Tech Over Tea - Flatpak, Wayland & Future Of Linux | TheEvilSkeleton

Episode Date: September 28, 2022

Today we've got the one and only TheEvilSkeleton, Hari Rana on the show, you may know from his blog under the same pseudonym, if not he commonly discusses, flatpak, linux packaging, bottles and various other Linux related topics ==========Guest Links========== Blog: https://theevilskeleton.gitlab.io/blog Github: https://github.com/TheEvilSkeleton Gitlab: https://gitlab.com/TheEvilSkeleton Mastodon: https://fosstodon.org/@TheEvilSkeleton ==========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 back to Tech of a T, this is episode 133, and today's guest isn't a YouTuber like I would normally have on. Today we have someone involved in, I guess, Linux blogging and Linux development, flatpack stuff. Welcome to the show, The Evil Skeleton. How you doing? I'm doing good, and you? I'm doing pretty good. If anyone goes on to places like r slash Linux,

they've probably seen your articles floating around every so often. Cause I noticed that you do tend to post like pretty much every single one of them. Yeah. I used to like my first and I think like another article I did post it, but after that like someone else was just posting every single article well they're all really well written so it kind of makes sense why they get posted thanks uh for anyone who may not know your name um just do a

brief introduction of what you tend to do so um my name is harry i'm also known as diva skeleton online um so like i'm mostly involved in gnome and flatpak related stuff i recently became a member of an open source project called bottles which is a an application to run Windows software on Linux and like I've been writing a lot of a couple of articles about GNOME and Flatpak especially like refuting a lot of the misinformation

around these projects and I'm trying to be like as objective as possible and also like as how can I say this? Like, uh, just objective, and then also, like, uh, professional and less, like, opinionated. Yeah, yeah.

With certainly a lot of, a lot of very opinionated, I'm, like, I'm definitely at fault for this as well. There's definitely a lot of opinionated takes out there on whether it's flat packs especially snaps um and well the app image guy does that to himself so that one he can't really complain about but i think the first time i ever heard about flat packs i think it was in the context of flat kill which is an interesting way to get introduced to them that's there's probably probably better ways yeah um i used to dislike flat back way before a couple years ago um but after I started making some research and things like that, I realized that, hey, there's so many misinformation. And after that, I started to study even more about it.

And I realized that we're actually trying to fix a lot of issues about the Linux desktop that we desperately need. Well, a lot of people when they see a project, they'll usually see it in the context of usually like a couple of years prior to where it's at, or at least

a couple of months, because unless you're going directly to the source, you're trying to find the most updated information. The information that tends to propagate is the stuff that's been sort of out there for a while whether it's uh you know flak packs when they you know did have legitimate issues with the sandbox and what they still do but it's getting better over time whether it's um snaps with their

performance issues which they still have but they swapped over to lz compression and that helped out with that a bit all All of the stuff that people tend to talk about with anything out there, not just in the Linux space, but pretty much anything else out there, tends to be at least a little bit outdated, especially when you have a project like Flatpak, like

Flathub, like things like that, where they are trying to improve the user experience, improve the usefulness of these tools fairly quickly from what I've seen. Yeah, it's especially like it's like we could say it's like

X-Wayland. The only reason why IPAC's permissions suck is because a lot of applications are very legacy-oriented. They don't rely on XTG portals. IPAC's permissions suck is because a lot of applications are very like they're very like legacy oriented or like they don't rely on xtg portals and because of that you have to we have to implement hacky workarounds this case being the like file system home or

file system host and things like that yeah i know that sorry yeah sorry i was gonna say i understand I understand why people are concerned about the use of file system actually why don't you just explain what the issue with file system home is before we go on from that the issue is when you give it access to file system home you're giving the application read write access to the entire user directory.

So not entirely, but most of it. Yeah, yeah. So for example, like 5Q.org, they do make a pretty good point about BashRC. So I do think it's overdramatic because distribution packages, they don't sandbox. There's no protection whatsoever when it comes to running the application at least flatpak has something

like that you can use in a third-party app called flat seal to manage those permissions and like if you use uh the f1 key or you look at the source code, there's going to have documentation that explains these permissions, what they're for, and how to use them as well. Yep, yep. Well, yeah, I always find it very amusing when people bring

up the issues with the flatpack sandbox and the flatpack, yeah, I guess sandbox would be the correct term and then say that flat packs are bad because of it but what they are using instead doesn't have a sandbox in the first place so it doesn't matter if the sandbox is leaky like your solution in the first place didn't have any protections there so like we're in the same position in the end but only one of them is going to go towards a more a more secure solution it is an issue but it's also an issue that is being addressed

exactly and plus um the distribution packages like you have to use a lot of back-end utility like app armor or um well, not backend but mandatory access control like AppArmor or other backend utilities like BubbleWrap. I think at least it helps you a lot and it should use permissions. I don't know if this is still the case, but

I heard a while back that AppArmor doesn't really function properly outside of Ubuntu. I don't know if that was just people misrepresenting it or what the deal was with that. I don't know if you have much experience with AppArm particularly. OpenSUSE is another distribution that uses AppArm and they use it really well.

The only issue is, just like SE Linux or I guess Flatpak, if the distribution they ship an App app armor that has no protection whatsoever, like Manjaro, then it's going to be almost useless. The only reason why Manjaro has app armor is because of snaps,

because snaps rely on app armor for sandboxing. Right. If we look at something like OpenSUSE or Ubuntu or even Debian, they have separate packages, they have more, they enforce more, they have more profiles for enforcing. Hmm.

So, actually, what led you towards Flatpaks instead of Snaps? Because when you would have been, actually, when did you first get into Flatpaks? It was a couple of years ago. Now, it's been a while, so I don't remember exactly. I was actually using...

Definitely back then, Snaps pretty much had the... Because of Canonical, pretty much had the main stranglehold on that segment of application distribution. Yeah. main uh stranglehold on that segment of application distribution yeah and like i first started using flatbacks ironically on gen 2 um and that's when i started liking it because like i would run into a lot of issues with gen 2 like for example uh our circular dependencies like where one dependency depends on the same dependency and things like that. It was a mess. And I saw that Flatpak was actually making it

really, really easy. I also realized that I could have no full 64-bit system without any 32-bit libraries and still have Flatpaks on top and run games like you know from steam yeah i could have a very minimal base and still play

my games and run a lot of applications and things like that that's when like i really started to uh understand the uh the idea of flat packs why they why they exist, and so forth. And the reason why I dislike Snaps is because they really haven't addressed a lot of these issues at all. Sure, they switched to LZ4, if I remember right, or LZO, I don't remember. I want to say it's LZO. That sounds right. Sure. Right. Yeah, so say it's LZO. That sounds right. Sure. Right.

Yeah. So they switched to LZO. But there's still a massive performance hit. And there's been articles by Canonical themselves explaining that, like, hey, there's an issue. Like, when they launch Firefox, it takes six seconds or something like that. like when they launched Firefox, it takes six seconds or something like that. On a really, really recent, on a like high performance computer, which is like really, really bad. Whereas with Flatpaks, it's not really like that.

It's no compression whatsoever. And it's just save space by deduplication, which doesn't, which there isn't any performance hit when when you run them i don't know if this is just i'm not sure what the cause of this is because i tend to notice that a lot of applications when i install them as a flat pack do load like noticeably slower i don't know if that's just some weird configuration issue with my system or what the deal is there but i can say for a fact that at least on my configuration i have noticed not as bad as snaps definitely not as bad as snaps but it's been it's been something where if the the way i usually choose if i'm going to use an app

is if it behaves properly just as like a regular system package regular like you know from the store for pac-man I'll usually go with that because generally I don't have those issues but do you have any insight into what could possibly be causing that issue I would say XTG portals mm-hmm that's um I'm really like I don't know much by XTG portals how to troubleshoot them i think if you use this uh journal ctl you might see some errors and things like that but i would say that xdg portals are really really fragile if something goes wrong you can really like mess up your system or like degrade the um. At some point,

XGG portals weren't even loading, and my themes weren't working, an application would take as long as snaps, essentially. It was really bad. I would say that maybe it's XGG portals, but that's

the only idea I have. Since you do use a window manager, XEG portals, but that's the only idea I have. I don't know if, since you do use a window manager, maybe it might be caused by something like that. It's weird because I did try Sway on a virtual machine and it kind of ran

fine. I never really had to troubleshoot XEG portals and I don't really have a lot of knowledge i have noticed if you want a quick way to break xdg portals actually just uh load up an x session cancel out the x session and go over to a wayland session and things just behave terrible like it's not just xdg portals a lot of things just behave terrible. Like, it's not just XG reporters. A lot of things just behave really weirdly when you switch back and forth between those sessions

without, like, cleaning up all the stuff that X has gone and said in the background. Whenever I have to, like, do a video on something in Wayland, I tend to just... I don't want to deal with the cleanup, so I tend to just reboot my system and just boot straight into the Swage session just to be sure that everything's going to be good like especially things like qt aren't very happy about swapping but that's not a normal way to use your system

from just general user experience i've noticed that portals actually make a lot of snaps especially like the obs snap on uh onland, actually usable, obviously that's partially due to Pipewire as well, but... Portals are this... Honestly, how Flatpaks existed earlier without Portals also existing, I can't even imagine trying to use some applications back then. Yeah, Exigiportals are like the like the savior when it comes to those stuff. But unfortunately, like, even though it's like one of the coolest things,

it's, it also, it's also one of the biggest limitations because like, it's like your global key binding. If an ex, if a portal doesn't exist, then like, oh shit, you know? Yeah.

And especially like, there are a lot of niche cases where portals just do not, will not be enough. And at least like portals, they're really good for general use cases. But especially like, as I said before i i am a developer of bottles and we've actually stumbled upon those limitations um at least like uh developers are starting to like work and address these problems but so like right now there's a lot of limitations with portals you've definitely been um how would you say it uh very very vocal about the way that bottles should be uh

should be i guess you might say bottled um because you have a lot of distributions from what i've seen that have... Like, it happens with OBS as well. Whenever you have these really complex applications, they will package them in a way where they mostly work, but it's not consistent in that way. And Bottle seems to be... Bottle seems to be the thing that is mainly brought up whenever we talk about packages just not being consistent across different distributions. Yeah, that's because bottles, since we're using wine, there's a lot of work around.

And not just that, with bottles, right now we're using a really scanty version of Libet Weta, which most distributions don't ship that. And obviously people will say it's our fault for shipping a alpha version of the library and things like that at least with flatpak since everything is sandbox and containerized um use that release candidate only for bottles so it doesn't affect any other like libadwaita application and it's not just that like we are also like it's a really really fast paced like we we bring up features really quickly we fix a lot of bugs really

quickly like every two weeks there's a release and there's always like a major feature or like a lot of bug fixes or ui improvements and things like that. And unfortunately, distributions, they can't really catch up to us. And that's why we decided to just abandon all additional packaging and went straight up for flat packs because at least it let

us continue in our own pace. We can push things to the really scanty version of Libadwaita, we were able to port bottles to GTK4 and Libadwaita. And if we didn't port it, then we'd be delaying everything else along. We'd be delaying everything else as well especially like um for many applications now like gnome apps that use gtk4 and libadwaiter it'd be way behind now and when we look at complex applications especially something like gimp are still porting over to gtk3 and that's

a really like yeah and that's that's a problem it is a problem yeah the gta 3 version is it's mostly there it's just gimp is a massive project it would be like you know taking blender or something like that and putting into new ui framework blenders got the resources to do it gimp you know not so much

it's i i can understand why it's taken when did the gtk 2 version even come out like 10 years ago something like that probably yeah um i didn't know gimp much back then so like oh yeah and that's like the problem if like for something a massive project like gimp if they started porting to gtK3 way earlier then they would have GTK3 unstable already and probably even porting to

GTK4. We've seen this and we don't want to repeat the same problem so we're like, you know what, fuck it. We're just going to use the... Actually at the time we used an alpha version of the bad beta and then we updated to beta and then really scanned it.

And we didn't want to make the same mistake as GIMP. And we're like, yeah, screw it. We're just going to go to LibAdVita right now. Abandon all the other packages and stick with Flatpak because they let us do that. And not just that. It's not like theming, like distribution theming, where users will believe that the application is the problem when it's actually the theme. Like here, people will know that it's actually our fault.

Like if we pushed an alpha version of Libby Adueda and we'd take all responsibility for that and actually address these problems and things like that so we were completely fine with maintaining uh with pushing or like porting to gtk4 and now like even then when they even then like we kind of finished there are still a lot of issues with like with bottles on gtk4 like it's still not mobile friendly and i know like wine doesn't work well on mobile but at least like we're trying to future proof it so and developers will start focusing on like wine on mobile at least like we'll have the ui we'll have like a lot of things like uh already figure it out and there's still

there's still like some issues with a mobile support with bottles for example and like we still need to fix these issues but even then like just imagine like the bad way to 1.2 i think it recently got released um imagine if we started porting right now then we'd be like like we'd be delaying all of these updates like two more months right right right i've heard some people discuss i i know you've responded to some of these people but people discuss like why don't you have like an aur package for example and because like part of your issue is having you know the these. Why wouldn't AUR package not work in this case? Because we're using a release candidate.

Before that, we were using an alpha and beta version of libadwaiter. If we use an alpha version of libadwaiter from the AUR, this will apply to the whole system. This can affect all your libadwaita on from the eur this will this will apply to the whole system this can affect all your like all your libadwaita applications and you can run into issues that only happen in libadwaita alpha and not on the stable version this is really unsafe especially like a merco like the maintainer of bottles sorry voice crack the maintainer of bottles um he tested the eur package and it was really, really bad. Like, it broke his system because it was using the bad way to git.

And there are times when it just wouldn't build at all. Like, just one package would cause maintenance burden for us. At least with Flatpak, since everything is isolated, we can make sure that it doesn't break any one system and it will just break the application itself. Especially when we make sure that we use SDG standards.

The only file system it has, the only directory it has access to, I believe, it's downloads folder. Not even your home folder, nothing like that. Because we, I believe it's downloads folder, not even your home folder, nothing like that. Cause we make sure that it's like secure and especially how those apps can be very like insecure and unsafe.

Well, at least it's sandbox by default. And we tell the user, like, if you want to open up the sandbox, that's do it or not. Like it's not on us. Yeah.

That makes sense. I know that for certain i'm sure that part of the reason i come to this conclusion is because some issues can be addressed through having say like a a application specific package for example with um a uh obs on the aurR. So the main OBS package on the Arch repos is terrible. Don't use it. Awful idea. But the one in the AUR, they package a separate version of, I want to say

it's CEF Minimal. And they install that alongside the version you'd have installed for other applications that might need CF Minimal. But I'm guessing that with something like LiveAdwaiter, it wouldn't really be viable to do that. You'd need to do weird workarounds in the application to probably get it to function like you'd expect, not overriding your system libraries and things like that.

Exactly. And obviously like that can cause a lot of problems and we just don't want to break users computers or well existing systems there's also like i guess the opposite idea on creator for example it's like it's a massive project you rely on an older version of qt5 and And the reason, like, if we look at the Flatpak manifest of Crata, they're building an older version of Qt 5, because that's where it works best. If they tried to use the newer, like the most recent version of Qt 5, then they'd actually run into problems.

And obviously, like, if you use a distribution like Arch Linux, and that's not going to work well, obviously there are going to be people who are just going to say that, hey, I used it from the AUR or whatever and it works fine. Obviously, if you're trying to do some basic stuff, it will work, but

it will run into a lot of issues if you're, for example, a professional artist or you use extensively Krita. And it's not just that. Thanks to, at least for Bottles, for using Flatbang entirely, we were able to really increase, add more

like more and more like fun features and things like that and like you know each release like a lot of people ask like on reddit or whatever like oh how do these guys like have so much time well it's not that we have a lot of time is that we put the time

developing the app itself and not on packaging or fixing distribution related issues. Or sorry packaging. I think this all sort of circles back into the sort of the idea that

Tobias what's the GNOME guy? The theming guy. What's his name? Tobias Bernard. That guy. Yes. Tobias Bernard. The theming guy. What's his name? Tobias Bernard. That guy. Yes, Tobias Bernard. The thing that he brings up with theming, he doesn't care how an application is made.

Like, if you want to make it to be a GTCAM, if you want to make it to be a QT app, it doesn't really matter. What's important is getting the applications out there in the way the application was QA'd by the developers. If that is QA'd with this theme, If that is QA'd with this theme, if that is QA'd with this set of dependencies, that's going to be the best way to use the app. Not everything requires this. Like, you know, a terminal for example. Most terminals function generally how they should function as any sort of system package. Like, there are some edge cases, especially back when Python

3.10 came out, there was a lot of AUR packages that broke, for example. Like, things like this do happen, but generally with the simpler applications, you can get around. Like, they're not as touchy about, like, oh oh do you have glibc version let's just say 1.1 version versus at versus 1.11 for example like a lot of applications aren't going to be that picky but i think where flat packs shine and where they make the most sense are with those applications that can be a bit flaky like obs like bottles like a critter for example like a lot of the applications where they're very complex and i can understand why one the devs don't want to be dealing with packaging issues

and two why they just want to worry about a single package like there's sort of like two benefits here you don't have to worry about knowing how to package for all these distributions and you don't need to test them as well just you have this thing it works and that's all that matters really exactly and it's quite consistent across the distributions as well so like like even then like even with that there's still like even if we ignore that um there's still some benefits like uh reproducibility and also like um uh what's the other one again um one of the biggest upsides of app images i i forget the name. Like, there's a specific word for that. Portable.

They're very portable as well. And, like, the thing with Flatpak is because they ship with runtimes, like, because, like, they really force runtimes to users, that's why it's really, really, gosh, I forgot the word again. What's the word I said again? I completely forgot. Yeah, exactly. Thank you.

I don't know why I just forgot this word. Like app images, they require like, it requires a glibc. And there's no like standards with appImages it's very much anarchy package it however you want that's a great way to put it to be honest yeah and honestly I think it's a really bad approach

because a developer can ship the AppImage with a library or they can decide not to. If they decide not to, the app image has to assume that the distribution has the library. And if they do ship it, well, they

have to also make sure that it doesn't use a lot of space as well. At least with Flatpak, there's deduplication going on. So that's a really, really big benefit here. And especially another nice benefit of Flatpak is just like App Images, you can install them in USBs or external drives and things like that. It's just that to App Images credit,

it's a lot easier to do that because AppImages, you just have to drag and drop or just move the file and that's it. The Flatpak, you have to type Flatpak create-usb and do some things here and there. Then you have to install the application in that external drive or whatever it is. There's more work,

but the benefit you get in return is a lot more because it uses runtimes it is guaranteed that it's going to work on the majority of distributions so even if you use Void Linux

or Alpine Linux it's still going to run because the runtime ships to G-Lip C and that's a really really nice thing and that's also why AppImages don't run on Void Linux Muzzle or like

Alpine Linux because it depends on G-Lip C they also depend on G-LipC. They also depend on LipC's too. Which a lot of districts don't ship as well. Yeah, exactly. That as well. And another really like something that no one talks about it.

Is that Flatpak has its own macOS Rosetta thing going on as well. It uses QEMU to emulate architecture. So if you're using an ARM64 device, you can run x86 applications on it. I didn't know that. Yeah, that's what I mean by it's really portable and a lot more portable than app images because it's cross architecture as well.

And if I remember right, a recent version of a free desktop runtime i started shipping uh race 5 support as well there's that as well wow yeah i had no really cool why is this not a thing that more people mention i guess hardly anyone uses ARM, but even so, I'd never heard about this before. Yeah, I guess it's, maybe it's probably, like, Firepack's fault because you don't have, like, really bad documentation. Is that managed on, like, anywhere? Yeah, it's actually on Reddit, that's how I got to know it. Oh, no, I mean, like, any official distribution, like, their website or anything like that.

Uh, no, I don't think so if anyone who has development access on the flatback site is listening to this please go and add that that's a really cool feature yeah and yeah

I think you were going to say something no I'm just baffled that it's not listed anywhere. Yeah, and I guess that's the thing. FopX, one of the biggest downsides is that it's documentation. I've read the documentation and it's something. It's like your average open source documentation. It lacks a lot.

And I'm trying to improve it, but it's just that there's so many things I don't understand, so I just ask. And recently I took a break and started programming a bit. After that, I'm going to start improving the docs as well. I've talked about documentation a lot in the past. The thing about documentation is no one likes doing documentation. Like the reason why companies will hire people specifically for it is because none of the devs want to do it. None of the designers want to do it.

Nobody wants to do it. Like they want to be working on the thing they want to be working on. It's such an important thing but especially like especially because a lot of open source devs build something for themselves and you can kind of get away with not having great documentation then like oh if i know how to use it it doesn't matter i've done this myself there's plenty of programs i have out there that good luck working out how it works but i know how to use it but as you scale that up and as an app gets more and more popular it doesn't work that that way someone has to sit down at some point

eventually eventually lay out how this is supposed to work yeah exactly and this is coming from someone who enjoys writing documentation oh good i hate writing documentation when it comes to my own projects. So I just want to work on it. Like, as you said, I just want to work on it. And like, I don't, I don't care about documentation. Even though I really need to improve like sometimes. Yeah. Like when it comes to other people's projects, I'm completely fine with writing documentation but for my own

projects no it's it's boring I think everybody can agree with that yeah I was gonna go somewhere from here I had some sort of segue um I don't

remember what I was gonna segue go somewhere from here. I had some sort of segue. I don't remember what I was going to segue into. Something, Flatpaks, cool. You know what? We're going to shift. I didn't ask you this earlier. How did you actually get into Linux, and when did that happen? And where did you start? Yeah, that was in 2016.

That's when I first heard about Linux. I was just really curious, and I tried Ubuntu. And after a couple of Windows updates, it broke. And, yeah, and, like, I was really, like, I was a noob, so I didn't know what was going on. And just, like, anyone who is, like, very, who knows very, like, has very little understanding,

I would just blame on whatever like whatever is the result so i would just blame on ubuntu and linux say that oh it sucks it's unstable and things like that i mean the windows update caused it um and but after that like at some point i ran into a lot of instability issues with windows and i got so sick and tired of constantly reinstalling that I was like, you know what? Fuck it. I'm just going to install Linux and I'm going to deal with everything

and all its problems. To be honest, I did the meantime, well, not in the meantime, but later on, I got a job and I started doing tech support. And when I go back like in 2016-ish I think I think that the issue wasn't Linux the issue wasn't Windows it was actually me because I learned

like after doing a lot of tech support I learned that most issues come from users right and like and then I'm just like of course i don't have like a fond memory of what happened exactly but i i'm a bit skeptical now i think that i was probably at fault even then i'm glad of the results i'm glad that i'm a linux user now um yeah like at some point, when I tried Ubuntu again, I remember Calculator app was taking 10 seconds to load. It wasn't

even following the theme. I didn't know about snaps back then, but the issue was actually snaps. I heard about Pop! OS. I just switched to it. I'm like, wow, this is a lot better and yeah after that like i just wanted to learn more and more about uh linux and i started to like try fedora like i just went to that district hopping phase uh

right try fedora manjaro horrible experience but uh um arch linux then gen 2 i even tried exurbo a bit and like i tried many distributions and like that's how i gained a lot of my knowledge it's by reading documentation and things like that and after that i tried nixos and i fell in love with it because like i started to understand like immutability and things like that. And that's when I also started using flatbacks even more. And then I was like, Nix is like NixOS is a bit complicated to use. And so I was like, yeah, I want to try something a bit simpler. And that's when I pulled upon silver blue you

went full circle you started the buntu you went all the way down to gen 2 fuck this went straight back up yeah exactly um yeah it's like you know at first you just want to learn you just want to customize and then you're like you know what i want to have a life that's fair i'll get that if you'll ask me like how i have so much time to like spend customizing my setup do all of this i turn customizing into a job like that's that's the only way you can functionally do it don't try to like i i totally get why uh someone wants to use silver blue someone's to install a boon to someone wants to use anything popper west like whatever they want to use Silverblue, someone wants to install Ubuntu, someone wants to use anything, Pop OS, like whatever they want to use. I get not wanting to waste all your time

customising your setup. All you want to do is like, hey, I'm going to fire up VS Code or whatever. I'm going to do some coding and then I'm going to watch some Netflix. And that's your day. And that's fine. The way I use Linux doesn't have to be for everyone

but I've made it work yeah and even I went to this very customization phase and all and like obviously I'm proud, I'm really happy that I did go through that phase because that's where I learned a lot about Linux not just the kernel but the overall

desktop also started learning a lot of its issues as well and I'm really glad like yeah like even though I went back to Silverblue at least I kept a lot of the knowledge with me and I think that's okay but yeah like

now I'm just sticking with Silverblue for now and since I've been really, really interested with Fedora and how they lead the Linux desktop, especially with them pushing Wayland really hard. You see every distribution, they're like, oh, we'll just stay with X11. Fedora's like, no, we're going to full Wayland. If you have issues, well, sucks to be you.

Go back to X11 type of thing. I know Fedora has its own issues and especially how they're pushing Wayland and Wayland has a lot of issues. At least they're contributing upstream with Wayland, fixing a lot of issues and things like that. So they're doing a lot on their part as well.

That's why I was really, really interested in Fedora. They want to lead the Linux desktop and push the newest technologies and things like that and it's thanks to them like a lot of newer technology like well systemd it's not new anymore but back then it was really new butterfs um pipewire like they were one of the first ones to push these, and everyone followed up later on. Yeah. In some cases, especially Pipewire, I think they did it a little bit too early.

When I used Pipewire, I was using Pipewire before they initially started shipping it. It was a bit flaky back then. It was still like a very... There was a lot of cases where you'd get a new update and there would be a pretty big regression like the way that uh one thing i bring up every so often is um so you know how in your audio mixer you have like your individual uh

process volume and then your master volume. Yeah. So, in OBS, if you capture your audio interface, the audio source, you capture your system audio, that will always be at the same volume, regardless of what level

you have it at, regardless of what level you change the processes to. In one version of pipe why that swapped around so it used the process volume and ignored the master volume so i had to like have things really loud for me and i couldn't actually like hear anything else in my stream properly that was when i initially went back to pulse audio Audio since then. I think probably in the past, I want to say not long after OBS started working on Weyland, Pipewire I think has been like mostly good.

I've had a few issues here and there, but besides that, like if there are distros that are still not shipping Pipewire, I think it's a much harder argument to make they should be doing that yeah and that's like the thing with fedora is they push things way too like way uh way sooner and and they kind of like to be honest like you have to do that like when you want to like sure sure someone has to do it yeah exactly that's the thing um especially with wayland like i think they started shipping wayland in 2016

or 2017 if i remember the like sources yeah um right and obviously that's way too early but of course like someone has to push in and some and least Fedora, they're really big and they contribute to upstream a lot. And the fact that even though they break a lot of users, well, quote unquote break a lot of user system, at least we get those technologies a lot sooner than if Fedora wouldn't exist. If you know what I mean. No, I like what you're saying. I think a lot of people sort of forget about the fact they bring it up when it's convenient, but

forget about the fact that Linux overall is this community effort. Sure, there are these corporate interests that are trying to push the desktop forward, but it's not like you just have one entity. Let's say you have like, you let's say red hat was the only the only developed like the only like company in the linux space it's not like they're the only ones who can do something but someone has to put that stuff forward so that everybody can be like hey we can use this and

let's see what's actually wrong with it let's contribute back to it and give even if we can't uh code stuff we can at least give our feedback on it sort of explain the issues we're having the different hardware configurations we have to bring things you know eventually forward it's going to take a while it's's not gonna be like, hey, Weyland is perfectly ready, everything is 100% next year, but I think it's much better to have someone slowly pushing the ball forward, rather than sort of individuals choosing to go and use that software. Like, look at Arkan, for example. People keep bringing up Arkan to me me nobody knows what that is because no distro ship it so like it's not going to get better because it doesn't have that

that interest in it that any like any users actually trying it out and honestly i never heard of that yeah i didn't think there's a few people in the discord who like to be like hey have you tried arkin like no no i haven't tried arkin no one's tried i'll try mir before i try arkin like you say it's another like display protocol yeah yeah okay i see oh i see okay i know i've never heard of that. Yeah, no. I know Mirror. Yeah, like, even Flatpak, like, technically, they're not ready. But to be honest... I don't think anything's ever ready in the workspace, though. That's the thing.

Yeah, exactly. Like, even distribution packages, they were never ready or meant for, like, for GUI applications. They have a lot of limitations and they have a lot of issues. People just got used to it. Just like many people who have issues with X11 screen tearing,

they just add the tear-free flag or whatever it is in the configuration and they call it a day. For them, that's fine, But obviously, that's not something an average user should actually suffer from. And

that's the thing. It's just problems that many people ignore. They're just really used to it and they don't see it as a problem anymore. Yeah. I think that's a great way to put it. Have you actually messed around with the tear free option then compared it to like you know sway or something like that because at least for my gpu tear free is it okay it works great on the desktop but anything

where i have a fluctuating frame rate, like I'm playing a game for example, I get horrible stutters, over on the Wayland side, it's smooth like, it works like you'd expect a, you know a v-sync solution to actually function

mhm yeah, um like, I only use it once and i remember like uh only time is because like i was using firefox and it was screen tearing on firefox right it was yeah so i just enabled tear free but i wasn't really playing games at that time so i never i probably never encounter any issues. Or maybe I just got used to it. So what is... Sorry. No, no, it's okay. Go ahead. I was going to say, so what has your experience been on Silverblue?

It's been a thing that a lot of people have been talking to me about, saying, hey, try this out, try that. But first-hand experience, what's it been like? Okay, first, I'd say it has its own issues because Dora has its own philosophical, like, or ideological issues. Right. Since they're very, very free software oriented, they don't want to push Flathub

because Flathub contains a lot of proprietary apps. Yeah. They have their own Flatpak remote called Fedora Flatpaks. And I don't want to bash on Fedora developers or contributors, whatever. I think that's a horrible approach to push Fedora Flatpaks. It's a great idea because FlatHub and Fedora Flatpaks, like Fedora's Flatpak remote, they have different backends. So Flathub, they use OS3 to...

OS3 is a backend utility, and you can think of Git for binary files. So Flathub uses OS3 to redistribute programs and things like that. But Dora Flatbacks, they use OCI containers, which means Docker containers. And again, as I said, it's a really cool idea. But because of Dora's, I guess, very free software-oriented ideology,

they're ruining a bit the experience here because if they would push flat hub like fedora would automatically have been a lot better i didn't actually had their own uh their own thing that's that i get it like think the, I think the most important thing is the, is user freedom. I know the, the free software guys like to talk about user freedom, but like, if we're being honest right now,

the user should be in a position where they can make that choice for themselves. Like that's actually user freedom there. Like I get it from, you know, software freedom, whatever, get it from you know software freedom whatever but yeah i know i agree like it should be pushing flat hub it's not a bad thing to have their own thing like that's okay if you want to do that you want to have things specifically catered for your for your system but yeah i think it's it's weird yeah i think like the fact that they're just pushing fedora flatpaks is

a really bad thing because i'd hope because there are times when both fedora flatpaks and flathub have the same application for example gimp so like there's gimp built by fedora on the fedora flatpaks remote but there's one from hub which works a lot better but then like yeah that's like another issue like when you open gnome software like and when you want to choose your sources like fedora flatpaks what the hell is that like i thought they were just like taking the the flatpaks that fit theirs and then just put it on their own thing that okay no i changed my mind that's a stupid idea oh no no unfortunately they they actually act as a stupid idea that's

exactly what they're doing with they come with a filtered version of fat hub that has literally literally seven apps um out of 1500 ish right it's it's such a terrible approach uh and like i don't know if they fixed it or not but there was a time like actually i think like last release of dora 35 or it would go under the name flat hub so when users like would try to install like gimp yeah um from flat hub they're like they would be like okay that doesn't exist but when they So when users would try to install GIMP from Flathub, they would be like, okay, that doesn't exist. But when they tried to add the Flathub remote, they would say that the remote already exists. But again, Fedora was using a filtered version of Flathub,

which has seven apps. And this really misleads users. And that's one of the issues like I have with Silverblue and just with Fedora altogether another issue I have it's not like for me personally it's for NVIDIA users and like as I said like Fedora

they're really really free software oriented so they try like they purposefully give a bad experience for NVIDIA users. And you have to add a third-party repository, and then you have to install NVIDIA drivers. Whereas on Pop!OS, they come with NVIDIA drivers pre-installed if you want to do that. I think that's the better approach.

But again, since Fedora, they're very, very free software oriented. They don't want this at all. Use the open source drivers. Use them. You know you want to. You don't want to. No one wants to. Yeah.

And at least recently they've made it a lot better like where you can install the nvidia drivers completely via the gui like from gnome software that's a really really nice approach unfortunately most documentation online are outdated and and some documentation from fedora have terrible search engine optimizations. So they're like at the bottom of Google, perhaps even the second page. Oh. Yeah, it's really bad. Yeah.

Yeah, it's really sad. That's awesome. I love that. Yeah, and there's also, I remember, actually, you know what? that yeah and there's also um i remember actually no actually oh yeah yeah the third problem anaconda holy shit anaconda is like the worst installer i've ever used in my life um like i mean i don't want to rant too much because I don't like shitting on developers' work. I think that's a bit too rude.

Okay, well, why don't we talk about what's good about Silverblue then? Yeah, so it's the immutability. I think that's a really, really nice thing, especially like they use OS3. As I said before, OS3 is basically Git or binary files. And I a it's basically git or binary files and I think it's a really really

nice approach because they use a package manager called well actually a hybrid package manager called RPMOS3 so even though you install like even though the system is immutable you can still install rpms from rpm

os3 and what it does it generates an image and then you can boot and have the package installed and another thing that i like a lot from silver blue is like as i said before they push newer technologies. I like being that edge where I can report issues. I can report issues and help here and there and things like that.

Right now, I'm using Fedora 37, which comes with GNOME 43. I think that's in beta. It's not even stable yet. I can 43 and that's i think that's in beta it's not even stable yet um so i can um i can report issues to fedora and they can have you know really stable um more stable like upgrade and things like that and another thing that i like a lot about um silver blue is that they try to make it very user-friendly at the same time like if we ignore Anaconda and Fedora Flatpaks,

they use a utility called Toolbox. And Toolbox is a little container utility for development. And it's really easy to use. You just have to type in Toolbox Enter. Or actually, I think at first it's at first a toolbox create it creates a fedora um container completely mutable and then you type in toolbox enter and then you can install applications using dnf which is like fedora workstations like the mutable distributions uh package manager and especially when, at least in my opinion,

even if I use immutable desktop, like Door, Workstation, or even Arch Linux, I would still use Toolbox because I don't want to install development packages on my host system. I want it in a place

where I can easily get rid of if something ever goes wrong. I don't want to deal with dependency hell and upgrade issues and things like that. Schoolbox really facilitates development because it's a front-end for another utility called Podman, which is a drop-in replacement for Docker.

The issue with podman is that it's it's very technical it's really it gets hard to use just like docker yeah i'm not using myself but i when whenever i did anything with docker like there's a there's a bunch of websites out there that have like the the command to run to open an application and it's usually like 100 or 200 characters long for anything remotely complex it's like how do i how do i do this what do i need to do like exactly yeah exactly and toolbox like it automates that and yeah so when you type in toolbox create well it pulls the right image then when you press uh when you type in toolbox enter uh it does all the flags, everything like that.

Yeah. So you can use it. There's also an alternative to toolbox. I think it's more popular. It's called DistroBox. And I think Intel, the name of their distribution, Clear Linux, I think they started using DistroBox. And it's a really popular program now. It's heavily inspired by Toolbox,

but it's more powerful. You can create desktop entries within containers, so you don't have to install the application on your host. Just install them on your container and launch them normally like you would like any application or flatpak whatever um it also runs like it works with both toolbox uh sorry uh podman and docker yeah the toolbox only runs on podman i keep hearing about like i don't know what happened i don't know if like some outlet picked it up and then everybody

started talking about it, but I've been hearing so much about DistroBox recently. I really need to cover this at some point. Yeah, it's a really, really cool utility. If I remember right, it's an entirely Niposix shell, if I remember

right. Yeah, it's really impressive. Yeah, it is! I'm looking at the game right now, 100% Shell. What the hell? Yeah, it's seriously good. I believe Toolbox is written in Go. I might be wrong, but

I don't think it's a bad thing, but whatever it is, the Showbox is in POSIX Shell, and it's really really fast in development and plus like you can I think there's like an installation script so you don't even need a package on your system you just have to

install it using like some command yeah there's a yeah there's an install script you can just download from the they reckon oh I don't please stop recommending this. They've got a curl installation method. I just... I don't like...

I don't like devs who include a curl installation method. This is all a flat pack, which is experimental, apparently. Yeah. My friends worked on that and then like the maintainer Like of this Roblox they liked it and you know the yeah with the flat pack as well and I don't know if he's gonna include it in Fathub, but probably yeah, well, and it would be cool if it does I've considered like the main reason I want to try this out is because I don't always want to fire up a VM just to try out something on whatever distro, like Ubuntu or Fedora or anything like that.

If it can be achieved in a system like this, then it's probably, it's going to be easier to do it like this, for sure. Like, obviously, it doesn't use as much resources as um and it's a lot easier to set up as well and it uses a lot less space and like there's just a lot of benefits of using a distro box even like if your distribution is mutable there's still a lot of benefits especially like the container part where like you can kill like you can destroy your container it won't affect your system because well like it's a container you can just remove it afterwards and create a new one well even just simple things like let's say you're on ubuntu, and there's something on the AUR you want to be using. Like, just that alone seems like a good enough selling point.

Mm-hmm, exactly. Or, like, maybe, like, the application you're running, like, from the AUR isn't working well, but it's officially supported on Ubuntu, so you can get the Ubuntu container as well. That's also that. Like, even better, like, as I said before, you can also create desktop entries with DistroBox. Like, there's literally a command for that, and it will just generate a desktop entry,

and you can just, like, launch it from your application menu. I don't, I don't, that's cool. I don't use your application menu i don't i don't that's cool i don't use an application menu like my um the way i run an application is i have my d menu it lists out every binary on my system and i just type in the name which is great for me but you know i get it if you're using like a desktop environment for example you want it to be in that application list and all that fun stuff. But I think that's cool. I think it's cool that it's there. Also, about the curl thing,

is there an alternative? Well, you can always just download the script and then run the script like that. Okay, that makes sense. The method they have there is just the curl into pseudo SH

yeah I guess yeah that's pretty insecure yeah like I get it it's just simple way to do it I just personally would prefer to download the script and then do the pseudo SH afterwards so it's just that's the way that I like to do it I get that people don't really think

it's that big of a deal and maybe you shouldn't just be downloading scripts if you think it's going to be insecure anyways there's a lot of reasons why you might not consider you know this that big of a deal but it just it it robs me the wrong way especially considering like you know coming from using windows and knowing how simple it is to just download random toolbars, for example, like, I don't want to be in a situation where I'm encouraging that sort of you know, reckless way of

installing software. Yeah, I completely agree. I think, like, using sudo is not the best approach. But there's always a system package there if you want to use those as well. There's a lot of ways to install this. Yeah, and since

it's just a bash script, it's really easy to package. You can just download the repo if you wanted to. Yeah, exactly. That's cool. Another thing I like a lot about Silverblue is that it has really good GNOME software integration.

I try to avoid the terminal as much as possible. I know how to use it. I'm quite confident with it. I just prefer using GUIs. And GNOME software, I think it does a really good job at integrating RPM OS3 and it updates all my applications automatically. I install applications using software and

well, to be honest, since I'm mostly installing Flatpaks, it's not really related to RPM OS3 itself, but when it comes to updates, it uses RPM OS3 for that system updates. Right, and it does list all the software out in the same place.

Not like you need to go and use the GUI for this software and then the terminal for that software instead. Yeah, exactly. It's not like Ubuntu where you have, like, three applications for updating apps. Yeah, it's really something. And also, like, obviously, like thing one of the things i really like about silver blue is the immutability like it's the atomic updates i really really hate it like when there are times like i don't know why i think i have a faulty cpu or something but there are times when my c my pc just crashes out of nowhere. It just locks out. And then it goes in a boot loop,

so I have to force shut it down and then turn it back on. If I used a mutable desktop, I mean, sorry, mutable distro, and I upgrade it and it crashes midway, that would be a massive problem. It would be a really big risk. At least with atomic updates, since it's either

it's upgraded or it's not there's no in between like where it's in the process of upgrading also maintains like a couple of prior versions as well doesn't it yeah and yeah that's also one of the cool things about silver blue is that you can just rebase to another version of Silverblue. Like you can upgrade, downgrade, whatever. If I have a lot of problems with Fedora 37, I can just type in things RPM, OS3, rebase, Fedora colon, and Fedora something 36. I would bring back to Fedora 36 from 37. Or if I want, I think there's like even Fedora in the 20s,

like, I don't know, 28. I could rebase to Fedora 28 if I wanted, and then go back to Fedora. If you for some reason wanted to. Yeah. Yeah. When did 28 come out? Fedora.

Let me check. Whenever it's like more than once a once a year release cycle, I have no idea when things came out. Look, I... The End of Life was in 2019, but when did it come out? I don't know. Doesn't matter.

What I was saying was I like Ubuntu's way of naming stuff. This is when it came out that's it that's all you need to know just go I think that's a really nice approach as well right now I'm checking the branches

versions of Fedora it goes all the way down to Fedora 37 which is really cool but of course like you just had a good idea to downgrade to 27 27 you said 37 i was confused for a moment there oh sorry sorry yeah 27 you can downgrade all the way to the beta version. Nothing more. Yeah, whoops. And another really cool thing about Silverblue, or actually just like OS3 in general,

is that you can make your own, like you can custom, actually you made a video about it. I did briefly mention, I've not seen good documentation on how to do so, but I know there are some projects out there and I think... was it Oro in the discord who kept saying about theirs?

I think? Yeah. That's something I really want to look into. I completely cut you off there. Go on with what you were saying. Swapping out images, things like that. Exactly, like you can create your own images And you can switch to it Very quickly If you have issues, you can swap back And things like that

Whereas with mutable desktops If you want to downgrade, well, good luck Because glibc is a massive pain And things like that At least with Silverblue, since it's immutable, you're always getting a fresh image or fresh

system. Excluding your user directory, but the root itself is really, really clean. I had a repo here that demonstrated it really well, but my history is not showing up in

Brave right now. Can I find it? Because there was a guy who took... Yeah, here we go. I mentioned this in the video. OSTreePity Workstation, which basically takes... It makes its own image that has Sway and a bunch of other

things like that on it using like toolbox and things like that for running other stuff this is not at all like the way you would normally have silver but you would have you know gnome or if using um how do you say the the other one is it people believe me the way i say it um because i've only ever seen it written out. But you can do things like this. You can ship Sway if you want to. You could take...

You could make an image and then use like DWM, have DMenu, have the Suckless Terminal. You could make this really weird system that's nothing like the way that Silverloo would be and it just

swaps right in which I think is actually really cool and the best part is when you swap to Silverblue for example or you swap to something a lot more minimal as I said you're going to have a clean install

or a clean system. So there's that benefit as well because if you use mutable distro and you uninstall GNOME, well, it's going to leave out a lot of things or you might even have problems

with, oh, like this dependency depends on this dependency which depends on this one things like that it's it's a massive pain if you want to remove a desktop environment well these would be fully clean would it because you have a lot unless the config files are handled differently wouldn't the config files still be located in your home directory uh Uh, yeah. That's why I meant, like, excluding user directory, because, like, I was... Yeah.

Yeah, like, I was talking about the system, like, the root partition, like, yeah. Yeah, of course, like, Plasma, I know they're notorious of, like, fucking up GTK, because, like, uh, if I remember right, when you install Plasma

it uses its own theme its own GTK like Breeze theme and when you switch to GNOME it will use like Plasma's cursor then Breeze and there's going to have a lot of issues switching to GNOME because when you switch from GNOME to Plasma it's pretty

it's seamless, it's almost stock and there's like also another thing you switch from GNOME to Plasma, it's pretty, like, it's seamless, it's almost stock. And there's, like, also another thing, like, I almost forgot to mention

with Silver Blue, is that they ship, like, a custom theme for Qt, and I think that's

a really, really bad thing, because they use another, like, platform, like, called QGnome platform. It's to make Qt

applications look like GNOME or look like Adwaita. Yeah. Since it's claiming, well, it's bound to break a lot of applications. And I've run into, like, I ran into a lot of issues with QGNome platform

because, like, especially, yeah, like, recently there's an ongoing merge request. Like, they're, as my developers, are adding a console, like, the terminal app on Flathub. And since QGnome platform is its own theme, but it actually messed up the whole application itself. It looked horrible.

And I can show you a picture later on. It was really, really bad and literally unusable. And obviously, if it's an average user who uses like who doesn't really understand how theming works and they look at the qt app and they're like what the hell that's entirely broken well they don't think that the problem is application and in reality the the issue is theme itself I think that the fact they're using QGnome platform is a really bad approach and I know

like like if you use Qt on Wayland it uses its own CSDs which doesn't integrate with Gnome so I can see why they took that approach by they I mean Fedora, I can see why they took that approach.

By they, I mean Fedora. I don't like the fact that they're using their... their styling to add Weta instead of just replacing the style bar. Yeah, like... I get it. I get the idea of wanting a uniform... a uniform desktop. unless unless you're going to be putting in that qa effort to make sure that everything that is available is working like it

should it's probably not the best idea to be doing and you know no distro is going to be doing that effort they just don't have the resources to really divert into that so it's probably better just to not do it exactly and that's like even though even if they do have the QA

resources to test well there will always be edge cases and that's like also a problem and you know the website stop theming my app right yes that's exactly like gnome's like you know tobias bernard's points like stop like that users make their apps like break our apps however they want. But like distro is stop fucking up our apps. Like you're giving users a bad impression, but our own applications,

when we're just trying to like give the best user experience. And the fact that, sorry, the fact that Fedora is using QGnome platform is in my opinion, not really a good approach. It's not really a good thing either when it comes to theming. I think... What was I going to say here?

Shit. What was I going to say? It's like 9.30 in the morning. Or 9.50 in the morning. Why am I already struggling to think stuff? Gnome. Theming. Something.

Lib Adwaita? Oh, fuck. What was it? Something about Lib Adwaita. GTK? GTK. Fuck, I don't remember. Lib Handy?

I don't remember what I was gonna talk about shit um edge cases um you know oh shit there we go well yeah uh i i think if we're gonna be dealing with this qa on the distro side it it's probably better, rather than trying to QA, you know, specific setups that distro is going to be doing, helping out with the upstream, because maybe if, like, you know, maybe if you actually help out with the upstream, there'll be some consideration about, you know, bringing it in more in line with the way your desktop looks like if you're going to be shipping like maybe maybe if uh maybe if you do help out with upstream that you could actually get into a position where the devs are aware of what you're trying to do on your platform

and are willing to help out maybe they won't be but i it's probably better to do that then that's what i was gonna say that was me bullshitting for a minute i got there now uh users don't care like users don't see and don't care about the distro when they install an app what they see and what they care about is they see the app and if the app is behaving weirdly it doesn't matter to the user if that's a problem because of the of the distro they're going to see the app is broken therefore it logically is the app's fault it's exactly you know it doesn't matter what the dev what the distro has done like this is like what i talked about in today's video. So like 12 years ago,

best way to stream Netflix on Linux was emulating a Wii. And when someone sees Linux from the outside, they're going to see Linux is bad. They're not going to say, oh, the problem here is, you know, Netflix. It's a website. Websites should work on Linux. You have a web browser. Clearly the problem is the thing that seems most obviously to be the problem. Even if it's really something completely different.

Yeah, exactly. completely different. Yeah, exactly. And it's not just that, like, a lot of the time, they will just open issues or contact upstream developers and upstreams are just

going to be like, what the fuck, that's your distrust problem, go contact them. And if I remember right, I think canonical is actually like actually contributes to GNOME like LibAdwaita and things like that

which is really surprising and if they are like good for them because it's nice to see canonical actually contributing to like upstream developers if we look at other like in another side like uh system 76 um like i respect pop os i don't really like system 76 because they're really really like

um aggressive towards upstream developers they're just going to be like oh yeah but like the issue is your style sheet or like you guys we want to merge like we want to give you this code and you guys don't want to merge it well style sheets are extremely like complicated to maintain and if sorry if system 76 wants to like integrate their own style sheet in in like gtk or whatever and you're asking them like the gtk developers to maintain a lot of css like you know a lot of stuff and that's obviously a really really big problem because like yeah i think the gtk developer is getting paid but basically like he has a limit as well sure um he can't maintain every single thing in the world either.

Oh, that's like Libadweda was out. So distributions can properly brand applications without actually breaking them. And if I remember right, a good... As you suggested before, distributions or developers, whatever, whatever companies they should be working with upstream, it's exactly what Red Hat

and Purism have been doing for a long time. I don't know how you managed to bring something useful out of that ranting I'm impressed Sorry? I don't know how you managed to like bring something useful out of my segue rant

and actually, like, turn it into, like, a useful point. I'm impressed with that. Yeah, it's... Because, like, I always talk about this to people, like, who don't understand how this stuff works. And, like, there are still many people who, like, link Stop Theming My App and, like, miss the entire point of Libaduera. There are still many people who like Link stopped theming my app and

missed the entire point of LibAdwaita. I have to explain to them they're five years old, how it works, and things like that. Purism and Red Hat are one of the few companies that actually work directly

upstream. LibAdwaita is maintained by Purism, as far as I know. And the idea was also suggested by Purism. And we can see, thanks to Purism and their work to upstream, how much GNOME apps have improved a lot performance-wise, even consistency. And now we're going to see things like recoloring API, like where you can recolor a lot of aspects of everything about GTK and Libet Weta. And yeah, that's a really, really nice thing. And I hope System76 start doing that as well,

that they start contributing to GNOME upstream. Instead of playing a 2016 canonical and creating their own desktop environment that will likely fail later on. Good luck with that, because they're already doing that. I know. I didn't know they actually had something set up for Weyland but someone posted in the discord yesterday they are

they're fully committed to taking this path I mean I respect that issue is like since 1976 I believe they are a lot smaller than what canonical was back then

and like if they have to maintain because like maintaining a desktop environment smaller than what canonical was back then. And, like, if they have to maintain, because, like, maintaining a desktop environment is really, really difficult. And GNOME developers, like, they have, they continuously have to, like, make a lot of compromises. They have to, like, remove features

because, like, and you can see, like, GNOME, they went in that phase where they removed, like, they kept removing each and every feature. But they, like, and you can see like GNOME, they went in that phase where they removed, they kept removing each and every feature. But the important thing is that they're actually re-adding them again, but properly implemented. Because, well, it's difficult to maintain legacy code in a very, very core, in a desktop environment.

Otherwise, you end up like Windows, where apps look at Windows 7, some apps look at Windows 10, some like Windows XP and things like that. If they keep maintaining legacy code, well, they'll have to make a lot of compromises. And yeah, like libad, like, Libadweda, even though it's limiting in a lot of places, it brings a lot of things, like, a lot of benefits as well.

I think getting away from that legacy code is also a big drive behind the push towards Weyland as well. Like, my favorite example of an XORG legacy code is there's a print server like in xorg to print not just not print to a screen like generically print to things a screen a printer why is this here why do we need it if i remember right they actually got rid of that code oh they did okay my example's outdated but true but like your point is still fans like they have they still have a lot of legacy yeah well yeah there's other examples out there's it was just my my my personal uh my personal pet

example yeah um there was a video like about the history of X11 Xorg and Wayland. It's like an hour and 30 minutes long. I might be misremembering, but at some point they removed the printing server part. Which was 200,000 lines of code or something like that. Fucking really crazy. GNOME, a lot of the stuff they did like they started from scratch and i think that's a really good thing um and that's exactly what we land did and obviously like there is going to have a lot of issues with that and we've seen that there are a lot of issues but as a long term like there are a lot of benefits to it. Well, one thing that really annoys me about the anti-Weyland crowd

is the idea that Weyland is going to be ready next year. It's going to be ready two years from now. Like, oh, Weyland came out. Weyland has been around for, what, 10 years? It's been around for quite a while. Probably longer than 10 years. Whenever Weyland came out... Weyland's been around for, like, what, 10 years? It's been around for quite a while. Probably longer than 10 years. Whenever Weyland first came out, it's been around for a while.

But a lot of people remember X386 being a thing. So I think that Xorg sort of came about... When was the swap? 2007? Something like that x36 it's not that old so people think that the swap over to xorg it only took that long to xorg to the state that it's in now but xorg is a continuation of a bunch of projects for like 30 or 40 years. 2008 is when X36 stopped. Yeah, so X.org looks like it's this really young project,

but it's really this like 40 plus year old legacy code base. And obviously Wayland hasn't had that dev time yet. Like it's going to take maybe another five or so years to get Wayland to that point where it's basically perfect. And this isn't just a Wayland problem. This is also a developer support issue. Like, you know, the remote desktop stuff, for example. Like, that's not just a Wayland problem to address.

That is also developers writing software and updating their software to use any solutions that might come out onto the wayland side but i think that from i've i i was using whale i was using sway all yesterday i actually forgot i was on sway until i like looked at my uh my header bar and i realized it was different from my regular ones like oh shit this has just been good all day like it most

most like the core Weyland composers like you know the Sways, Rivers Gnome, Katie's things like that they're mostly good enough they're not perfect but they are mostly good enough and that I think is much more than could be said when Fedora started shipping Wayland back

in 2016 oh yeah for sure yeah and I think in 2016 was also when GNOME like that's when they started pushing Wayland really hard like yeah as it's still

really important that we do push these technologies and if no one pushes it, it never gets mature I think SteamOS doesn't use Wayland by default but I know that they use Flatpak and Flathub by default and that's a really really nice thing to push Flatpak's

adoption unfortunately and that's a really really nice thing like push flatpaks adoption yeah like unfortunately there are some many applications that still need to update their toolkit or framework like discord yeah like discord

yeah it's like a really really big problem to be fair on discord's case they barely is a really big problem. To be fair on Discord's case, they barely have screen capture working, so they're never going to update that. Yeah.

They just care about Nitro at this point. Yeah. Well, they care about Windows and mobile. Mostly mobile. I don't know. I mean, they did change, like, completely changed the mobile experience. At this point, I don't even know what

they prioritized. That's what we can say for a fact. Yeah, exactly. And it's the same thing with other applications. There are many applications who are just like, oh, we have a Linux build, and they don't care about it anymore.

DaVinci Resolve. Zoom. Yeah. No one cares about Zoom at this point. They had that two-year period to really build up Zoom, make it good, and then, you know, people start going back to

work, people start going back to school, well, the money's gone now. Uh, what do we do? I don't know. Yeah. Oh, good for them, they were really sketchy anyway. Well, yeah, you know, end-to-end end encryption except the middle's

not encrypted so yeah yeah you know end to end both ends are encrypted they weren't technically lying they're just a middle man and I don't know what they were doing with the data to be honest like

yeah yeah who knows I don't know what they were doing with the data to be honest like yeah who knows and the I do kind of wish that during that period tools like why am I blanking on the one the really popular open source

video conferencing tool Jitsi Meet sorry yeah jitsi meet yes yes i kind of wish that tools like that did sort of gain some sort of traction but people said they went to the easiest solution that solution was zoom that solution was cisco webex that solution was um google has a offering that i'm forgetting the name of google meet sounds right uh yeah i think like one of the um if i remember right zoom has a couple of features that jitsi meet doesn't have yeah and oh like, oh, like, there's that issue as well.

Yeah, like, I really wish that people, like, would actually adopt, would have used Jitsi Meat instead of Zoom. Look, if we have to choose between terrible proprietary solutions, I'd prefer Zoom over Cisco WebEx. I don't know if you've ever used it, but it's not a good solution.

Basically, it takes everything that's good with Zoom, gets rid of it, and just focuses on the bad stuff. Ah, I see. It works, but you don't want to use it. What's the pain in the ass to log into and a whole other

array of issues? Yeah, I never used it. It's my first time hearing about it, but yeah, I guess I can not consider using that ever in my life. It's mainly for the serious corporate business people, because it's Cisco. You've probably

got a thousand Cisco routers throughout your office and Cisco access points. You know the name Cisco, so you keep using cisco that's pretty much it yeah that makes sense just branding we're talking about um red hat and they're like how much effort they put into improving the desktop i think a lot of people there's a lot of people that don't like systemd for example and don't like red hat but don't realize how much of the desktop actually is red hat even if not directly red hat like heavily funded or started from Red Hat developers. Yeah.

What I like a lot about Red Hat is that the employees, they're just like Fedora contributors that are the fan base. They're very free software oriented. In case you didn't know, they're very anti-CLA. They're against CLAs. They even have their own anti-CLA movement They're against CLAs. They even have their own anti-CLA

movement because they know the risk of how much of a risk it is for open source projects with CLAs. And instead, they adopted DCO. I forget what it means, but whatever Linux uses.

Develop Per Certificate of Origin. Yeah, something like that or probably that because like they actually do care about free software and obviously they do care about their own benefits as well because like you need a profit after all

One thing that I really really like about Red Hat is that they want their things to be community driven SMD it's very like actually not SMD, Fedora. It's very, it's a company, well, I was going to say company. It's actually, like, politically independent from Red Hat. If, like, Fedora, like, if the contributors, if they wanted, you could choose something that would go against Red Hat.

For example, Red Hat XFS is developed by Red Hat. But Fedora, they use ButterFS and not XFS. That's one of the things. And if the community wants, they could choose something that is not even developed by Red Hat at all and whatever it is. What I'm trying to say is it's not Red Hat that decides what they do with Fedora. But a lot of the things they do, it's mostly infrastructure, like Wiki, that's where Red Hat comes in, like Bug Tracker and things like that.

It's the same thing with Flatpak. Flatpak was started by a Red Hat developer, but it's very much community-driven as well because it's not just Red Hat that is involved. There's GNOME that is involved, KDE, Labra, Purism, Enlist, many organizations or foundations that are really involved in flatback development.

Same thing with SystemD. It's not just by Red Hat. Well, now it's Microsoft as well. But yeah, it was... Well, now it's Microsoft. Red Hat, there's... If I remember right, there was also Intel involved in SystemD. And there was also a couple of Debian members of Debian

who were involved in SystemD and actually members of the SystemD project. Because it's not just red hat that um that controls well actually they don't really control it they just employ someone to work on it there are also like other members who uh would like to vote or whatever it is even with GNOME GNOME isn't really a hat project at all but people say it's company driven

I completely disagree with that I think it's very community driven it's just like they do lack individual developers a bit but if I want see like they do lack individual developers a bit.

But if I want, like, see, like, Red Hat could employ me to work on GNOME, and a couple of months later, I can apply to be a GNOME Foundation member, and they will accept me because, well, I contributed to GNOME, and, you know, I've showed that, look, I put work on GNOME. Right. If I was an individual developer, like, and affiliated with, like, I put work on GNOME. If I was an individual developer, affiliated with any company,

and I would contribute to GNOME, and I would apply for a membership, they would still accept me for that. And after that, I can vote, I can start maintaining software and things like that.

The only issue is, if you like that. Only issue is like, if you see that GNOME is too corporate, well, you might as well try to like, start a bit contributing to GNOME. When you become a member, like you can have your say, you know, you can voice your opinion and say like, oh, we should probably focus a bit more

on the community side and things like that. I think the two corporate things kind of amusing, like there's, there's a lot of red hat employees that work on this. There's a lot of, you know, there's a lot of, well,

the reason why there's a lot of red hat employees that work on, on a, on a project like that is because red hat hires a lot of people and they hire a lot of good developers. So like, and they hire people to of good developers so like yeah and they hire people to work on on projects like it's not like they're just making this money then going and buying mega yachts like they're putting money into the projects you want to see improve I think

I think a lot of people romanticize this idea of completely free of corporate money, completely community-driven. Nobody is getting any money to do any of this work. It's just doing it for the joy of the code and doing it because free software is the morally good thing. But people have to pay their rent. And if a company like Red Hat has a model to make money

from corporate customers and bring money into this space I don't necessarily see that as a bad thing. It can be a bad thing in certain cases where developers say like a single entity gets a lot of control and then pushes other people out

but that doesn't seem to be the way that Red Hat and Red Hat employees are managing any of these projects. Yeah, and also a lot of the projects they work on, it's like that as well. Like, Gnome, as

I said, like, there's a couple of companies and organizations that contribute to Gnome, like, maintains LibAdwaita. GTK is maintained by the GNOME Foundation. Then there are some applications like GNOME Boxes that is maintained by Red Hat. And if I believe, if I remember right, WebKit GTK, like the engine, is maintained by a company called egalia if i remember right i might be wrong but we can see like our companies work on different aspects and it's not like one company

can just abuse it because if they do abuse it they're going to like affect other companies as well like that work on a gnome and plus the other members who are independent. There's going to have a lot of issues, so even then, Red Hat can't or any company can't really abuse inside GNOME. I think one thing

to make clear there, though, is these projects do... GNOME has their... the way they want, they've got their vision for the project. That's what I'm trying to say. They've got the goals they're trying to achieve. They've got their vision for the project.

And even, I think that's separate from Red Hat being involved. Because GNOME, for example, is very hard on the CSD side, for example, and doesn't really care about SSD. That's part of the GNOME vision. And even if that doesn't align with other projects and people try to contribute things into the project and they're not really receptive to this,

that's not necessarily a bad thing on the projects front. What you're trying to do doesn't align with what they're trying to achieve, if that makes any sense. You mean like it affects other desktops and things like that? What I'm trying to say is like, the Gnome project has its vision. And even if people try to contribute things in if that doesn't necessarily align with

that vision it's not them trying to keep you out it's more like what you're trying to do doesn't really fit with what they are trying to achieve.

Exactly and and there's a lot of the time there's like more than just contributing there's a lot of backlashing like a lot of the time, there's more than just contributing. There's a lot of backlashing. A lot of the time, potential contributors will just demand. And I think that happens.

This often happens, actually. I often look at the issue trackers. I often see people just demanding GNOME to do whatever the user wants. And obviously, who the fuck are you? You're not contributing to my... Oh yeah, maybe you submitted this merge request, but even then, you didn't say it in a nice way. There's nothing respectful about it.

Why should we respect you? Why should we merge you like um the way like uh um why should we merge your like uh merge request or why should we like honor your uh opinion and things like that um like i've had a really good like um experience with gnome developers because like i've been forming like I'm formulating my sentences like in a way that's like um Can I see this like? Wording is key basically. Um, oh I have to explain why that approach is better or whatever it is like or why their approach is bad

Like it goes against like doesn't really align well with their um mission um for example um i opened an issue on gnome software's um like issue tracker and like flat pack with flat pack there are extensions so like when you install obs, there are a couple of plugins you can install, like extensions. GNOME software right now, they list those extensions really well, but before that, they used to have a bit of problems. For example, you know VKBuzzle? Do you know MangoHUD? Actually, MangoHUD.

A lot of people know that. So MangoHUD is a free desktop runtime extension. And the problem is, if you wanted to install the MangoHUD extension from GNOME software, it would be pretty difficult. It would go to the runtime entry, and then you would install Mango HUD. That's really difficult because if you try to just type in Mango HUD, nothing would appear. I opened an issue and I

said, in my opinion, extensions should show up when you're searching. I explained why that's really how they should do that. They considered my issue and they literally

implemented it the next day. That's the same thing. You need to word properly. If I go on ADE's issue tracker and I start saying, oh, do it like GNOME, they're just gonna be like

what? no sure yeah well okay with the KDE people they are very touchy about anyone acknowledging the fact their desktop looks like Windows they really don't like it when people say I know that you know they have their own

designers and they have their own design principles and they got to this design in a very different way but I think everybody can kind of acknowledge that KDE kind of looks like Windows. That's not a bad thing. But it's also, when you're submitting an issue it's probably not the best place to bring that up either.

Yeah, personally, if people say that it looks just like Windows, I'd take it as a compliment because I think that Windows, they put a lot of work with UI. Especially Windows 11, it's really consistent. It looks really good. Obviously, I'm not going to be using Windows. I've heard other things about its uh issues with it but it looks good i'll give it that yeah like windows issue is more about the legacy stuff like for apps that look like windows 7 if we just focus on the windows 11 like ui like the settings app or um the windows store app those look amazing like they're really modern. They have a lot of effects. They have a lot

of animation. They look really good. If you say that Windows is bad, that's whatever. I think that Windows has improved a lot in the recent years. What we should, on the latest desktop and mobile, what we should

do, we should only take the good aspects of Windows and implement them and leave out the bad. For example, for Dora, they have offline updates. That's a really good thing from a safety point of view because everything is done offline and there's a lot more safety and stability. Obviously, people will just say, oh, that's bad because it's just

like Windows, but that's actually a good thing about Windows and we actually need proper updates on Linux. If that makes sense. No, I get what you're saying. Yeah, no, it... I think when it comes to Windows,

when it comes to Mac OS, a lot of people are very touchy about anything, anything that could possibly be turning their favorite Linux offerings, their favorite Linux distro into something they don't like. But the nice thing about having this project here is you can pick and choose the things that you like um and like yeah that's pretty much it is this it just pick

and choose what you like and if you don't need to have a windows update system for example like doesn't need to be a thing that forces your system to reboot. You don't need to have that. Just because Windows also has that, doesn't mean that, you know, I don't know. The way that, like, RegistryEdit,

for example, is an incredibly useful utility. Like, that is a great thing that Windows has. All of these system settings are in one place obviously it'd be a lot more difficult to do on something like linux but that's a great part of windows but yeah having that doesn't also inherently mean the bad stuff as well exactly and like gnome like people shit on GNOME a lot but they really try like they try really hard to only pick

the good stuff of numerous like operating systems you'll just hear many people say oh GNOME is a budget macOS believe me I've used macOS full time during work let me tell you right now like it's nowhere close to macOS

I feel like it's an upgraded version of macOS because it follows some UX standards. macOS, you cannot snap, you cannot drag your mouse to snap windows.

Oh my god, I can't speak. You can't drag to snap windows on mac os wait you still can't do that no not by default you need a third party uh application for that yeah it's we just had this since eight didn't it how's mac still not have it yeah i don't know honestly i have no idea um i guess it's just their thing different bullshit they like sleeping on yeah but what you have to do instead you have to press and hold the full screen button then it asks you like oh snap to the left snap to the right that's such like that's so non-standard.

Yeah, in my opinion, it's really bad. And I saw a lot of inconsistency issues as well. For example, if you double-click the top bar, it actually behaves differently depending on the application. If you use Finder, like, depending on the application. If you use, what's it called? If you use Finder, like, the file manager, when you double-click at the top bar, it does nothing. But when you double-click on Chrome, it maximizes the window.

And then there are some, like, applications where, like, Safari, when you double click the top bar, it expands it to two-thirds of the window. It's super weird. I don't understand. There are so many issues with macOS. It's really sad. And I'm not too sure if GNOME got inspired by like

Aqos' Mission Control. You know GNOME's Overview, right? Can I share my screen? I don't have it set up right now to capture your screen. I'll rip. Okay. Just let me... Can I find it? I'm just going to send a... Oh, yeah. I probably know what you're talking about. I just don't know the...

Specifically what you mean. Hold on. I'll just send it on Discord. Okay, yeah. That's all good. yeah that's all good yeah oh that's the overview right yeah so this like mac os they have their own thing called mission control and it's really like it's a similar concept the difference is that gnome makes it they do that a million times better. Like, one of the things I really like about Gnome is that they

follow web standards. Like, you know, if you look at many websites, everything is like at the middle of the display, right? There's some maximum width.

GNOME does that as well. So if I open my... Hold on, let me show you. If I open the application menu, you can see how everything is centered, right? Sure. And on macOS, they spread everything everywhere. And that's not really a good experience when it comes to eye travel.

Because this causes eye fatigue. GNOME, since they're putting everything to the center of the display, even if you use an ultra-wide display, it's still going to put everything to the center of the screen. And it reduces a lot of eye travel, which means you can stay on your computer for a much longer period of time. It's coming from someone who uses computer

like for four or five hours, not any breaks or anything like that. And I'm still not tired when I use GNOME, because you do many things right when it comes to UI. About the rent and all that, there are many things where... Have you seen the new quick settings from GNOME? There's a lot of stuff that GNOME's doing

that has not crossed my crossed my path. Okay, that's fine. Let me just see another picture. I think I might be able to find it. Oh, yes. Yeah, that, like, I don't know if you show it in your stream, but that's like Android.

That's like Android. You know, the newer Android 13. Yeah, I'll check it now. That's like Android. Like, yeah, that's like Android. Yeah, nice. Like, you know, the newer Android 13. Yeah. That's what I mean. Like, they take the good stuff from other operating systems and they put them into one environment. You don't try to, like, follow the bad stuff.

And, yeah, I know there are some exceptions, like removing system tray. I think that's a massive, like think that's a massive problem and especially like many users actually depend on system trays this also brings to the same issue as I said before like you know you have to keep up

legacy code and then you have to work around all of these issues that you have to maintain you have to support you have to port over to each and every single version of GNOME. In that case, they're like, you know what? Let's just get rid of the code. And the community deals with that, like using extensions.

And in my opinion, it's better when passionate people maintain those types of software than people who are not really passionate. Do you understand? Yeah, I get what you're saying. Yeah, and... Yeah, I do agree that system tray was... System trays are pretty useful.

I like system trays. Yeah, personally, I hate system trays, but at the end, we're trying to focus on the average person, and removing system tray isn't really a good approach to that. At least we're creating a new protocol for system traits. Because one of the problems they had with older versions of current implementations of system traits

is that they're very legacy, they're not sandbox friendly, they don't work well a lot of the time. And like there are issues here and there and things like that. At least the fact that they're fixing that after 10 freaking years, at least that's a good step. Even then, like that's at least in my opinion, it wasn't that much of a good decision. So I hope there's a lot of long-term benefits to that. I think that a lot of the time, Gnome, they do a really good job

when it comes to taking the good side of the good UI from different operating systems. And yeah, I know that you're not really, you're not a fan of header bars at all. That's, that's, okay, my issue with header bars is when you stick

use, I don't think the only way to access some functionality in an application should be in the header bar. Like, if you're going to have something in the header bar, fine. But have some other way to access it, because there are a lot of desktops that don't show the header bars or will hide the header bars in full screen like if you're gonna like don't have all of the buttons in your application inside of a hamburger menu that's going to be hidden when i go full screen like that's that's all I have to say. Put it in a context menu, do anything else, like, give some sort of option, or if you

are on one of those desktops, then have it show in another way so it doesn't end up breaking the application. I've seen so many programs which do not function in full screen on my setup. Like, that's the issue I have. If you want to do it on GNOME, be my setup like that's that's the issue i have if you want to do it on gnome be my guest that's fine but yeah like and that's like i guess that's one of the like i mean i don't really have an issue with that but like i i understand why a lot of people are against that it's like gnome they only focus on GNOME that's it if you have issues on another desktop

you're just gonna like there are times when they're just gonna tell you fuck off like don't care this application is made for this desktop entirely so like like your problem I mean obviously I'm putting it in a really really rude way

like like GNOME Your problem. I mean, obviously, I'm putting it in a really, really rude way. Like, GNOME, obviously, like, was seen in a much more, like, professional way. And I don't really blame them for that. Because, like, when it comes to toolkit, it's really, really difficult to maintain. And if, like, what you suggest will require a lot of code to maintain like a lot of time a massive merge request and things like that so i can understand why like they don't want to take that approach because like like uh it's just too much to maintain my concern isn't like gnome core apps like if i get like things made specifically for the GNOME desktop are going to be designed in this way. But there's a lot of apps that are designed for GTK4, for example, that are designed as GNOME apps that sort of have the same issues.

Where it doesn't make any indication of the program that it's designed specifically for GNOME. But when you run it, it's really obvious what it's actually for okay yeah yeah that makes sense yeah i don't think there's like there's gonna have something like that anytime soon but yeah if they say like oh this this app is like made for gnome that would be really nice though yeah i can't deny that app is like made for gnome that would be really nice though yeah i can't deny that then it brings to another issue where like if people like newcomers they won't really understand what a desktop environment is they won't know what plasma is or gnome is and when they say oh this application is optimized for gnome they're gonna be like what like what's a gnome what actually no not even what's a gnome because they don't even know what gnome is.

Yeah, no, I I understand there's certainly issues here, but this goes back to the documentation issue, like I think it's better to have it there than to not have it there. Even if, like, many users aren't going to understand it, like if they're new to the

system, I think for the people who do understand it, having it there is probably, at least for me, it's definitely gonna be better. It's gonna save you a bit of time for sure. Yeah, that's where like, um, these, uh, software front-ends come in. They show screenshots of the application. And, like, yeah, I understand that this isn't, like, you're more of a CLI guy.

Well, I get a lot of stuff from, like, I check GitHub's stuff as well. Like, if, like, a lot of projects, I'll, like, check out the GitHub, just, you know, because a lot of the time they have their documentation there as well. Like, yeah, speaking of that like like it really bothers me when

apps don't show any screenshots in their like reviews I agree yeah that's one of the things I hate the most and obviously like it's not something you can just like submit a merch request and be like oh add these screenshots

I'd rather because there are times when developers have like such terrible takes oh better yet like just open an issue ask them hey can you add screenshots if they say no well you didn't have like any time wasted on submitting a merge request or taking screenshots and things like that but like so it's also a bad take on the developers and things like that but yeah it's yeah it really sucks it really pisses me off when when apps don't put screenshots in their readme it's one thing if it's a cli tool you can get away with it like that. I still prefer them, but when it's a GUI, just include one. That's all. You don't need a big array, you don't need a video demonstration.

Just a single screenshot. That's all I ask for. Yeah, exactly. I don't know what's so difficult about that, or why people overlook that well it's even when you're on github like

it's not like you're paying for the hosting anyway it's not like you can even make that argument just press your screenshot button just do it it also brings back yeah it also brings back to the issue like where no one wants to document so they just don't care

yeah yeah well that you know is as good a place as any to end off the show documentation the pain in the ass and no one wants to do it yeah exactly and exactly how we started off and so uh if i remember right you also asked like how i think about like how the future of the linux desktop will go if i remember right yeah yeah i don't think we got into it on the show but i did i did mention it on the discord yeah um i mean if you want i can share my thoughts yeah we can i i don't have to stop right now it's just uh yeah we can go into that that's fine like yeah like i think like because flatpack um is very like a collaborative it's a collaborative project between many organizations i think like that's where we're going to head

towards and and like since it also emulates different architectures, that's also a really big upside from using Flatpak. And plus, like, reproducibility and things like that. Or, like, it's not just like Flatpak. You can make one manifest, one for ARM64, like just one small section of the manifest, like the build script, I guess you could say it. You can make like one part of it ARM64

and the other side x86. So if someone uses an ARM64 device, it will automatically pull the ARM64 binary or archive, whatever it is. If they're using x86 device, it's going to automatically install and suggest the x86 app. And not just that, especially with the rise of XTG portals,

we're already starting seeing integrations electron uh gtk also has many like many um portals integrated already and you know those permission systems on android right yeah yeah or actually we're already starting to see that on linux as well um that's gonna scare some people they don't like any change change is bad yeah and i mean the upside is that this only happens with sandbox environments if you're using from a distribution package they're not going to ask any of those permissions yeah they would just gonna have exactly so like no uh when you open a camera app on android, it asks for, like, access to the camera, right? Yeah, yeah. There's already a portal for that, like, for your camera.

And there's an application called Decoder that already does that. Let me try to find a screenshot. Probably sent it. You probably saw it. I don't know. I saw the little icon you had for that. I've never seen this app, though. I did talk about it in your Discord server.

Let me check. A lot of things that talk about that server. There's a lot of things that I missed. Yeah, that's fine. Okay. I found it. You can see it has turned on camera. And it actually works.

And of course, this is because GNOME, they have a vision, and they want to push the Linux desktop forward. It's already integrated these portals already. And many app developers who do applications tailored for GNOME, they want to use portals as much as possible. And this application decoder, as far as I know, it has no access to your file system. It's an actual secure application.

Because it doesn't have any access to your file system. It has very little amounts of permissions in general. And it doesn't even have access to your camera by default. Because while the portal is in use, it will ask for permissions before actually using your camera. Yeah. And obviously like this is a really nice thing especially for um not only the linux desktop but the linux mobile space

i see it in a like in a way that linux mobile space is going to be a very like really big contributor factor for the linux desktop because they're already like very like very bleeding edge like they're pushing wayland already like many distributions already using wayland already using flat packs and flat hub like i don't know about pipewire like i never had a linux phone i i wish i hope i do at some point but like the they do push a lot of those like bleeding edge technology and that's a really really good thing

especially for the desktop because in many ways in Linux mobile in some ways already ahead of the Linux Desktop, despite being very new. Yeah. And, like, yeah, people will just say that Linux Mobile isn't ready and all.

Linux Desktop isn't ready, and it's far from ready compared to Linux Mobile. At least on Linux Mobile, we have WagerOid, On Linux Mobile, we have Wadroid, so you can run Android applications on your phone. And I know that you can do the same thing with desktop, but Android apps are less in use compared to on mobile space. I think we'll talk about Linux mobile not being ready it's some people do focus on the software but I think the main concern is the hardware side like the hardware is still very much

geared to even when you buy like a pine phone they very much say like this is for a developer like this is a developer system if you're not a developer if you're not someone who's trying to improve the Linux mobile experience this might not be the device for you

yeah and i don't think it's bad because like you gotta start somewhere and it was often like that like a lot of those experimental stuff start in like other places like server space or ssds for example they sucked at the beginning but now they're really really practical and it's the same thing like at the beginning they will always suck it's just a matter of time the price goes down um people start seeing like better hardware for cheaper price, better software as well, and better drivers, and just everything regarding Linux mobile. I don't have a Linux phone myself.

I don't really... I think the mobile space is interesting. It's just I'm already so involved in the desktop space. interesting it's just i'm already so involved in the desktop space i don't know if i can uh if i can justify spending that time on mobile as well maybe i could make it work somehow but i just don't really care about my phone i just have it and then i use it i don't even run like a custom rom or anything i just it's just it's just there it It's just a phone Yeah, I went to that like distro phasing well ROM custom ROM phasing Distro, oh my god some ROM hopping phase. I just had many custom ROMs and at some point like I'm bored of it. I just want my phone to work. Yeah, and I Think on Linux like I'll go back to that like phase where I'll just distro hop

again and like learn try out a lot of desktop environments and things like that but like it's good to see like a mobile space and it's not like you know GNOME like you know that GNOME they're focusing on like mobile right like

or making I'm sorry? I was just sure sure yeah so like because like at some point gnome like right now it's just it's a desktop environment but it's going to be more of a general purpose environment like they're working really really hard on mobile um have you seen any like uh announcements or things like that about GNOME Mobile? I did. There was a video that came out the other day where someone was scrolling through GNOME Mobile. I don't have the link on hand right now.

Yeah. And, see, like, a lot of those stuff, like, desktop, like, GNOME on the desktop end actually benefits as well. like GNOME on the desktop end actually benefits as well. Mm-hmm. And especially like, you know, I think it's called Samsung X or something like that? DeX. Know that? Samsung DeX? That's ringing a bell, but I'm not sure why. Samsung...

Like... Uh... Transform your galaxy phone into a powerful work ah yes yes yes okay i know this is yeah and since like gnome is shifting towards a um general purpose environment like and the fact that like i use a gnome and i can see like it's a really really efficient desktop if you spend the time to learn it and like once uh becomes very mobile-friendly on GNOME Shell, we will be able to easily just switch to the desktop environment of GNOME, and it's going to work just fine without any mobile limitation as well.

without any like mobile limitation as well that's so sounds a lot more compelling with uh with gnome than it does with uh with samsung's uh skin for android like i get it like you want to use one device but gnome sounds a lot more appealing for that. Like, it's a full-fat Linux experience. It's not like this weird... This weird Android experience. Like, not really made for the desktop. Like, desktop Android never really picked up until they went down the Chrome OS route, which is doing it a little bit differently. But desktop Android just never really made sense.

But Gnome starts on the desktop so if you have this experience that you can like share with one device and you know when we do get that powerful hardware like right now like you could do it like you could probably use you know the hardware in your phone connect to like a bigger display and then you could have an experience but like you don't really want to with the hardware we're seeing right now in uh in the linux phones but give it time when we actually do start seeing that powerful hardware i i could imagine that being a pretty good way to have just one device you can carry around and just use wherever yeah exactly and that's what i really really like

about it and it's not just that like gnome scales really well on a desktop it scales really well on tablets and on laptops as well because like with wayline you can use like touch like it has a really good touchscreen um touchscreen um experience and trackpad experience of course there's the linux kernel limitation where like if your drivers suck well there's no experience it's going to be trash unfortunately but even then like gnome by itself i have a dell laptop with me it runs so well on wayland um like i have the um trackpad gestures and things like that it works really really well and thankfully even Plasma they're making their

desktop environment to laptops and touchscreen devices and I think that's a really really nice thing I think the Linux mobile space is a space that it's kind of like looking back I think the Linux mobile space is a space that... It's kind of like looking back into the past of Linux.

Like, this is, you know, Linux in 1998, for example. Like, people know about it and, you know, some people use it, but it's still this very early thing that no one really is sure what's gonna happen with it, but it has this massive amount of room to grow, especially now, you know, we're not in 1998, we have the hard, the hardware is available, so you have this, like, massive, this massive, massive um existing range you can already grow through you're not limited by what is possibly available so i i don't know if it's ever going to be in a position where it's popular to use linux on on mobile but i could see it getting

to the point where it's kind of like Linux on the desktop where it's not the main thing by any stretch of the imagination but it is this thing that is entirely reasonable to use if you are more of a you know a hobbyist more of an enthusiast, then maybe from there, it can get to the point where it's, you know, anyone can reasonably go and use it. I think that's a while down the line, but I don't see it as

being impossible. Yeah. I think, like, now, I don't really, like, I'm not really sure if it will ever become popular. I do think it will become successful because Windows Phone failed because it had, like, no support. Yeah, that was Microsoft. They just released it.

Well, it's not the first time they've done that. They did it, like, earlier. They had a partnership around the time the iPhone came out as well. And they just didn't do anything with it and they keep doing it whenever Microsoft gets in the mobile space they just don't do anything with it you can't just release it and then

expect the devs to come they're not going to show up because no one's there already like what are you doing? Exactly but on Linux mobile I think the main savior we could say is wageroid because you don't need application developers to pack it to package like as a flatback like if it's available on android all users will be able to install android apps on linux and maybe like you know

there might be limitations here and there, especially with SafetyNet. If you try to run your bank application, it's like, oh, your device is supported or it doesn't have SafetyNet and things like that. But I think for the majority of apps, you'll be able to run them on the Linux mobile, especially because WayDroid showed that it's a really good project.

If I remember right, you can install applications from WayDroid, and it will automatically create desktop entries and launch it as if it's a Linux application. And things like that. This, I think, WayDroid is one of the biggest saviors of the Linux mobile, because it's basically wine of mobile.

Instead of translating from DirectX or like windows apis it's from android to linux well that's also much less of a barrier as well because you know linux kernel it's not it's not as big of a jump between those two because you still got that same base there it's a very modified base but it's not like you know the windows api calls are very very different from what is available on linux but the linux api calls in many ways are still going to be there in the android side this is going to be extra things that android needs as well yeah and even then like android is open source so yeah you can inspect the source code yeah yeah like i really think um like i'm going if tourism releases a uh liberum 5 second generation

i'll try to buy it as soon as i can because i really want a linux phone and you know i want to support the development and things like that. And if I remember right, Purism also, like, donated or funds Wageroid. But I'm not too sure. It's still really cool that Purism, it's sustainable as well. Well, at a bare minimum,

they would be developers on it, if nothing else. Yeah. I don't have any plans to buy a one anytime soon but maybe you know i i don't like to buy like early generation stuff i tend to like wait until stuff has matured a bit um but i could certainly see a day where like how i'm running you know arch on my desktop i'm running linux on my phone um i think i think we're in for a exciting couple of years maybe exciting like decade for linux like we've got this shift over to wayland happening that i would be very surprised if by you know 2030 for example we weren't heavily heavily on uh on Wayland by then we have a lot of people on Wayland now but I think most of the shift is going to happen by then and there will still be the people holding

on to X but it's going to be a lot fewer than we have now we have flat packs continually getting better we have portals and other technologies being made around them to make them you know generally usable on the desktop and we have the linux phone which i don't think is gonna hit like it means from appeal by then but very well might hit a point where the average linux user would possibly consider using it as a daily driver. Oh yeah, that's what I want, because I don't really like Android, and I don't like 50Net, and how Google is basically giving all the dependencies to Google itself.

And so like at least having a Linux device, it's something that I would love to have in the near future. And even then, like, I know it's not going to be ready anytime soon, but if you really wish something to be good, it's the best to contribute. It's best to like spend your time or money there. So like at least you can give something back to the developers so they can continue working on their own stuff.

Well yeah, when it comes to buying a device in that respect it kind of depends on how important you consider it because it's not like an iPhone 14 but it's still a lot of money

and a lot of people don't have that to spend, but you can still help out with things like documentation. That's not something you necessarily need to be hands-on with the device. Yeah, exactly. And even then, if you don't understand something you can always ask the developers like hey how does that happen how does this work and they'll explain to you nicely yeah well i think yeah i was gonna we're gonna end it like 20 minutes ago i think i think now

it's a good time to end unless you got something else you wanted to bring up? I think everything's fine. Okay, cool. Where can the people find you, find your work, things like that? On my GitHub. On my GitLab. And if you want, I can send you links, actually. Let me just send you links. Yeah, I've got your website open right now. So GitHub here, that is. On the evil skeleton. I've got your website open right now so github

here that is on the evil skeleton your gitlab is also the evil skeleton wow you actually have better branding than I do and then your website is on your gitlab so it's also theevilskeleton.gitlab.io yeah and there's also the

gnome gitlab where yeah so like when I do my contribution to gnome and i also have a mastodon i also have a twitter but like i prefer using mastodon and recommending mastodon users um that's like i didn't put my Twitter on my website because it's a secondary thing yeah on Mastodon I post updates as well yeah any other things you want to mention?

I think everything should be good now okay cool as for me the main channel is Brody Robertson i do linux videos and all that fun stuff there the gaming channel i is brody robertson plays i'm thinking of changing the name at some point i have not done so one day i'll do it though uh right now i'm playing through the world ends with you and some other mystery game that I haven't decided yet. So, show up for the stream and you'll find out.

It might be a roguelike, it might not be, I don't know. I decide these things usually the day before I start streaming. If you're listening to the audio version of this, the video version is available on the YouTube at Tech Over Tea. The audio version you can find anywhere you can find podcasts. There's an RSS feed, it'll be on iTunes, all that stuff, so you'll find it somehow. Give me a last word, what do you want to say?

I mean, I guess it was I enjoyed talking with you. I think it was a really fun experience. Especially like it's my first time being in a podcast like that. And, like, I did better than I thought. Because at first I was, like, I was a bit stressful. I'm like, oh, no, like, why am I going to...

Hopefully I'm not going to say anything stupid. But now it's... Don't worry, I'm here to do that. Nice. I think what I've noticed is anybody who worries if they're going to do well on a podcast generally so there's two people who do well on their first their first podcast with me the people who worry about doing a podcast and the people who have their own podcast which usually worry about it anyway if you if you're like fully confident like this is going to be great. I am going to be speaking perfectly.

Usually it's terrible. Well, yeah, that's going to be it for me. So I'm out. Goodbye.