that the thing keeps working even though we put out updates in their ideal world there would never be any updates ever but they don't live in the ideal world so there will be updates and we have to figure out how to do that and we have to figure out a way to do that in a way that doesn't put a burden on our maintainers who aren't interested in maintaining things that aren't the latest and we have to figure out good ways to do that so can I add one bit before you're asking another question I think we both actually or at least a bunch of us are interested in this potentially even being silver blue the problem is from a developer experience perspective it's like the worst case for silver blue so until somebody can start showing me why I'm not going to bang my head against the wall using silver blue trying to do development the other thing I'm interested in doing merge disk it from CentOS and Fedora possibly for this particular case of the hardware on Laptops actually building some of the solution from rebuilt CentOS packages or so that we can have those packages that are already being maintained somebody's already doing the work why do it twice kind of things we need to figure out how that will look and that's uncharted territory as well for the this updating issue Windows 10 deploy the solution where they just force everybody to update to their latest release every six months maybe we could do the same if we Kevin's comment is that Windows makes everybody update every six months now so why do we have to be better than that? We can be LTS, it's seven months LTS will mean seven months now Adam yeah where are we with rings 2.0 3.0 whatever iteration of rings we're on so Paul Paul Fields is the life cycle objective fleet and he is dying of the plague right now as people do at this conference and he's yeah so did you ask the rings question last year? yeah yeah so a lot of that and that is kind of enabling work for some of this this is like where are we with rings which has been my thing that I put on slide since 2013 and there are no rings the general concept of letting us focus more on the core of the operating system yeah the general concept of letting QA for example focus more on the core of the operating system QA on the base OS so yeah right so yes we don't need to call it rings we can call it base OS we can call it something else so part we have an objective that is defining that base OS and figuring out how that will work so that is something we want to see in the next year whether it will be rings or some other topographies to determine so yeah so we already have a Fedora core OS now so that is taken the question is will that be the return of the core and extras split I'm going to let Josh talk next unless you want to answer that specific question actually I want to ask Adam a question what is the point of focusing on that now we don't have that tradition on this one can you guys figure out what you want to test and test it Josh says it's all Adam's fault we're just testing this is what I'm testing if you have something in here it needs to be kind of a project white thing to touch it into other areas I mean it tends to seem to see eye obviously it's like you keep coming up against this thing if it would be easier to do this if we knew what were the central parts were you really cared about and to some extent it's like you can't just define it in the code the core should be this but in the sad world we have the core is this yeah so I should have given you the mic for this try and summarize Adam says that this can't just be a QA problem because it touches so many things across the project and that also it's hard to define with Fedora as it exists right now because you can define an ideal and then you get sad when you look at how that ideal relates to the reality to maybe define an ideal and a way to strive towards it I also want to address what Kevin said about isn't this just a return to the horrible core versus extras split we had so to me the core versus extras thing had two major parts and one was there was a separate core repository and then extra stuff on top of it I don't think that was a problem the problem was that the core repository was developed behind the firewall and then dumped into the public and it didn't have the rigor that the extras repository did which is weird and backwards like the better effort was put on the outside of the circle and yeah and the inside was only if you worked for Red Hat in specific roles could you touch it and that was the thing that I think was really the problem there was a big problem that packages in core couldn't depend on packages in extras right there are other distros at least the same problem with Wain and the universe that sometimes they can't enable features because the dependencies are in universe we had that with core and extras and I think that this problem is coming back into it yeah so the problem the technical problem Kevin identifies is that packages in core can't depend on packages in the outer rings it's kind of a one way thing that can be a problem I think there are different ways to answer it one way can be that we could actually have modules for some of those packages in core where there is a version of the package which only depends on things in the core but there's an alternate stream of it which can be enabled which depends on things outside of it so there's things we can do Dennis has some some of that can also just be solved by looking at it not as a absolute the core is this core and it's built as a core but you build everything in a big bucket and then the distribution the way you carve it up when you deliver can split things out which is something that in the core in extras you couldn't do because core was built over here and extras was built over here and never the two shall meet when you can bring them together you can do different things to solve the problems enabling things like extra functionality to be built in a sub package that is not part of the core and is shipped but also we're talking about different concepts here that are conflating one is about how do we separate what we ship and how we define things as we see it as a user and one what I heard from Adam was what do I focus on as the core what is important I think with CI we have what it takes to really define that organically it's not about let's go out and try to define what we really define as the core but it's about what am I willing to create on as we define that relation that's not something we can impose on everyone we have to agree as a community on certain things to say for example we all agree that the system should boot things like that if we have those stories those core concepts then I think that is also part of the definition of what the core is and that has to evolve and automatically everyone who contributes code contributes to that definition a lot more organic way of growing that definition I think we had a question I just wanted to mention that in rel 8 beta we actually have can you hear me okay in rel 8 beta we have an initial attempt at doing this split and it's not about who can do what it's just about being able to apply different policies to the different content sets so if you're looking for a starting place have a look at it at the final it's not perfect but you have to start somewhere and then iterate I think 10 minutes ago it might have been a different topic so yeah I think to add to what the meaning said I think we should define the core or the stuff that is being tested functionally and we already kind of have the definition of this that you wrote so let's just say that everything that influences the beta criteria matters and everything else is outside and we're talking about I mean Dominic said we should define it in terms of what needs to work and sorry I'm blanking on who you are I do know but Zbyshek I'm sorry I knew who you are like I said no sleep Zbyshek points out that we have this already we have the fedora release criteria which is what which does this it defines the things that fedora is required to do and that's great and that's a good starting point there's two two kind of problems there and the criteria cover a hell of a lot of stuff just the stuff that we are required to care about by the criteria there's a lot of it there's two desktops in it which is twice as many desktops as well like beta has for instance and then we don't have a focus on pairing down what goes into the products that we produce to meet those requirements so the requirement is the workstation must do blah blah blah blah blah and KDE must do blah blah blah blah if you look at what's in workstation and what's in KDE especially it's the kitchen sink it's huge so I mean the way you would go about doing this the way we should go about doing this properly is that we should have a system that knows what goes into each of our key deliverables right this should all be tracked exactly a product definitions and it would be a great way to do it that should all be tracked so that we can know instead of there being a big list somewhere these are the important packages or a comps file which says these are the important packages it should be what goes into what and this should all be tracked up and down the tree but we should also when we have that we should have some kind of focus on making that as small as possible to achieve the things we want to achieve and if something is not achieving one of those things then being able to say okay that's less important it's kind of where I hope we could go with it so yeah it seems to me that the core should be that the minimum viable bootable set for the architectures we support right that seems like that is the simplest way and just keep it as absolutely small and condensed as possible because these are the things that must work it doesn't necessarily fulfill all the missions of Fedora but that is a definable set it shouldn't change it shouldn't really grow or shrink a ton over time and it would make it easier to just get this done I agree it must be all of those for sure all the supported architectures for sure that's why you keep it real small because yes yeah yeah that's what I was trying to point out desktop is really beyond the scope I think it's also I think I think the other thing too is if we go back to the new charter right is that it's also a measure separation of responsibilities right for instance is that what Fedora the project promises is this really small set of things to boot or whatever you decide what it is right then we move the responsibility of defining okay what is a known desktop experience to the known desktop edition team right and then they are responsible for making sure that you know they're delivering that thing Adam the human may be doing some or a lot of both he will be but it's it becomes it shifts the responsibility of making that decision first of all off QE second off of trying to define you know the difference between and this like you keep hitting on this but I would hit on it a lot more which is that Fedora is a project not an operating system right and we need to stop treating all the things that we want to deliver as one thing right and so we don't need you know the desktop team you know or let's say sorry like the KDE team or the GNOME team or whatever they know what they think is important to deliver the experience they want to deliver and if they don't it's still their problem right so I just want to keep hitting that again go ahead I was just going to you should really read the post that the council put out from its hackfest for more on how these teams and all can work and what Lang did saying because that definitional component is critical and when you start to look at things like CI those teams will one day I don't want to commit for you be able to say hey we're building the XFCE desktop and here's all of our requirements and we may not get every package in every case against those requirements but everybody gets their information in advance and that's all well explicated in that that post so please please also beat on Matthew to write the follow-ups so I would like to point out that in fact we actually we already have some kind of definition of packages that are core packages that are important to test and that's the critical bath packages but it's yeah yeah but the thing is yeah yeah yeah the thing is I think this list is actually smaller than what we're talking about now and still it's already suffering from the problem that I was describing with the dependencies because it then drags in all the transitive dependencies and build dependencies and even build dependencies for the other sub packages and so then you end up I mean this is not the very latest example but at some point we wanted to have KDM is critical but KDM is a sub package of KD workspace and KD workspace build requires on Akonati and Akonati dragged in MySQL and some MySQL ended up in a critical path without anybody actually wanting to do that and this kind of things this is exactly what I'm saying that without losing features or building the same source dashboard multiple times it's hard to split it yeah just because it's hard doesn't mean we shouldn't try number one right but I think right now I think one of the problems with that is that the people who know the most about what it should be doing are far away from the people who are doing it right and so that's why like I said separation of responsibilities more right it's like you have to know how every single deliverable of Fedora works in order for you to be comfortable with the testing that's done on the super set right or fail right like so that seems like a separation of responsibilities problem we need to decompose that you know you in the abstract are the core person and you're responsible for this set and then the like I said the KDE group is responsible for the set that they care about right and the core OS set when the container set and the cloud set and the you know and all of those could be separation of responsibilities if we had people who were close to the problem via things like CI responsible for how they're delivering the things that they deliver and the things that they care about then they can do things like break up you know transitive dependencies or other kinds of pensies right that are dragging things that they shouldn't we can have with modules right we can have actually different sets of available rpms depending on which addition you want to ship that doesn't drag in something you don't want to drag like I think we have a lot more flexibility on actually you know did you put the charter up I was going to put the chart out fail but the end of the charter basically says we want to enable you know Fedora members to kind of deliver for their users right we have we're starting to get the tools between the modularity and the life cycle stuff and you know we're starting to get the tools that that's actually doable like executable and while it's getting executable we need to start to shift the responsibility on to executing it that way so that they can be responsible for whatever it is they want to deliver and our classic example what was it the sheep shearing company well I'm a different example because I don't think there are people from a large social media company that is usually confused with our logo who do a cool thing where they take CentOS packages and then they use raw hide system D and a bunch of other things to build their own kind of custom solution for their needs they're doing that often their own world their own infrastructure it would be super awesome if they could come to Fedora and do that in Fedora and also not generate more work for Adam but at the same time like do their packaging and do put their branches in our disk it wouldn't impact your critical path thing no it would make it easier but it would be right they would be contributing back and then they would be able to pull things out fork back and forth it would be awesome and Jim is looking really worried we want you to do more work Jim I think that would be awesome it would be really they could deliver that solution for themselves using our tools and in our space can you turn on and make that face of the camera yeah I was kind of hoping that we would not talk about this particular thing until we had documentation and testing and possibly validation just because it's hard doesn't mean we shouldn't start right oh this thing up here you're talking about this okay we'll talk about this later then I won't show he's making a face of what's on the screen here I want to point out this shirt here and put you on the spot this is why we need the logo redesigned right like wow we can't print our own logo on blue and have it look like our logo from like you have to come like right up to you and make you uncomfortable in order to see that it is our logo so anyways okay I'm not allowed to talk about this thing on the screen Jim we've got it okay there's a thing that we're not allowed to talk about yet but it's going to be awesome don't write that you are all down hahaha Adam the discussion about the rings thing quickly became very personal about me and QA I wasn't yeah that's Josh's fault I wasn't solely aiming at that but there's also the problem of we have 50 100,000 packages in fedora and we're packaging like huge parts of javascript you know ecosystems which does anyone ever use any of those packages for anything and it just makes me think we have all of this stuff and some of it needs to be not part of the whole conglomerate there needs to be more differentiation going on so one of the things a lot of that stuff like the node packages and things when there are upstream maybe node is a scary example because their ecosystem is frightening but yeah rust is a good example basically each of these language ecosystems have a way that they work at managing their source code their libraries their dependencies those kind of things and what we do is we take their thing and translate it to our thing and call it adding value and sometimes it is and it definitely was 15 years ago when everything was terrible but some of these things are actually getting pretty good node may be terrible but the rust ecosystem is pretty great for rust it makes sense because you're compiling yeah okay well yeah so compiled maybe is a difference there so it is much much better it actually does vulnerability testing so node is actually learning painfully from their mistakes as well it's still 10 minutes nice it was I thought it was 10 minutes like 10 minutes ago I don't know how this time works yeah no yeah so I'm looking at like doing IOT stuff for my house there's a project called home assistant which is very popular and it's a python thing which should be easy to package up in fedora theoretically but the upstream actually releases a docker image that they build of the thing automatically and that's like their oh I'm sorry Dan Walsh tells me do not say that word they may call it a docker image Dan so you have to go yell at them they release a container image in some popular format that is what they support as their image and they maintain that if there's problems with it like there's that that's their tree and so if we would if we were for example to take all of those python things make them into rpms package them up and then make a fedora container image from it I don't think I could go through all that work I know how to do it but I would really like question whether I'm adding any value I just now made an alternate stream for this that is less supported is confusing to the upstream and doesn't so I don't know we need to find a better way to work with those things and I don't have the answer to that but like I said I don't have to have all the answers so somebody come to the council and solve this form the problem with the first part is that they typically don't interact very well with system package things and sometimes you don't need system package especially for python they don't interact very well with system package things in the upstream so yeah there's problems for sure can we imagine that fedora container that we use can we imagine making fedora containers that don't depend on rpms but are made from upstream source trees yes I can definitely imagine that that sounds very easy to imagine I am and I think there's also cases where we're talking to the people making flat packs right now we've got an awesome system run of people into ones talk that takes existing fedora rpms and makes flat packs out of them automatically almost automatically which is neat and will be a way for us to like seed that flat pack ecosystem with trusted packages very easily but that is also going through this extra hoop that doesn't really provide a lot of value there and actually causes some problems so it would actually be nice to be able to go from source to flat pack source to other container images source to OCI so I think that would be a great thing for us to move into in the future because that's where the world is going and if we're keeping trying to put people into our square pegs or whatever we want to call it yeah right I know it's very uncomfortable they're not cheap that way yeah okay everybody's walking away so I think there's a party the closing talk is intended so alright thank you for coming everybody see you in good time