Tech Over Tea - Arch Linux GUI Is Dead, This Is Why | Demonkiller

Episode Date: August 3, 2022

Recently the Arch Linux GUI project was shut down so today I brought the maintainer of the project on to discuss what happened and why he made this decision along with what he plans to do in the future. ==========Guest Links========== Arch Linux Gui Website: https://archlinuxgui.in/ YouTube: https://www.youtube.com/channel/UCt1mvddXeU41hJmzrGYJF5w Github: https://github.com/arch-linux-gui Discord: https://discord.com/invite/NgAFEw9Tkf ==========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 126 of Tech for Tea. I am, as always, Brodie Robertson, and today we have the, I guess, maintainer, former maintainer, not really sure what's happening going on into the future, of something that you might want to call a distro, if it's not really a distro, the maintainer of the project, formerly known as Arch Linux GUI, how are you doing? Hi Brody, so yeah, I'm really doing well. Arch Linux GUI is discontinued, and it's now called Arca Linux GUI to settle the trademark thing, but yeah, I'm doing great.

What about you? Yeah, I'm doing pretty good um i guess we probably get like straight into the whole like what's going on with that project because i know a lot of people who are like i had some people in my comment section the other day who are still talking about hey this seems like a great project to use like and had no idea that anything had actually happened to Arch on this GUI so why don't you just start by explaining sort of what's happened and why you made the decision you made yeah so uh when the project started growing a lot of it came under

the radar of a lot of people that were closer to the Arch team. So there has been a lot of effort in the past to make a GUI installer for Arch Linux. And obviously that has never been put into practice because, you know, there's a common resentment within the, I don't know if I should put it like that, but there is a resentment in the Arch Linux, the core team somewhere that a GUI installer shouldn't exist. If you look at the FAQs, it's a debatable topic. Some people are for it and some are not. So it's, I would say, a 50-50 thing.

But basically, when I started the project, my main goal was to solve three problems with the main Arch installer. So first is obviously the text user interface. Maybe that can be like ignored if we talk about how good the wiki is. But when you talk about things like because the installation process is a lot online, you need to get connected to some kind of internet and then download packages. So it works for a lot of countries that really have good internet connection but where I have personally experienced this when like before reflector was a thing downloading packages sucked really you would have

kilobytes and and before pac-man 6 came out this was really a big, big problem. So you didn't have parallel downloads, right? So each and every package took a lot of time to download. So what if we had an option to install Arch Linux with all the packages frozen in time so that we could get the entire operating system running without an internet connection? So that was one of the main uses apart from having you know just a GUI installer. And then the other thing is that whenever you install Arch Linux on a

computer you usually need a second device. You're watching the wiki or I mean you're watching a video or a wiki, you need a laptop or a mobile phone something like that. So when you have a gui installer you have basically endless possibilities you could use something like calamaris or you could just use the wiki and i basically recently did a video on how you can do that so that was the main goal of the project now like i said there is a resentment whether a GUI should exist or not exist and then the whole thing that transpired to be is that someone or somebody just you know told the Arch Linux team hey a project exists like this and yada yada yada. So the Arch Linux team basically

says this you can't have something known as Arch Linux something you you know. Sure. There are projects like Arch Linux and Arch Linux 32. They are official projects because the Arch Linux team says, yeah, this is the official Arch Linux system project. So my project was basically violating that. So Arch Linux GUI kind of makes you think that this project is affiliated to Arch Linux. So that was one of their problems that they pointed out. And secondly, they said that anything that uses Vanilla Arch Linux in the description or whatever on the website,

you can't have Vanilla Arch because Vanilla Arch Linux is basically Arch Linux installed via the command line installer with everything that's official. So not even the AUR. So I had a chat with Levente, who is the, I think I'm pronouncing it wrong, but yeah, he's the lead developer of Arch Linux. Okay, yeah.

He told me that even an AUR helper like Paroo or Ye, if that's there on the system, it's not vanilla Arch so I can understand like coming from their perspective where they want to make sure the AUR is this sort of like weird separate thing they know that everybody on Arch

uses it and most people use an AUR helper but they still want to like keep their like that weird separation there I don't understand why i understand like the internal logic that gets them to that position though yeah well i think it's more to do with infrastructure because arch Linux as far as i know they do not uh operate on a donation they don't have any kind of like money making thing

right uh as far as i think they just accept donations i'm not sure if they have anything else actually i think they try to minimize a lot of things and weed out a lot of extra things that will in the future kind of impact on their infrastructure i was sorry i was mistaken they do have a couple of um sponsors but they're like uh there's sort of sponsors that like a youtuber would have like shells and things like that nothing like nothing like upwind who has for example yeah yeah exactly so i think that they are pretty uh they have a lot of constraints on their infrastructure for example a lot of packages that exist on the aur exist there because simply because if you were to put them on the official repositories,

they're going to take up a lot of space. So I think it makes sense from an infra point of view. But when you have a product that is related to you in some way, which the AUR is, trying to distance yourself from it is kind of sketchy, I believe. And you're going to expect users to be using... I mean, you have articles on that, on the AUR, on the wiki, right?

Yeah, yeah. If you would want to distance yourself from it, you wouldn't have it in the first place. No, I understand that. I don't... What does the actual description say right now? Something like, it's not officially endorsed by Arch, or something understand that. I don't... What does the actual description say right now? Something like... It's not officially endorsed by Arch or something like that.

Give me one second. AUR Arch Wiki. It says... AUR packages are used for used contents. These package builds are completely unofficial and have not been thoroughly vetted. Any use of the provided files is at your own risk." That's a fair warning there. There used to be a different one here, maybe it was on one of the other pages, but I do remember there being one specifically

saying they don't recommend using the AUR. Yeah, and that's kind of weird because if you have all the packages, for example, if you look at Debian and APT, most of the packages are there in their repositories. And anything like what Ubuntu does, what Ubuntu's AUR, you can say, is that PPA stuff. Yeah, Fedora has the same thing with Copper. Yeah, so every project I believe has some kind of constraint because when we talk about free and open source, it's free not in the sense of money but freedom, right? So money is always something that, because whenever you're running something on servers, it's going to be constraints on the infra and that's the reason why I use chaotic in some of my packages because I face that constraint myself as well

so that's pretty understandable but then when you like for example I have this repository called ALG repo if I tell people just stop using ALG repo you should not use it or let's say let's say Garuda Linux comes and say you should not use chaotic it kind of sounds weird because it's your stuff

and then you're telling people not to use it. So that's the position I have with AUR. That's the kind of opinion I have with AUR. I can see, okay. I think there's sort of two sides to this. I can understand having a warning there about the fact that it contains community-driven content because they have moderators there

that will get rid of things that are blatant malware, but it's not like they're constantly combing the AUR to make sure it's always going to be safe. So making sure people actually understand what they're actually getting into when they go and use it. But yeah, I don't see... I personally personally use the aur for a lot of things i don't really see any reason to discourage its use just make sure people know what they're actually getting into before they start using it exactly exactly anyway um before we go like way too far down the tangent um i think the idea of a arch maybe not

an arch dewey but an arch installer that's not just like the manual installation process is slowly starting to change because what was it like a year was a year ago maybe a bit less than that i don't know time i've lost all track of time in the past couple of years. Arch brought in Arch Install, the script to automate the installation process, which obviously isn't a GUI installation, but it is a step

towards a much simpler installation. Yes. I don't think there's ever going to be a point where the main Arch project fully endorses having a GUI. I just don't think that fits with the goals. No, I think it's in the works because like someone, if you look at the Arch ISOs issues, I talked about those recently. I talked about recently someone has suggested a GUI like not in GUI installer as in Calamares but having a graphical user interface like plasma and

then you have a browser like Firefox because the official installer has links which is a TUI installer right you know you have when you are in the TTY you have to navigate press alt and your arrow keys back and forth between the window you're installing on and then the one that has links up and running. Like to be honest, links is not, and not just links, any TUI web browser is not the prettiest, right? It's not, it's just going to render HTML. It's functional for the Arch Wiki, but the other problem is you don't have something like Tmux, so you have to jump back and forth you can do it on one device so if you have like just plasma running let's let's just ignore calamaris for the moment if you have something a plasma and then you

have a terminal window and then you have a browser like firefox you could just spawn these two on one screen and then one it allows copy pasting of. Let's say you want to discourage that as well. It's fine. You can type the commands out, but you have everything on one screen. Oh, so what you're saying is like bootstrapping it from a graphical environment rather than the two environment. Yeah. So that is something that is suggested on the Arch ISO issues right now.

And the really positive thing about it is that David who is the maintainer of Arch ISO, he hasn't said no to it right. So I think this is a very positive thing to consider. I believe that like in a couple of maybe months we could have potentially a separate profile which because it makes sense to keep the tui uh interface yeah yeah it should be there right because that's what arch linux is and that's how it all started well it wouldn't be out of like out of the ordinary to add an extra mode into the installer because a while back they added in the um what was it They added in features from

I can't remember what the project was called, but it was an installer for visual impairment. The project, that original project is dead now. It's called Braille TTY. Yeah, yeah. They had a separate mode for that. I think it had a screen reader and a couple of other things

built into that mode as well. Yes, you could boot up with either just the TTY and then one with the screen reader and a couple other things built into that mode as well yes you could boot up with either the just the tty and then one with the screen reader yes yeah so what has been like suggested on the issue is that basically you have the main profile uh which is the tty which is for the tty installer and then you have a separate profile which is with some kind of gui so now the thing is we have a lot of options on linux you have someone separate profile which is with some kind of GUI so now the thing is we have a lot of options on Linux you have someone likes Plasma someone likes GNOME my idea is basically just get something that's running on X or Valent it doesn't have to be something really

fancy you just need to spawn a terminal window and should be able to spawn some kind of install some kind of web browser and you're good to go. So a separate profile makes sense. So one of the things that I like kind of proposed to Levente is that if part of Arch Linux GUI could be made official, so that's the whole conversation was about like I have a lot of stuff from the AUR which shouldn't be there and then this is not vanilla arch Linux and this and that so if I have to make the project if I have to try to make the project official then I have to make certain changes which obviously requires removing all of the team settings customizations everything yeah and then removing all of the AUR packages including the AUR helper that's also not allowed and then obviously removing Calamares

right so that's one way I could make ALG into an official project but the only caveat here is that I would have to convince everybody in the team to like tell convince them that my method of doing this is what should be adopted and yeah so the reason i decided to discontinue is because this i know that since so many years this has been a 50 50 opinion whether a goo installer should exist or not exist i know that someone in the team would veto this out right so all the legwork legwork would do to even try to make this official, I don't think it would be worth the effort it would require to just do it. Because at the end someone says that no, this idea just doesn't fit in the philosophy because when we are talking about art we have to talk a

lot about philosophy right so it doesn't fit this philosophy or that philosophy then you know right right all the work is waste basically i yeah that makes sense i would love to see like an, an official, like, arch. I don't know how, like, I understand not wanting to use Kalamaras, because that brings it a bit too close to things like Manjaro, and maybe they don't want to have a fully automated graphical installation, but at least having that graphical environment. I don't think Weyland, just because of the NVIDIA problem.

Yeah, absolutely. You would have to do it on X.org, which, that's fine. It wouldn't be out of the ordinary for the Arch installer to ship extra things that don't get shipped with vanilla Arch. For example, it uses ZSH. So, also shipping extra things, it's not that weird.

They won't end up being a part of the thing you install at the end anyway exactly and i think you want to uh you know uh if you want to uh worry about infra at the end what you can do is keep the official arch uh installer which is the tty obviously uh to get it from whatever mirrors you have right now and the rest of the isos let's say you want to offer gnome plasma and xfce or maybe just one you can probably provide it via source for joe sdn or you don't even have to like spend extra for infra like all of these things can be taken care of like

pretty easily so like infra isn't an issue i think it's more of the philosophical thing that exists with the yeah there's the whole idea of you know keep it simple stoop and things like that where i think if it was going to become an official project you'd need to have support within the team before you took it on just in case you know someone did come along and say this isn't going to work. If you already have that initial support from at least some members, then that gives you a bit of extra,

I guess, leeway to push that forward. Yeah, absolutely. If you happen to have those, the links to those issues or know what the issues were called, can you just send those to me afterwards? I would actually love to look at those. Yeah, absolutely.

It's, by the way, Arch ISO 170, but I'll send you the links, no problem. Yeah, no problem. Awesome. Yeah. So you've spoken a bit about Arch Linux GUI. So temporarily, and I guess for now, the name's going to be Arca linux gui even though you've discontinued it uh just let people know what like why you want the name arca linux gui

so arca i was basically looking to keep the alg acronym right because a lot of people uh they don't just keep saying arch Linux GUI they call they call it ALG yeah yeah so I want to keep that acronym and uh Arca Linux is already taken so I was trying to find uh the like a derivative term that means Arch in another language so I did a bit of research and Arca is basically Arch in I think Ukrainian or like Eastern European languages Lithuanian or whatever. Yeah. So it's ARCH in another language, and then I can keep the ALG acronym. So that's why I chose that name. So what are you thinking of doing with the project? Is it just dead forever? Are you thinking of bringing it back at some point

or adjusting to something else? Like, what exactly is your plan for now? Or are you just not really sure just yet? No, I do have a plan, actually. I don't know if I can say this at the moment, but I've got some of the developers of the other distributions. They asked me if I wanted to join their project.

Okay. So I'll talk about that if I do. I just told them that I want a break right now. You don't have to give me specifics on which projects, but that's fine. Yeah, I mean, I'll just... Because I haven't joined any projects. So if I just take their name and then it doesn't materialize... Yeah, yeah, of course.

It doesn't make sense, right? So if I do join them, then I would obviously reveal it on my channel. And I'll tell you, obviously. Yeah, we have. But as far as ALG is concerned, for the foreseeable future, I think it's gone. Because unless and until there is some kind of effort to make it official, there is no point in continuing, given the kind of... I mean, the community is great. The community around ALG is great.

And when I announced that the project is like being discontinued at least like, I don't know if it's forever, but yeah, for a foreseeable future, it's gone. So a lot of people were like, yeah, I mean, we use this and some of us are daily driving with it and it's sad, but then, you know know you've got to respect the parent projects policies and whatever philosophy they have so there's no point in really

you know grinding out again and I think as far as what I wanted to achieve with ALG and the message that it gave like to the general arch community is that a GUI installer is feasible. Out of all the, you know, by the way I use Arch and all of those meme things, the whole point of having the GUI installer at the end is time. Installing Arch Linux takes time, and if you want to do it again and again, because given the rolling release nature, something's going to break and there will be a time when you have to to do it again and again because given the rolling release nature something's gonna break and you know then there will be a time value when you have to reinstall it in some way or the other or maybe you have multiple multiple machines so GUI installer does that quickly and this is

like one of the most common messages I received when the project was discontinued is that wow like we had this means of installing Arch Linux, which was really not only simple, but efficient in like time point of view. So yeah, that's what I've decided about ALG right now. Yeah, I've been very big on the pushing forward for installers of Arch. Like I get the, oh, hey hey you got to install arch manually you learn whatever you learned i i think the whole you learn a bunch from going through the install process is kind of nonsense like you if you dig down into each individual step sure you can start

to understand a bit more about your system but once you've done that once there's not really much you get from doing that again again again again so i don't think it's just necessarily with the gui and so i do like having a gui installer as an option because you know there are people out there who prefer to have a gui but i've been pushing just generally forward with having some sort of installer like there's no reason to just have a manual installation. The only other thing, like, the only other distro

that has only a manual installation that comes to the top of my head... Actually, sorry, there's two. Bedrock, because the whole... Bedrock, the whole point of that is doing your own thing. And Gentoo. And Gentoo, you know, Gentoo is Gentoo.

Gentoo will do whatever Gentoo does. Yeah. But for anything else, I think having the option of a automated installation process, whether that is a TUI or a GUI, is a good thing because most people, the

actual install is pretty much the same. Like, you're gonna set up your fstab file you're going to set up your file system it's going to be one of the standard file systems you're going to have your petitions you're going to have you know the standard stuff you would do to have your system functioning the thing that differs is the post install i'm not a big fan of i know that alg did have some of this but i'm generally not a big fan of any post-install stuff in the installers because that's usually a bit a bit more of a

personal preference thing yeah i mean people are free to have their opinions so some people like that's that's why i had these pure and themed editions so people want just the like vanilla arch plus um okay when you say vanilla just people want arch plus some kind of you know stock de you can use the pure editions and if you want all of the nvidia stuff configured and you want the browser installed and theming and stuff like that basic stuff you can use that so you have i think linux is all about options right you shouldn't be like just stuck to, no, I should only install via this method. I should only have these repositories

and only these packages. The whole point of Linux and the options we have here is that I can have, I can have a system on, I can have an operating system on my computer whichever way I want it. And I can choose whatever packages I want

as long as I obviously trust the developer, but I can choose whatever I want on my system. And that's also one of the core philosophical points of Arch. You build your system according to yourself. But the kind of thing that people argue about, you learn about Linux in general, I believe that's not true to a great extent as far as Arch is concerned. because if you look at the installation process It's what, Packstrap is you partition, you generate a first tab and all of stuff. You get Packstrap to create your root file system and then you install the bootloader microcode and stuff like that And then you create your hostname and stuff, but that's not really learning a whole lot about Linux. Anyone who has used Linux in the past, whether it's Manjaro or Ubuntu,

they would know this stuff pretty much like in some way or the other. The package manager might be different or maybe the default file system might be different but in some way or the other it's kind of same because it's all Linux right. If you really like want to learn about how your system or how your operating system works I think Gen2 and its installation is the way to go if you want to like learn about how everything is being made how everything is compiled because as far as repositories in Arch are concerned we have packages we are still in Arch talking about packages when we come to 2, we are not talking about packages anymore because

then we are locally compiling everything on our machine. So I really, and this is my personal opinion, that I disagree with people who say that if you don't install Arch the wiki way, you're not going to learn about your computer because at the end, for example, if you have Arch install the the script that's doing the same kind of thing a GUI installer would do for you as well so it doesn't matter if you have it if you want to keep the installation the wiki way it doesn't matter whether you are doing it on a TTY or a GUI like plasma because as long as you have stuff like arch install or any kind of automated script, it's basically going to

leave you as clueless as a GUI installer would. Even just going through the wiki and just running the commands blindly is going to, like, you won't learn anything from that. Yeah, exactly. Exactly. One thing I would disagree with you there is with Gentoo. I've gone through the Gentoo installation process. It's not that

different from Arch. Like, it can be. If you want to go through that process of like, you know, compiling the kernel yourself and customizing all of the compile options, things like that, you can absolutely go and do so. But if you want to take the easier path, you can

basically do an arch style installation on gen 2 with no problem whatsoever the one thing gen 2 does have much better though is the documentation for the installation is just miles ahead of arch like everything has like two or three paragraphs explaining what every little thing does. Arch, it's like, hey, do this. It sets your time. Do this. It sets your keyboard. Do this.

It sets your file system. That's basically it. Exactly. If you really want to learn about how Linux works and how it's all pieced together, honestly, the best way to do that is just sit down. this isn't what you would do for a daily drive system but sit down and go through linux from scratch it will take you a long time but if you really want to learn about linux that's the best way to do it yeah absolutely because that will also like most of the like, for example, Arch Gen 2,

you get a package manager out of the box. When you start from LFS, you don't even have a package managers to start from. So everything that you want in the system, you're compiling it manually, and installing it manually. And like a very good example of this is KOS, if you know that, KOS, KOS, whatever it's called. They're also, I think, an LFS system, but then they use Pac-Man as a package manager.

So, if you really, like, want to learn all the hard code internals from scratch, yeah, LFS is the way to go. You did mention, like, I don't remember if you mentioned this earlier, it was just on the Discord, that you sort of felt, like, I don't know if you mentioned this earlier, it was just on the Discord,

that you sort of felt, like, burnt out from the project. How much work were you actually putting into Archling's GUI and sort of what really went into maintaining it? So, well, I started the project in the pandemic. It was, I think think December of 2020. Wow, I thought the project would be around longer than that. Yeah, like the planning and everything was because the whole purpose actually to begin with was to make an ISO that would run on RISC-V. It was this project that I was working on separately. I had to get an operating system so I tried to get Garuda but

obviously like that wasn't feasible. So I reached out to the Garuda leader Srinivas who happens to be a friend and I asked him like how do I get into building my own slimmed down ISO and he was the one who pointed me to ArchISO and then obviously got a lot of help from Eric from ArcoLinux. So when I had the project it was running well on RISC-V but then I just got bored out of that project and then I really enjoyed building ALG. So because it was the pandemic I initially could spend a lot of time. Like if I have to give a number, it would be like 12 hours a day. Because I was still figuring things out. I was reading a lot of documentation.

And I was obviously not the best programmer out there. So I had to learn a lot of like, I was learning shell scripting, for example. So things like that, like reading documentation and figuring stuff out, that took a lot of like I was learning shell scripting for example so things like that like reading documentation and figuring stuff out that took a lot of time but once I had Calamaris set up as I wanted it and it worked on x86 the way I wanted it I got it registered I think in January on source forge and then people started like using it and reviewing it and then it started to grow gradually but so so that point of time it was chill because nobody knew the project existed and people who knew it it was a small number i had like 10-15 downloads a day and it doesn't really matter so now what happens is uh armano from EFLinux made simple.

He covers the project. And suddenly I have this boom. And so many people start coming in. And when any project expands or it scales, you get new issues and new stuff coming in. And then new requests. So with that, the kind of pressure started increasing. And I was like, I don don't know lucky or unlucky i'd say lucky that armano covered uh the project very early on like i think april so that was when the entire thing like really took off and then other youtube channels including

yours covered it so i got started getting a lot of requests about hey you have plasma right now why don't you get no why don't you see so now I have to like read the documentation of gnome and the documentation of XFCE and i3 and a lot of these things like maybe gnome and XFCE I've used before I've used plasma before like but I get like BS PWM and that's something I've never used before. So I have to read the entire documentation, see how SXHKD, that's how it's configured, how the BSPWM.5s are. So that really took, like, a lot of times it was all-nighters, so I used to sit all night, and it would be like 15, 16 hours daily.

Wow. Yeah. night and it would be like 15 16 hours daily yeah so it was yeah and this happened uh continuously for a year until uh like the pandemic eased off and lockdown was over so like we finally had to get out of our homes go to college go to work and stuff like that so then obviously i had to uh i couldn't spend much time but by the time this happened, the project was already like I had 12 editions and everything was stable. So now it wasn't about coding anything new or, you know, like building anything new. It was all about maintaining it. So I could just maintain a bit like the given source code I have. I just had to maintain.

So if anything changes recompile it and anything new is there included or excluded so that was like I think from March 2022 that was pretty chill but yeah before that when the entire thing was being developed when the request came in and then I got a lot of like obscure hardware like hey your project's not working on my hardware and then I have this archaic intel CPU and stuff like that so I that really took a lot of time so the 15-16 hours that went daily into it uh were because of that reading documentation figuring out people's requests on uh like support their hardware which i don't have obviously and i think that's the

reason why the project grew because i tried to help like each and every person out there and i was doing all of this alone so by the time like i reach here and then you know i have this trademark thing on on the project uh i really feel that i hadn't like, and I haven't been like, for example, on a holiday since like 2019, before the pandemic. So yeah, I just felt that I, it's a lot of time and energy and effort put into the project. And I also like put my YouTube resources, like my ad revenue to get hardware, stuff like that. So I think it's like at this point in time it's pretty much good to say goodbye to the project because I want to like explore new things as well and yeah the project served it purpose what it was supposed to do and then like I said earlier

it there is a like a notion about a graphical install in the community now that people will now talk about this more whether art should have a GUI or not have a GUI something which wasn't too much talked about in the past so what you're saying is the YouTubers are at fault what?

the YouTubers are at fault for giving you too much work no no I think what people sort of misunderstand is your fault for giving you too much work. I don't know. I think what people misunderstand with a lot of projects, especially when a project is doing something really well and people are giving it a lot of positive coverage, is

a lot of projects, not just distro projects, but software projects, just general software projects as well, a lot of them are maintained by a single person and when you have, I guess when you have that spike in users, it is going to

massively inflate the amount of work you need to do yes especially with the distro I keep calling it a distro how would you classify Arch Linux GUI just as an installer for Arch, as a distro. It's not really, like, I don't know. How would you classify Arch Linux GUI? Just as an installer for Arch, as a distro? What would you say?

I mean, I'm open to opinions. I call it an installer, but many people call it a distro, so they're free to call it whatever they want. Mm-hmm. I think with whatever we're going to call it, when you have some sort of installer like this people expect whatever it is that they use is going to be available whether it's you know one of the main things like kde you know which generally you you know you'd expect to be there but then you start getting into like more obscure

stuff like vspwm and you can go down the window manager rabbit hole for an entire year and still not run enough stuff to go and include in there. And that's without even getting into the stuff that a few people use. Yeah, absolutely. And as far as window managers are concerned specifically, I got so many

requests. Some of them, some window managers I even hadn't heard of, right? So I had to like search the ArchWiki, what is this? And then search YouTube, what is this? But then I like I thought I had thought about, for example, I got a lot of requests for DWM, awesome. But then I had to draw a line somewhere because like this, the number of additions would just exponentially increase so then i just decided as far as window managers are concerned i3 and bspwm would be it and then i i had to like tell

people that hey i mean i love your enthusiasm but you know for me one person it's not really feasible to you know maintain so many and and like you said it's just some people who are going to be using it as compared to Plasma or GNOME, where thousands of people are using it. So, at one point in time, I had to draw a line, obviously. It also came with DEs, like, for example, dpin and this budgie. People were requesting. this budgie people were requesting in fact i also uh phased out mate just for the sake because like a lot of the amount of people who are using mate was really less as compared to no so it was just a sentimental kind of thing those who are just used to mate they are using it but yeah i had to draw a line there as well yeah, I think that's one line

that's easy enough to draw. When it comes to the hardware issues, that... I don't even know how you would go about approaching that. Especially when... Because there is this... Linux has this wide array of supported hardware. I think

I want to say if the oldest thing the current kernel supports is the Intel 486SX. So everything from there up until now is supported by the Linux

kernel. And there's a lot of things in that range, and there's a lot of things that could definitely break. And for a single person, I don't even know where you'd begin trying to look at that. Yeah, and then we also have to look about graphics, right? So, for example, NVIDIA drivers. So that was another big, big, like, I wouldn't say a headache.

Like, initially, it was a lot of work, because I wasn't wasn't like really abreast with all the nvidia nomenclature and how they named their products so it took me quite a time like a week to understand like how and then i like someone someone i know works in nvidia so i just got in touch with them just to understand what are the different types of cards on desktop what are cards on laptop and stuff like that because like for example recently someone said they had a 3050 ti and i was like what this card exists because on desktop it doesn't right but uh then they told me oh it's on the laptop so okay when you're considering laptop use of like hybrid graphics to worry about as well and that's a whole nother can of worms to open yeah that's another another big big thing that i had to worry about well hey you got out early because uh give it a couple of months and you have another set of uh gpus to worry about when uh

intel desktop cards roll out i guess the laptop cards are out right now yeah but uh as far as intel is concerned it's pretty easy. Everything's in the Mesa package and anything that you need outside of Mesa is usually the older archaic stuff. So Intel is not really something that most of us distro developers or project maintainers we worry about and I think the same has to be for AMD but as far as whenever Nvidia comes into the picture then we have to for example when I introduced Nvidia the 700 series was supported which is the Kepler architecture yeah but soon it was phased out so then anybody who has like a 710

or what 730 or even a 760, you suddenly lose all the support. And the funny thing is anybody who has a 750 Ti, they have a support because their architecture is not Kepler. So it's like all of those minute things that came into play and that's like a lot of work that usually people don't see. That's behind the scenes. The main concern I was bringing up with intel i know they obviously have the integrated graphics but whenever you launch a new product skew it's always

going to have some sort of like early teething issues obviously it's going to be using those mesa drives which already existed but there's going to be like specific issues that show up when those desktop cards drop like it's probably not going to be as bad as like a whole new company doing it but there's gonna i i guarantee there's gonna be something wrong on day one yes there's going to be but um intel is one of those companies that has uh like contributed to open source a lot yeah yeah it's not gonna be as big of an issue as nvidia doing it yeah even if there is for even like not even day one let's say a couple of months in the beginning intel is going to we

we know that intel is going to you know find a fix yeah at least six months yeah so that issue uh isn't really really big even with amd i believe they are also like contributing back to the open source community so even with their drivers like recently when the 6000 series launched Wasn't much of an issue because with people did have an issue I believe someone on the discord or telegram group was saying they had a 6700 X or something and the AMD GPU

Driver just didn't work. And then all I told them is pretty much I can't do anything about this I'm pretty sure AMD will release an update and they did which fixed the issue so as far as Intel AMD are concerned they do it like in a very short amount of time yeah yeah what are you running on your personal system? I'm running the ALG Plasma Edition at the moment. No, I meant hardware-wise. Sorry. I have a 12600K which came out

and as far as GPUs, I have a 750Ti. That's mainly for testing. Ah, right. Okay. I'm using the 12600K's iGPU because I don't need anything. I don't do a lot of gaming. That's a weirdly skewed system.

Yeah, that is. Actually, I was looking for something in the 30 series, like the 3060. That would be like more than enough because I also do YouTube videos. So that would help me. But the prices are really like not sane at the moment like even uh right now the gpu prices have gone down really well but it's still not you know in that msrp kind of thing at least in my country so yeah it's about the same here um for i was looking at buying a 6700 XT and from what I've seen it's like maybe a hundred to a hundred and fifty dollars above MSRP which is still it's much better than it was because this time last year we were looking

at like double the MSRP yeah and and right now for example 3060, if you are looking at the OC cards from maybe Gigabyte or Asus, they're selling at the price of $3070 at the moment, which is not a lot. But yeah, last year it was definitely double of what it is, like the double of the MSRP. So the situation has improved right now, but it's still not where we want it to be like post-COVID. right now, but it's still not where we want it to be, like, post-COVID. Part of the situation improving is NVIDIA

kind of, they banked on the crypto boom, like, continuing longer. They had a, um, I don't know if you heard about this, but they had a contract with TSMC where they said, you know, manufacture all of these chips.

They were trying to get out of the contract and they were like, no, you're going to buy these GPUs whether you like them or not. So it's probably going to continue with the pricing at least probably for at least another year and hopefully that's enough time to get back to like somewhat reasonable supply chains i would hope yeah actually the uh yeah nvidia is at blame but uh as what what problem in my country is that the people who uh not the scalpers the people who sell these cards, right? Yeah. They bought it in excess to sell it to scalpers. Yeah. Or to the miners.

Now, what has happened is the demand, because crypto crash and all of that stuff happened, so the demand has gone down drastically. So they have these cards which they bought for these very high prices, and they just don't want to sell it for like the near MSRP price. So they're still holding the cards which they can sell, but they don't want to sell it at a loss because they bought it at like double the MSRP. So there's no shortage actually. The cards are in the country and they have it, but they don't want to sell it for a loss. Yeah, I've noticed the same thing as well. There's a couple of cards here and there out of stock, but as a general market overview,

there are cards in stock. It doesn't really make any sense, the pricing that we're still seeing. It's better than it was, but I would like it to be just a little bit lower. Yeah, absolutely. So one thing you mentioned on the Discord was thinking about, like, if you were going to, like, continue on with the distro thing,

you wouldn't want to go and turn ArchLinks GUI into its own, like, separate distro. If you wouldn't want to go and turn ArchLinks GUI into its own separate distro. If you wanted to do that, you would go and actually make a whole new project to do that goal. So have you still been thinking about that? Or has that sort of been replaced by potentially joining one of these existing distros? It would be joining one of these existing distros? It would be joining one of the existing distros and the reason for that is basically we don't need more

and more distributions out there. There are plenty of them, right? So contributing to the ones that are well established is the way forward and I believe Linux doesn't need more distros. Linux needs more applications and application support

And we also need some kind of standardization because like there's so much option earlier Like for example on Windows we used to make fun that hey, you don't have a package manager, right? You have to go to your respective vendor sites to download apps But now Windows has the Windows Store, which is you know, not the best obviously it sucks But yeah, you have now everything in one place. So they're trying to do something like Linux. Now, Linux is not perfect either. We have to standardize things because if there are a ton of options available,

there's no way that you can make Linux mainstream anyway. Because there are going to be factions, right? Someone's going to like APT, and someone's going to like Pac-Man, and someone's going to like DNF. And there's always going to be these wars like we have with SystemD and OpenRC, for example. You've also been bringing things like Flatpak

and Snaps as well. Yeah. So there has to be some kind of standardization that we need to be doing in order to you know so if you introduce more of these things more distributions and more package manager and more more more just desktop environments and everything more options you're just adding to the chaos rather than trying to you know subsidize the or bring the chaos to calm.

So I can make a new distribution, and probably people who used ALG could probably migrate to it. But I don't see a point in doing that because they're really well-established distributions that are doing great. So Endeavor OS, for example, they're doing great. So if I get a chance to get there, you know, I would consider that as a better alternative to starting out a new project from scratch. Yeah, I think with ALG itself, that while, you know, adding in extra distros for the sake of it doesn't really make much sense,

but ALG did something different. I think that's why people treat it a little bit differently. It's not like, hey, here's just another distro that's doing... Like, all of the different... Ubuntu is a really great example of this. We have different distros of Ubuntu that are just Ubuntu with a different desktop environment. There's no reason for those to really exist and not just be, like, options in the Ubuntu with a different desktop environment. There's no reason for those to really exist and not

just be like options in the Ubuntu installer. Yeah. Yep. But the thing about the Ubuntu based, like the Lubuntu and Zubuntu, whatever they are, the thing is they have really

separate developers and separate websites. No, I get that. But it just feels like, from an outside perspective, it just doesn't really make much sense why... The only difference really is the environment. Yep. But the only... The problem with trying to work on standards is

you need support from people, because if you don't have support, then you have the problem with... You've probably seen the XKCD comic about standards, where you have 14 competing standards. Now we need to make a universal standard that covers

everyone's use cases. Then the problem you have then is now there are 15 competing standards. Yeah. That's the reason why i believe uh there had because uh i think someone mentioned this on their channel that there was efforts to for example standardize uh the package management in linux and yeah and out of all the package and I think there was a committee set up by Red Hat to do this, and obviously Red Hat is going to choose DNF of all of the package managers. And when they said that we are going to be using DNF, everybody backed out immediately.

So there was a huge response to standardizing the package management. Everyone was in for it. But the moment the committee said we are using DNF, everybody said, okay, goodbye. We're not using DNF. So I think one way to standardize anything in Linux or in the open source world is to, you know, everybody, every project has a good thing. For example, the really biggest selling point of of arch which arch team or anybody says is the installation project of process I disagree completely the biggest selling point of arch Linux is pac-man and

Pac-man is I believe the best package manager of all time the way it handles dependencies the way it just functions pac-man is like I believe my personal opinion is better than apt dnf or even what emerge whatever that is so if you want to standardize let's say a de or a package manager try to get the best one that handles every aspect every use case um in like you can't obviously satisfy everybody yeah but in like a 90 percent uh use case scenario it satisfies like 90 percent of the people and that could be standardized and the others obviously they can exist together so anybody the rest 10 percent who don't want to use the standard uh they are free to use whatever they want to use but what happens when we free to use whatever they want to use.

But what happens when we standardize this, you can directly compete with Windows or platforms like Microsoft or Apple. And the only reason I think that on desktop, we are not able to do this is because we have this greatly varying opinion. And I think ALG is a very good example of this as to whether we should have in graphical installer for arch or we should not have a graphical installer for arch we really have

like a lot of opinions if there would be like some kind of way to standardize any opinion that we have out there or any component that we have out there then we would really be on an advantage because all of the proprietary companies they are, if we are going to take out open source licenses out of the picture, they're going to collapse. Most of these companies, the reason why they spend so much money on open source each year is because in return they benefit from it a lot right yeah so that's one way we as open source users or developers or the community in general we can try to like get all of these things streamlined and then obviously i think richard stallman will be really happy well i think that issue with like standardization goes back to

what i was saying earlier with uh having the different options in your installer where everybody is happy to agree that we need some sort of standard we need something that everybody can agree on but the issue is that everybody has their ego tied up in their specific option and they want their option to be the one that everybody else is using whether that's with you know the graphics card brand you buy yeah i'm an amd guy i'm an nvidia guy whether that's with your package manager i use pacman i use emerge i use I use that, or programming languages or anything else. If you're going to have something be the standard, everybody wants it to be their thing.

But then if you make something new and try to get everyone to agree on that being the standard, now you just have more things that are getting in the way of there actually being a standard. It's this problem that is really difficult to solve unless everybody wants to solve it exactly and like i believe not a lot of people are interested in that because everyone's happy with what they are on yeah i think every like there's a lot of people that make a relatively compelling argument to say that it doesn't really matter if Linux gets any bigger.

Like, and when you as long as you have those people, there's always going to be that I guess barrier in the way to try to push that ahead. Yep. I don't know what the best

idea is. I think there is definitely compelling arguments to both sides. I'm just sitting back here and sort of watching what's happening and I'm just going to talk about what's going on, basically. Yeah, we all are going to be doing that. Actually, the thing is, a lot of open source developers that contribute to really important projects, for example, the kernel or some kind of drivers, they are working in big MNCs. So they might not get something by doing a commit to a certain project, but they get paid for doing it by other companies.

For example, Google has their own open source team or this department that pays all of the developers that develop the Google open source projects. Similarly, Facebook has it. And now even Microsoft has it, surprisingly. So many people, they have two sides to the story. They want to contribute to open source, but the companies, the proprietary companies that they are working in, the salary pays their bills, right? Right.

So they want to keep both the aspects going on. They want open source to thrive. And they also want to be able to pay their bills and, you know, survive. Because, like, not everyone is Richard Stallman who gets sponsored. But you go to a country, give a talk, and you get sponsored stuff. So everyone has to run their families and run their like pay the internet bills pay their bills so they kind of keep this proprietary software thing up and running so that proprietary software remains so that they get they are employed

basically and as a side business as a side activity or side project they can always they're free to contribute to open source so that's the one reason why i believe that there's this like barrier kind of thing that people don't want to make linux like uh or or any open source project for that matter to be like really really mainstream people still want proprietary to thrive because at the end of the day it runs people's houses so i think that's one of the reason probably i think that makes sense hmm i'll just think about no i yeah i think that that would make sense i just have to think a bit about that um because i i believe i was looking like all of the Linux kernel contributions after 2010.

And majority of these contributions, like 99% of them, all of the individuals were working in some organization. These are big organizations. So whether they are paid to do this or not, they have a salary that they are earning each month, right? So that's the thing about open source right now a lot of the projects that really matter to the proprietary companies they are the ones who are primarily directly or indirectly okay yep yeah I think that's something a lot of people who are like hardcore free software don't really want to admit like we can say microsoft bad intel of no more intel bad google bad like facebook even

but all of these companies do have a vested interest in making the projects that they care about actually better and one of those main projects they care about actually better. And one of those main projects they care about is going to be the Linux kernel. Yes. Yes. Anyway, you said you couldn't really do much past an hour. We're getting close to an hour now, so it's probably about time to end it off.

Yep. Let people know where they can find you if they want to like I don't know ask about future projects you're going to be involved in I'm really active on my YouTube channel and I plan to like do a lot

of stuff there so like if I join another project that's where I'll be telling people I'm also available on Telegram, Discord. What is your YouTube channel, just so people know? DemonKiller. DemonKiller. D-N-K-I-L-L-E, yeah.

I'll bring that open right now. And on your Discord, you say? Same, DemonKiller. Yeah. 9375, I believe. I will leave your Discord server and your YouTube. Anything else? If there's anything else you want me to link, I'll put it in the description down below just feel

free to send the message to me yep sure I think uh it's more than enough uh and I think we've had a great conversation today yeah I I loved it um if you want to come back on and talk about I don't know whatever it is you're going through through in the future, if you end up joining one of these projects, I would absolutely love to have you back on. Yeah, sure, no problem. If I join the projects, I'll get the other developers as well, if I can. Oh, awesome.

That would be really cool. I've been bringing on a couple of recently a couple of people who either maintain distros or are involved in distros and I think it's fun. I think there is a lot of people who just

work in the background and don't really have much of a spotlight on the work they do that absolutely deserve a lot more attention. Yeah. Well, one thing I'd like to highlight before we end is that no matter what the parent project Arch Linux they think or whatever their philosophy is, one thing that is really, really good when I talk about Arch derivative distributions and projects is that we have

a very common goal and I think we are very united. Talking about projects like Arco Linux, Endeavor OS, Garuda. So all of the developers are in touch. We constantly share knowledge with each other and we constantly have dialogue basically. So as far as the community around Arch is concerned, outside of like official Arch project, we have like a lot of friendships with other projects as well. So when I was like maintaining ALG,

I was constantly in dialogue with the developers of let's say Endeavor OS or Arco Linux or Garuda Linux or Manjaro. So I think the community outside is like really really vibrant so i'm pretty sure people would want to like come talk about the stuff we do behind the scenes that people don't know about well that's a really positive message it ended on um yeah well as for me um my main channel brody robertson i do Linux videos, Linux news, Linux apps, things like that. Occasionally, I'll look at Arch installers as well.

Yeah, go check that out. My gaming channel is Brody Robertson Plays. I stream there twice a week, upload five or so YouTube shorts alongside that. If you're listening to the audio version of this, it is available on YouTube as a video version. The video version... Sorry, if you're listening... If you're watching the video version of this. It is available on YouTube as a video version. The video version, sorry, if you're listening, if you're watching the video version,

the audio version is available wherever you can find podcasts out there, iTunes, whatever the other things are. You'll find it. There's an RSS feed as well if you want to use that. So, you know, you'll find it somewhere. Do you have anything else you want to say?

No. No? No. No? Okay. We'll just end it there then. I guess see you guys later.