Tech Over Tea - Fedora Linux With The Project Leader | Matthew Miller

Episode Date: June 28, 2023

I've talked a lot about Fedora Linux on my channel and today we have the one and only Matthew Miller the current Fedora Project leader on the show to discuss the current state and future of Fedora. ==========Guest Links========== Fedora Linux Website: https://fedoraproject.org/ Website: https://mattdm.org/ Mastodon: https://hachyderm.io/@mattdm Github: https://github.com/mattdm Gitlab: https://gitlab.com/mattdm ==========Support The Show========== ► Patreon: https://www.patreon.com/brodierobertson ► Paypal: https://www.paypal.me/BrodieRobertsonVideo ► Amazon USA: https://amzn.to/3d5gykF ► Other Methods: https://cointr.ee/brodierobertson =========Video Platforms========== 🎥 YouTube: https://www.youtube.com/channel/UCBq5p-xOla8xhnrbhu8AIAg =========Audio Release========= 🎵 RSS: https://anchor.fm/s/149fd51c/podcast/rss 🎵 Apple Podcast:https://podcasts.apple.com/us/podcast/tech-over-tea/id1501727953 🎵 Spotify: https://open.spotify.com/show/3IfFpfzlLo7OPsEnl4gbdM 🎵 Google Podcast: https://www.google.com/podcasts?feed=aHR0cHM6Ly9hbmNob3IuZm0vcy8xNDlmZDUxYy9wb2RjYXN0L3Jzcw== 🎵 Anchor: https://anchor.fm/tech-over-tea ==========Social Media========== 🎤 Discord:https://discord.gg/PkMRVn9 🐦 Twitter: https://twitter.com/TechOverTeaShow 📷 Instagram: https://www.instagram.com/techovertea/ 🌐 Mastodon:https://mastodon.social/web/accounts/1093345 ==========Credits========== 🎨 Channel Art: All my art has was created by Supercozman https://twitter.com/Supercozman https://www.instagram.com/supercozman_draws/ DISCLOSURE: Wherever possible I use referral links, which means if you click one of the links in this video or description and make a purchase we may receive a small commission or other compensation.

Good morning, good day, and good evening. Welcome to episode... I don't remember. 173, I'm going to guess, of Tech of a T. I'm, as always, your host, Brady Robson. And today, we have a guest that doesn't really need much of an introduction, but I'll let you introduce yourself anyway. Welcome to the show, Matt. Wow. Matthew Miller, the Fedora project leader.

and Matt. Wow. Matthew Miller, the Fedora project leader. I actually go by Matthew, despite having my handle be MattDM everywhere. And this is a story of me being old. When I was in college, our usernames for the VACs were limited to eight letters and the system was first name, middle initial, last name. And so I had the choice of Matt or last initial. I had the choice of Matt DM or Matthew DM. And I was like, Matthew DM sounds silly and it's too much to type. So Matt, Matt DM it is. And I've got that everywhere. And so people call me Matt.

It's fine. I just might not answer just, you know, while I'm, while I'm on here, we can get important, important information. I've got, gotten i've got yours matt on the uh the overlay if i if i knew that i would have put matthew on there maybe again like i said it's my own fault i've i've brought it on myself with the next though um whatever it's fine i did uh i did spot on your website you said there are a lot of people called uh matthew miller i i just did this because like you know i when i bring someone on especially someone someone who has some sort of history behind them, I can usually

find your LinkedIn or something like that to see the things you've done in the past. I type in Matthew Miller, and it's like seven other people before I get to you. So I'm like, Matthew Miller, Linux. And luckily, you're the only Matthew Miller in the Linux space, at least of note. You're the most notable if there is another one.

That's good. I'm glad I've gotten there. There was somebody, a writer for ZDNet or something that was often causing confusion for a while. And yeah, at Red Hat, there's actually three other Matthew Millers. And the funny thing is we've all been there for like 10 years. I expected the number to go up or down, but it's like we got the four of us.

If I just kept going like more and more specific, honestly, if I just started with Matthew Miller Fedora project leader, I would have got there from the start. I don't know why I went through the whole hassle of doing the other stuff. I wish I would have like done this over time. Maybe there's because like at the hate,

like at the geek peak of like the, the internet, like, you know, when Google was new, and the search results would all come back being the most nerdy things possible because only nerds were really using it.

Like, I was Matthew Miller, you'd get me. And yeah, I don't know where I am. I don't actually track this, where I am in the Matthew Miller search rankings. I think there's a politician the Matthew Miller search rankings. I think there's a politician called Matthew Miller. Yeah, a politician. There's several journalists. This is not a I want a Wikipedia page for myself. I don't need one.

But I mean, everybody would like a Wikipedia page. Come on. I want to be famous. But I find it kind of frustrating that a lot of people in open source who have done huge amount of things are not seen as notable by the standards there. Whereas everybody who's played like minor league sports gets their own entry because, of course, look, they've been in all these newspapers. They must be famous. Exactly. So you'll find plenty of those Matthew Millers too. No judgment against them. Some judgment against Wikipedia, because come on, that's not a good standard for notability. And I really do think there are a lot of stories of people who've done amazing things in open source who are underrepresented and don't get the recognition that they deserve. So actually having people on podcasts, that kind of stuff,

doing journalism where you talk to people, it's awesome and important. Well, look, I know you are mentioned by name on the Fedora Wikipedia page. So I don't know why you wouldn't have a page. Okay, I don't know. I haven't looked at that. I have gotten in trouble before by trying to edit the fedora wikipedia page which as you know is uh is a no-no but some things were so just wrong on there it still could use some help if anybody's watching this you want to like fix the wikipedia page a little bit i'd love that but yeah on the topic of of websites you're mentioning before that um there's this new

website for fedora i wasn't aware this was a new website but it does look really sleek um yeah i am sort of not doing it justice because i've got it like covering half the screen so it's sort of like the mobile view right now uh but it it looks really good yeah this is actually a a really good community story too Because so we had the first, sorry, the most recent major redesign of the website was really back in 2014. Although there was a kind of a big like engine redo somewhere in the middle and a lot of actually work. I don't want to discredit because I said this wrong before. And then people were like, wait, what about this? But the first, like the real like conceptual change was back in, yeah, like 2000, I don't know, 14, 15, something like that. A really long time ago. And that was a lot of work by some, mostly some people at Red Hat.

And then also we had someone who had been like doing Fedora website stuff for a long time. And there is a pattern that we've seen in Fedora and it can happen in a lot of places where there is a team of people. But eventually, like one person has been there the longest, whatever, kind of becomes the person. This person who was really good was kind of, you know, was the websites guy. And there was sort of a thing where instead of it really being a team, it just became an individual. And, you know, there's some pride in that. And there's some things that are good about that in some, like people should be recognized, like I was just saying, for what they do and get to own things. And that's amazing.

But eventually, and, you know, sort of suddenly he got different job. And turned out in his day job, he didn't have all this time. And so I went back to Red Hat. And I was like, okay, we don't have this volunteer anymore. Obviously, we need the website updated. Can we get resources to update the website? And it turned out the answer was no. And there's a whole complicated thing about how Red Hat provides resources to Fedora. Yeah, I did want to get into that at some point.

We'd have to talk about it. But yeah, a lot of it is a lot less than people think. And so we were kind of struggling for a really long time. And just recently, though, someone who was a community volunteer, not a Red Hatter, I like to make sure that Red Hatters are also included in community. We really try not to make a, that's the community, this is Red Hatter's distinction, because that's not Fedora at all. But he was actually hired to work in the community platform engineering team on Fedora after being involved in the project from other days.

And he was like, not by himself at all but uh was able to get like okay i can put a little bit of my time into this and then with some other amazing people and i'm not going to name names because what will happen is i will forget somebody and this is the recognition thing but i like every everybody is amazing and there's so much good work on this and people really came together um to both people have been around the project for a while and some like net new contributors to like figure out how to make this website, you know, the design, the content management engine behind it and like the, you know,

honestly the philosophy of it all work together with our design team and the websites team to make what's really a great, amazing website for the project that is truly community powered. And also behind the scenes has a content management engine so that instead of needing to contact the website team for updates, if you are, for example, the IoT working group,

the team working on that can actually own their own content and messaging in their section and update that and change what it says and how things are laid out there without having to go through permission or bureaucracy from some other team. So that's really nice. I really like it. Is there a repo somewhere I can point people to to find out who actually worked on it? Because if there is, just want to know who worked on it, look at that. Yes, in fact, let's see. Just so you don't forget names, just look at the repo and you'll see who's done it. Yeah, absolutely.

In fact, if you go to the bottom of the website there, there's a line, help approve this website, and it links to the CMS and to the get lab repo and see all of the people there On CS get over here we go Yeah, I will I guess leave this in the description Shout out to Nico who made the most recent commit. I don't know how much yeah One of the people who have done a huge amount. It's yeah, but really great. No, it definitely does look really good Absolutely, like it's yeah but really great no it definitely does look really good absolutely like it's a real sleek design it's it's not like you know the problem with a lot of you'll see a lot of modern

website design where there's just no content it's just scrolling and scrolling and scrolling like everything pretty yeah like the apple website is the worst example of this like it's just a product picture you scroll for like five minutes and then you eventually get to something important yeah so we didn't have that much scrolling before but one of the one of the problems we actually had with our previous design was um a failed project which is something which is so at the time um the seat the search engine optimization um wisdom was that it was better to have lots of separate domains all interlinked, which is we're back to one domain is better. I don't know, whatever. That's not my area.

Yeah, I'm not the person to ask about that. But that was part of it. And also, we were looking at things like Mozilla, where they had get Firefox. And so we wanted to have kind of a a are you a user or are you a contributor how are you coming to the project and for the user facing thing we wanted to have this pretty crisp get fedora what do you want to do you want to get download it get going and that's so we had get fedora.org and then fedoraproject.org was going to be a thing called fedora Hubs, which was one of the problems we've had for a long time in Fedora, and I think it's getting better. But a lot of our activity, when I joined the project, it was on IRC.

And even then, when I joined the project, it definitely was. But I'm sorry, when I joined as Fedora project leader, which is... So I've been at Fedora for 20 years. I joined as Fedora project leader, which is... I've been at Fedora for 20 years. I've been Fedora project leader for nine now. 20 years. You've been involved since basically the start then, pretty much. Yeah, I was actually involved with Red Hat Linux. Oh, okay.

Before, in some ways, as a community fan. And so I was involved very early. I won't say I was one of the most instrumental people at the beginning. I was just kind of around though. You could find me whining about things on the mailing list. Some of them early posts, very embarrassing. But I think I was helpful some as well. I'm also pretty easy to distract. What was I talking about? Oh, yeah. So it actually, we were kind of in a bad place just for various reasons. But there actually was a huge amount of activity. But most of it was, you know, on these, I think, IRC, a lot of like just chat all the time. And then even like formal meetings.

There were, I think, that year,, 1100 Fedora IRC meetings held, which, you know, that's like there's a lot going on. But if you looked at our website, you know, our online presence, we were just it looked like, oh, is this a dead project? Is anybody here? Right. You know, and we have a wiki, but our wiki is not like the arch wiki. we have a wiki, but our wiki is not like the Arch wiki. It is a do-whatever-you-want wiki, which is very useful, but means that the levels of curation are hugely variable. Some things, people keep up to date, and then you can click on something, and it's like a page copied from a decade ago that no one's maintained,

but looks important, but actually isn't. It's like a trap. So there was that, but that didn't that didn't look it didn't look well maintained um and yeah then you know mailing lists um there's a whole thing about mailing lists but you know that's also invisible to a lot of people so the idea was we would have this thing called fedora hubs which wouldn't be a replacement interface or anything, but would be kind of like a dynamic feed showing all of this activity and showing the different teams. You'd have a hub for the design team and for the desktop team and for the server and all these things. And

each hub would show all of the activity of that team and help you get connected and be all really exciting. This was great. We just overreached, didn't have the resources to do it. And a lot of postmortem could be done, but it never happened. So we were then ending up basically with Get Fedora being our main website. And you'll see a lot of places, it was never meant to be our main website.

That was meant to be the brochure download page. So finally, we're getting around here. This isn't nearly as ambitious as the hubs thing never meant to be our main website that was meant to be the brochure download page right so finally we're getting around here this isn't nearly as ambitious as the hubs thing for all of that you know being a social media feed kind of thing but it's meant to be more dynamic and we do want to actually have some of that social feed things there i think maybe we'll hook up some activity pub stuff or something i don't know um that should should be that crazy difficult to do right yeah um you know it kind of show show what's going on and keep it keep it something that's or something, I don't know. That should be that crazy difficult to do. Right, yeah. Show what's going on

and keep it something that's live and moving. It is back at the main fedoraproject.org home site there as well. I don't even remember what question I was answering. We're talking about the new website.

You stayed on topic there. It was fine. you did briefly mention um like irc and mailing lists in there like so how much of the communication with fedora nowadays is done in those sort of um systems how have you like migrated away from that i'd actually don't i obviously i'll see a lot of stuff from on the mailing list like but i don't know yeah, I'll see a lot of stuff on the mailing list. So, I recently just threw a gigantic grenade in Fedora Devel list, which is our main mailing list for people to interact.

So, I'm a huge fan of the discourse forum system. I know not everybody is. And I didn't grow up a forum system. Like, for a forum person, I hate PHP, BB style things. Oh, it's awful and terrible design.

You know, honestly, I'm using that like they think that got what? You cut out really badly right there. Oh, sorry. Usenet. Is that censored somehow automatically? I don't know. I don't know what happened there.

Weird Australian internet thing, so don't worry about it. You were saying Usenet, and then it cut out. Yeah, Usenet, they got a lot of things right with the design of that as a conversation platform, except for it has no authentication and spammers and other things basically destroyed its usefulness. and, you know, spammers and other things basically destroyed its usefulness. Oh, well. But forums have not been really very good. And, you know, I've got a soft spot for mailing lists.

I've been communicating in mailing lists for a long time. But they also have some big disadvantages. And one of them is that they're really hard for new people. And just the sign-up barrier is difficult. And just mailing list culture and how to interact. And now I've got to sign up, and now I'm on a list forever. Oh, no. I think that barrier is also bigger today as well,

because a lot of young developers, if you're getting involved in a lot of newer projects, they're doing communication on Discord or a matrix or something like that where it's a very different style of like getting involved you just join a thing you probably have an account in some other server already is just join it it's good to go yeah um so uh there are a lot of people there are a lot of things that people like about the mailing list and one of them is that you know people have a workflow set up and it all comes to you and you can skim it and so on so that's been pretty important and people people don't want to lose

that right right um but um i yeah i feel like the downsides are pretty high of having that as well they also are fairly easy to archive as well. That's another really big advantage. Oh, yeah, right. That's another thing. And having my own local archive, having it offline access. Yeah, that is an important thing. Or just a public archive.

Just anyone who's not involved in the mailing list can just, you know, easily scroll through it and see what's going on. Yeah, yeah. It's something other than, like, the main site or whatever. The site is down or whatever. Yeah, that's definitely true. So the grenade that I threw into the develop list was that I really think we should close down that list and move everything over to the discourse forum,

which is a very big change in how things work. And I want to make sure that we make it so we bring people along as well. I don't want to, you know, lose people just for that. And I would hate to. It is a change in workflow. And, you know, it actually causes pain in your brain to have things moved from where they were. And I'm not kidding.

It's real. And it's hard. And, you know, it's, I'm not kidding. It's real and it's hard and you know, it's, it's gonna, it's gonna be hard for some people, but we're, we're going to try it out at first with like our changes process and then maybe, you know, with other things as we go. And I think that will really help make it a lot more accessible, make the, make the activity more visible and give us some better tools for better conversations where often mailing those threads you you can moderate mailing lists only with a very big hammer um right

i i will ban this person and it's often after the fact and once the message has gone out it's there so forum gives us a little more ability to uh not just you know block things but to encourage people to you know rephrase or to split topics into different things if something goes off topic it's okay the off topic thing might actually be important but let's have this topic over here so all those kind of things modern tools are nice so um i really would like us to move more to that and that's a challenge it's one of the how do you get there from here things um i think you know on my side in that is the general collapse just like like usenet failed before of email as a usable platform which i know people are disappointed by because it's, you know, it's decentralized and so on and all these things, at least theoretically.

But really, there's a handful, a half dozen of big mail providers. And for a while, you know, fighting spam was a big deal. But actually, at the scale they are at, spam is not a problem. It's annoyance, but they have the visibility into everything and exchanges with each other to control it for their systems. But I don't know how many of you are nerdy enough to run your own mail server like I still do. It's insane the amount of spam that comes in. And I can run Spam Assassin and whatever, but I don't have the scope to pick that up and therefore other mail the big mail providers have less and less incentive to even care about mail from smaller providers and i know um and it's it's uh it's not going to be long before basically mail has to go

through a major provider or else it will not be delivered. I know that sounds like doom and gloom, but that's the way email is going. And that also matches people's use of email changes. I can't remember the last time I've exchanged an email with a friend. We used to do that. It was back then we used to, remember when people sent letters? Now we have email. Ha, ha ha ha. Now email is like, like sending,

you know, I write, writing out a letter, a long head and putting them in the mail and sending somebody an email. Well, a personal email seem about the same kind of antique to me. Honestly, for me,

most of my emails I send are for the, like the people who prefer to use email to talk about doing the podcast that's pretty much it yeah right yeah there's coordination stuff like that can happen and then then it's like business exchange yeah exactly or like my you know my my my receipts go to my email and my like you know login information and you know and the thousand of websites you've given your email to that keeps yeah the things they say that i've opted i get things because i opted in and i did not opt in oh i'm tired of arguing right yeah right so um so email's falling apart and i this wasn't supposed to be a podcast where i rant about email that's fine like we need like it's we need we need to move to other things but i would actually like

to make sure we can have some of those strengths. And Discourse is an open source platform. It's got a, the API is not a beautiful API and it's under-documented, but it does have one. And so I'd actually like, there is actually an activity pub plugin. It's not, it's not good yet, but I would like to explore our things for making other archives. Actually, somebody had,

this is one of those things that I should not do, but I'm very tempted to work on. It would not be that hard to make something that takes, that uses the API and the webhooks of the system to take every post and mirror it to a read-only news server, and people can connect to it with their news clients and read it that way if they want. That's actually a good idea, though. I know it's a good idea. I just shouldn't do it because I've got so much other things to do.

No, fair enough. If somebody wants to work on this, I've got thoughts. Talk to me. Because I think it would be really useful. And I think actually other people running discourse would be able to use it. And one of the things people were complaining about, which I think is also valid, this is actually Jonathan Corbett, who does Linux Weekly News, was saying that now everything, he's kind of having everything in one place. And it feels like having everybody on their own uh discord server somewhere kind of feels now like everything is separate and um you know you can get notifications in your email but

the email and it we have people who basically use it only by email but explicitly in the design from you know the project the emails really are to get you to the website. The website's where you should be there. And people maybe don't want to do that. So if we could do this Usenet bridge, we could actually set up a thing where we could have a bunch of different... It could support hierarchy, as Usenet does, of different things. We could have org, Fedora, whatever, and then have our stuff there and GNOME and whatever else could all be in a mirror on the stage

server so people could look at it all together. I don't know about posting from there because of the authentication problems, but at least reading it and then having links to go reply and interact would be cool. So yeah, there's a project

for somebody who wants to work on something. If anyone happens to be interested in setting it up um yeah getting contact with matthew see what he see what he wants to do um earlier you're mentioning like there's not really many ways to deal with spam with like a mailing list and like people trying to just basically just waste your time like i would imagine i don't really seen it myself but i would imagine a project at the scale of fedora has some amount of people doing that like how many people really are they're just trying to just you know waste time on the mailing list just posting absolute nonsense like not just like taking things off topic but just genuinely bad faith actors? Only a handful, really. And, you know, it's... There are...

There are more people who have trouble interacting in a healthy way who actually mean well. And I think that's the hardest thing, because you know, there's this thing that I subscribe to, that, you know, one toxic person in your community i i subscribe to that you know one toxic person

in your community is never worth it no matter how much they've contributed um the damage they do is worse than anything they can add which right i i definitely believe um but i also believe in the fundamental good in people and that people can learn and change. And so I want to address all these things with that in mind and that, you know, people don't have to be, a lot of people who care about Fedora, about the project, about open source, you know, may not know how to communicate their passions. Obviously, I think in tech in general, and maybe open source source especially, we have a lot of people who are not neurotypical. I have ADHD, which is a struggle for me. And, you know, sometimes it's fun when I can do a podcast and I'm going in every direction.

And, you know, ADHD is great. That's totally fine with me. Yeah, if you want to just go anywhere. But it can also, you know, it's a challenge in getting things done and being productive sometimes. And, you know, and sometimes in communicating with people as well. And so we have a lot of people with many different kinds of ways their brains work, certainly not just ADHD. And, you know, a lot of times we have people from different parts of the world with different worldviews, different ways of communicating, just different approaches. views, different ways of communicating, just different approaches. Yeah, like we have one thing my boss, when I started as the federal project leader, gave me a book about communicating

in different cultures. I forget what the title is. I think it's bow, kiss, or shake hands. I think that's what it is. Kiss, bow, or shake hands. Yeah. And it's an interesting thing it's got two different countries and a little bit just like a quick cultural summary and one of the things i learned there um red hat has a big office in the czech republic and one of the things that i found from this book and that validated people that i know is that um making a request where you make it a personal request like hey could you please do this for me, is very strong. Like that is taken as a personal obligation favor. And in America, we say that very casually.

And so it's very easy to accidentally ask for a strong personal favor for someone when you just meant, hey, here's an idea. So like being aware of those things is really important. And another one in some cultures, it is saying no is seen as very rude. So it is it is the appropriate thing to do when you when you mean I will not do that is to say, yes, I will do that and then not do that, which is much, much more polite than saying no. Right.

And again, as an American, like that's shocking and can be very frustrating because someone is just trying to be polite to you and now they're just not holding up to their commitments. Right. Exactly. So understanding all these things about how people interact in mailing lists and you know things that may seem to be in bad you know seem harmful can sometimes just be misunderstandings which i hope we can work out um on the other hand we do have to look at the impact like if it is causing trouble we need to act it doesn't matter if it was meant well it matters what happened and

so that's a whole really i don't know it's an ongoing challenge in any community we especially especially a community that is very like worldwide like if it was just like hey this is this is the project that americans work on like it would be pretty even then there's still like some you know differences between states but it it's a much easier thing to manage when you're just dealing with like a region like that but i can see how there would be also just language barriers like when you are dealing with people that may not have english as like a native language they might phrase things in a way or might read something in a way that they the person saying it to them wasn't intending it to mean yeah we often on like ask fedora help for them we often we get we sometimes

get people who um are interpreted as being very demanding about their response but you know really they're just in panic because their system's not working right and um sometimes we see people like hey stop demanding anything. You're not going to get an answer from us if you act that way. And really, they didn't mean it that way. And so there's a little bit of charity and interpretation, which helps.

On the other hand, please don't come to Aspidora and demand answers. There have been people who have been doing that. I'm sure. This makes Red Hat look bad i'm gonna you know something ibm stock no it's every time i mentioned every single time i mentioned anything to do with fedora it goes i talk about fedora comments then go red hat then ibm and everyone's just like ibm ibm ibm like can we. Can we talk about the thing I was actually talking

about? Let's not do that right now. Yeah. I'll say quickly, IBM, obviously, there's some influence. They're a big company and Red Hat's now a big company, but Red Hat is really operated independently. And even before IBM bought the company, it was a publicly traded company with shareholders and, you know, in part of the American capitalist system. And the things that IBM is asking from Red Hat are really no different from what the stock market was asking from Red Hat as an independent company. So when I say, trust me, it wasn't IBM. It's really true. It wasn't. Sometimes it's Red Hat being a company with 20,000 employees, multi-billion dollar revenue company. That's a big company. Sometimes it's just like, that's what's happening. We can mess things up all by ourselves. It would be nice. Actually, it's tempting to be like,

a scapegoat. IBM's fault. But let's take some ownership here. Well, it would also be nice if you were directly integrated in Red Hat and had the entire Red Hat budget to work with on Fedora. Yeah, right. And so that's... We mentioned earlier, Fedora, like, so I do not have anybody who reports to me. I report reasonably high up into the Linux organization at Red Hat, but I don't, like, report to the CEO or anything. And although there are a lot of people who are paid to work on a job where Fedora work is a component, where they may maintain some software, and so working in Fedora as part of that is how they do it.

There are not really very many people paid to work full-time on Fedora at Red Hat. There are people in, I mentioned, the community platform engineering team that basically supports the infrastructure for build infrastructure for CentOS stream and for Fedora. And then Justin is our community architect paid by the open source program office. There are a number of other people like that. We have a QA team that is full time as well.

I'm sorry, I'm supposed to say quality team now they wanted me to stop with the assurance because um it's you know we're trying our best here so we have that but that is also you know um we try to make it so that these roles are not the role doing it the quality team is there to enable community quality team and make sure that there's a backstop, that someone's there, but mostly to work on enablement. A lot of companies have a mission statement that is a bunch of BS.

And Red Hats, though, was actually developed by a bunch of Red Hatters working together in a kind of crazy way in public. And I will not be able to quote the entire thing because it's got a few more words than it should because it's done by a committee. But the point of it is to make better technology with open source by acting as catalysts in communities and with partners so rather than by doing it like what redhead is trying to do is find how we can multiply the effort and so i think that that's the investment in fedora when it's at its best is working that way as well. I think I just found it. Where is it? Catalyst. To be the catalyst in communities of customers, contributors and partners, creating better technology the open source way. And then a lot more words that I'm not going to read.

Yeah, right. Yeah. But yeah, it's good. So and that, I think, is what Red Hat tries to do with their investment. I'm really I'm happy that they pay me full time to work on this. It's amazing. But really, the bulk of work in Fedora is done by people who are not employed by Red Hat and Red Hat works by working in that community. And there's a lot of times where people are like, you know, Fedora exists only as a rel beta or upstream for Red Hat or a testing playground for Red Hat. And that's really not true. It exists for its own sake. And Red Hat sees value in it. But there is value to Red Hat.

They wouldn't pay me otherwise to do what I do, honestly. We wouldn't need my job if it were just that beta. But we want it to be a functioning actual community with its own interests. And that's completely vital. And I feel like there actually aren't a lot of... It's somewhat unique in open source how this works. And I think it's worked pretty well. It's worked well enough that Amazon is now basing Amazon Linux on Fedora and working with Fedora. And. And it's not just a Red Hat backdoor thing. So I think that's been some nice validation that we're doing things well.

And that's true, obviously, Amazon, very large. But it's true of smaller companies and individuals. It's that same, you should get the value for yourself in working in Fedora and helping build an operating system that belongs to all of us. I think that's kind of what I love about it. You can be a Windows fanboy or fangirl or fan whatever you would like to be, or an Apple diehard. But even if you work for Apple, like that doesn't belong to you. That belongs to, you know,

the ghost of Steve Jobs and the shareholders, right? But when you work on Fedora, like that, Red Hat may own the trademarks, but like when you work on Fedora, you are part of it and it is your thing. It's like having, you own it and you can do what you want with it and you can, you know, help build it and it is your thing it's like having you own it and you can do what you want with it and you can um you know help build it and make it better and that's amazing

and powerful uh and yeah like i'm happy to be part of that i think that's making making the world better definitely sounds like you're you're happy involved with it i don't know you got a skeptical there what's no no maybe that's just the way i speak i know that you definitely said like you've been involved with this what 20 years you said yeah yeah you wouldn't you wouldn't have been involved for 20 years if this didn't like catch your interest if it wasn't something you really wanted to work on yeah um and yeah and it is released nine years as project leader here, which is... If anybody is calling for my resignation

or my head, they're doing it quietly in the background. I haven't heard any. When is your 10th anniversary as project leader? So it'll be... It's the beginning of June, so it'll be a year in June. You got a little bit longer left until someone

can get rid of you then it's 10 is the limit it may be um you know there should probably at some point some fresh ideas should come in but I really try to to listen and not um you know I I don't have the ability to dictate but also that means it wouldn't work if I just tried to tell everybody what to do. So I want to make sure there's room for people with new ideas and tell me what they think and come and shape things. So as your role as project leader, what do you actually,

like, what do you do in this position? Right, yeah, like, what is my day today? Yeah, basically, yeah. Yeah so um there's a lot of meetings with people um just like video calls and chats and whatever um and you know it's hard to that's very useful the one-on-one kind of things it's hard to figure out how much that scales uh there's some some internal Red Hat stuff. A lot of trying to be active in the forums, on our mailing lists and seeing what's going on,

talking to people there, and then writing up. In general, my job, it's to make sure everything goes smoothly and to help the project have a coherent strategy that we're working on. So that's kind of my main thing. And I think you've seen I've been kind of working on what our next big strategy is with the Fedora Council. The recent layoffs and things, which I don't want to talk about. Yeah, no, that's totally fair.

That's kind of shaken things up a little bit. So it's not like it's ruined anything, but it's kind of just put my's kind of shaken things up a little bit. So it's not like it's ruined anything, but it's kind of just put my interview for that on pause for a little bit, but I'll come back to that soon. It's very important. But yeah, and some of this, what I'm doing now here,

talking to other people outside the project and kind of sharing what we do and who we are is part of it as well. But I think overall, what we do and who we are is part of it as well. But I think overall, the nature of any sort of community volunteer project without a really strong hierarchy, and there are some kind of things where there's a,

like this is our charity, this is our mission, do this or don't do anything. But where we let people kind of explore things, it is to diffuse in every different direction it's like a you know releasing a cylinder full of gas molecules into a room it'll go into every corner and you'll you'll explore the shape and that's actually really powerful that's like where a lot of innovation comes from because you people often some weird corner of the room that nobody would have looked at and that's amazing but also um you know if you let if it keeps going it just becomes you know thin air and nothing and everything you don't

want to stretch yourself too thin right so there's some amount of figuring out okay how in all of these people going in their different directions what's our commonalities what do we care about and what are the things that we can decide to focus on even if they're not the things you know individual people you know, individual people, some people might not care about that thing so much, but if we can get people to agree, okay, here are going to be some of our focuses, and then we can kind of intentionally move there

rather than kind of staying in one place because there is... So helping make sure that that happens is really my job and um there's it's a thing that um i don't know i'm not sure to explain how to do it but it's uh um there's a there's also there's a one of my favorite futurama episodes is the one where bender is drifting through space and becomes accidentally the god of a small civilization.

Oh, yeah. Okay. Yeah. You know, and everything he tries, he messes it up. And then finally, the punchline is he basically learns that if he's doing his job well, nobody notices that he's done anything. That's not a good way to get promotions and raises. But it is something, you know, when you're in a position like this or any kind of leadership position in community, like, yeah, there's something to that careful light um because it can really get a lot done

and making sure that people people are empowered to do the things rather than um taking that for yourself well okay let's go all the way back to when you first got involved in fedora so you said you were doing some stuff with red hat linux at the time the thing prior to for anyone who hasn't heard me talk about it before red hat linux is the thing that exists before red hat enterprise linux they are two separate things red hat linux no longer exists yeah so um i was assistant admin at boston university um there's a whole story there but but I was a Linux fan and my job was for commercial Unix, Solaris, Irix, some beautiful SGI, really fun, terribly insecure systems. And all sorts of weird stuff.

The main campus servers were AIX, which was out of scope for me. That was the central. I got to play with the ones all around the university. Those were expensive. This was 2000. I'm starting to

see how old I am. Departments were starting to be like, well, we can't afford $20 know, whatever, $20,000. We're the foreign languages department, but we want to have a web server. So, and back then you couldn't just like have somebody host something for you in a reasonable way. It was like, so people would, you know,

buy a PC or have an old PC, have it be under, literally under some grad student's desk, they would put Linux on it. And then, because also back then, every IP address in the university was exposed directly to the internet. Oh, that seems like a good idea. And not only was that just the architecture,

because there wasn't the firewalling NAT stuff to make that not happen. It wasn't scalable yet. That's how it was. It was also seen as that's an important part of Internet freedom and academic freedom. Okay, that's fair. People strongly believed in that. You connect to the Internet, you connect to the Internet. You're not some sort of reduced client kind of thing you are

on the internet right okay that's fair it's fair but also around that time one of the linux magazines did a study where they plugged in they installed the latest distributions plugged them in and waited till they were broken into and like at the time i think red hat Hat Linux won because it took only, it took 15 minutes before it was compromised, right? So, like security just wasn't a focus. And there were things like, you know, there'd be an account named printer with the password printer, and then that account basically having root equivalents, like, great, great. So our security team at the university was basically spending more and more time running around to people, to departments saying, you can't run Linux. It's not secure. Stop that. And I thought, well, and I was young and in my 20s and full of energy.

And I thought, this is open source. We could fix these problems. How hard can it be? And so I looked at some of the things out there for hardening. There was a thing called Bastille. I didn't actually end up using those things, but kind of looked at what needed to be done to make it so these systems could survive. Added some scripts for automatic updates. There weren't automatic updates. Things like Yum wasn't invented yet. Those things weren't there. So added some scripts to do that and

made some things to hook it into our central account system, the Kerberos system, all those things. And took that and actually pitched that to my boss and then to the VP of IT, who was a wonderful, geeky, wonderful person who had actually been at the university since he was a student in the 70s when he had actually written OS that they ran on their mainframe at the time. So he had a soft spot for this project. And we ended up launching, actually actually have Boston University Linux here. Here's the later one. So here's the first one here. One of the first ones. There we go. Which... Yeah. So we actually made this distribution and rather than doing things from scratch, because it's open source, I took Red Hat Linux as it was and modified the RPMs to have the changes I wanted, rebuilt them and recomposed it all to distribution. And so that was going along pretty great. And we actually started running some of our central servers on it because that cost savings was useful to the IT, not just to the poor departments.

So that was a pretty big success. And then the whole thing with the RHEL Fedora split happened. And it was hard to figure out what to do because where they're positioning Fedora, it wasn't ready for being the base for what we're doing. These days, honestly, I would happily recommend, anybody wants to do something like that, we're in a great place. You can base your thing on Fedora.

You're a completely unbiased source to say this. Yeah, absolutely. But we've worked hard on it. It's been some years. Sure, sure. But back then, things were pretty rough. And I've got a talk. I'm actually going to be giving this talk at DevConf in Tucker Public in a couple weeks here. And I did the same thing at Linux Summit a little while ago.

About 35 Fedora releases in 30 minutes. We're going to go into the history of some of this. And I won't repeat it all here, but there were some rough things at the early times, but I kind of got involved as part of that. And through that, I met Seth Vidal, who was actually the guy who wrote Yum. He was an admin at, oh no, was it Duke?

I think it was Duke or NC, somewhere in North Carolina. And then he was a wonderful person who was just very caring and friendly and really encouraged me to work together and collaborate. Don't just do this thing, you know, over at BU, we could share all these things. So that kind of helped me get, let's put all our changes upstream that we can. And those kind of, kind of things get involved directly in the project. So, yeah,

that's, that's kind of how I got involved in the early days there. That, well, okay. So when did you, you're saying you'd already had like some unix experience at that point when did you first get involved in that then yeah so that was um uh again yeah i had that account on a vax in my

college in you know small town indiana and i had a friend a friend who's a couple years older than me he graduated and discovered that suddenly he no longer had email because, you know, any Internet access. Again, email was a key thing, but, you know, all the other things, Usenet and this emerging World Wide Web thing. He didn't have it anymore because there weren't Internet providers. because there weren't internet providers. So I was in the process, I was working in the IT department at that small college as a student employee and had kind of set up some,

helped set up their dial-in services because that was a thing they wanted to consider. It was actually for the students and faculty, not for general, but so my friend who'd graduated, no good for him. Just for context, what is a dial-in service? This is a bit of my time.

Oh my God. Okay. I was, for reference, I was born in 1998, okay? I'm like 25. No, no, no. That's a fair call. You just made me feel extremely old. Right.

So, you know, now we've got wires that go to your house that bring you the Internet back and forth. But previously, people just had their phone line. Right. Maybe they had cable, but actually cable, even like cable for TV, was kind of a luxury thing. People mostly had broadcast TV. And then, you TV, was kind of a luxury thing. It wasn't. People mostly had broadcast TV. And then, you know, they'd have a phone line. And that phone line was used for, you know, talking on the phone, which people did a lot. And then you could have a modem.

And so, you know, there were bulletin board services. You could dial up and connect your computer directly to another computer and that. And then the internet providers, which would actually provide internet services over that dial-up connection. And then you would tell your parents, I'm going to be using the phone for the next three hours. And then you would consume the phone line and then they would pick it up or their friends would go and everybody would be mad because you're using your phone line for it. And, you know, that would cost something like, you know, $20 a month, $50 a month, whatever, for very, very slow. This is like 14.4 K kilobytes per bit. I forget. OK, now I'm nerdy people in the comments,

please tell me how wrong I am about whether it's bits or bytes. But it was a long time ago. Very, very slow. So we basically bought a T1 line, a fast interconnection to the accounting firm where my friend's dad owned that. And we started up a little small business

making our dial-in service there. And we originally were doing it on Windows because that's what we knew. That turned out to be terrible. And so in the middle of something being broken in Windows one night, we were like, what if we tried this Linux thing and ordered some CDs from the back of a magazine,

installed Linux, and kind of just started playing with it and learning that way. Which, by the way, is still the best way to learn Linux. Install it, play with it, break things, try to build an internet provider. I don't know, try to build something. Probably something that would be useful these days. But that's that's how i learned got into it i thought there's some additional context i was missing you're about a dialing oh i know what dial up was i thought sorry something else i was missing there

um that's that's fair also my first there's probably are going to be people listening who are even younger than you who that will be like wow i had no idea well also for reference um i grew up in rural queensland with two parents who grew up on farms i didn't have an internet connection till 2005 we got my first connection was dsl um my parents didn't know what the internet was they didn't want the internet we didn't have a computer at home until around that same time as well right but d, of course, is luxury because that's always on and at least 10 times as fast. Right? But, yeah.

Look, I wouldn't have said that in like, Australia in 2005. I would be lucky to get one megabit, which is definitely faster. Again! Oh my god, that is so many times faster. Top end of dial-up, you can get... 56k. Okay, I thought... Forup, you can get... 56k.

Okay, I thought... For some reason, I was thinking 256k. We could go back to... This is definitely a pointless argument. Oh, no, it doesn't matter at all. Yeah, no. So, yeah. Anyways. I carried my internet

uphill both ways in the snow. Yeah. And now, you know, you can get a fiber connection in your house and have symmetrical gigabit if you live in a place like South Korea. Yeah, or faster. Yeah. Insane. So, let's see.

So, you're saying Fedora's had issues over the years with, like, especially early on, but like one of the things that Fedora I've noticed does fairly often is adopting new technologies probably a little bit before they're ready. Like Wayland, for example, was Fedora 21 when that was brought in with Gnome, then PipeWire back in 2021 you guys were looking at it yeah it's a careful balance i mean we definitely have in our mission and this

partly comes from um the split from rel where this news like fedora is supposed to be fast moving but it's also it's not you know um i i think we'd keep it because it's a good place to be. People kind of want new things quickly. But we try not to be, whatever somebody says bleeding edge, I kind of cringe. I would like to be leading edge. I'm an art user. I feel the same way. Right.

I like to not be, you'd be bleeding. Let's stay bleeding. And that's one of the reasons we do still put out releases rather than having being a rolling release, because we want to have kind of control over all this integration we do of all this change that's coming in. in. But there's also a thing where if someone doesn't do it at scale, it will never be ready. And so, you know, with Weyland, you know,

I think it wasn't perfect when we put it there, but we tried to make sure that if there was a problem, it was easy for you to fall back. So we had that option there. And I think that without that wide exposure, it wouldn't have had the ability to bake and improve. So yeah, we kind of take that for everybody else and that's okay. So the problem is you know the position you're in and you're a bit

too nice with how you described the project. The answer is it was terrible then, but you're not going to say that. Yeah, you know, some of it might have been. But we try to learn from each of the things. So I think one of the big things, and this is kind of a central thing in my talk that I'm pitching, the DevConch talk. pitching the dev conch talk um when we did uh fedora like number 15 that was i went to both switching to systemd and gnome 3 at the same release and we lost half of our users um that was just too big change and yeah i think you know in retrospect those are both great projects great technology and were the right things to do but wow wow, too much all at once. And so when we've made other big changes, both Fedora and like GNOME as an upstream, so like the change in GNOME 40 to the way switching to, you know,

horizontal rather than vertical desktop layout and like a pretty big UI change. We put some work into doing UX research with actual human beings rather than like a theory and getting feedback. And, you know, it's not a perfect change for everyone, but that went very smoothly compared. And I think it didn't cause that same distress. And I think, you know, we obviously in retrospect, we could have made the system be changed, go a lot more smoothly. Yeah. Yeah. But. You know, there's some excitement for the new and, you know, things things are going to break sometimes. One thing you can do, and I know people who do this, we do keep two releases current. This is part of what I was saying about the release model. So, you know, we just came out with Fedora Linux 38, but Fedora Linux 37 is still supported. We have basically seven months of overlap there. So you can basically stay one release behind,

upgrade every year and have a much less dramatic experience, but still be kind of close to the newer things if that's what you prefer. That's what I like doing with my hardware. I always buy a GPU the generation behind, then all the driver issues and everything else is going to get sorted out by the time I upgrade. Yeah, right. Yeah, it's kind of a similar thing to that. And if that ends up still being too fast for you,

there is CentOS Stream, which I think is actually a good option for a lot of people. I won't go into too much depth about that because that's a whole other talk. Or RHEL. RHEL is available. You can have 16 licenses as an individual to use for whatever you want in production. I think that's a great choice, especially if you're looking at a career or if you're just a small thing and you want to have a production server. There's other distributions that do other things. Those are out there as well. It's all Linux. It's all good. But I do think we want to make Fedora... People who are using Fedora Linux should have a good, confident experience in it,

and it should feel safe to use as your daily driver, and you shouldn't expect it to brake on you. I think that's fair. And we should expect that big changes are handled in a way that makes them, that at least you have a way to adjust things to fall back, even if there is something that is new well speaking of big changes i know there's been some like discussions about possibly gtk5 being like

it's still a while away but like gtk5 being wayland only um and when that if that happens that means gnome at that point is going to be Weyland only this is probably a 5 or so plus years into the future but I feel like that is going to be a really difficult transition, because Weyland is eventually going to happen

this is just the direction we're going but it's going to be another one of those System D sort of changes where there's going to be a bit of resistance but I don't yeah i think i think if it's five years if you put it at five years in the future i don't think they'll be i don't think anybody will notice honestly um i i used to you know when we made that change i used to try wayland every new release and then three days later i'd be back on x and somewhere along line, and I don't know when, I stopped doing that. And now I've just, oh, yeah, I guess I'm a Wayland user now.

Because a lot of that work has been done. And right now there are maybe some like network, remote network use cases and things where people use X. But X is actually not very good for that because it was designed to be used on a university network with no security and all these different right um and and not the way people want to use it yeah one of the comments i usually get comments about like you know network transparency every time i do i bring up wayland but i get a lot while i get a lot of people saying i don't know how many people that's actually a use case for, especially in the X sort of format of doing it. Right, yeah.

It's actually, unless you have a very high speed link between your systems, it's a very chatty protocol and not great. I don't know if the no machine thing is still around but also just the VMC kind of approach to things is actually better and I think that's probably what's going to happen with there's work on things like that for Weyland I don't know the details

I want to say I honestly believe that if it's five years from now, no one will notice. I think even three years from now, it'll be so smooth. It won't be a... People will be whatever. Now, yeah, if it's like one year from now, everybody's there,

then some people will be grumpy. But even then, I don't think it is that kind of gigantic system d kind of change um more like a pipe wire sort of change where it's yeah maybe yeah um and hopefully a more more smoothed out pipe wire than the first couple releases i sort of migrated fairly early on as well and a couple of those updates that came, especially being on Arch, a couple of those updates came through I was like, you know, today we're going back to Pulse Audio for a little bit

Yeah, I didn't have much of a need to go back but I did have a lot of restarting my audio Yep, just going to have a little script running in the background restarting my audio every five minutes just in case I My main concern was because of the video production stuff i had some issues with obs early on um that was the my main concern uh because i would have sometimes like audio tracks with like

the different audio sources would merge together like my mic and my desktop audio would just be the same thing and i didn't know why interesting it's what i need to find out now i have i have um three monitors here yeah i do as well and uh one of them has speakers on it which i sometimes use when i'm tired of my headphones but the other two do not but they're both hdmi you know things so they show up on my audio list and i just need to figure out how to tell it never show me that and i do not understand the syntax uh what is um fedora use as it's what do you call it the the it's wire wire plumber yeah you can get some lewis script you can do for that uh yeah there is but wow i can't figure it out. I know, the Arch Wiki is so good.

I mean, Arch Wiki is amazing. I need to find somebody at Arch and just be like, could we also just use your Wiki officially? The detail there, and a lot of that comes out of, it's not something that can be easily replicated because it's not that it's a Wiki. It's that there are people who are it it it's the same thing that could be irritating about wikipedia when it doesn't go what you want

but there are people who are very strict about how this is how the wiki will work this is what it's for this is what things will look like and they you know curate it and keep it there and that's the energy that makes in a lot of ways i think arch is a documentation project where there happens to be a distribution that you can run. That's fair. At least, you know, that might not be completely fair, and I'm probably being unfair to people working on the bits, but that documentation, that wiki is so amazing. Yeah, I definitely look at it I think one that doesn't get as much credit is the Gen 2 wiki is also really good. I think one that doesn't get as much credit is the the Gentoo wiki is also really good um the people usually when they talk about the the wikis in Linux usually the the Arch wiki is

the one that gets the main focus but Gentoo's wiki for a lot of things is really really in depth I will maybe maybe add some Gentoo search bookmarks then too a lot of the stuff you know a lot of it is obviously going to be especially in still section very gen 2e like here are the compile stuff that yeah yeah but like for the configuration stuff for sure it works quite well as well i don't know if you know the people who work on the arch wiki as a project or if anybody's listening if you want to be hey let's collaborate with fedora and just add a if you're on fedora linux section to some of those pages um i'd love that i mean why duplicate it's um i have heard some people suggest maybe like it would be cool if the arch wiki was more of just like a general linux wiki because for a lot of

things it really is like take out the the arch install guide and for a lot of the stuff on there it's just the same everywhere yeah exactly um and again you know um why duplicate we could try and recreate the thing but that would be silly we could you make something that tries to scrape it i assume it's under an open license i um i don't know yeah i don't know what it's under, but I would imagine so. Better. If it's not, someone, surely. Come on, ArchWiki people. Yeah. GNU free documentation license.

Yeah, there we go. Sure. You know, we use the CC by SA 4.0, but, you know, whatever. but you know whatever that's it so on the topic of not duplicating effort recently uh red hat is going decided to stop maintaining the libre office rpm and there is this sort of also there was an article from george castro about this change in the distribution model on linux there is a lot of push right now to flat pack. Also, you know, Canonical's doing their thing with snaps as well. But there's a lot of this push to flat packs right now. And what do you...

One of the main points I wanted to get your opinion on was... George... I'll see if I can find the exact line. It was something like, distros need to stop pretending that repackaging uh repackaging office suite adds any value or something along those lines was uh office suite let's find the exact line i'll quote him correctly because if i don't he might get in my DMs. Let's see. Where is it? Here we go.

People don't want to hear this, so I'll just say it. Distributions can no longer afford to pretend that they add value by repackaging office suites. The real value is in the hard stuff. We need a well-maintained kernel, graphical stack, desktop, and associated core tools. Yeah. So I was saying before we started, I think that there's some of these things, I see all these different opinions,

and I kind of agree with all of them in different ways to some amount. I definitely come from this distro packaging background for many years, so I think there's actually some of the hard work that's been done in the process of packaging that gets overlooked. And it isn't necessarily in the packages themselves. But one of the things we take for granted, ARM systems everywhere running in the cloud and ARM notebooks, the new Mac coming out, like, oh, yeah, of course, it's Linux. Stuff will just run on arm. Right. That wasn't necessarily true. In fact, even like when we went to X 86, 64, like not everything was actually so much,

embarrassingly so much code was not 64 bit clean. It just wouldn't work right. And so by doing that in the distro where we work to convert everything we could, we touched a lot of software we you know including you know improving the compilers and and you know finding all these things in the software that made that software work you know cross distribution whereas the developers you know usually have a much smaller smaller scope and they may even have all their own targets there may be some developers who are like yeah i care about x86 64 and i love risk 5ISC-V, or, you know, I've got a power PC. Yeah, you've got that, right? Like,

that's like, you've got, but trying to be like, we're going to run on all these architectures, like, that's a big project and hard work. And it really adds a lot of value. You know, people are going to expect you know when they run their apple m1 system with vidara linux that you know labor office has the arm built that they need and so that's like um the work we did as a distro to make sure that work and not just us but you know debbie and other other products that do that at that distro level that's actually really huge and uh we try to you know when we make fixes and improvements we make sure those go back up to the project we try not to that shouldn't just be you know a fedora thing um and some of that stuff doesn't like we could do that kind of thing

outside of packaging but it kind of loses its center. So putting it in the package makes a way to do it, a flow there. And so if we have all the different distros packaging, one of the things I was thinking about, so in Fedora, we talked about the CPU architecture baseline. And that basically is, as Intel and AMD make new processors, they keep adding new features. And if you compile your code for the newer feature set,

it can be, especially in some cases, a lot faster, but it won't run on the older hardware. And so this is actually, a RHEL is moving to a newer baseline, but they can, it's an irony, because Fedora is fast moving, we can't. Because it means if we would we only you know our releases are supported for 13 months so somebody with you know five-year-old hardware

might find that they're really that stops working whereas if they're getting using enterprise operating system they can keep running whatever rel you know version on it until they are done paying exorbitant amounts or whatever for this long lifetime life cycle right so we've decided we're holding it at this baseline but there's nothing like that's that you what we do for our pack our package set and that's consistent but there's nothing that says people who are building their things to go on flat hub have to do that so somebody might decide i'm building a graphics processing you know app of doing you know photography raw conversion i need this higher baseline the performance is so worth it they

provide that and then people find out that they're two-year-old systems that mysteriously doesn't work um right uh like because so that kind of consistency and and it is deep hard work across all of these things that um people do and and there are across all of these things that people do. And there are a bunch of other things that Flatpak makes better because it also used to be that developers would have interesting opinions about where configuration files go and all of these things. And doing something like Silverblue means that those have to be consistent. They need to go in your home directory in the right place. They need to be doing flat packs. They need to be consistent.

But, you know, people used to be like, our config files go in, you know, a new top-level directory they've made up. And so distribution spent a lot of time, like you say, okay, look, we're putting this in a consistent framework here. And so, you know, there's something to be said for letting developers experiment. And, you know, shoving that stuff into a container, fine. You can do whatever mess you want inside your container, developers. That's a nice way to draw a line and let people have flexibility, but also make the whole system consistent. But there are some things that kind of go outside of that still. but there are some things that kind of go outside of that still. And without,

without there, there's something to this project of integrating everything together that we do in the distros that has some value. Now, I think that with some of the bigger projects, you know, like, you know,

LibreOffice, Firefox, those kinds of things, you know, we can kind of, we can work together and, you know, people are going to be understanding about these things, but just with the wild west of all the packages, it's hard to figure out how to do that. And I think, I don't know,

one of the posts, so I saw recently, which, you know, it was kind of muckraking and I won't, won't point to it, but it said, you know, Flathub is a distribution and in some ways, therefore a distribution competitor. And I think that's really true. And it's interesting because it's been a distribution that's chosen to be very free with how people do things. Like there's very little, you know, things about supply chain security,

supply chain management is is a big thing you hear a lot about um you know with flat hub a lot of that stuff is just like yeah whatever don't get in the way of the developer building their thing uh and so like that we have all that stuff in in our distro model where we kind of try and track that you know when you build something in fedora project it's built from something that's checked into our get repo. It is built. You don't have network access while it's building. So you can't,

you know, do you pull down all your dependencies from whatever thing. And so there's, there's a consistency and traceability that you don't necessarily have with people doing whatever they want with flat packs. That doesn't mean things can't be better. But I think that, I don't know. with Flatpaks. That doesn't mean things can't be better, but I think that, I don't know,

we've got distros and Flathub as a distro or as a meta distro or something, have a conversation to have about what is important for those things for users. Because, yeah, it's kind of easy to set up like a developers versus distros thing. And both developers are like, distros are getting in the way of our relationship with our users. And distros are like, yeah, but distros are usually built by people who are users or are sysadmins. And developers don't always have the best interests of users in mind

or the same idea about that. And I don't mean it in a malicious way, just that developers can have their own mindset about what they're doing. And some of this may be with data collection and whatever other, other things. And one of the other points that,

you know, people who are criticizing flat pack, I think rightfully make is the sandboxing is not very strong by default. And it is sandboxing that is voluntarily defined by the developer. If you just like, Hey, access the entire file system. Like that's an option. Yeah. developer. Yeah, if you're just like, hey, access the entire file system. That's an option.

Yeah, right. And basically the developer says, okay, here's how my app needs to be constrained to be secure. Which is fine, unless you know, people don't want to do that. And I think as this all

gets more popular, there's going to be people who actually are bad actors who are actually trying to do malicious things. And so I think that FlatHub, they're adding a thing where it verifies that the thing is from... Yeah, that's being released now. Which is one step, but I think there needs to be... And they've got some other things, but I think there needs to be kind of an increasing series of verification about those things. And one of the ways that I've been arguing that we could do this is we can make flat packs from what we do in Fedora. And right now we've got a system that is a Rube Goldberg machine, which involves making RPM packages. And it's honestly, it's crazy. But, you know, there are reasons that it was done that way, which I will not apologize for, but it's not beautiful.

work in the way that we do as a distro. I don't know if that's the right thing to do. I'm willing to be argued out of it. But I think that we need to have the same kind of wherever it comes from, reproducible builds and verifiability, traceability, supply chain things. When there is a CVE in some dependency in a thing that's packaged in fedora linux like we trace that and we you know we we can know um and that

introspection isn't all there for the dependencies of things in flatpak and you know with the sandboxing not being so tight that could be a problem um i i don't think that it's wrong to have it the way it is because this is the way android worked too like the early android releases like there was not very much security and then they've made that tighter and tighter and so i think that's probably what needs to happen with flat pack and the same kind of thing where you know now uh flat pack or sorry android apps ask you for permissions as they're needed and also if you haven't

used them for a while, they go away. That Flatpak and Flathub need that kind of thing as part of it, as it goes forward. And I think it'll get there. But I think people are using that as

a gotcha. And I don't think it's a gotcha, but I think it's a that as a gotcha. I don't think it's a gotcha, but I think it's a valid criticism. Oh, Flatpak's crap. It doesn't have security. I think it's... Somebody in the Debian project said a long time ago that working on a distro requires you to be able to accept tepid progress for the somewhat better. And I think it's hard for a lot of people to accept that this is not now the best thing.

This is just we're moving a little step very slowly. And you make those little steps over time and you get somewhere much more effectively than if you try to drop that thing in especially going back to the whole system d thing sure oh yeah absolutely no um and i love system d it's great but like um there's some hubris right um that that first version is not that you think is so great, maybe not as great as it actually could be with some of the battle scars of experience. So where am I going with this? Yes. Right. Like people should be more accepting of small changes that are like, OK, that's not everything everything but it's getting us there right and right um sometimes you need to make some leaps but uh incremental change can be very powerful

yeah if you you know you improve upon something one percent every day you know eventually yeah yeah exactly yeah compound interest it's exactly um so on the topic of lap packs and all this there's the immutable i would you call them spins how how do you describe them in the project yeah let's say spins um okay i we don't have a term naming things is hard it's just immutable desktops on here that's kind of gotten out of control with with all the different fun names people like. People like Silverblue, then we had Kinoite, and then Serika, and then... So, we're gonna have to reign that in. I did talk about this in a recent video.

There's Odix and yeah... Yeah, part of... Naming. It's one of the famously hard problems. But Immutable is also not really right for what what silver blue is so i don't like that either and uh colin who's the author of os tree you know he really hates it i'm seeing a lot more people use like image base now which seems like a better term but the thing is like that's not what OSTree is. It's different from the image-based ones. It's not an image-based at all. So, like, image-based, like, that's a whole different thing. Like, that's actually, like, Leonard and now at Microsoft is working on image-based. Like, that's, like, they're seriously image-based. Like, that's the whole thing, right? Like, there's different approaches that are really are image-based so this is something different as well but it's sort of in the same the same family of problem solving i

don't know definitely a better term than immutable because like the problem with immutable is i've my comments when we're talking about immutable desktops every time someone's like oh it's immutable therefore nothing changes like this is the problem with immutable. Like, it's such a bad term. It has so much meaning attached to it. So before Red Hat bought Coral Hess, the company, we had this thing called Project Atomic, which is where all this is done in.

And I am trying to convince the people at Red Hat who are the holders still of the Atomic name. Wouldn't it be nice if we'd just be able to use that? But I've got to do some more sweet talking to get to there. Because I think Atomic is a fine thing. It describes kind of how it works. The updates are Atomic in a technical way. And also it's a really cool logo.

Yeah, we can just go full circle with Silverblue. Just... So, we started at Fedora Atomic Workstation, now we can go Fedora Atomic Gnome, Fedora Atomic KDE. It's straight to the point.

I understood why Silverblue came about as a name. Fedora Atomic Workstation is like, you know, it's very much Fedora. People who don't like Fedora are going to have a weird opinion of it. Silver looks like this new thing. Yeah, and there's something to that,

although I would like to, you know, we want to get Fedora in there to make sure people know that we can... But, like, it definitely helped early on, but now that the immutable desktop has this appeal to it, I don't think it's needed as much. Yeah, and I think doing that helped break the idea that Fedora could only do things one way. So once people understand that, we don't actually need the forcing function quite so much

for it. But But yeah, kind of going back to this whole thing, I'm going to interpret the question I think you were asking, which was, are we going there? And so this is part of the strategy proposal. I talked about setting strategy. And one of the things, that's a direction we could go in. But part of the problem is, from our workstation and our traditional model, it works pretty well. So it's, again, hard to make a big commitment to it. And I think we have to make a declaration and that's what we plan to do. So this is our strategy.

We are going towards this approach for our systems. This is going to be the main way Fedora will be consumed, possibly ending the naming problem because it will just be like, that's how Fedora Linux be consumed, possibly ending the naming problem because it will just be like, that's how Fedora Linux systems are. And, you know, I don't, I don't know what the target date is for that. There are certainly some problems to solve, but I think that's the direction. That's the direction everybody's going. I think the direction we'll go there as well. And, you know, that also helps with the, the problem with sometimes things are broken.

With this, you can roll back and go back to an earlier version. And OSTree has a really neat thing where you can actually bisect, which is basically, you know, this worked last week. It doesn't work now. So you can say, okay, let's go back to three days ago. Does it work? OK. And then just by dividing in half, do that binary search and very quickly find here's the change that caused the problem. And then, you know, the person can report it to that particular package and then they can do you know, they can find the bug. So it actually really makes finding what caused an issue much easier as well. I can't promise there will never be issues. But I think that's going to be really powerful as well.

For your personal day-living system, what are you actually running? Is it just like mainline Fedora Workstation? Yeah, this is mainline Fedora workstation here and on my other desktop um and i'm planning to reinstall with silver blue on my laptop after i get my talk uh at the next conference i have i've played with it a few times before but i've never kept it on a system and as i'm saying we need to commit it's time for me to commit as well and then i will probably also move these other systems as well well it's it's definitely i i'm sort of intrigued by it especially with tools like distro box being available now where it's it's made that whole uh like it's made that sort of bridging the gap

because the problem with a purely immutable system is you know you got your you got your flat packs and the flat packs will address most things but sometimes there are things that like can't be done in the flat packs yeah so and we have your toolbox as well yeah yeah right um and i don't know i i'm not involved in directly in. I wonder if maybe we should just merge in with Distrobox and go with that. We would like to encourage people to use Fedora for their thing. But whatever. We should also go with what's more powerful

and what people are using. But I would actually like it so that... These DMs are open. You can certainly have a discussion and see what's going to go on go on yeah work something out yeah um i i there was something that uh one of the uh godome designers who works on fedora did a while ago uh where it was actually a terminal replacement that was aware of the environment um and so terminal ui that understood what you were in and i really think

like i think that's something that would be worth working on similar to what um windows does with wsl possibly i'm not i'm not sure i i haven't run windows okay well on on windows with their their new terminal like if you open up like an ubuntu environment it has like a little ubuntu logo being like you're in ubuntu you feel like Ubuntu logo being like, you're in Ubuntu. If you go to PowerShell, it tells you you're in PowerShell. Yeah, but not just that, but being able to switch environments from menus and actually have control over what's going on as well. But then I would also, as the next step of that, the default when you open up a terminal window should be inside a toolbox environment rather than on the host system going to the host system should be a oh yes i actually want to mess with that level of thing right right um and i think that would that will kind of help the whole thing work together

one of the big unsolved problems is how to make things like vs code um and ides that don't run inside you know the terminal work nicely with that kind of environment i think we've we obviously need to get that solved nicely in order to really make this the mainstream solution um but i think we'll get there i'm sure like you know something we worked i know that with um with Distrobox at least like VS Code doesn't seem to have any issue with it there's people have been running like really stupid things with it as well like hey I'm gonna run my entire desktop environment in Distrobox like what it works don't do that right yeah so well so yeah so in that you're like installing GUI applications into that and running them there, we'd like to be able to install

something as a flatpack and then have it interact with the environment and there's ways to do that It's not as convenient Yeah And it should be made not just as convenient but you know

I'm trying to find a nicer way to say it than idiot proof because there's always, you can't like it. It's just now. Right. I know too late. No, it's,

that's not even really the right concept. It should just be like, that's what happens. You don't have to do anything. Right. Do it. It's, it's,

it's beyond obvious. It's just the way. Rather than something you kind of have to act. Oh, I mean, it's beyond obvious, it's just the way, rather than something you kind of have to act, oh, I mean, weird quirky mode to make it work. Well, on that note, what right now do you see as the biggest limitations of Fedora that you would like to see addressed? I'll put you on the spot here. Yeah, no, I'm just trying to... I think we have a hard time finding how to put people who show up

interested in doing something into the right places. So we often have people like, hi, especially actually, honestly, with discourse there and the matrix for. So we often have people like, hi, especially actually, honestly, with discourse there and the matrix for our chat, we have people showing up who are interested and there's just not enough people who are prepared to absorb someone new coming on board. And, you know, that onboarding someone can be a lot of work. And so one of the things, again, in the strategy we're focusing on is a mentorship and having that and onboarding is kind of a key part of that. So when somebody shows up and is interested, have not just a, here's a, here's a list of tickets you could get to, which, which works okay for some people,

but Fedora is so big that it's so overwhelming and we've got so much history and so on that like that approach it it filters people out it you only get people who have a certain mindset and are able to dig through that which amazing but there's a lot of people that we could really you know be valuable members of the project, have a lot to contribute, get a lot back themselves who we don't know kind of how to get that hook. And we have a really, there's a join SIG that basically kind of does a handholding approach for people. But that, I think we need to scale that up. And also that's very much a, here's the intro level. And there's not quite the ongoing, like, you know, support that people might need to get involved. So I think that's really,

that's really a thing that I think we could do better as a project. And if we can do that, then we can do everything else better. This is why, you know, of the strategy, the guiding principle of what we're trying to do in five years is double the number of contributors who are active because um i think that's that's a metric that is um it shows the health of the project um and i don't know where technology is going to be in five years it's going to be somewhere that neither of us have predicted um and so having people who are there and involved like that i know is going to be valuable, no matter if, you know,

now all of our code is written by AI and I don't know. Someone rewrote the latest code on Rust. Yeah. You know, they use chat GPT to write it in Rust, obviously, right? They wouldn't do it themselves.

What is this? You know, 2022. Yeah. So, yeah, I don yeah, I don't know. But having more people involved in whatever capacity. And that is, it's not just coding. It's documentation.

But also talking to people, helping run meetings, doing design, answering questions, welcoming people into the project, coming to Fedora Social Hour and just chatting about things. All of that stuff is really actually important project contributions. So, more of all that. You touched on something very important there.

Documentation. I think a lot of people when they think of a software project or a distro project, that you have to be a developer to get involved in some way. But there is a lot more like, you know, if you have I be a developer to get involved in some way. But there is a lot more. Like, you know, if you have, I don't know, maybe you're bilingual, you have some ability to do translation,

you have, like, there's more than just that. Like, what are ways that, obviously those are two main ways, like, what are other ways people can get involved with the project? Actually, translation, I want to highlight that a little bit because I think that's something really interesting and special about open source projects. Also, interestingly, what AI is bringing to all of this. But a commercial project is going to be limited in what languages they care about. And there are thousands of languages in the world, and some of them are spoken by only a few people in a small community.

is in the world. And some of them are spoken by only a few people in the small community. But that small community can have Fedora Linux and all this stuff translated for their language. You can have it local to you. And that's really powerful and important. It doesn't matter if that translation is commercially viable or mandated by law somewhere or whatever. You can have that. So it's part of this. This belongs to you, OS. So I think that's a really important thing that I definitely would like to call out.

I think there are a lot of other ways. We have a great design team in Fedora. The Fedora desktop wallpapers, every time we're done with the community process, I think we've got great, fun wallpapers that I really love. And that's

done in the open. And our design team is really great. One of the easiest things to do well, wait, let me get another one. QA, or quality work. I'm not going to say quality. quality work i'm not gonna say quality it's like um just 100 certain it's quality that yeah during during the releases you know as we go through it's

like going through just going through and the list of tests and checking off checks um you know sometimes that's gonna be very satisfying uh to like okay look at look at all this i'm getting towards 100 completion on our matrix. That's one thing. But also, as new package updates come out, those are constantly coming, and there are regressions. Test them, find the regressions, comment, and that helps. But then the thing I was going to say, ask.photoroproject.org, which is now a subsection of the discussion forum. Like, go there, look at people's questions, provide what you know. You don't need to be an expert to know something. And even if you don't know the answer, you can look at people's questions and be like, okay, I know that they're going to, they didn't say what graphics card they're using and they're talking about a graphics problem.

So you can say something like, hey, so and so, welcome. Hey, could you provide these details so we can provide a better answer? Like just being friendly like that, like that, like and then, you know, you'll help build some expertise. At the ISP that I helped start, like, you know, I did telephone tech, frontline tech support, and I did that also at my university job actually helping direct people and there's no way to learn more faster than trying to help people with their problems like it's so good so that's not just like a way you can help like you will learn a lot doing that so

i really recommend that as a way to get involved as well i think it's a good answer um well on the topic of improving things someone did ask me about uh about working with desktop environments on accessibility because that's a really big focus of for a lot of people a lot of people desperately need that and like there are issues with wayland right now where a lot of the accessibility tooling people were using in the past may not be working or as consistent as it was. What sort of work does Fedora do with the desktop sort of improving that?

Yeah, so I mean... Excuse me. I mean... I have been talking for a long time and I don't have tea, despite the name. My voice is starting to go. We may have to call this soon. But this is also one of the things where it actually came out really early as a focus on our strategy.

We want to work on this. And I think broken it down into three different things. So one of those things is accessibility of the OS itself. And that's, I'm going to put that in a little box for a second and then come back to that. We also have accessibility of our own tooling and of our websites and things like that. We need to make sure that someone can contribute to the project despite whatever abilities they have. They're not limited because we didn't have the accommodations we should. So we're going to work on that kind of thing. Other things like the desktop itself, we are an integrator. So

those changes are going to happen in GNOME and we'll work with GNOME. I think we can probably help the best by providing user feedback and information that channel back into the upstream projects to GNOME, KDE, XFCE, Sway, whoever else wants to hear from us. at Red Hat, which does a lot of work on Fedora, has actually hired somebody to work on this in GNOME in particular, which I think will have a pretty big impact. And so we want to make sure that we're showcasing those and leading with those features in Fedora Linux. Because that is really, yeah, I said that this should be an operating system for everybody

and that absolutely should include everybody. I don't even want to say that. I don't want to cut you off. No, sorry. Firefox decided that a meeting in a half hour needed to be notified and stole my window from me and suddenly vanished. That was the guess.

But I was also going to say, as I'm getting up towards 50. Amazing. And, you know, I can my eyes aren't what they used to be. And so I am, you know, reminded of. Yeah. Ability is a temporary state, right? Everybody, everybody needs accommodations in some way and our lives change. And we should we should putting this in a central place. It's not a it's not a special case. It's the main case. And there are so many ways in which accommodations benefit everybody as well. That's going to going to going to Europe from the United States.

You asked we have a very strong disability act. And so, you know, public buildings have ramps, all these things there. And, you know, going and realizing like, oh, wow, people with wheelchairs can't get in this building. Sorry, this is just a building that you can't have. Or, you know, and that also, you know,

but that also means, you know, people with strollers can't get in there or it's hard to bring your packages in, or all these things. There's benefits from these things that continue on to other things, and I think we'll see that in the distro as well. Well, I've got a couple other questions on here that I definitely want to address. Let's do them before my voice goes. I'll see if we can rapid fire. definitely want to address um let's do them before my voice goes i'll see if we can rapid fire so a while back for like a year or i think it was whenever it was a bunch of youtubers started to be like fedora is this great distro that everybody's running a bunch of people started

swapping to it um and it's sort of being started being described as like the new ubuntu for by some people. What did it feel like having that sudden... Obviously, Fedora was already a massive distro at that point, but suddenly getting this extra attention and these new eyes onto the project that may not have been interested before? I'm pretty happy about that, I've got to say. I think it was nice because we didn't do a marketing campaign and tell everybody to say that it was just kind of a grassroots thing where people had

noticed i do kind of think maybe that silver blue thing got people's attention a little bit but also um honestly some of the work on audio on pipewire and all those things even though you're yeah i i feel your pain but like it's good now we we had, there was a funny effect of this where, you know, the target wasn't that we would make life easier and better for streamers and broadcasters, but you know, if you do, they're going to talk about it. So that, that, that effect certainly helped a little bit, but no, I really do think we've been, we've just been, you know, kind of quietly making things better and better. And yeah, you there were some times you know where it was kind of rough running the distro um and uh we've really really worked hard on making upgrades easy to make you know things run smoothly

to not be bleeding edge and people started noticing that and i think that's um that's that's what happened and i think people people also noticed, you know, hey, this really is, you know, despite the Red Hat link really is a community distro that is doing real important things. So yeah, it's definitely felt very validating to hear that. Of course, you know, then it starts being the, you get the pushback because whenever that's somebody's got to be the naysayer you got the the one person on rotten tomatoes who doesn't like the pixar movie uh wow they've got they got to get there you know so we you know there's some of that and and it also means you know you got to you got to keep executing and it's also one of those things

about um the incremental change is somewhat better. We don't always have something exciting every release. And so for that enthusiast thing, just like in the mainstream press, if we don't have something splashy every release, then we start getting back to being the has-beens again. So that is something, speaking of things we could use contributions to, there's a lot of good things are going on and things we could talk about. We could use help marketing it. That's an area we don't get any funding from Red Hat for.

Red Hat's funding goes into the products. That's fine. But we have stories to tell and it would be nice for each release, everything to come with those stories that we can talk about to, you know, people like you and enthusiasts and you know, mainstream journalists

so that they have that, they kind of keep the, keep that story of, yeah, this is there. And I want to make sure we don't lose that. No, that definitely makes it keep like the, keep the candle burning keep people paying at least a little bit of attention to it yeah yeah um hopefully in a good way

i want to want to say the good news is good news not yeah so one other thing was um i don't know how much you've looked into george castro's project universal blue which takes what fedora is doing upstream and then you know it's silver blue with batteries included. Do you have any thoughts on that project? Yeah, I talked with him a little bit about it. I think it's really exciting. And it's something that's kind of in line with our mission. I think whenever possible,

I would love people to come and do things in the project. I would like you to make, you know, a Fedora variant or spin. You know, if you have a new crazy desktop idea, why reinvent packaging and all the distros and try to build your own marketing team and docs? I mean, you can, fine, but it's open source. But we've got that stuff. Come work with us and we can make it a really cool place

to have a showcase for your thing. But there are other things that are sort of outside of project decisions. For example, that architecture, baseline architecture thing, like what if you do want to compile everything for the newest? If you've got resources you want to bring to the project to help us do that, awesome. Thank you, Intel, if you're listening, please. We'd love to do Clear Linux as part of Fedora. But I sidetracked myself. Not everything can be done that way. And I think that's a lot of what's going on in Universal Blue. It kind of gives them freedom to do things another way. And this is actually kind of part of Fedora's own

mission. We want to enable people to make their own solutions based on Fedora. And hopefully, we want to have a good relationship. So, you know, like, you know, RHEL as a downstream where we, that feeds back resources into the project and people working on it and things we would like to have things coming back from Universal Blue. And they're really interested in working that way as well

because it makes less work for them. The more things that are further upstream, the better. Yeah, same Amazon Linux. Those downstreams are an important success metric really as well. i'm glad actually really glad to see that too and i'd love to see more of that well okay another one so obviously the main focus of fedora isn't being like this is a gaming district it's a district that everybody if you're if you're running fedora you're running fedora like you if you're involved if you like doing gaming you're probably going to game on your system what sort of focus if any does fedora have in helping accommodate that sort of use case

yeah so i mean we definitely have people interested in that we have a gaming uh sig special interest group that kind of works for that. And we have a lot of people on the workstation team and on the desktop, you know exactly what you said. Like gaming is a fundamental thing. In our interviews with people, there's like, what keeps you on Windows? Well, I got a reboot for my games. I got to have a Windows system for my gaming, right?

And so improving that has definitely been something that's just kind of a, this is one of those, sure, we want to maybe focus on developers, but developers play games kind of thing. So there's some general interest there. There's not a lot of resources to devote to some of the fundamental work. But I think we may end up benefiting from things like GP, GPU computing. You know, that is basically, you know, the whole AI all runs on graphics cards kind of thing, where, you know, NVIDIA has finally been convinced that they need to do something different with their driver. And it is NVIDIA is by far like I referred to AskVidora before, I don't know, I may be biased,

but I would just guess off the top of my head that half of the problems there are something that was with NVIDIA, right? And there's very little we can do about it because it's like, I mean, even if we tried to make it as slick as possible, but like it's their driver and their code and we can't fix it. We can't do anything, you know, and we could hold back kernel updates so that they match. But like that's holding back a whole lot of other things, including security improvements and um you know i i don't feel like we should

hold the whole thing hostage to the one person who doesn't want to play nice with the upstream so finally anyways the gpu stuff has basically made enough big companies tell nvidia your way of doing this is not working for us we need this to be open source like like AMD has for a while and Intel. And I have to say, this is not a software purity thing about open source is the only way. In fact, from that point of view, it's not that great because AMD and Intel require a pretty big firmware that doesn't, it's not, it's running, it's in the GPU. a pretty big firmware that doesn't it's not it's running it's in the gpu this is a kind of art distinction in fedora space to draw a line on you know where's our hard line of freedom the firmware that doesn't run on the main system uh if it's redistributable uh it doesn't need to be open source and so basically nvidia is moving a lot of stuff from you know the driver to being

outside of the driver and maybe and there may be some other user space optional things. I don't know the whole details of how it's going to be. But so that won't, like, it doesn't advance the cause of, like, we're building this OS that we all own together in quite as far as I would like. And I would like to see more of that. And hopefully we'll get there. But it does solve our practical problem

of every time the kernel updates nvidia breaks everything um and so i i went when that getting there it's going to make a big difference from gaming because let's face it nvidia has a very strong um you know market yeah if not you not almost monopoly. Even though, especially in laptops, your options are very small. And also, whether or not... Being able to go pick what you're putting on your hardware is a luxury. And so a lot of people

have the system they have, and it's got the hardware it has, and we'd like to make it work. So I think that hardware enablement is kind of the biggest the biggest thing um i i don't know i am my my gaming is more of the like civ 6 and city skylines right and i'm very addicted to stolaris um which you know uh stars actually is very beautiful for no good reason because the amount of times

that I like zoom in to watch a gigantic space battle going in 3D, like normally I'm just looking at the high level view and, you know, I, yep, but it has those graphics, but, you know, it doesn't,

it's not intense. It doesn't need a high frame rate. So I am an easy to please gamer with a lot of these things um so other people with more difficult problems um you know i think uh there's going to be some work about you know high high monitor refresh rates and those kind of things that's happening but um you know it's it's not something that we have a really large investment in and we're mostly going to work on it's going to be the upstream projects. Yeah, just bring in whatever changes happen.

This does go to the, we're going to try and bring them to you quickly. So there's that. Well, last one is a very simple one. What do you, maybe it's not a simple one, actually. I can make anything complicated. What do you think is the most exciting thing happening in Linux ecosystem right now? Like maybe like upcoming tech that's being worked on maybe i don't just something

that yeah see i told you it sounds easy but it's not actually easy no i mean and you know it's one of the things that's easy to get wrong i do think the the rust thing is interesting i think um yeah moving to bpf for a lot of things that that's basically technology that went into the kernel for firewalling, but can actually be used to write little programs that do things. That's actually like an example, the desktop team right now. So you've got input devices like this that are all supposed to be just using a standard USB connection. But it turns out a lot of them are quirky and they have their own thing. So each of those things ends up meeting its own driver, which is a lot of work and kernel and whatever.

So there's a project basically to instead of doing that, use BPF to detect these quirks and just as a loadable thing, make all these things, make all of these things, just devices just work, which will be so much faster, easier, less work. So that's, that's really cool innovation. And I expect we'll see that being used for a lot of crazy things that we've never thought of in the kernel. I think that's, that's exciting. Obviously the whole immutable thing. Yep. Whatever we call it. And along with that, you know know there's the supply chain security

reproducible builds um all of those things uh it'll be interesting to see the impact of the large language model ai things i'm i'm very skeptical about a lot of those things but um there's um there's it's also important there's stuff going on it can do amazing things I think the trick where you can try and you know you can make it act like it's a human speaking conversationally and then get code out of it and stuff

that's kind of fun but it's that's really just very good autocomplete and it's there's interesting things that can be done with that but that's really just very good autocomplete. And it's, it there's, there's interesting things that can be done with that, but that's not really the key thing. There are a lot of other just things that can be done with the models that

are amazing. And I think we'll see more and more of that. And that's going to affect Linux, just like it's affecting going to affect the rest of the world. I don't know. I don't know what that'll look like, but I it's definitely, it's definitely happening.

This is, again, I'm going to leave out something important in all of this. Yeah, you know, that thing that we're talking about with, you know, Flatpak, FlatHub, that where distros focus kind of things. I think that's an important conversation. important conversation. And I think I, just for travel reasons and whatever, and COVID and whatever, I want to get to the Linux Application Summit to kind of talk to people about some of these things, because I really do feel like the value that distros have been providing is easily overlooked because of just the way we do it. And, you that, that value in packaging, it's a two-way conversation because I think we need to find a way to provide that value that isn't necessarily shove everything into five different package formats or a hundred different package formats or whatever we're doing.

But you know what? Maybe it is actually, it's not the worst thing. It's worked so far. And maybe we can make that easier and more automated. And I would like, there was once upon a time, a Red Hat spinoff company called RPath that had a build system called Connery. This was a before it's time thing. But part of their whole build systems philosophy was that everything would be, instead of having packaging build systems philosophy was that um everything would be instead of having packaging guidelines there was code and everything that was different would

either would be an exception so your packaging would just be what's exceptional and then ideally the the package definition for most packages would be blank. And they would be detected, okay, this is a Python package. It fits these normal things. Just follow the Python rules and build it. And I think there were some problems, not perfect, but that idea would make things so much easier if we don't have to have an art of crafting a package. So I don't know so i don't know i don't know something going going in

that direction would be would be useful i don't know not that anybody's that i know is particularly working on that i've not heard anything either yeah um you know and maybe we'll just have ai generate all those things and worry about it but uh when someone says the most exciting thing usually that means one thing that's fine no that's fine you mentioned seven different things there's a lot of excitement for sure um and because someone asked me to do so uh we have this as well yes fun yeah i've got um i've got two hats here you know this is the traditional one i won't go so. We have this as well. Fun. I've got two hats here. This is the traditional one. I won't go on over my headset very well. These headphones are falling off now. This is a bad idea. Yeah, that's the traditional red hat

hat. And then I also, some wonderful folks in the Fedora community, I got promoted to a fancy sounding title of Distinguished Engineer and they thought that the fedora didn't quite rise to distinguish. So I have this as well. It's actually, it's not, you know, it's not a costume hat. This is a very nice hat that I will not wear very often because it's not quite my style, but I happily display on my shelf up here. I, I do,

you know, we, we picked this name, you know, for the project back before. Yeah. When fedoras were neither cool nor uncool. And then there was a brief period where people thought they were cool.

And then there was a long time where people thought that they were very cringy. And, you know,ingy and you know um for like uh you know uh the linus tech tips couldn't help making jokes about the meme right and i'm like you know what we're older than that meme we will outlive that meme so when he when he first heard about fedora like isn't that sounds like a joke distro like yeah right and that's i that's just a funny point of time of how old that person is that meme was it made an impression on them you know

and you know uh i think the name is originally from some from like a russian play um that's where it comes from originally um you know so you know a thousand years from now when we're at fedora i don't know fedora linux i don't what version would be at a thousand years from now when we're at Fedora, I don't know, Fedora Linux, I don't know what version we'll be at a thousand years. 500? Yeah, you know, twice a year. Twice a year. Yeah, you know, Fedora Linux

2000, right? I did my math the wrong way around. 2000 something, right? People will not remember that name, but they'll still have it. Well, on that note, I think that's a... We can end the show here. That sounds good, because I think I cannot talk anymore,

so I am going to show them. I do this all the time, and I do a stream early, say, for like four hours. I've gotten used to talking for a long time, so I kind of forget, like, normal people don't do that often I actually do but

I had COVID in January and there's still some residual that so that's the world we live in but I'm glad this was fun it was worth exhausting my voice on so thank you for having me on you're always welcome man

this was really fun I definitely enjoyed this it one thing i do want to say is i i would like to see more people involved in like in the sort of distro space doing things like this and sort of putting a face behind a project sort of humanizing the like you you see a distra as just a thing it's a project it's it's hard to well it's easy to forget that there are actually people working behind it and maybe you don't agree with every single perspective they have but there are still people that have these these things they've thought about maybe they've been involved for like 20 years and i think i think that's one thing I do like doing with this show, bringing those people here,

just letting people see that, hey, there is someone actually here. Yeah, absolutely. That's incredibly important and useful and good for our projects, good for community-built software, good for the world. So, yeah, thank you. And I'm sure there are other folks in Fedora space we could have on your show as well.

Awesome. To talk more about their areas and what they do. Well, if you ever want to have a free marketing budget for Fedora 39, feel free to come back on for that one or whatever you want to do. Sounds good. I'll be here awesome um let people know where they can find you anything you want to promote the fedora project obviously yeah yeah you know fedora spaces i try to get the username matt dm most places um i am no longer actively using twitter you can find me on uh on mastodon um go to MattDM.org. That's the Matthew Miller, that is me

of the websites. And yeah, of course, in Fedora spaces, chat.fedoraproject.org, discussion.fedoraproject.org, MattDM. Your website seems to list pretty much everything, doesn't it? I tried to there.

This is a website that's made out of hand-built HTML. I love it. It's the 90s, so there it is. Absolutely does the job. Anything upcoming that you specifically want to direct people to? I pitched it before, but I'll say it again. devconf.cz. It's in the Czech Republic.

It's a lovely conference. It's normally in February, which is a terrible time to be in the middle of the Czech Republic in Brno. But June is a lovely time. It's a beautiful little city. And I assume things will be streaming. They'll be recorded otherwise. There's a lot of fun talks.

And my talk about 35 Fedora releases in 30 minutes goes kind of into history of the project and lessons we've learned. And I've tried to make it be fun for people who want to know more about Fedora or who have been around and want to reminisce, but also for other projects who could maybe learn something

from the mistakes we've made over those years, as I talk very, very fast and will definitely lose my voice during that process. Well, hopefully it doesn't go too bad. Hopefully you can still speak. Yeah, you know, I'll make sure I bring some honey tea with me. That's definitely a good idea um is that everything you want to mention i think that's the main thing i'm sure there's something i'm forgetting but let's put that as the highlight and if there's anything that you feel like you definitely need to mention now just send me an email message or something i'll just put it in the description or something like that sounds good sweet as for me the main channel is oh wait one more thing another

conference fedora conference flock to fedora right we've got our own conference right and so this is a contributor focused conference it's not necessarily a user conference but um people are welcome to come if you're just interested that's in ireland at the beginning of august um it's flock to fedora.org is a brochure website for that. And the call for papers or proposals or whatever is currently open. If you've got something interesting to say about Fedora or to tell Fedora developers

something that we ought to know, submit and come to, it's in Cork, Ireland. And it's going to be a fun time. Our first in-person conference for a while. So yeah, I really, really should have put that like a big sticky note. Don't forget to pitch flock as my thing. So I'm glad I remembered it in the last second here.

Yeah, let's send people to that. There's nothing else you want to mention. Oh, I'm sure there's something else I want to mention. Nothing else in my mind. I'm not queuing anything up right now as I'm doing this. Oh, let's stop. It'll go on forever. That's calm.

Time for calm. As for me, Linux Channel, Brody Robinson do videos there six-ish days a week. I don't know what's going to be out when this comes out because I'm way ahead in my schedule. So this is going to be out in two or so weeks. It could be anything. I don't know. Well, the call for papers

may be closed by then, but check and see. Yeah, that's a good point. But it's going to be open, I think, until... I don't know when it's open until... Check anyways. And you can... The conference, we have a nominal small fee for attendance, but we want to make it affordable for everybody.

So it's not expensive. Come and see us. Awesome. I have my gaming channel where I do gaming stuff. I guess when this comes out, I'll be playing Final Fantasy 16. So pop on over for that if you want to see me play that. I'll also be playing through Yakuza 0.

Very good game. Highly recommend it. If you're listening to the audio version of this, you can find the video version on YouTube at Tech Over Tea. If you're watching the video version, you want to to the audio version of this, you can find the video version on YouTube at Tech Over Tea. If you're watching the video version and want to hear the audio, that'll be available on any podcast platform, RSS feed, chuck it in your favorite app,

and you'll be able to find it. Or you can close your eyes, I think. That also works. That'll see if you can do that again. There are people who tell me they go to sleep watching my videos. Like, how do you do that? I don't understand. Like, i know they're trying to some of them saying i'm boring but like if you hear my voice and you go to sleep like i don't know how you managed to deal with that noise in the back of your head you're trying to sleep yeah i i won't i'm just gonna not touch that one that's where we're gonna to... Sorry for interrupting your closing.

I just couldn't help inserting. Totally fair. I'll give you the final word. What do you want to say? Thank you for having me, and thank you for doing this to talk to all the people in the communities and supporting everything. It's really valuable, and I appreciate it. Thank you. Thank you.

I definitely appreciate that you came on here and gave me your time for this. Awesome. I'll see you guys later then.