LINUX Unplugged - 413: Community of Enterprise Linux

Episode Date: July 7, 2021

Linux server admins don't know where to turn next; how the cult of personality might be shaping Linux's most important market. Special Guest: Jack Aboutboul.

Congratulations! A big congratulations to the Linux Foundation. They announced a game engine today. Did you see this, Wes? I did. Very exciting. Apparently donated from AWS. Oh, very great. There's no version for Linux, though.

Not yet. I mean, they're working on it. Of course. I mean, at this point, it's kind of the corporate-friendly software foundation, right? Oh, hey, friends. Welcome back to your weekly Linux talk show. My name is Chris. My name is Wes. Hello, Wes. Is that a cheesesteak you've got there? Unfortunately, none for you.

No, no. I thought we talked about this. Well, this episode is brought to you by the all-new A Cloud Guru. They are the leader in learning for the cloud Linux and other modern tech skills. Hundreds of courses and thousands of hands-on labs. Get certified. Get hired. Get learning at acloudguru.com. Well, coming up on the show today, we all know it's true. We hate to admit

it, but the Linux server is what makes the money. It's not the desktop. In fact, it's where all the corporate-sponsored, for the most part, nearly all corporate-sponsored development time goes. But there has been a change, and something we've covered on the show before. And Linux server admins don't really know where to turn next, especially those that got themselves all caught up in the Red Hat ecosystem. They are stuck between some hard choices.

And if you go by apparent community size, Rocky Linux has a lot of momentum at this time. But this choice is a critical one, and we thought we needed to consider the entire picture. So last week, we ran a new segment on the show that we actually cut. And we cut it for completion's sake because we decided we wanted to play it in today's episode. And it provoked a strong discussion after the show that took place after we stopped the official recording. And we thought we should take both the new segment

and the response to that new segment and play it in one single episode for you. This week, we're going to play that cut information, that cut story, and that response. And we hope that after this episode, you're going to look at the Linux server landscape, which is exceedingly important for something that we all love. We hope you're going to look at it a little differently. Even if you're not an admin yourself

or in the Red Hat ecosystem, I think there's some questions that get posed that we should all listen to and all consider and stuff that those of us who've been using Linux for a long time know at our core is probably extremely critical. So we're going to get to all of that. We're going to have some community news. But first, we have to say time appropriate greetings to our virtual lug. Hello, mumble room. Whoa, there's a lot of you in there. Great to see all of you. Thank you so much for making it. Shout out to Popey and Wimpy.

We haven't seen you for a while. Welcome back, gentlemen. Nice to be here again. Absolutely. And we hope that you're having an excellent day and had a great weekend. I set off for a little adventure this weekend, and I have some good news and some bad news. It's going to be another little quick story time before we get into the show. Stay a while and listen. So I decided to spend the 4th of July out in the woods, let Levi kind of get away from the fireworks,

because the little pup doesn't really like all the explosion. And, well, I'll tell you the bad news first. The wife lady and I were heading off, going to go, you know, get set up, and we kind of just, even though we know better, even though we're not noobs anymore, we made some critical mistakes. And as we were leaving from our home base, we kind of were in a rush and didn't pay attention to a couple of things. And, you know, like a good boy, I thought I had everything figured out. I knew I was in a rush and I knew I was in a time crunch. So I went around and I tried to make sure everything was all put away. I collected all my

tech gadgets and put them in a little basket and made sure everything that was stowed need to be stowed. And I thought, OK, I was doing everything. And my wife, she's in a car behind me. I'm going down the road in Lady Joops in the RV. And I'm thinking this is going to be a great trip. Because we're going to test some stuff for the road trip to Salt Lake City in Denver, get an idea how the power system is still holding up after a year or so, really put everything through its paces. And I'm about 25 minutes or so into the drive, and I'm pulling off the highway into a gas station.

And I hear a rush of water behind me. And I turn around and I find an indoor tidal wave rushing down the center of Lady Chips. Like a wave of water coming at me. Thankfully, I was already pulling off the highway because I had to jump into action. So I pull off. I park the RV immediately. There's like nearly standing water in the RV at this point.

Total worst case danger zone for RVs. So I was rushing into the back. I can almost instinctively figure out like what's happened. I think I already know. And as I'm getting back there, I hear the which is our water pump as it's pumping away and pumping away. I open the back door. It's kind of like this one move. I open my back door and I flip off the water pump, which should have been off a critical mistake. My eyes knew exactly where to go. I look right at the sink that I'd put my tech basket. And I look up and my old 2013 MacBook Pro had slid across the counter. It runs Arch, by the way, and hit the water handle and turned the water on.

And it's a heavy old laptop, a big old, big old MacBook, you know, from the older days when they were thick and heavy. And it it slid off and hit the handle in just a way that it managed to turn it on. And it must have been going for almost 15 minutes because we managed to pump somewhere between 15 and 20 gallons of water into the RV. I mean, first it filled the basket, then it filled the sink, and then it just started overflowing and overflowing. And we spent about an hour trying to get all that dried up. And man, all of the gadgets are pretty much wrecked.

In like one fell swoop, I basically took out nearly every gadget I own. Not my phone, not my ThinkPad, but everything else essentially got wiped out in this one moment. It's like your own water-based EMP device and betrayed by your own MacBook, no less. This is horrible. My old trusty 20, you know, the one I walk, the one I'm like, this is how they used to build them with a good keyboard and ports and it runs Linux. And now that bastard, it wrecked everything. And of course it's still, it's still good to go. Oh, it survived. No way. Yeah, it survived. Yeah. it's it got a little bit of water on the screen, but I let it dry and it's working just fine. I can't believe it. Five days later, everything in the RV seems OK.

We got it pretty dry. We thankfully already pulling over. And my wife was in the RV with me within minutes, getting everything wiped down and all of that. So, wow. But just to have like, you know, like I had a fire tablet that I had been carrying around because I use it for home assistant and I was walking around turning off things and setting automations and I set it in the basket. So this guy just didn't want it to be loose. Gone. Wasted. All of it. Just wasted. So that's how I started my 4th of July weekend. But I do have some good news. Good news, everyone. While my slide issues are not fixed yet, and I was hoping to get them fixed before the Denver trip, thanks to Virtual Lug member Colonel's

help right there, he turned me on to the Vroom system, and I've contacted the creator of the Vroom slide system, who is an RVer like me, who had the same slide system I do, and it broke on his first trip. And he's a mechanical engineer, and he just said, screw it, I'm going to build a better system. And now he has. I don't know if very many people have ever installed these before. It looks pretty new. So I'm kind of rolling the dice here, but I'm hoping it's going to be like the final solution. But the issue is, you know, with the way the timing worked out with the upcoming road trip, we're not going to get in there until after the road trip. So that means all through Salt Lake City and Denver and until September, we're not going to have a living

room, but the road show will go on and we'll still have a great time. We'll still make content from the road. It'll just be a little more cramped. You want to come and stay in there, right, Wes? Of course. While you're beta testing a new slide system. No problem. Bring the dogs too, Wes.

We'll just have a big old sleepover. So, yeah, there you go. That's what's going on. People have been asking for an update. I wasn't really going to cover much, but that's some progress at least. I want to get to the community news this week, and let's start with IBM's president and former Red Hat boss, Jim Whitehurst, stepping down just about three years on the nose since the announcement of the IBM acquisition. And that president position at IBM is the classical position someone takes in the company before they become CEO. So a lot of us expected Whitehurst was in line to be CEO. In

fact, the board had arranged it so he would become president. So it kind of seemed like the board was preparing for Whitehurst while there was an interim CEO. And Wall Street seemed to like this idea as well, because when the news was announced on a Friday, IBM stock dropped 4%. But now IBM's chief and interim CEO is apparently becoming the permanent CEO. Aviv Krishna has managed to work at IBM since 1990, 31 years. How about that? Old timer, yeah. Yeah. And it's kind of wild.

The Financial Times, which sometimes has a paywall, depending on how many times you've clicked the link, but they have an interview with Krishna that is, wow, is this telling. He was going to be the transitional exec, but then it seems by what this Financial Times interview has, we'll have a link to that, it seems like maybe he worked with the board to work Whitehurst out and become the new CEO. And there's some quotes in here that are remarkable. And the article says that IBM is now doubling down on IBM's current path under Kirshner's leadership. Quote, this is Kirshner.

He's done amazing for us, but he wasn't going to be here forever. Speaking in the interview with Financial Times, he added, acquired CEOs normally only last a year or two. And then he went on to suggest that while he had been in the running to succeed him, that kind of any hope of that had been dashed as the company's board has now backed him in the next stage of IBM's overhaul. And then he points in the interview to the fact that the board had named him as chair as well as chief executive earlier in the year as a sign of that was backing him and that the shakeup now is about building a leadership team to take IBM where I'm leading it, he says. He's got an idea. He's got a vision. And it wasn't Whitehurst. It's his vision. And it's all about the cloud. This Financial Times interview basically

makes it sound like Krishna worked Whitehurst out. It does, doesn't it? I mean, it's quite telling. And especially when you think of Krishna being, you know, one of the people internally who was pushing for the Red Hat acquisition, you got to wonder, like, maybe this was part of his personal plan for a long time. And it seems to be working out for him, at least. Yeah, everybody always says that Krishna was the one that was pushing the CEO at the time for the Red Hat acquisition. But that, if you've ever worked in a big company, you know what I'm about to say rings true. Big companies are really good at grabbing onto a narrative that's out there right now and spinning it to make the new leadership seem competent. And so the Red Hat acquisition was still fresh. It was still new

when he was the new interim CEO. And I think it just seems natural to me that, you know, 110 year old company that knows how to message would latch on to a narrative like that and use it to promote the new boss, regardless of how true it may or may not be. So I don't put a lot of stock in the idea that Krishna was the man behind the merger. It's definitely possible. But I feel like that's gotten a life of its own that I don't completely agree with. I'd like to see some proof of it, because what I've seen here is a 30 year long company man who is well liked and has done good in the cloud division. But he seems very much like a traditionalist in this space.

The mayor, I know you want to jump in. I want to give you a chance to sound off. I helped Jim Whitehurst when he transitioned into CEO at Red Hat. And one thing, one time I was traveling with him and one thing that he told me was, and I think this is kind of the same thing that's going on here. When he was at Delta, he sort of thought that he was in line to be the next CEO as well. And then once it became abundantly clear that they weren't going to tap him for that, he was basically out. And he's just he's a very ambitious guy and he has his vision and he has his leadership style. And I think that he just doesn't want to, you know, he doesn't want to be in any other position aside from the top. So I think that's pretty much what happened

here exactly again. Yeah. And it's telling that three years on the nose, which kind of maybe suggests there was some deal in place. And when that had wrapped up and he wasn't going to get the job he wanted, he decided to leave. And I think that the mayor, thank you for that insight. He's if you don't know, because it was cut from last week's episode, but it'll be in this week's episode. The mayor is the community manager over at Alma Linux. And so he has some insight into this world. And when Jim was at Red Hat, after the acquisition, a lot of us in the community thought, OK, well, he's kind of like a firewall between a 110-year-old company

and a young, modern Red Hat that has to be super flexible. And I worry that with Whitehurst gone, that leadership firewall that defends the culture of Red Hat is has to be super flexible. And I worry that with Whitehurst gone, that leadership firewall that defends the culture of Red Hat is also gone. And I think that the traditional model of IBM is not compatible with the innovative model that Red Hat has to use. And I grabbed a couple of clips of Whitehurst from a TED Talk, just a couple of short clips, we'll link the full talk, that really encapsulate what I'm talking about here. And this is a clip of Jim Whitehurst explaining the culture shock that he felt just a couple of weeks after he left Delta and switched to Red Hat. It was not like any other traditional large company he'd ever been at. I began to realize something significant was happening my second week on the job at Red Hat.

I was in a meeting talking to the team about our strategy for an area called virtualization. For those of you who aren't in technology, virtualization was the thing everybody was talking about before they were talking about the cloud. And through this, I had the senior team and a bunch of the engineers kind of walking through our strategy and the what's and the whys of it. And about halfway through the meeting, one of the engineers, one of the most junior engineers, just took a pause and said, look, I know this is what we're doing, but this is absolutely the wrong direction. And he didn't stop there. He kept going and going and going about why we were wrong and how wrong we were. And I'm looking across the room, because this room had his boss and his boss's boss and his boss's boss's boss waiting

for a reaction. And there was a reaction. I would call it a point and counterpoint, which is a very polite way to say we had a huge argument. But at the end of the meeting, everybody walked out like this was normal. Now, I'm two weeks out of being in a very traditional organization, and I have to tell you, I was sitting here thinking, this is the most bizarre thing that I had ever seen. I would hazard at Delta, and frankly, most large organizations, and many of you know what I mean here, if a person to the CEO in front of their boss, their boss's boss, and their boss's boss's boss said what these people are doing is completely wrong, that person would have clearly been fired by the end of the day, and to be honest, maybe killed before the end of the meeting.

But everybody acted like this was normal. And I remember going home that night and talking to my wife, and I said, I have come into this most bizarre, strange company. I've clearly been brought in to clean it up. And it wasn't just this. A couple weeks later, I asked about doing a research report in an area. And a couple weeks later, when I asked about where that was, the team very happily and merrily said, yeah, we thought it was a bad idea, so we decided not to do that. So at first, I'm really like, we're just like, do I have like a mutiny on my side? But they're so friendly, and they're like positive about not doing what I told them to do. And so before I had a chance to impose kind of the traditional management structure, which I had learned so well and honed so well, I thought,

well, you know what? Red Hat's been really successful before I got there. I at least need to step back and observe. And so my management consulting kicked in, and that's what I did. And I have to say, I felt a little bit like an anthropologist tossed into some odd ecosystem, and I really had to hold myself back from wanting to inject myself. And I have to say, the first few months at Red Hat were, I think it's fair to say, shocking. It was just so different than anything I'd ever been used to. And I realized my role as a leader was going to look nothing like what I'd been used to in the past. My role as a leader would look nothing like what I was used to in the past.

And the other thing he says in there is that when I came in, I started to think my job was going to be to clean this up. I think that is such a typical manager reaction. And I don't think it is a typical reaction to think maybe I'm the one that needs to change. I think that's antitypical. And so when I look at IBM and I look at their leadership and I look at their culture and I look at the fact that they're a 110-year-old company, I think maybe people running IBM are not the right people to run Red Hat. When I look at Red Hat, they're a young, innovative company that's had to be extremely adaptable, that's had to respond in Internet time. IBM has failed to do that. That's why they don't have a stronghold in the cloud right now. That's why they needed to buy a company at the expense that they paid for Red Hat.

Like they were desperate. It was a desperate move. And Whitehurst touches on one more thing. He follows up on that story. And it's a demonstration of the kind of agility you need to have to run Red Hat. Let me come back to that story about the engineer. About three months after that initial conversation, so I was about three and a half months on the job,

that same team came back to me and said, you know what, it turns out he was right. Circumstances have changed in the marketplace, and we need to now go pursue this other direction. And by the way, we need to go buy a company to do it, and we need to go do that quickly, and so let's get going. And that all happened without me asking it to be revisited or from some planning process. It just emerged as circumstances had changed, and so we needed

to change direction. I think it's important that the people in that room, not a single one of them was apologetic or defensive, right? They weren't like me. They had been there for a while. They recognized that in a volatile, more dynamic environment, when circumstances change, you have to be ready to change. So success isn't about having the best ideas or making all the right decisions. Success is making a lot of decisions, but being open and transparent and non-defensive enough to be willing to change those decisions when circumstances warrant. That's what's key. And I fear, although I have no guarantee, I have no way to actually know for sure, but I fear that perhaps IBM's leadership won't be up to that particular kind of leadership

challenge. And if that's the case, it gives me concern long-term about what kind of impact that could have on Linux and particularly some of the projects that Red Hat's involved in that are the harder bits of work on Linux. I want everybody to get a chance to jump in. Wes, did you have any thoughts on that before I go to the virtual log? It's sad to see a strong and eloquent person who can kind of describe open source and

how open source can work with business and all of the interesting insights that you just highlighted. You know, that was a big voice within IBM, or at least people that had to take him seriously while he was there. So I think you're right that there are at least fewer of those advocates now. Right. And I do worry that as mergers go, that could be a bad trend, but maybe not. We'll see. Perhaps IBM will simply just have too much going on to care what Red Hat does. Neil, what are your thoughts? So I want to bring up the counterpoint that you can't be 100 plus years old and not be able to have an adaptable bone in your body, right? Companies don't survive for more than a century by staying still. They learn to react. They adapt and do things. While I'm a little concerned personally about Jim Whitehurst leaving, there are all kinds of unrelated reasons why he might want to leave. But it might also be, you know, the great resignation effect. There are lots and lots of people, you know, changing their focus in life. But I will say that about about IBM is that being over 100 years old means you have to make pivots and changes and adapt at a level that most companies can't do.

And I haven't met Arvind Krishna. I don't know anything about him. met Arvind Krishna. I don't know anything about him, but I would imagine that being at IBM for 33 years, that means that he's seen IBM shift in major ways over the past three decades from doing mainframes to personal computers, to software services, to business consulting, to the cloud and managed services and all these sorts of things like there's a ton of pivoting going on in those three decades maybe they weren't all successful but i think the more important aspect is that ibm is still here and they still keep trying and yes there's still some cause for concern i'm a little worried myself because

you know red hat is a fantastic leader in free and open source, and they do a good job of adapting to the realities on the ground and working to provide high-quality solutions. But that's not to say IBM can't do that, too. I think it's more of a question of whether Arvid, as being the CEO of IBM, will actually learn from the Red Hat folks about how he should handle the Red Hat business unit. Because, you know, like it or not, Red Hat is a business unit of IBM, semi-autonomous and could run on its own. It has its own legal, has its own HR, has its own finance, has its own executives.

IBM runs it somewhat independently, like a lot of the other business units. That's why they're business units. So I would be mildly concerned, but I'm also more of the wait and see camp and we'll see. We'll see how it goes. Like IBM has been around for over a century. They've got to be doing something right. Yeah, I think that's pretty reasonable. It's a reasonable position and it kind of makes me hopeful that they will let Red Hat be Red Hat and continue to do their thing because I think the whole ecosystem thrives when we have really strong, strong companies that are like, I think of,

I think of canonical in that group. And I think of Sousa in that group. And I think of Red Hat kind of in this class of companies that are doing some of the really hard work that has to get done to make Linux a contender. And we just want as many companies in that mix as possible, I think. And I hope that Red Hat can continue to do that. And so that's why many companies in that mix as possible, I think. And I hope that Red Hat can continue to do that. And so that's why what we're about to talk about is just,

it has so many knock-on ramifications. And this is part of, I think, why the choice has gotten harder for server admins. And I go back to just how critical the server space is for Linux. So it's with this change in leadership that we're going to continue the conversation in this episode. But first, Linode.com slash unplugged. Go there to get $100 in credit for 60 days on your new account and support the show. Linode's also making our road trip possible for our team reunion. We're going to all get together in Denver. You're invited to Salt Lake City and Denver to meet up with us. Linode's going to be there in Denver. They're going to have a spot. They're going to be giving out some swag and perhaps some treats and goodies too. So if you're

going to be in the Denver area, come out and say hi to Linode. They'll join us out there because they are helping make the trip possible, including some of the tech we'll be using that we'll be talking about in a little bit. I don't really actually, I don't know when yet, but I'm working on it. I'll tell you soon though. And we'll tell you how they made it possible. So go to meetup.com slash Jupiter broadcasting for that and go to Linode.com slash unplugged to get the $100 in credit for your new account. We're hosting everything for JB 3.0 on Linode. I started using them almost three years ago now, personally, because everything for JB was on different providers and different infrastructure. And I wanted to set up something for myself, you know, for like game servers and SSH and just like my own set of things.

And I had seen Linode around forever making community events like our team reunion possible. And so I knew that they were in the community events that were like actual real true community events, you know, not just like events that are put on by like large corporations, but the whole range of ones were like, it's just a small community and Linode had a booth there. And there was an event I was at where like, they were the one booth that was like really kind of taking the event seriously and had good stuff and making great conversation. Well, some of the other vendors weren't even staying at their booth. They were just like off snacking on barbecue.

And I remembered thinking like, you know what? Linode's really put an effort into this. And they had a good looking booth and they had it down. And I thought, I'm going to try them out. And so I did. And I didn't get $100 in credit. I paid with my own monies. And I loved it. I thought they had the100 in credit. I paid with my own monies, and I loved it.

I thought they had the best dashboard out there. It was nice and clean. They had the fastest VPSs out there. They got 11 data centers to choose from. They've really got the system dialed in. People often ask, like, what's the number one reason you switched to Linode? And I suppose maybe I might say it's probably because of speed.

Maybe. But now that I've used it for a while, it's really because I trust running our production stuff on Linode. Like when it has been crazy hot here in the Pacific Northwest recently, I am really glad that the stuff that makes it possible for you guys to get our shows and all of the services that support the production of our shows, I'm really glad they're up on Linode. And, you know, you can get a system for like $5 a month,

and they're 35% to 50% cheaper than the major cloud providers out there. And it was like 120 degrees recently in the studio garage. I just had to shut down our NAS server. I just turned it off. I turned off the modem too. I mean, it was like 120 in there, right? Meanwhile, the Linode systems are sitting there and they're perfectly temperature controlled data centers

with their 40 gigabit connections just humming right along. You know? Like, there's a lot of reasons. So go try out the $100 and, you know, see what you get. You know, play with it. Just experiment. Like I did originally. I was just curious.

I wanted to try it out. But now you can do that with $100 for 60 days. That's great. And you can really put something up there and really bang on it and see what it can do. So go to linode.com slash unplugged. Get that $100 for your new account and you support the show. There are a lot of different options for hosting, but Linode does it all and they do it better than everybody else, including their object storage, their interface, and their

fantastic customer service. Linode.com slash unplugged. So let's get into the story that we cut last week. We didn't cut it because it was like so spicy and, you know, Wes, I mean, Wes did swear a lot. That happened. Yeah, I know. I came in a little hot. Whoops. You get fired up about enterprise distros.

That's your thing. I do. Yeah. And we actually were just going to play it on the show as normal because we were just talking about Rocky Linux's 8.4 Goldmaster, and they had hit their main release, and that was a good moment. We were happy for them and thought, okay, let's get it in the show.

We'll have this. We'll document the moment they actually hit the release because I felt for a long time that the long tell is what's really going to reveal if these enterprise distros are standing up. What I mean by that is it's like a three- or four- four or five year game before you're going to have a really good sense which one of these is a RHEL replacement. Right. This is the enterprise we're talking about. But we wanted to cut the segment anyways because we felt it was better in completion to play the segment itself.

And then the reaction we got a little bit after that that we felt was like a really good transparent conversation that was informative. And so we started the coverage. I'm going to play it for you now, talking about the initial response that Rocky was seeing from their release. Yeah, sounds like within 72 hours of that launch, Rocky's assets have been downloaded nearly 70,000 times. Hey, that sounds like some excitement out there. Yeah, and that's from the mirrors that they can get numbers from. That doesn't really count trackers for torrents or maybe other mirrors that are just providing downloads. This post also took some time to clarify a little bit about the structure of the Rocky Enterprise Software Foundation, which is a public benefit corporation

formed in Delaware. Where else, right? Backed by a board of advisors with access control policies that utilize the principles of least privilege and separation of duty so that no action can be taken unilaterally, not even by the owner, Gregory Kertzler. Sounds like there's some work here to try to say, look,

yes, we're trying to do this above board. Here's how our structure is. You can decide if you like it. Yeah. And so there's two key bits of information in what you just told us there. Number one, they're a public benefit corporation. Now, that's different than a charity, but it's notable.

We'll come back to that. But it's notable. We'll come back to that. And number two, they are clarifying that Greg is the legal owner of that foundation and of Rocky Linux. And so the way a public benefit corporation works is after you've spent the money on what the project is, the additional remaining revenue, the way I understand it, goes to pay out the staff, including the owner. That's just important to note because when we're looking at these different distributions, there's going to be several of them, we need to understand how the entire structure around them works. How does the support model work? How does the public governance model work? And what is the revenue angle for each participant involved? And so we start now to get some insights into what maybe could be the revenue model.

That doesn't mean that Rocky Linux isn't great and that it's not going to be a fantastic distribution because don't you want the people involved to be living happy, well-off lives? I mean, if they create a well-loved product that is successful in the market, it seems fair that they should have a comfortable living from that. So I don't really have a problem with the structure, but I felt like we just didn't have a lot of clarity in the past around the structure. And now we do. And now I feel like all the cards are out there on the table. Well, now it matters, I think, how they actually communicate, right? We've got this little clarification here, but really what it comes down to is like, if you're asked to give money to this foundation,

if you're trying to support things, is it still made clear to those in the future, right? That's a great point, too. And as long as we're all clear about it, I say more power to them, move forward. And I hope we continue to see this pattern of transparency. And I think we've also witnessed a bit of a tweak in the Rocky Linux's approach to community.

They seem to be pretty passionate. I've heard from a couple of them who wanted to kind of elate some of my concerns. And they approached me with a very reasonable, friendly dialogue. And, you know, we exchanged a bit of it back and forth. And I had a good experience there. And so I like that they're taking those steps. But at the same time, we have to sit back and kind of watch and see how all commitments are met, how this thing plays out for a long time before any one particular winner is declared as the traditional CentOS replacement.

Because really, the game is still afoot. And this is just the starting line right now with Rocky Linux hitting version 8.4. We're just getting started. Yeah, we'll see you in like four years, I guess. Yeah, maybe, right? I mean, Enterprise 7? I don't know how long you wait. I think you just sort of sit back and watch, I suppose. But it is a tricky one.

Now, you might not expect it, but that topic provoked quite the response that we thought needed to be in the show. Alma Linux's community manager joined us and shared his perspective after we wrapped up last week. So the following discussion took place on the live stream after the show.

And just a note, we have reached out to Rocky Linux. I emailed their community contact and invited them on the show. I haven't heard back from them yet, but that invite does remain open. I think that it's kind of important to note that the situation there seems a little bit problematic

because you're not any better off than when Red Hat owned CentOS, right? I mean, it's still owned by one entity, by one person, and they can decide to do whatever they want to do with it. Now, they can say that, yes, we have a board of advisors. By the way, advisors are not directors. So it doesn't mean that they really have any particular say in what happens. And then, you know, they can decide to do whatever they want with it tomorrow.

So you're not any better off than when Red Hat owns Santos. I think that's a very big problem. Yeah, I suppose people would just argue that they have more faith in the mission of the Rocky Foundation than they do of Red Hat to steward it, I suppose. I guess that's their decision to make. But, yeah, it's a fair point, right? So in the contrast with Alma Linux, just for an example, how would the decision to end producing the distribution came? Say it's 25 years down the road and everybody's using Fusion now. How would that process work with the governance

model of Alma? So we have a board, a governing board. It's an open board. We're going to announce additional membership guidelines so that additional people can join that board. I think the plan is to ultimately have that be at 13 members. So it'll be a nice mix of people from within the community versus corporate interests as well. And really just anybody that has any interest in the distro. And then, you know, let's say 25 years down the line, we decided that we want to stop making the distribution. That's a decision for the board to make. You know, I'm assuming it would be proposed and there would be a vote. And if the votes came back, you know, the majority or whatever it is that we decided to do to vote on it, and that's what the board wanted to do, then, you know, things would wind down. It's a vote process, basically there. Yeah. I mean, it's a real

community process. They're people that have an interest, have a say. They get to decide what happens. We're all in this debacle because that's not how Santos originally was or not originally. Well, you know, once Red Hat took over, that's not how it was. There was one party that held all the keys and that's not really changing at all with Rocky's governance model. So some people might say, you know, I think it's great that they're trying to make money off it and that this is the way they set it up. But I just think that, you know, as a community, we need to look a little bit more broadly and think about what we value. And to me, at least, that goes back to community ownership, stakeholder ownership, the community gets to drive and decide what the future is. That's really a very, very important point that I think many people may not realize yet. And they may be, you know, purposely turning

a blind eye. As you touched on there, doesn't the composition of that in a manner, I could imagine a version where the corporate interests were dominating that if they had the voting control. I can kind of speak a little bit about this because the OpenSUSE project is structured similarly, though it is of course owned by SUSE and not legally separated, blah, blah, blah, blah, blah. The board is actually set up in such a way that it is not possible for a single corporate entity to have majority membership in the board.

And whether that's good or bad, one thing that that rule basically preys on the thought pattern that companies don't actually like working together very much. They have a hard time figuring out how to collaborate at a business level. And by making it so that there is not a path for a clear majority for any particular company, they would essentially have to build a coalition to make something to happen, which means effectively the community would have to have a say and would have to agree to it anyway. Is there kind of a protection against one particular entity, perhaps Cloud Linux, having significant influence on the board like that? Yeah, so we're actually in the process of putting together the membership guidelines for participation on the board. And that is definitely one of the principles there.

I mean, our plan is to completely dilute Cloud Linux. Like I know when things first came out, like everything in the press said, Cloud Linux, Cloud Linux, Cloud Linux. Well, that's because Cloud Linux started it. But the plan is to totally dilute Cloud Linux and exactly like Neil was talking about, it should be based off consensus

and based off what the community wants. So that's totally what we're planning on doing. OK, so it sounds like it's in it's not fully firm yet. And then it's but it's getting worked out. Let's say the board now was always meant to be provisional because, you know, we need to. First of all, we need to form a structure. You know, paper legal documents need to be filled out, bylaws need to be set up and accepted. So that was all meant to be provisional while we work on the membership, essentially. And then once we have the membership

in place, that's when we'll have elections and we will, you know, invite everybody to participate in those elections. Okay. I track your point earlier too. I think when you see the reaction that we saw when Red Hat announced that they'd be discontinuing the traditional release model of CentOS, people reacted in a way like they were shocked at the arrangement. Like all of a sudden, like all of a sudden the reality of the arrangement hit them right in the face. And it does feel like one of the lessons learned when you're looking back at that initial shock is,

oh yeah, there was a single entity that was in total unilateral control ultimately that could make decisions like this. And if we are going to move to something else, and I think that's why a lot of people looked at something like Debian, even though it's a significantly different, you know,'s why a lot of people looked at something like Debian, even though it's a significantly big transition. I think people looked at Debian and said, that's a governance model that perhaps insulates from these kinds of surprises.

And I think, so your point that you're making is, if you're going to move to another system besides CentOS, and you're looking for CentOS-like compatibility, the governance model is something you need to consider, because that was what ultimately bit everyone in the butt originally. And if people are able to look at the board structure and see that while Cloud Linux is involved, they don't necessarily control things, that I think could be a really big deal. But you're right in the sense that you have to make people aware that it's something they even need to consider. Because right now, what Rocky Linux has, I think, going for it in a pretty powerful way is this narrative about one of the original co-founders. And because an original co-founder of this original idea is involved, that makes Rocky Linux significant.

Now, I can't really throw shade on them for leveraging a positive spin, a narrative that they've been given. Like that's, you know, that's, they've been given a gift in that regard and they're using it. And I can't really give them a hard time for it because that's, you know, probably what anybody would do. Yeah, exactly. Well, I don't think we should give way to any spin because I think the truth is the truth and i don't want to get into that right now but um there's certainly been a lot of spin which um i don't i don't want to um that's not what i'm about i'm about building bridges and about doing the right thing and i think that for anyone that cares about free and open source software and you really care

about this movement that you know took close to two decades to build to get to where we've gotten i think you need to really take a solid look at what are the leadership structures what are the ownership structures and you know ask yourself what does that really mean to me you know if you're all about just getting free software and just taking it and running with it and doing what you want to do with it, then I hear that argument too. But I think if you really care about things, you need to take a deeper look and really ask yourself what values you stand for and what this all means to you and then make a decision. Yeah. And like, this is something that has, you know has bothered me a lot about this whole CentOS conversation as well. Right. So I work in Fedora as it is today. In general, I am fairly comfortable with that because if you understand the dynamics of what's going on in a particular, in the

developer side of the community, then it tends to be fine. Like, so in Fedora, like we don't really have those conversations anymore about Fedora being, you know, its governance being problematic or whatnot, because it's clearly not. It's clearly been fine for almost, you know, 18 years now. It's not an issue anymore. For CentOS, the thing with them is throughout most of its history, and I don't want to name specific examples, but throughout most of its history, it has had a very troubled governance, even before Red Hat. It has had a very troubled governance. It's been a very difficult project to maintain and bring up. And, like, you know, you can go back and see the specific examples.

But today, you know, something that's been consistent with the CentOS project is that the CentOS project wasn't a project for contributing to. Like when Lance Davis created the project with the others, including, you know, Greg, they made it really to be a platform that they would consume to bootstrap Chaos Linux. And Chaos was supposed to be the real platform in the end. And nobody was interested in Chaos. And so CentOS won out because people just wanted something that they would use and didn't really want to contribute to. And that particular mindset stuck through as CentOS broke out of Chaos. And that also opens up the question, like as Jack said, you know, what are your values? What do you stand for?

What are you trying to achieve here? Right. The biggest problem that I've always had with CentOS, the reason why it was never interesting to me up until a couple of years ago, is that there's nothing I can do with it. It was not meaningful to me as someone who wanted to make the free and open source software better and more accessible and all those sorts of things. The CentOS stream effort changes that dynamic. It makes it possible to do that sort of thing without really breaking the stability promise that they offer. And the problem that I have with how all of this stuff has been going down is that nobody is thinking of it in the sense of what

would make the community healthy. It's all about, you know, what did they screw up for me? Because it is not sustainable to have one and maybe half a person building an entire distribution by themselves for everyone to consume and then complain about when something isn't exactly right or whatever. Building a Linux distribution is super difficult. And I think people had expectations that they shouldn't have had in the first place. And I know from personal experience that the substitution effect is real. And that substitution effect has created a lot more damage than I think people actually know. Like, I don't want to get into specifics about it because that's really complicated and kind of gory. And that's not what I'm here to talk about. But I think this, as disruptive as it

is, is a good course correction. Like, I sit in both the Rocky and the Alma communities and as well as, you know, being in CentOS, working the hyperscale SIG and stuff like that, which hyperscale SIG, a thing that was never possible before now. Like, that's a huge win. You know, I've talked to the Alma Linux folks, Jack and many others, and something that stood out to me was Cloud Linux and other people participating in the Alma Linux community, first and foremost, are actually engineers that know how to solve problems and they write code and they have a business that supports all this stuff. And they have set up themselves so that they will be able to contribute directly to CentOS Stream and to RHEL whenever something is broken with something from Red Hat. And that is something we have never, ever, ever had with a RHEL clone.

Ever. Really a new era, huh? Guys, that's a driving principle for us that everyone should know that. I think that there is a ton of good that can be done now that could not be done before. And I think that we're not only looking to benefit from, you know, the downstream portion of the relationship, but we really solidly feel, and Neil has seen this and he's spoken to people about it, that needs to be a driving force about what a good CentOS replacement distribution needs to become in order for the community to maintain its health.

And I think that that's one of our guiding principles. I hadn't really considered the fact that your role could be a significant contribution to what ultimately will land in RHEL. And by doing so, you'll be able to solve problems for not just yourselves, but other RHEL users. And because the way the CentOS stream model now works is people can now make an impact at that point that will ultimately end up in RHEL. And that being open like that, I think, is the aspect of this transition to CentOS stream that the community hasn't realized yet. I think until we've moved past, like, December 31st or whatever the end date is, I think the community just won't wrap its head around the fundamental shift that's happened here. And so, Jack, this is my analysis, if you will. Number one, Rocky Linux got a great start with that narrative. Number two, Alma Linux, just because of the timing and just how things just naturally played out, really initially got brand recognition

as Cloud Linux. And I think that's a potential headwind. Then you combine that headwind with people haven't actually realized the value of having significant engineers and contributors in the CentOS stream ecosystem yet. And so it seems to me like it's just going to be one of these, you'll have to grind it out and get there and sort of earn attention over time because it's it's sort of the initial cards that were dealt perhaps aren't as favorable as they could have been but ultimately when you look at the entire picture it seems clear i think what will draw people to alma linux but it seems like it's going to take a little bit for people to appreciate all

the value you're definitely true in a certain sense. And I think that the whole spin with the press was a little bit one sided. But I mean, that's nothing that we can control. Right. We can only do what we can do. And, you know, I think on the on the other end of the spectrum, you know, we certainly have a great narrative as well. I mean, like we have core team members that have been rebuilding rel for almost 15 years. Like that's, you know, that's pretty, pretty significant. We have a ton of experience. Like we were able to rebuild eight, four in like three and a half days. That's phenomenal. You know? And I think that's one element that was also missing from the narrative overall. And, you know, I'm very comfortable with us having to earn our keep. I think, you know, we're not trying to rest on our laurels. We're not trying to spin anything. We're not trying to, you know, have people trust us blindly. None of that.

I think we're here to work. And the best thing we can do is continue doing that good work, continue being members of the community, continue building for the community, and just earn people's trust and build bridges. That's what we're all about. I think now CentOS has earned its name more than it ever has before. It is the community enterprise operating system because the community has the ability to contribute, develop, and deliver an enterprise Linux platform. We've never had that before. And I'm going to just spread a little bit and talk about OpenSUSE too. OpenSUSE Leap is doing the same thing. Now we have both top tier enterprise Linux platforms moving their models

so that they're more community centric. The Red Hat one is a little bit more painful because the user base is just worse in the sense that they are not used to actually working in the project. And now that's going to have to change. And CentOS stream is fine, even for the average person, not really doing much of anything with it. But it's also now, if there is a problem, there's an opportunity to fix it. That is powerful. That is important. That is how free software survives. To Neil's comment about it being actually honoring its name and being a community enterprise distribution, now we finally have a setup where all real derivatives can actually work together in the CentOS community and make the entire ecosystem better, not just their own derivative. That's something we've

never had before. And also following up from Neil's other comment about just having a handful of people working on CentOS, that's what a lot of people don't seem to understand is that there was never an incentive for RHEL to staff up the CentOS engineering effort at all. It was, you know, they were kind of competing with themselves in a way. But now with the transition of making RHEL derived from CentOS in a way, now with the CentOS 9 effort, you're seeing all of these RHEL teams come together and work on this. And there are orders of magnitude more people getting involved working on it

than ever before. To me, that's crucial to CentOS's long-term health, even though it's a change that not everyone likes. That was just the start of the conversation, really. The community of Enterprise Linux is changing, and I think coming out into the open more than ever before. We hope Greg, or anyone from Rocky Linux, will take us up on our invitation and join us on an episode to chat about this more. shop local in the Linux community when you can. And Entropy is a development shop that puts desktop Linux first. They're an industry-leading mobile and web development studio based in the UK with

global clients including Mars Wrigley, Walmart, Mediacom, The Financial Times. I mean, they're just trusted by the biggest names in media and content. But it's not all about those big global corporates either. Entropy loves working with smaller businesses and startups to realize some truly outstanding mobile desktop and web applications. They can do it all. And yeah, they run it all on Linux and open source. They're a team of distro hoppers and Linux hardware nerds. Everything they do goes back to love for the Linux community. Some of their key tooling includes Ruby on Rails for web development and Flutter for mobile and desktop. Yeah, yeah, that includes Linux desktop. They've pioneered some of the most complex Flutter apps ever built, and they're shipping globally.

They've integrated machine learning and image recognition at the back end of these Flutter apps. They really know their stuff, and I think they're pretty excited about where this all is going. With Flutter, Entropy rapidly builds apps for iOS, for Android, and for the desktop platforms like Windows and macOS, and of course, the Linux desktop. And they can do it in no time. In fact, they're already one of the global leaders in Flutter development with dozens of deployments in the last year alone. So, if you're thinking of taking the plunge into Flutter for Linux,

maybe for Linux desktop or for the other platforms, or if you're a small or large business that just want to get into the game with an outstanding app on any number of platforms, Thank you. even ask them about their latest favorite distro. They love all that stuff. So to get started, Entropy is offering listeners of Linux Unplugged 15% off their first project. A mobile app, a web app, or yeah, build a Linux app, anything. You'll get 15% off. Use the promo code LUP13. Just head to linux.entropy.works, fill out the contact form, use promo code LUP13, and they'll get in touch. And there's never any pressure. They're always happy to just discuss your project, chat about Linux, or maybe even chat about your favorite Linux podcast. That's linux.entropy.works, promo code LUP13. And a big thank you to Entropy for coming on here, being part of the community, and helping us all shop locally for our app development. linux.entropy.works.

Promo code LUP13. Well, speaking of Flutter, we've had some feedback. Mr. Light Fixture wrote back in. Hello, Chris and Wes. My initial response to Flutter as a whole had kind of been, oh great, another one of these. But as time goes on, I'm warming up to the idea of a ubiquitous cross-platform toolkit for designing apps that isn't dependent on the browser. Because yeah, definitely better than Electron and that whole mess.

My first time hearing about Fuchsia, I was skeptical. The discourse at the time was about replacing Linux as a new developer darling. And so I sort of went on thinking about Zircon as how it would stack up directly against Linux. For a long time, I thought of it like, well, how could the Zircon kernel possibly catch up with the software integrations and hardware compatibility in Linux? I mean, enough to supplant it anyway. But as you said on the show, they don't really have the same goals. Linux kind of tries to be everything for everyone. Zircon was never really meant to be a drop-in replacement. It has no intention of supporting everything that Linux supports. It has no intention of supporting everything that Linux supports.

So while this new platform diverting developer time at large companies away from Linux is definitely a shame, I'm not sure I see a future where Linux will ever really be neglected. If things continue like this, Linux's quality of trying to be everything for everyone, I think, is almost guaranteed to have developers far in the future motivated to keep it alive, because it'll be the only thing that does support all this esoteric stuff that just isn't profitable for the other big platforms to maintain. In the past, many technologies have died because they lost or maybe never had a common thread from

which to maintain compatibility with what came after. But today, I think we're assured in a way that we'll still be able to support so many things, as long as we still have Linux. Yeah, I've been shifting my opinion on this too, Wes. Mr. Light fixture points out that, like we talked about last week, we kind of started kind of skeptical

and have started to shift, and part of it was when we were playing with Fuchsia, but the other part that honestly made a pretty big impact on me. And I think it's telling is I tried it myself. And I think that probably had trying Flutter and building a stupid little hell hello world app made me realize, oh man, if people had access to this,

we could actually get some pretty good, pretty quick applications that weren't all Electron. And that I think is like the, like the underlying thing that was really motivating me. And it's, and I don't actually still feel like Flutter is a perfect solution because some of the things that it relies on are getting a little long in the tooth. However, that seems like a solvable problem to me. And the fact that you have Google and Canonical behind it, it makes me reflect on that conversation we had with the Lutris developer.

And you recall he said, for a little while, it was like Canonical was really leading with the Linux desktop. And so you just kind of followed what they did when Unity was kind of at its peak and developers didn't really know how to come into Linux. But I remember I'd have these conversations with Mike from Coder Radio too. You know, you could kind of look at what Canonical was doing design-wise with Unity and take clues from that.

We kind of lost that recently. And he's still like in that mindset, the Lutris developer. He didn't know where to go from here. And I thought perhaps we're seeing Canonical recapture a little bit of that mindset and a little bit of that direction with their announcement and endorsement of Flutter.

And then when I was talking with Entropy about the applications they've been building in Flutter, I started to realize like they're actually like building custom apps for businesses that want to like to have some data workload or something like it. Maybe it might be in this case, it was machine learning and image recognition, and they want to be able to run it on Mac, Windows and Linux because they have all three in their environment. That's not a necessarily uncommon thing. Like, that could be a scenario that's kind of common in the enterprise environment.

You've got multiple, multiple OSs. Maybe it's just Mac and Windows, or maybe it's just Windows and Linux. And you've got mobile in there, too. And you want to create something that's essentially native on each platform. You look at your options out there. You've got Electron,

but I don't know, I feel like things are awkward in the Electron world. Microsoft seems to be in love with their edge-powered web view thing, and I'm kind of done with Electron apps. I have been tolerant for a long time, but I'm kind of ready to have something smaller, faster, and lighter, and something more approachable. So I agree with Mr. Light fixture. I think it's worth reconsidering. No, I think you're right. I mean, there has been this question of, faster, and lighter, and something more approachable. So I agree with Mr. Light fixture. I think it's worth reconsidering.

No, I think you're right. I mean, there has been this question of, well, you want to make a desktop Linux app. How do you do it? There are a lot of options. It's hard to feel a sense of any momentum in one particular area. And then you're going to have to go dive into that thing. Maybe it's GTK. You're going to go in the Qt route and figure out how do I make cross-compatibility work

and which versions do I target and how do I get it to run? And Flutter has a lot of answers there. And yeah, there are probably a lot of old-timers or traditionalists out there, to some extent myself included, that view it with some skepticism, especially because of the big names involved. But if there does sprout a legitimate community

of open-source users developing apps and if it makes these apps available on Linux, I think it could be quite the pragmatic win. Something has to give. I look at what Microsoft is doing, and they've opened the floodgates for the Windows App Store. Any app, any payment model, any way you've built it, you can sell it now in the Windows App Store, in Windows 11. And then I look at the Apple model, and it's not as open, but it's, if you want to build an app for the iPad and the iPhone, you can also now ship it on the Mac, and you can run iPhone apps on the Mac. And so it opens up for developers, when you're looking at potential customers with wallets,

it opens it up to a lot of customers. I don't really see anything that really enables that for Linux right now, without some sort of enabling technology like Flutter. I'm curious, Wimpy, if you have thoughts, maybe I'm way off on this topic. Yeah, obviously, I was involved in the early sort of stages of canonical working with Google on Flutter. And a lot of what you're saying is obviously very true. flutter is the fastest growing developer platform right now and it provides an opportunity to deliver a developer experience on the linux desktop that has a lot of developer traction behind it and a lot of interest and i think it will become what electron has become you know electron has been a great equalizer i mean you were just saying that you know you're you're over electron

but because of electron we have a lot of applications on linux that we wouldn't have had were it not for that cross-platform compatibility that it provides, even though it requires some sort of heavy lifting. You know, even things like Steam are based on Ceph for all of the UI, which is the forerunner to Electron, and Spotify uses Ceph as well. And love it or hate it, you know, big brand apps like, you know, Slack and what have you and Discord have enabled computing parity on the desktop that we've not had for a long, long time. And I think Flutter will be the extension of that because there are so many large organizations building their apps in Flutter now and they can build their apps in Flutter for mobile

and desktop and web. And I think Linux will continue to see fair treatment as a result. and they can build their apps in Flutter for mobile and desktop and web. And I think Linux will continue to see fair treatment as a result. We have seen more games when game engines support multiple platforms, and we've seen more apps as Electron has come around. But Popy, I'm curious if you think maybe we're trading one evil master for another, because it's always possible that Microsoft whips Electron into shape with their edge rendering engine, and maybe they'd be a better overlord than Google is with Flutter. I think it's not unreasonable to hedge our bets and diversify a little bit. You know,

we have toolkits ourself. You know, we've got GTK, we've got Qt, we've got others to develop our own things. And yeah, I don't think it's unreasonable to welcome these other, as you say, corporate overlords at arm's length at all and make it possible for developers. Like Martin said, these technologies are popular, insanely popular. And I think it's rather blinkered not to open the door to have those technologies enabled on the Linux desktop. Because if we don't, then we will continue to be an also-ran and we will continue to have no killer apps for the desktop. Because right now, we don't have any. And we don't have robust professional toolkits to enable people to develop them in my mind.

And I don't think it's going to take away from a native GTK application, if you will, or a Qt application, because the developer that's going to choose Flutter or Electron likely was never going to choose those other toolkits simply because of the requirements of the application. Maybe. I mean, without speaking to each developer, I mean, you can certainly surmise that it's not a, it's not an unreasonable and outrageous accusation to make that someone probably would choose a technology that's backed by a very large company with a lot of comprehensive developer documentation and mindshare and developer

advocates over something that's been around for 20 years and keeps changing and is not well documented and there isn't a huge community. And by the word huge, I mean different from the size of the Linux community, like much larger. So I don't think it's an unreasonable suggestion to make that the people would go for those technologies, just the same as developers went for Electron, because it was quick and easy to bootstrap up an application.

You know, okay, negatives aside, it was quick and easy for people to get started. Yeah, really. I mean, I do backend development, mostly some web development. I know not that much except as a Linux desktop user. And Electron is the thing I've used to make handy tools for myself. And I could see Flutter really being that because there's not a ton of money in desktop Linux. I can't take the time to become an expert in doing that

unless I want it to be my main hobby. But if there are tools that are well supported that let me leverage what developer time I have and also get it on the Linux desktop, that seems like a nice tool. We want to get your emails, your feedback. Let us know what you think on any topic or what you'd like us to cover.

Linuxunplugged.com slash contact. Last week you asked us about Landlock and we responded and that coverage is in this week's Linux Action News along with our thoughts on the situation around Audacity and several other stories in the world of Linux this week at LinuxActionNews.com.

So really in this, I just wanted to say just the basics. We'd love to hear from you at our contact page. We have the Telegram group where the conversation is going all the time, jupiterbroadcasting.com slash telegram. And last but not least, the Leplug. I popped in there for a little bit last Sunday as I was out for a walk. We had a great conversation in there and the Leplug goes every single Sunday at noon Pacific. We have it on the calendar that gets converted to your local time at jupiterbroadcasting.com slash calendar.

And if you'd like to join our mumble room, get in the conversation as they're happening. We have information for our mumble room at linuxunplugged.com slash mumble. The show's live on Tuesdays at noon Pacific. And that's also at jupiterbroadcasting.com slash calendar to get converted in your local time. and that's also at jupiterbroadcasting.com slash calendar to get converted in your local time. Well, earlier Neil mentioned the CentOS special interest group for hyperscalers, and we wanted to follow up on that a bit here in the pick section. Davida Kvelka of Facebook presented at this year's CentOS Dojo at Fosdum

all about this SIG's efforts, why it's being founded, and why Facebook wanted to get involved. CentOS runs everywhere at Facebook. It runs on all hosts and all the containers. 75% of the fleet, as of yesterday morning, is currently running on CentOS Stream 8. The rest is running on CentOS 7 and is in the course of getting updated.

We use CentOS as a stable base. And then for components that we want to track more quickly, we take backwards and we update them as needed. So for example, we do a lot of work internally with systemd. So for systemd, we will have an internal backward that we maintain that tracks systemd master and that we deploy. We also use things like Appel

to get access to packages from Fedora and we contribute to Appel when needed. We don't run the right kernel. We run the upstream kernel from Linus with an internal configuration for our hardware, and our kernel folks try to contribute to the upstream kernel directly as much as possible. That's a lot of CentOS servers running at Facebook. We put this in the pick this week to just kind of keep this episode on a theme, because now it's beginning to look like there's something to at least keep an eye on, maybe grab an ISO soon. This is an interesting problem that Facebook had.

They were taking CentOS and then essentially slapping in a kernel from Linus, some backports from Fedora and trying to get it really modern. They were essentially solving the problem I was trying to solve when I switched to Gentoo in frustration over a decade ago in the enterprise. It's funny, it hasn't really been solved yet. But now with the way that CentOS is being built and the way it's being composed, there's kind of an opportunity here for different groups to build up these alternative ideas that maybe get adopted, maybe they never do. But the CentOS hyperscaler effort is building a CentOS with faster moving packages, some backports in there. We're mentioning it this week because ISOs are out, but we're keeping an eye on it long term as a possible workstation here in the studio. So we'll let you know. We have a link in the show notes for information about that. I say at this

phase, keep an eye on it, see where it goes. It's pretty neat to see a version of CentOS that maybe has some more current stuff in it. Could kind of sit in that sweet spot, especially with the Plasma desktop. If you'd like to get a hold of us on Twitter, this show is at Linux Unplugged. The whole network is on Twitter at Jupiter Signal. And the entire network is at jupiterbroadcasting.com. We got a website now, Wes, brand new. Fancy. I think there might even be SSL. Oh, gosh. I don't know about that.

I think it's just a fad. So I didn't bother. Instead, we're just doing all of our stuff over HTTP forever. Send us your PGP keys and we'll get back to you with a signed MP3 in a couple of days.

Yeah, good enough. Good enough. Also, join us live. We're here every Tuesday. See you next week. Same bad time, same bad station. And of course, if you haven't caught Linux Action News, you've got to go check it out this week.

More stories, stuff that before the show, people were like, are you going to talk about this? No, we already talked about it in Linux Action News. We already talked about it, so go check that out. Links to everything we talked about today over at linuxunplugged.com slash 413, including our Mumble Room, our contact page, our subscriptions, if you want to get in your podcast catcher of choice. All of that's over there at the Linux Unplugged website. Another website.

You know, they're just cropping up everywhere. Somebody's going to need to build some sort of way we can find them all. Anyways, thanks so much for joining us. See you right back here next tuesday jbtitles.com Let's go boat. Let's pick a title for this thing conversations continuing on in the chat room i hope it hasn't devolved in the chat room to the point where we can't have an open transparent conversation on the show that would be a shame but perhaps uh dildan i know you had a point you wanted to make about uh maui plus and dot net or Maui and.NET, maybe being a contender to Flutter?

Yeah, it's a multi-platform app UI. It's an evolution of Xamarin using all the native toolkits. Does this mean I can have Tomboy back for my notes? Possibly. I missed some of those.NET applications. What was it? Who was the one called? Banshee.

Banshee, yes. Banshee was a great music player. Even today, I pulled it out with just a week or two ago because I got the nostalgia goggles. And it is still, I think, the best jukebox that I've ever used. Yeah, I loved it. I actually was so happy with a lot of those early.NET apps. And I never really quite agreed with all the mono hate, but I did understand the uncomfortable position and with all of the IP crap that was going on back then. So you got to remember that stuff was really in the air back then. I understand why it went down the way it did. And it makes me wonder if there isn't just some built-in resistance to MAUI and just.NET apps in general still on Linux.

Well, at least with MAUI, the main complaint people have is that it steals the name from another toolkit that already exists. Right, right. Yeah, there's that too. You know, there's a lot of names out there. It's hard. They could have picked a better name. It's not like they don't have an army of lawyers to figure this stuff out. Right. And a search engine to just do a kind of a quick check. Right. Like it's before Microsoft announced it being called Maui, the original Maui from the KDE

project was the top result. And now it is not. I know how they feel. There's a group out there that have stolen my name and they know who they are. And I'm still salty about it.