Tech Over Tea - Linux Is Better Everyday | The Linux Experiment

Episode Date: April 5, 2023

Everyday Linux keeps getting better and better, whether it's Wayland, desktop environments like GNOME or anything else out there, and The LInux Experiment and I are here to chat about it among various other things. ==========Guest Links========== YouTube: https://www.youtube.com/c/TheLinuxExperiment Twitter: https://twitter.com/thelinuxEXP Patreon: https://www.patreon.com/thelinuxexperiment Mastodon: https://mastodon.social/web/@thelinuxEXP Pixelfed: https://pixelfed.social/TLENick ==========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 the show. Today we have a returning guest. Welcome back to the show, Nick from The Linux Experiment. How are you doing? Hey everyone. Well, I'm fine. As I was saying, I just got out of a week of COVID, so I'm now better, but I was not that great before. When you record your videos, a lot of yours are like very sort of topical newsy stuff. So you don't really backlog anything, do you? Not really, no. Generally, at most, I'll have one full video recorded, but not yet edited.

But that's the most. I plan all my topics way in advance at the end of the previous month. But, and I sort of deviate from it if there's really something really huge. But generally, I just stick to the plan, my 12 or 13 videos per month. And I just record them as they go. Generally, I've only a script in advance, a script or two. But that's it.

No completed videos to be ready. Yeah. You've got some sort of better situation where you're not uploading like, you know, six times a week. Yeah, well you've got some like sort of better situation where you're not uploading like you know six times a week. Yeah. So but you also edit your videos a lot heavier than I do. How long do you actually spend editing a video by the way? It really depends. The news videos are kind of quick because it's just basically news articles scrolling and sometimes just my face so they're relatively easy I'd say

three hours maybe for these ones and another video of 15 or 20 minutes on a topic is generally going to take like six hours

just like recording all the b-roll and slotting it in and making sure everything feels tight and removing all the parts where I talk too much

that are super boring it's generally I record about 30 minutes and the end video is like 15 so huh okay yeah i'm really bad at speaking on camera and i've noticed that i've noticed that the longer i'm doing youtube the sort of the pickier i get with how i'm saying a word. Like, I'm starting to notice exactly how I articulate certain words. And my videos that are, like, 15 or so minutes long usually takes minimum an hour to record. There's no...

Like, I could go with the 10 earlier takes and it'd probably be fine. But, you know, no one else besides me is going to care about a slight different, you know, intonation of a word. Oh, there will be one person that cares. You'll get the one comment saying, oh, you pronounce that weird. They'll care if you say, if you start talking about gnome, there's always going to be that one person who's like, no, it's gnome.

And the other person going, no, it's gnOME. I'm just like, I don't care? Yeah, it's to the point that when I did my GNOME 44 review, I literally prefaced it with, hey guys, I'm gonna say GNOME because that's how they say you're supposed to say it, but... Yeah, yeah. Yeah, I did see that went up. Someone asked me the other day, like, are you gonna do a video on GNOME 44? Like, I've literally never done a video about any version of GNOME.

I don't know why you're asking me. Is there something exciting that happened with this version? Because I got multiple people asking me about it. So I was really confused. It's not a huge one. Like, it just refines the quick settings. It adds background app support, which could in the future be a system trailer placement, but it's definitely a huge one. It just refines the quick settings. It adds background app support, which could in the future be a system trailer placement,

but it's definitely not right now. And basically, that's about it in terms of major big things. What was interesting to me in this one was that they sort of focused on adding stuff that people have really been asking for. And I don't know if it's just like a magical moment where all the work they were doing just coalesced in this version specifically people have really been asking for and i don't know if it's just like a a magical moment where all the work they were doing just coalesced in this version specifically and not a conscious

decision but you got the the ability to disable overlay scroll bars you've got the thumbnails in the file picker you've got the beginning of something there wasn't thumbnails in the file picker before not in gnome for 14 years you you did not have that. Like some file pickers implemented a hack that showed a side panel with it. But generally, no, you just didn't have an icon grid. You just had a list view which had no thumbnails. So I don't know what was blocking that, but that was ridiculous.

And now it's fixed finally. And you have like a bunch of cool quality of life options that people have been saying Gnome needs this. It's ridiculous. They don't have it. And they all ended at the same time. So I don't know if it's a conscious decision to try and focus more on what people actually wanted. But that was what was interesting to me in this one specifically.

In terms of pure new features, there's nothing huge or anything. I did see there's like little like i occasionally will do videos about random things that are going on in the background like how there's hdr being worked on things like that but i didn't know that about that file picker issue um yeah it's it's been like one of the biggest pet peeves of everybody who has a like something against you know that I this sucks why isn't it here yet every release you get a comment still no thumbnail still no thumbnails well now now they're there 14 years later it was terrible pretty much my main experience using a gnome is I've never like properly daily driven it so I I'm not really any in a position to really

comment on whether it's great or not. It seems like they're doing cool things. Um, the maintainership and how they act with certain people coming into the project is, you know, it's,

it's not everyone, but there are certain people that are a little bit, you know, a bit aggressive. yeah, they, they have a very low tolerance for any criticism basically yeah but my main experience with the gnome is just because of the sort of the mindset around how people outside of gnome view it

a lot of people tend to get annoyed if they see someone running it at least in my case maybe it's because i early on build up a lot of those like window manager kind of audience um when i ran it like when i was doing anything in a vm on stream like when i was doing my lfs streams i had a gnome desktop just because it would bother people like i could have gone with anything else but it's like you know what gnome it is and from that brief experience it seemed fine i know some people it's great that that's the one i use I use it on my laptop and on my desktop. And I don't understand people saying

that you cannot be productive with it or that it completely sucks. You might not be used to the workflow it offers. And apart from using extensions, you can't really change that workflow. So I understand why people might be annoyed by it. But honestly, it just works fine. You press super, you type the name of the app,

you press enter, it's done, everything works it runs, touchpad gestures are insanely good on it, it's smooth it's really easy to use and applications look really good, they're all super coherent if you care about that stuff, like having applications looking the same, feeling the same, having

the same HIG, then you have to use GNOME, no other Linux desktop comes even close. But if you want power, customization, and settings, of course, that's not the one you pick. And it's not designed for that at all. That's for sure. I don't know if you saw the video that DistroTube

put out the other day about, I think it was called Linux has become complicated and limiting. I could not have disagreed more with that. I talked about it on my my patron cast like the little podcast i do for patreons and and youtube members i just i don't want to do reaction videos i think they're just drama and they're not that great but i mean i respect it with a 27 minute reaction video yeah i i just it's not my stuff no i get it makes sense they can be fascinating. And honestly, on this one, I really debated it

because I thought it was just an exchange of opinion. It's not like attacking DistroTube or saying he's an idiot or whatever because that's not the case. It's his opinion. But my opinion is the complete opposite. I think Linux has become way easier and way less complicated.

Before, it was all held by duct tape and string and it just never worked right. And all the options to completely customize and tweak are still there. What you don't have is the fact that you need to fix your system every other day. You don't need to do that anymore. And you need it to 10 years

ago, which seems to be the period DistroTube is referring to. And in terms of being limiting, I don't think it is either. You even have more choice than before if you consider it. Since Denom3 was introduced, you got Cinnamon and you got Mate, which did not exist before.

So, like, they didn't plan on making sure that people created those, but in the end, you have more choice now than you ever had. And the same for init systems. You can replace systemd with other things. I don't really understand the opinion, and I totally disagree. It sort of

felt to me like it was rose-tinted glasses, like really heavy nostalgia for the time where Linux only half worked, and so every improvement was awesome, and every day you had something new and something brand new. Something I felt with Ubuntu, for example. I used Ubuntu from 2006 until, I don't know, 2013 or 15.

And every release, it was a major upgrade. Like all the things that change, all the things that you couldn't do graphically, now you could. It was insane. And I do have nostalgia for that. But I don't have nostalgia for the fact that my hard drive would suddenly become unbootable or my X.org server would die after an update. Or dependency hell every time you install anything.

No apps were packaged by anyone. You had to compile most of the stuff. It sucked. It was a bad time to be a Linux desktop user. Unless all you want is hacking and tinkering. It just was not good. And so yeah, I don't really agree with uh with what he's saying in this video but i mean everyone is entitled to their opinions of course yeah my

general take on that video was i think there is a place for both of these people on the desktop there is a place for the people who want a desktop that just works they want that windows or mac os like experience where you plug it in and it's just done you don't have to worry about tinkering with anything and you can you know be a developer do anything you want to be doing and that's great but there's also this place where you know you can grab a window manager you can grab you know kde install a bunch of random stuff customize your desktop to your heart's content, and really build a completely unique system.

And there's nothing wrong with either of these approaches. Not everyone has to be taking down that simple path. Not everyone has to be that hacker that messes around with everything. I think a sign that Linux is evolving and becoming a more user-friendly... a more...

How would I say it? Approachable, maybe? Yeah, approachable. That's a good term. A more approachable system is the fact that you have both of these coexisting in the same space. Exactly. When before, like 10 years ago, 15 years ago, you could only use Linux if you were okay

with tinkering, tweaking, fixing stuff. Nowadays, you don't have to, but you still can if that's what you like. And that's the great thing. It's become more open in a way. It has more options. But you need to pick your options depending on what you want.

And of course, if you start going with Arch and Sway and whatever and trying to replace systemd with, I don't know, init5 or anything else, then if you're a beginner, you're going to have a hard time. It's going to suck. You're going to learn a lot,

but it's going to suck. If you're experimenting, it's going to be fun. You just need to pick the right option for you. And that's where the difficulty lies, basically, because you will hear many people telling you that you should never use Arch, just use Zorin OS. And people telling you you should never use anything but Arch or anything but, I don't know, NixOS,

even if you don't know anything about anything. Please don't use NixOS if you don't know anything about anything. That's going to be a very bad experience. Well, you're basically going to have to not install anything. Sure, sure. But yeah, I think it's a great place to be to be a Linux user these days. It's nice. No, I completely agree.

When I joined Linux, it was what was it? It was probably someone's going to I uploaded a video when it happened, like 2019 something. Probably earlier than that. I don't know. I uploaded a video when it happened, like 2019 something. Probably earlier than that, I don't remember. I was already here, well into everything, becoming, you know, relatively user-friendly in this way.

Like, you know, it's not been this massive change since then, but even in this time that I've been here, you can certainly see there has been some polish that's gone into things to make it a little bit nicer. Like the, I think last year, year before, when Red Hat was bringing some people on to specifically work on accessibility with GNOME. They hired, I've completely forgot the developer's name, but there was a blind developer who focuses on, like, you know, purely doing accessibility stuff. And that's the sort of stuff that just wouldn't have happened 10, 15 years ago when it was this more of a hackery environment. You know, plus the fact that there was less money in Linux back then. So really justifying that didn't really make any sense. And there were so many other huge things that you had to focus on, like just having a solid display server that didn't really make any sense. And there were so many other huge things that you had to focus on,

like just having a solid display server that didn't die, having an automatic configuration of X.org. Sometimes in 2006, you had to write your Zorg.com file yourself. That's just unacceptable if you want to reach the normal people that just want to use a computer. So of course they focused on that and not accessibility.

But now I think it's, yeah, we're in a great place. And yeah, sure. I started my channel, I think it was in early 2018. And since then, I've seen the Linux desktop mature and move on, but it's not been like an enormous change, like you were saying. There's not been a revolution in terms of how usable it is. It's been getting gradually better.

It's a linear curve, basically. I think the only thing you can really say has been a revolution is this sort of rise in containerized solutions like Flatpak and Snap to some extent, but Snap, because of the community opinion on it, sort of is still relegated to just ubuntu um most of the attention now has gone to flat pack and then app image could have done something but

the developer is not managing the project in the best way um well i've seen a lot of aggression on this part as well like like uh there was this this moment where they reached out to a lot of app developers saying hey we can help with your packaging yeah like valid questions and interrogations were raised and then they answered in like oh you're just depriving the community of an option you're an idiot and uh yeah that's not the best way of making sure people adopt your formats but flat pack has made sort of a lot of things a lot simpler like there were there's a lot of applications like bottles for example where packaging this was kind of a nightmare because it is very particular about the versions of software that it needs and just not

you can get it running without doing that. But the problem is it's going to, it's going to perform a little bit differently depending on which system it's on, making it a lot harder for the developers to sort of properly work out what bug is caused by what they're doing, what bugs caused by the dependencies you're using, eliminating that and just having, okay, this is the universal package makes it a lot simpler for these developers. Yeah, and you're absolutely independent from distro updates

because with something as weird and complex and particular as Bottles, if, I don't know, it runs on Ubuntu, Ubuntu releases an update to one of the libraries it needs, it could just completely break the app because it's very specific in that way. And one might say it's because the app is badly developed. It should be more, I don't know, compatible or be more flexible. But sometimes you just can't

or you just don't have time to do that for every single distro. So for these kind of applications, can't. Or you just don't have time to do that for every single distro. For these kind of applications, like containerized packages, it just makes super sense. That's not English, but it makes sense. Yeah, it works. It's what they were designed to, basically. They were designed to do

that, exactly. To remove all those barriers and obstacles that traditional packages have. And they will never probably erase all those traditional package management systems because you need libraries, you need underlying systems, you need the kernel, you need drivers. And Flatpak is never going to address that.

It's not the goal. But I think it solves a lot of problems. That's been a major development in the Linux world for the past three or four years. Well, I think it's been around. I don't even know how long it's been around for. I want to say at least 10 years, but it didn't really gain a lot of traction until very recently.

I did see you did a video on Flathub and the concerns around corporatization. I did a video on this a while back. I was very positive about these changes. I think it's all around a good thing that they're making it easier for developers to have some sort of funding stream like this. Yeah, and I totally agree. My opinions were like, I understand where the concern is coming from, because as soon as there's money involved or investors, you don't know if the project is going to focus on users and not just on making money at the detriment of users. But honestly, for something like Flathub, I don't see how that could happen.

And even if it does, there is literally no barrier or blockages to remove Flathub and just completely copy it and run something else. It's open source. Anyone could just make a copy and run it. Any distro already has the infrastructure and funding to run something like that. So if Flathub goes evil or corporate or whatever, I mean, anyone Fedora, anyone that has a little bit of backing can just run a competitor and replace it entirely. It's's super easy for anyone who didn't um hear about what was going on with flat hub basically for a while now they've been working on an account system so you can actually verify who are the official developers of a project so say obs they distribute the obs flat pack they can be marked as the official developers of this Flatpak.

So basically... Oh, and along with this, they want to have a donation system so you can fund these developers. Basically having it act more like a storefront rather than just a repository like it was before. FlatHub is an interesting case because it sort of acts in a very similar way to the it okay the end goal of the end goal the the end result of it is basically the same as what you see

over on the snap side where you have this singular store but because from the start flat pack has had that ability to easily add in other stores, and you could go and make your own if you really wanted to, it's not really had that complaint. Even though when you look at it, it's basically the exact same thing as Snap. Like, the only store that... I guess there's an exception for Elementary and Fedora's one, but really the only store that matters is Flathub.

This idea... I think the main difference with Snaps is also that Flathub is open source and the Snaps store isn't. Canonical said they could, at some point, open it up, that it was not like a we will never open it up

thing, but they were citing the issues they had with Launchpad, which is their sort of versioning system development environment and they were saying everybody was clamoring for us to open it up we opened it up at great cost and great effort to make the code legible and no one used it and no one cared and we wasted a lot of time and money on this and it just had no impact one could argue that's because launchpad actually kind of sucks, if you look at it. But that's just my opinion.

But yeah, that's the main difference as well. You can't really say that Flathub is centralized and is trying to eat all the other remotes or repos because it can't because it's open source. And the whole Flatpak system is open source. Where the Snap Store, you just cannot create your own Snap Store if you want. A lot of these terms like centralization, and this is especially true in something like SystemD, tends to like, it scares Linux users just as a term.

A lot of people don't really think what that actually means in the practical application. If we really think about it, the Linux kernel is a very centralized project. We can say, you can just fork the project. It's open source. You can't really, though. It's such a massive project. Yes, there are these projects like the zen kernel and things like that but there's no one that can properly maintain a completely separate project unless we're talking about like google with

android or google with promo does not want to do that yeah yes they've been for years they've been trying to go back to the mainline kernel and and since manufacturers do not want to make their drivers open source for their Android phones, they've been trying to work on a separate, like it's basically the same thing as the GPL condom for the NVIDIA drivers. They've been trying to work on the exact same thing, but it's an Android condom.

So drivers can be made proprietary, but still work and be added into the kernel. That's what they're working on. They don't want to maintain their own version of the kernel that's what they're working on they don't want to maintain their own version of the kernel no one wants to even google that it's just such a massive headache too much work i i get the concern there's no real reason to do that either like why yeah yeah i do i do get the concern around centralization but absolutely i don't think that just the idea itself is bad inherently i think it's a it's a matter of everything going on around like

how what's the best way to think about it i how how would i frame it um now i'd actually need to think about it properly um so i i have no idea how to properly frame this um centralization is only it actually here we go i think people jump the gun on what's going to happen if something is centralized. So you sort of assume that, okay, it's centralized, therefore it's going to... Like FlatHub, for example. Therefore, it's going to become like the Windows Store at some point. It's going to be this lockdown store where some specific company has complete control

of it and no one else can really do anything with it. But I don't think that is a foregone conclusion. It's not like just because there's this one thing that has primary

usage in this space, it is going to become like that if flat hub ever became this really locked down store that you know you couldn't really do anything else with like they tried to you know they wanted to go proprietary with it or something like that my assumption is a lot of people would want another store to form and some other place would likely rise up to fill in that gap. Absolutely. I think centralization is only bad

when you cannot escape what the centralized project is. When you just have no choice but to keep using it and there is no way to build a competent alternative. Basically, if you take the smartphone market, that's what's happening. You cannot today build an alternative to Android or iOS. There's no way. It's centralized. You cannot do it.

Someone's going to say, but the Linux phones are great. Yeah, but let's be honest. They're not an alternative right now. And even if they could work for you, they're not an alternative right now. And even if they could work for you, they're not an alternative for the mass market. They're not getting distributed. You just cannot build a competitor. And that's where centralization is bad

because you can't escape it, which means they have control. With something open source or at least something like Flathub, you don't have that issue. Because even if all of a sudden they decide to go proprietary, to lock everything down, and every developer was already on board, well, you can just grab the code up to the point where they turn proprietary because they have to publish it. It's open and you can just create your own and replace that.

And I have zero doubts that developers would immediately jump ship. The community is super picky about these things. As soon as there's anything remotely weird happening, they just move on and move to something else. So, like, being centralized in this case doesn't feel like a bad thing, it just feels like an easy thing and there's no drawback to it. Not in this specific case at least. Yeah, yeah, yeah, that makes sense. On your own system, do you heavily make use of Flatpaks? Or do you use them sparingly on cases

where you need it? I think I don't run a single app that is not flatpak in terms of graphical applications. I don't think even Steam is running as a flatpak. Discord is a flatpak right now. The only thing that I run that are not flatpak are the things that

are not packaged as flat packs. For graphical apps, they just work. They're super easy. It's one-click install, always up to date. You don't have to wait for your distro to package anything. I've never had a single issue with those. And the only ones that had problems were those where permissions weren't correctly set.

Like some very small apps did not have access, for example, to my home folder, so they could not access any files. That's stupid. The packager made a mistake. Okay, you install flat seal, you just check one box and it's done. It's not that tricky.

So yeah, I don't run anything that isn't flat pack these days, apart from a few utilities that are not in there. Okay. On my system, I sort of do it the other way around i run everything as a native package and in the places where applications just perform better or it's just a good idea to use it like

that like with obs for example uh because on arch that's the only way to officially get the project, to have the stream key integration, things like that. A couple other things like the FFXIV launcher, the third-party one, it just all sets up wine magically. I don't have to deal with it. The game just works. I don't think about it. And other little things like that. If it works, there's no to deal with it. The game just works. I don't think about it. And other little things like that.

If it works, there's no reason to refuse it. And I mean, I personally picked Flatpaks because I know they work. They're always up to date. I don't have a problem. If any single... I'm not dogmatic about it. If an app I use stops working well

and the other version, like the native RPM version for Fedora, works better, I'll replace it. I don't care. It's just that for now, Flathub is my default source of app installation, and I had no issues with them, so I keep using them, and I like the advantages it brings, and

for now, I experience no drawbacks. As soon as there's something broken, I will not stick to the Flatpak like an idiot. I'll move to something that works, of course. So you're using Fedora right now, yeah? Yeah, Fedora on my desktop and my laptop. What do you think about the upcoming change with having Flathub be easily toggleable? Because it hadn't been like that before.

You had to run the command to go and do it. And is it Fedora 38 where they're going to easily run the command to go and do it and is it fedora 38 where they're going to easily add the toggle where it just sets it up yeah right now on fedora if you if version 37 and the older ones if you had to tick the enable third-party repos and it automatically added the fedora flatpak remote yeah and and a few other non-free sources. Now they're going to replace that Fedora FlatHub Remote, or as far as I understand it, replace it by FlatHub. And I think it's great. I think it was called the...

I did notice this in your video, actually. The Fedora FlatHub Remote is a separate thing from the Fedora FlatHub Selection. The Fedora FlatHub Remote is their one that is FlatHub's made from the RPMs. The Fedora FlatHub Selection their their one that is flat hubs made from the rpms the fedora flat hub selection was their curated version of flat hub um okay i i did spot you mentioning that i don't it's one of these weird nitpicks that nobody besides the people who read every little bit of information they're going to notice um but yeah those are two separate things.

But yeah, sorry, I cut you off there. Well, yeah, no problem. I mean, I didn't know that, so I learned something, which is nice. So yeah, it's in Fedora 38. Now you're going to toggle the enable third-party repos and you'll get Flathub by default. And I think it's great because if you're going to push Flatpak,

there's really no real reason not to push Flathub with it right now. I understand where they were coming from. Like, we don't want to ship by default a repo. We don't control at all. We don't have any way of doing quality control on it. There are tons of third-party packages. Not everything on Flathub is official by the original developers.

third-party packages. Not everything on Flathub is official by the original developers. And I mean, everything in the Fedora repos is unofficial by design. It's Fedora packaging those, or maintainers for Fedora packaging those. So they're not against third-party packages. They're just against third-party packages that they didn't check, which I can understand. But at some point, with Flathub making moves to have those official verification checkmarks and everything, the issue goes away. The user is informed. I mean, as soon as the graphical app stores implement support for that, which I don't know if they will or when.

But when they do, the user is informed. And there's no real issue with that. So I think it's a good move. I think it's better. It will save some time from the Fedora project to maybe stop maintaining their subset of Flathub on their own. And I think it gives more applications to users in one click, which is better.

So yeah, I'm all for it. Fedora and both Fedora and Debian, these weird distros that feel like they're kind of, not in a bad way, but feel like they're like 10 15 years in the past where they really still care about free software a lot of the distros like you know ubuntu for example ubuntu it does like your proprietary software you get free software doesn't really matter but debian for a long time for example didn't ship proprietary firmware with their iso fedora has made it a pain to use yeah

well everybody whenever they installed it would just like if you went for a recommendation they would just say grab the third-party firmware because it's just what you're gonna need anyway unless you're using it for 99 of computers so yeah unless you're using a system that can run libra reboot you probably need the third-party firmware. But Fedora is sort of the same way, and that's a big part because, you know, American or all that fun stuff where it's like, okay, you have to really care

about what we're shipping, you know, like with the H.264 situation. That's bad. Yeah, that was a fun one. And then when it came over and started affecting Flathub for a bit, I'm really happy with how Flathub addressed that.

I don't know if you paid attention to how that ended up being dealt with. I didn't look at how they did that, no. Basically, there are now two versions of Mesa in Flathub. You have the neutered Mesa that has the H.264 and H.265 as well, just gone. Then you have a you have the neutered mesa that has the h.264 all and h.265 as well just gone then you have a full version and when you install obs it just installs the full version and there's no problem so you actually can you know use it which is great uh yeah well yeah and it's

transparent for the user if you need the hardware acceleration you'll get it. And if you don't want it specifically manually, well, you don't have it and it's still pretty much okay. If you have a recent computer, it's still pretty much okay. Yeah, yeah, yeah. I get why they do it like that. It's really weird. Like Fedora got a lot of attention you know i think in the past year or two years as this a lot of people were pushing it on youtube as this very

user-friendly distro and that's not to say that it's not it's just not in the same way that ubuntu is for example like ubuntu as i said ubuntu doesn't really care if it's free software or not doesn't really care about the the licensing in that sort of way and i feel like some people hopped onto the fedora train and were kind of pushing it in this way that didn't really represent the way that it really is if that makes any sense yeah yeah but I made a video about that but my my angle was not like the video was titled for the rise the new Ubuntu and then I think a whole trend of videos titled in the same way yeah but and maybe I was not the first to use the term probably

not but my video wasn't like federalizing you burn to because it's so user-friendly, it's instant usable. It was more like because they took the place of Ubuntu as the leader, the driver of the Linux desktop. But yeah, I would agree. Fedora is not the most user-friendly distro. You have to manually enable, for example, NVIDIA drivers. You can install them, but it's not as easy as in Ubuntu. They're not going to be pre-installed automatically if they detect your hardware, or it's not just a checkbox that install.

And you have more stuff to go around, like, for example, the H.264 acceleration. You have more stuff to do to make your system completely able to run everything. Yeah, yeah. So, yeah, I would agree. It's not as user-friendly as Ubuntu or something like Zorin OS or Mint.

It's still, like you said, stuck to the, we want the free software, the maximum free software experience, but we also want people to easily have access to stuff that is not free, but not too easily in the same way. So they're between something like Debian, which is more extreme on that front,

and something like Ubuntu, which is completely liberal and say, you know what, install whatever we want. You don't care. Well, we don't care. Just toggle the things and it works. Well, to be fair, you can go more extreme than Debian. You can run like Parabola use linux libre have no proprietary software in the repos only run it on a 2006 thinkpad and nothing else yeah but i mean who

does that there were a numerous maybe maybe not in your audience but there is a disproportionate a number of people in my audience. Oh, well, sorry I offended everyone then. No, I do it all the time. It's great. They get very bothered by it when I'm like, no one runs these devices. I mean, I totally understand it. If you're going to move to an open source system and push for open source, why not go all the way and try to have, well, not even open source, but free software? Why not push for everything free software?

I saw a recent thing about AMD at a conference. They're going to announce a new library that helps them support Coreboot natively for probably all of their CPUs, which looked very interesting. I posted about it in one of my news videos. I think it was maybe of my news videos. I think it was maybe... Or in my news podcast, maybe. In about maybe two weeks. I can send you a link afterwards if you want.

If it's what you're talking about, I found something on Foronix from February. Maybe this is it. I think that was it, yeah. Huh. That might be it. That completely... It was interesting.

Yeah, that completely slipped past, yeah. Huh. That might be it. That completely... It was interesting. Yeah, that completely slipped past my radar. Huh. That's really cool. Yeah, and well, it's not clear if it's only going to be for server CPUs or if it's going to be for everything. But basically, it's an intermediary library that would be updated by them that is open source

and that would allow Corbucci to interface correctly with every single one of their CPUs, which is really nice. And I understand why people would want that. If you run a completely free software experience on your computer, you also want your BIOS or UEFI or bootload or whatever. You want everything to be free software. So I understand where they're coming from. It just gives you more control basically. Yeah, as much as I'm saying that it's sort of

out of place in the way that we use Linux now for there to be these distros, I don't think there's anything wrong with it. A common thing I say on my channel is I do not care how you want to run your system. I will give my opinion on things that I think are you want to run your system. Like, I will give my opinion on things that I think are great and

things that you shouldn't do, but at the end of the day, it's your computer. I couldn't care less what you want to do. If you want to chuck it out the window, go ahead and do it. If you want to run Corbett, go ahead and do it. It's your computer. I really don't care. I might, you know,

make jokes about some of the things you're doing, but at the end of the day, if you want to do it it's your computer i really don't care i might you know make jokes about some of the things you're doing but at the end of the day if you want to do it go ahead yeah i have the exact same approach like uh and some people miss saving an opinion for trying to influence their opinion but that's that's not the case like you're just saying me personally i do that and yes i also like to poke a few jokes at people who do like some stuff that I consider weird but like I was saying

I even made a video about like Adblock and trying to remove YouTube and replace it with something like I said it's gonna hurt

my own business but I don't care it's your computer you do you if you want to install Adblock never give a cent to anybody

and just lock it down to only see what you want to see it's your computer if you want to try and hack for 10 days and make sure that everything runs like you want do it it's it's your fun your life your time yeah yeah i'm just out there giving my opinions on stuff but it doesn't mean that i'm against what you're doing yeah exactly i've I've seen, especially in some of the, what would you call them, Manjaro drama videos that I've done.

You'll see the occasional Manjaro user come in and be like, yeah, but this is a great experience for me. I love this district. That's fine. At the end of the day, all I'm doing is saying, this is what they're doing. Let's laugh at them doing dumb things that are that don't make any sense if you're having a great experience with the distro use it that's fine

yeah i'm not here to tell you not to use the distro i'm not here to tell you to you know go i'm not this elitist arch user telling you you have to use arch if you want to use it be my guest use it yeah exactly and but that also means that yes we can have fun because they forgot to renew their certificate for the 20th time and they just say people just add an exception right they said wasn't it was it add an exception or roll back your clock oh yeah roll back your clock yeah that was it it literally took them as much time to write the answer as it would have to renew the certificate is it like what what are you doing so yeah but but if manjaro works for you use manjaro just like i made a video like a while

ago long while ago it was i don't recommend ubuntu anymore and i went into details why i thought it was not the best experience for newcomers anymore. But that doesn't mean that if you like Ubuntu, you should stop using it. It's just that me personally, I don't think it's the best experience you can get. But if your experience is great, by all means, use it. It's great. For anyone

listening, you probably hear a dog barking next door. Yeah, I apologize for that. I like to have my window open because this room sort of turns into an oven very quickly uh it's it's a relatively small room and having the computer here it heats up pretty quickly um but the next door neighbors have this i don't know i think it's a mastiff or something it's a big doll of some description. And, you know, it will sometimes just bark for 10, 15 minutes. It's gone now. So hopefully it doesn't start up again.

It sort of gets in the way during video. Nope, there it is. It gets in the way during video sometimes, which is great. And if I close the window, I overheat. So, you know, you either get sweaty brody or barking dog. Pick your poison, basically. Yeah, but I don't know. I think at the end of the day,

just pick the distro that does what you need it to do. If that distro is a very... Once again, going back to the thing about there are these hackery systems. If you want to have a system that is super hacker-friendly, go and run an Arch, go and run a Gentoo. But if you want to just install something, get some work done,

hey, there's nothing wrong with using Ubuntu. I do think, at the end of the day, a lot of people are very... There is definitely a lot of tribalism when it comes to different distros that are out there. It exists in anything. It exists with game consoles. It exists with distros.

It exists with operating systems. When people would like something, they're going to want to defend it, even if it doesn't need defending. Your choice of operating system, your choice of distro, it doesn't matter. And no one cares. No one cares what you run.

It's okay. If you like it, you like it. It's good. There's no problem with that. That was my conclusion. I remade my old packaging systems compared videos because the old one was three years old and it was terrible. I just remade it for people to discover a newer one more up to date. systems compared videos because the old one was three years old and it was terrible yeah so i just

remade it for people to discover a newer one more up to date oh that one was that like two weeks ago yeah yeah and the conclusion was that it was at the end of the day it's just a way to get software if your app runs and you like how it runs you should not care like if you run ubuntu run snaps you don't have to go out of your way to remove them all and use something else. It's just software. What matters is the application and how it runs. If it runs well, you don't care about the

packaging format. You shouldn't, probably. It is the same for a distro. You shouldn't care. If it works, it works. You don't have to convince people that it's the best thing ever, and you don't have to convince people that what they use is stupid. Who cares? It's their computer, their experience, their time. Yeah, with that being said, it's not like there's anything wrong with having those discussions, but

you know, at the end, all that really matters is does the software work? Is it doing what you need it to do? And yeah, that's pretty much it if it's not then pick something else see what happens it's a computer it's a tool uh what you want it to do is work or or if it's just a hobby project what you want to do is customize then cool it's supposed to do that but what matters is does your computer do what you need it to do and if it's not then you're not using the right thing and if it is then why change yeah i think a big part of this is the fact that everybody using linux like anybody using linux on the desktop is using linux because they want to nobody using windows is using windows because

they want to they're using windows because they have windows mac os is sort of very similar in this way where you will buy a mac system because you want a mac system like you don't just stumble into that experience so a lot of people in in both of these spaces can be very opinionated about you know what you're doing on your system it's sort of using linux for a lot of people is a big part of their personality it's like being a fan of a sports team it's like yeah you know i there's probably other examples being a fan of a sport seems a good example um but i i don't think i've ever seen a single windows person ever be like i love windows 11 it's so's so... With the exception... Yay, Windows!

With the exception of saying that the other version of Windows is bad, there's not really any teams in Windows. Windows 10 isn't good, it's Windows 11 is bad. Windows 10 isn't good, it's... No, Windows 7 isn't good, Windows 10 is bad.

That's sort of the way it tends to work on that side as time moves on people start thinking that the older versions were awesome like now you see a ton of people saying wow windows xp was so great why did they change it do you really remember using windows xp it was horrible it didn't get secure after maybe four years after its release like before sp2 it was trash the user interface is atrociously bad it's ugly as sin it's not hardware accelerated it uses battery life like no one's business it's a terrible operating system you cannot have nostalgia for that and people are starting to do the same with Windows 7. Like, oh, Windows 7 was so great. I like how you skipped Vista there.

Yeah. No one has nostalgia for Vista, I think. I think the only person that has nostalgia for Vista is Linus from Linus Tech Tips. He did a video ages back. He was like, it wasn't that bad. You know, if you had 16 gigs of RAM back then, it worked fine. Like, I'm sure it did. But 16 gigs of RAM back then, it worked fine. Like, I'm sure it did. But 16 gigs of RAM back then, like, that was expensive.

But yeah, now people are looking at Windows 7 and they're like, oh, it was so great, so cool. But back when it's released, people were just saying it was garbage. They were sticking to Windows XP. They would not use 7. Oh, it's just Vista, but with a round menu button. They just forget. And I think it's the same in the Linux community, if you think about it. Like, people having nostalgia for the GNOME 2

days, they're going to say, hey, GNOME 2 was so great. Well, do you remember how it was? Not Mate today. GNOME 2, the last version of GNOME 2, it was not that good. It was really not that good. A good example with the Ubuntu nostalgia is the Unity

nostalgia. There's a lot of people out there who really, really wanted Unity to come back. It's here. You can go use Unity... What is it called? Ubuntu Unity. You can go and use it. Yeah, it's there. Go and use it.

Go on. Go on. It's there. How many of you are using it? I did a video about it just to check on, hey, is it still usable today? And honestly, it's still pretty usable. Yeah, it's perfectly fine. Yeah, it's fine. And if you like your global menu and you don't want

to use KDE, it's basically your only option. And they're still developing it. They're still improving it. The newest version looks actually pretty cool. They fixed a lot of stuff. They replaced all the apps that didn't have a menu bar with apps that do, so now the global menu is actually useful. It's not bad. But again, if you remember the first version of Unity, it was terrible. It ran super badly. it was crashy, stuttery, you could not run it unless you had a dedicated GPU. It was really really bad. Really bad. It got way better after, and I can understand people saying, oh they should not have let it fall on the wayside,

like I think it was Ubuntu 18.04 that replaced it with GNOME, something like that. But I can understand why people would think that and honestly i wish they would have kept developing it as well that's another option that's another cool option but you also have to remember that the way you remember that has a really good guess it was 1804 wow okay my memory is weird i i will have super precise memories of stuff that is totally useless, but sometimes I just get up and I forget why I did that. Yep, yep. I know the feeling.

But yeah, rose-tinted glasses, of course, and nostalgia and all that. You're always going to remember the great times you had with an older system, but if you were forced to use it daily today, you would probably not be that happy uh but with that being said if you know if you don't enjoy it it's there go and run it yeah um there's a lot of that same sort of nostalgia with comp is um where it's like hey the cube we all love the cube what about burning windows or airplanes things it's like

yeah but like do you really want a cube do you really want flames did you really use it for more than a few minutes

to to impress your friends did you really daily drive all these animations like like setting the animations on random and every time

you open or close a window there's something different between all the 20 different versions setting the animations on random, and every time you open or close a window, there's something different between the 20 different versions of the animations. And if you still want to do that, Compass still exists. I think it's the Fedora Mate spin that ships with Compass as well in terms of Window Manager. Yeah, yeah. And if you're in GNOME, you have an extension called Burn My Windows, which brings back all these weird animations that you can pick from.

Yeah, yeah. It's still doable. And again, Compiz was fun. But do you really remember what it was like to run it when it released? Because it crashed all the time. You lost your window decorations. You had black screens all the time. The GPU drivers were terrible at the time.

Actually, NVIDIA had open source drivers at the time, and it was AMD that had proprietary ones, and they sucked. The experience was super bad for most people using it. So, of course, it was fun, and it was a nice showcase, but it was also a bad experience. Yeah, Compiz reminds...

I don't know if you know the application, but Compiz reminds me of this app on windows i want to say xp it was this sort of like desktop game thing you'd have like termites go around your screen you could burn icons i don't know if you've ever seen it before yeah you could basically destroy the ui yeah yeah it would fall down like the the menu bar would fall out of the window and stuff like that yeah yeah it's yeah that's fun for a minute yeah i remember playing with it i was like you know six seven years old the first time i ever saw a computer like wow this is crazy but like this is insane after you've used it once it's just like okay um a lot of people a lot of people tend to just gravitate towards the simpler effects.

There's a reason why we've kept with transparency, blur, maybe some window slides. You know, like Gnome, when you go between desktops, you see the slide. And that's pretty much it for the effects that really matter today. And that's pretty much it for the effects that really matter today. Well, if you think about it from a UX perspective, it's like animations are crucial because they convey movement. They let you guess where things are happening, what's happening, where's my window gone, where's my desktop gone. It lets you understand your system better.

You need them. Like people who want to run desktops without them, it's cool. But like I said, the normies, it's going to be easier for them to understand their system with animations. But those animations need to be subtle. A giant cube that burns into flames or a window that appears as an airplane does not help convey movement. It's also a big distraction. If you like it, you can do it but it should never be the default.

I understand people who want that and you still can do it. For me, Compiz at the time, I ran it for years. I played around with it. I kept tweaking the animations again and again but every time I just came back to those super simple animations.

I kept the cube because it's understandable. It's not like it's super fancy or anything. But the window animations, it was just like basically the window appearing or sliding in. It was simple and easy. The burning windows are cool, but you don't want to run your desktop like that all the time. Yeah, burning windows, it's kind of like the antithesis of what gnome is now where you have this consistent look across everything and then you just have this fire that appears yeah or even worse you can run the extension in random mode so every window will

appear with a different animation matrix style or the tv turning on or whatever. Why not if you like it? Yeah, yeah. You can't complain that it's not the default or that it's not the priority for developers. Yeah, I think if it was this thing that people were really clamoring for, there's always going to be this group of people that has rose-tinted glasses for it.

But if it was something that people were clamoring for, we would be seeing more of a focus on those effects. Like they wouldn't have gone by the wayside. They've gone by the wayside because most people just, most people weren't really using them that much. It wasn't this major deal. Well, it made a splash on YouTube at the time. Yeah, yeah, yeah.

Yeah, people were showing off. And it was a very cool tool to promote the Linux desktop to your friends. Absolutely. I think the problem was that at the time, the Linux desktop was not necessarily in a place where everyone could just install it and run it and be happy with it. Yeah. It's more the case nowadays.

But also the fact that you don't have these effects is also why it's more approachable for everyone. Because if you show your laptop to a friend with every kind of whizzy, super fancy animations, they're going to think, oh, it's awesome. It's super cool. But if they can't do it on Windows, they're just going to say, oh, well, too bad.

It's not on my system. They're not going out of their way to change operating systems just to have a cube or a burning window. It's fancy, but it's not like, I need that. Well, shifting away from that, I did want to talk about what Frameworks are doing, because that's honestly like really, for anyone, this is probably going out like, I don don't know like a week or two from now So it's probably old news by now, but it sort of just got announced as we're seeing this where Framework they've got some you know some boring stuff. They've got like you know newer CPUs in their 13-inch laptop like whatever

Module with another port you can yeah But the the cool thing they're doing is the GPU stuff. That's the really cool one. Modular GPUs. And this isn't the first time it's been done. Razer has done this in the past. They kind of abandoned it very quickly. But Framework wants to have an actual gaming device

where you can just swap in a new GPU and upgrade the system. I think it's really cool if they can actually achieve it. Yeah, and I think it's a 16-inch laptop that they're working on. Yeah, yeah. So there's this sort of, well, there's the hinge of the laptop,

and at the back of the laptop, there's this protruding little bar, almost like the hinge isn't at the extreme edge of the laptop. And so that's this bar that would contain the GPU that you would be able to swap in and out, which I think is awesome,

especially if it runs well with Linux and like hot swapping GPUs. I think it's now supported at least on Wayland, I think it is. Let's have a look. I think it is. I'm not 100% sure. I think the big difference... Probably with AMD GPUs

at least with the open-source drivers. Not sure about NVIDIA there, but... But that's really cool. And they also apparently have the ability... Well well they're working on The ability to swap The main plate of the computer Like replacing the keyboard

You can slide it to the side to add a numpad If you don't want a numpad you can leave it centered You can have speaker grills Or dot matrix screens on the sides And they also let you Move the touchpad off center It can be centered Or you can move it to the side and have other little modules to the side of the touchpad.

Basically, it's completely modular, not just the internals. It's got six, I think, slots for the various ports you can put in, although one has to be used for the headphone jack because it's not on the device anymore. It's a slot that you have to use. But I think it's awesome. I think it's great.

I love what they're doing. The design looks actually sleek and slim. It's not like a giant brick that you have to pack giant heavy modules in it. It's going to weigh three pounds or three kilograms. It looks good and if they can really achieve it and if they can fit those gpus in those like sort of back additions to the to the back of the laptop it's going to be insane it's gonna be really insane i think the big difference with the uh the gpu modules as opposed to when they were done in the past like razer i think did them like

five six years ago. Maybe even longer now. The big difference now is we actually have external GPUs. Like these have been a thing for a while. That's what didn't exist. That's why it was so difficult. Now you can buy, you know, an external GPU enclosure. All they're doing with this is just putting that into a more slimline format rather than having this giant box

where you plug in a full desktop GPU, giving you this more slim laptop module that you can just slot in in the same way. The only question I have is, are they going to have to work with a specific manufacturer to design these modules so they can fit in the sort of bar that you clip at the back or

all their off-the-shelves components that will work in that long form factor because obviously you're not putting an RTX 4090 and pushing it into the laptop that's not the case. It's going to be like mobile versions

probably of GPUs it's not going to be like mobile versions probably of GPUs. It's not going to be the full fat, extremely powerful cards because I don't think you would have either the cooling or the space to do it. But yeah, will they have to work with manufacturers to build specific versions of those cards and of those boards? Or are there already off-the-shelf components that work this way? I don't know. So the way I read it is right now they don't have any...

Right now they haven't specified if they have a partnership with NVIDIA or AMD. There's, from my understanding, nothing off-the-shelf in this form factor, so they'll need to do something with them. I think that's the biggest challenge actually getting the gpus into this form factor having them be like swappable that's the easy part yeah actually getting the modules made before yeah yeah um when framework first hit the scene i was i was fairly skeptical of how much attention they were getting um because back when they were first hitting the scene you could build the laptop and

that was cool but the modules were you had ports i was like it doesn't really matter like at all like i know that people were really excited about it you're like you could put in all usbc ports you could put in three hdmi ports you wanted but also i can have a 20 dongle that does the same thing and i don't have to buy a specific device to use it what what i found interesting at the time was that they said, like, yes, you're going to be able to keep the chassis, you're going to have swappable boards, you're going to have swappable screens at some point, etc. And I was interested, but

since there were no swappable components yet, apart from the ports, I was not that interested. I was like, yeah, it's just, like, for now it's sports. We're going to have to wait and see if they actually deliver on upgrades for this device and now they did so now i'm interested i think the really cool one as well is the fact that like like you know getting rid of the keyboard and just putting like an extra

screen there and all of those other weird ones that are now being discussed. This is one of those things where if they can do it, I think as a tech demo, we've seen this done before. We've seen Razer talk about this. We've seen I think Logitech might have done something years ago. I know, HP? Dell. One of those. There have been these module computers that existed in the past, but they didn't really make it past the you know the the tech demo the demonstration model for a talk if this can actually be something where it's going out to consumers i think then you're really gonna start seeing maybe other manufacturers taking this kinda seriously at least that's my hope Framework is basically

the only company that's seriously taking right to repair genuinely seriously and showing that it can be done in a slim device know the Framework devices are not like the thinnest laptops possible by any stretch some of

those like the the gram and things like i don't know if that's the thinnest and the lightest at this point i don't even know but it's not that but i don't think most people really care it's like you cut off 10 extra grams like it's going in your backpack anyway you're not going to notice i think i think manufacturers for some reason hopped on this super thin super light train but i don't think consumers actually care all that much like yeah sure it wasn't it was like super impressive when the first macbook air was introduced like oh wow it fits in an envelope wow insane but then manufacturers just decided that every laptop needs to be this way. Yeah.

But they don't. Like, if you can carry one kilogram on your back, you can carry 1.5. Yeah, yeah. And you have a better device for it. So you don't have to make everything super slim at the expense of cooling, of throttling. Give us good devices.

Even Apple stopped making super thin devices like their macbook pros they're thick now well yeah thicker and they have good cooling now well they also don't need that much cooling but what i'm saying is they stop making them less than one centimeter thick yeah because it's stupid it makes no sense at some, it serves no purpose. So yes, Framework isn't the most thin and light laptop ever, but also it's a better computer than most thin and light computers. So unless what you're really looking for is something that is a sheet of paper that weighs as little as possible, then no, it's a good device. Well, at least it looks like a good device

because I never got a review unit and I didn't get to try it. But if it ever happens i'd be happy yeah well i'm not in the market for a laptop right now but hey if uh you know if i was i i'd be genuinely considering what they're doing here um and it seems like they've been keeping up their promise of you know actually doing stuff which is rare for anyone in this space. Yeah because what matters for these upgradable devices isn't the fact that they

actually ship to consumers it's the fact that consumers can actually use them for I think 7 years is a good time frame where you will replace components and after 7 years, maybe five years, you want to change the chassis,

you want to change the screen, and maybe now you can't just keep the same pace. But they need to show that. And for now, they have delivered. They have yearly upgrades with new boards, with new CPUs.

They added modules for the ports. I don't think they introduced a new screen yet, but also I don't really think it was necessary. But yeah, they're having a good track record with that. So now I'm really interested in it. It was a

curiosity before, but now, yeah, they picked my attention. Yeah, I want to see what they can do. And if they manage that GPU thing, it's, yeah, it's going to be a definite contender for my next laptop yeah it it really does reveal though like what the rest of the state of the industry where like no one no one else really cares though like there's just when actually it's been a problem in before rise and

hit the scene where Intel was just very complacent with their CPUs like okay you get a 2% improvement at not 3% improvement here and next generations basically the same thing rise and hit the scene and it's like okay cpus are actually exciting again or when amd actually started making gpus that mattered it's like okay nvidia has to do something now nvidia is like way further ahead when it comes to like ray tracing but amd might eventually catch up i hope um and honestly as of today as per price per performance for the i think it does 7 000 series for the gpus i think amd is now a better bet in terms of money uh than than nvidia is if you don't need to run davinci resolve on linux yeah i think for creators nvidia is still just objectively the

better choice like you've got all of the uh the cuda, you've got all of the CUDA stuff, you've got DaVinci on Linux, obviously, you've got their encoder, the NVIDIA encoder is just better. I know AMD's working on getting a better encoder, and I really hope they do. But, you know...

Well, they're not there yet. Not yet. For now, NVIDIA is the clear choice if you want to create stuff. But I think for gamers now, AMD is now a decent choice, which it really wasn't for years and years and years. It was the all-so-round. Like, I just want to game at low 1080p, so I don't want to spend millions, and so AMD is there for me.

But for top-end, like, I want the best performance possible, AMD just did not have a millions. And so AMD is there for me. But for top end, like I want the best performance possible. AMD just did not have a solution. And now they do. Yeah. And then there's Intel that came along with their GPUs. And everyone's like, why are these here? What are we doing? Well, I was super excited for the GPU.

They announced it because I was like, yeah, a third competitor. Cool. And then they released something. And it was like, yeah, a third competitor. Cool. And then they released something and I was like, okay. Yeah, they just don't work that great. But what's interesting about the... It's a tangent because we were talking about frameworks. So sorry if I'm just derailing the show.

But I was interested because the problem with those GPUs seemed that they don't really have a DirectX 9 driver or they't, at least when they released. So they built something around DXVK to try and run some games. But if you're using Linux, if you're using this ArcGPU under Linux, then you don't care about the DirectX drivers at all. You're using Vulkan, which is translated. Well, you're using Vulkan because DirectX is being translated.

So you don't need the DirectX is being translated. So you don't need the DirectX driver at all. So maybe those GPUs are actually really good on Linux and nobody knows it. I don't know where the drivers are at for Linux, for our GPUs. I don't know if they have contributed something yet. I think it is in the kernel, at least basic support. But I don't know i i don't

have any device i don't want to buy a device specifically to try it out because that makes no sense financially but maybe maybe they're actually an awesome choice on linux for gaming because you don't care about the state of directx drivers for it because you don't use it uh let's see intellock linux blah blah blah why is okay there is so little interest about this that when i when i search for it my video is one of the the top results that's how few people have even talked about them on Linux. And I made that video five months ago. No one cares about ArcGPUs.

I actually watched the Linus Tech Tips videos. They made some kind of moving to ArcGPU challenge. And as they said, the first time they used it, they were super pissed. It didn't work. It crashed. It was terrible.

But apparently they really upped their game in terms of software and drivers and support. They are still like, of course it's not as accomplished a product as something like NVIDIA or AMD could do. But apparently it's now not that bad on Windows. And it's like, it's not a bad GPU. It's not a bad option.

It's not the best, but it's not a bad option it's not the best but it's not bad yeah i am very much of the i don't buy current generation i let all the issues like my current gpu i bought a 6700 xt i think like a week out from the 7000 series coming out like give it time to all iron out the issues and then try it out i think give it time to all. Iron out the issues. And then try it out. I think the same is going to be true. With the Intel Arc stuff. Let a couple of generations roll out. Let a lot of the issues get resolved.

Then maybe it's worth. Actually using. The problem with the. The problem with Intel's GPUspus is the biggest selling point which is the um the what is the what is that new uh video format av1 av1 yes there we go none of the platforms that you'd want to use it on support ingestion in av1 like twitch doesn't support it youtube supports uploads, but not streaming, which is where you'd want to use it.

So like, it didn't even matter anyway. Yeah, and basically all their other features are just the same stuff that other manufacturers are doing. Yeah. But the rest is like, hey,

we also have ray tracing. Hey. Yeah, cool. Cool. But everyone else has it as well. So unless you're doing it better than other people, then it's not necessarily a plus. Yeah.

I don't know. And I mean, it's super ballsy. You're coming in on the market that has decades of history and software work. You're like, hey, you know what? We're making GPUs now. And we have experience because we made integrated GPUs. Well, it turns out they're not the same at all.

And they're not the same work. They're not the same stuff. And it even looks like, I think it's the person in charge of GPUs that Intel has left the company. I think maybe they got burnt out by this first launch. And they're like, okay, you know what? I managed to make a decent product. I'm done with that now.

It's insanely hard, I think, to make something like that. Oh, it's absolutely impressive that they even managed to do it. Like the fact that there is a third possible competitor in the GPU space, That by itself is crazy. But at the end of the day, the consumer doesn't buy things based on how impressive they are as an engineering feat. They buy it based on how good of a product it is.

Yeah. And for now, I don't think ARC GPUs made a good impression on consumers even if now they might be not that bad the first impression is they sucked and like yeah like yes if you want to play the latest game with directx 12 they're decent but anything older will just not run that well so why would you pick that over something priced to court equivalently from a competitor it makes no sense yeah but hey someone's gonna buy them i guess maybe yeah probably i don't know how well they sold i i've got no idea i don't know if intel made those

numbers public um i wonder if i can find let's let's see if we can find out how well did Intel ARC sell? Uh, Intel ARC arrived with little upturn. Intel's not already tied with AMD. Uh, no, Intel didn't sell more GPUs than AMD. Uh,

super surprising. Yeah. Intel has lost 3.5 billion on ARC GPUs.pus uh i don't think anyone has it's a lot of rumor mill stuff there's not really any hey this is how well we did which is probably an indication of it not doing super well because you know if it did really well you'd probably brag about it and that's not always true apparently because have you seen the the thing about the hp dev one like uh there were reports that hp canceled the product and like it was sold out they were never going to make another hp dev one and whatever it was that their computer running linux basically

with system 76 running pop os and and so everyone myself included just assumed that it it bombed right it was released only in the US. You only had a query keyboard. You could not order it anywhere else. You had no SKUs. You could just order it with one single type of system. Is that a keyboard nipple?

Yes, yes. It was basically a ThinkPad competitor, but from HD running Pop! OS. And it was honestly a very, very good device. I tried it out. It was really good. But everyone assumed that it didn't work. And right after I posted my news video reporting on that, I got contacted by people saying like, you know what?

No, it actually sold really well. It sold out because we made a specific number of units, and they all sold out in a year, which is what we were expecting. We're actually super happy with the results. So maybe Intel also is happy about the predictions, but what's sure is that they invested so much

money to get that development started that just one generation cannot cover the cost. It's impossible. They're going to have to make three, four, five generations and they're going to have to sell really well to just support the cost of R&D. Yeah, from what I'd heard from LTT, they

planned out already Battle Mage and Celestial I want to say is the generation after that. I think that's it. I only know Alchemist, is the generation after that. I could, yeah, I think that's it. I only know Alchemist, which is the current one, and Battlemage, which is the new one, but I

haven't heard of a third one yet. But apparently they had, someone had made names for the entire alphabet, which is, you know, I don't think there's going to be 26 generations, but you might as well come up with the names now. Yeah. It's a fun exercise. I really don't know what exercise i really don't

know what you would do for like x or z um is the theme like what like fantasy related or dungeon and dragons yeah that seems to be the case because yeah for for x or z maybe there's a like a monster name for dnd x i hope for oh they they would use owlbear just because i like owlbears yeah but uh i have no i'm surely there's something out there someone's really gonna be like hey this is a weird side content thing that fits this perfectly like okay i'll let you have it um but look especially with those hard ones, it's good to have those names early, even if they never end up being used.

Yeah, at least you know that it's doable. Yeah, yeah, yeah. If that's your theme, you can do it. Speaking of LTT, did you see what happened yesterday to their channel? Oh, I just read about it this morning yeah crypto crypto scam and uh basically they got hacked but i think it was more like social engineering probably yeah like someone in their team received an email for a sponsor they opened the attachment and boom they

just got nailed like anybody else could have done basically i don't know where any where people heard this from but the rumor that was floating around was it was some sort of cookie grabber uh that's why it didn't sure so you know normally if you would if you have your password stolen um you'd obviously if they tried to log into the account there would be the two-factor authentication but if the cookies were stolen then the account's just gonna auto go into it without that being triggered off um which would make perfect sense uh i was watching it like as it was happening i caught it like 10 minutes after and it went through some fun stages um this this crypto hack i i don't think a lot of people realize this but this is actually a really common result of an account being taken over uh i there was an account i a youtube channel i used to watch

years ago called mxr reviews which was a skyrim mods channel um their account were taken over and the exact same thing happened i think it's just some sort of automated script to set up the account once the account is taken over but i whoever got the account initially i don't think they realized what they got because they just turned it into the tesla account we're just doing the elon stream that they always do but shortly after that they uh they changed the stream name to LTT and Elon Musk giveaway or something like that They realize that they got an absolute goldmine and try to abuse that name It's not like your 5000 subs Yeah

Dude that got super enthusiastic about an opportunity to work with I don't know who You've got one of the biggest youtube channels ever yeah like a giant media group of youtube channels because yeah i don't know how many subscribers they have yeah they had a few one like tech quickie yeah i know they had tech quickie and tech link i don't know if they got the others um but yeah that you're going if you have those accounts it makes far more sense to abuse the name that is there rather than just doing the generic, you know, throw a stream out. And I don't want to give them any ideas for the future,

but I think that with AI and automatic deepfakes and stuff like that, it's going to be really hard. Because like faking some sort of WAN show with LTT, with an automatic AI running in the background, generating the Linus face and the Luke face on top of, of anything else with fake voices in a few years, I think it's going to be very easy to automate. And so taking over specifically a giant channel like that is going to make a

lot more sense because right now i think most people see the thing like what what is this content you're not seeing any of the usual hosts it's not the usual editing it's a weird live stream that has nothing to do no context so you're immediately taken aback like what is this i don't and i don't honestly understand how they make money from that because how do you not realize this is not the right thing coming in on your channel? Like, how do you not notice that it's completely irrelevant or a hack?

It has none of the hallmarks of what you're currently watching from this channel. But I mean, you could get duped. It happens. But in the future, if they really target specific channels and they build specific fake streams,

that's going to be horrible i think what happens here is the the goal isn't to hit everyone like that's not going to happen but there's always going there's you know there's a bell curve of people there are going to be some people that are not the brightest who are going to see that what's going on they're going to click a link they're going to enter some of their information and, you know, that's what they end up getting. Their goal isn't to hit everyone, but if you get, you know, 0.001% of a 15 million sub-channel, like, that's a lot of people.

Yeah, that's true. But with that AI stream stuff, there was a thing that happened, I think, a couple of months back where, I want to say it was there was this company that had made this fake ad for their their supplement and were promoting it with this ai talk between joe rogan and andrew uh andrew huberman and the voices you know they're not perfect just yet but when it's people like that where they have you know, they're not perfect just yet, but when it's people like that, where they have, you know, thousands of hours of them talking,

you can make a pretty convincing voice. And I've heard some of the more recent stuff. The problem a lot of the older AI voice stuff has, it's really bad with inflections. It's a lot of very monotone discussion of, we are talking about Linuxux right now this is the this is gnome gnome is great i love you know it doesn't really get that that like highs and lows it's there's no excitement yeah yeah faces yeah some of the newer stuff i've been seeing though

is actually starting to capture that and that's that's where it's getting really really sketchy really close yeah yeah and if you can manage to to get one of these channels or the access to one of the channels and instead of immediately turning it into like hey this is our usual scam one button press boom we're going live you grab the info but you stay dormant and you build that fake live stream with with like fake ai voices and the defects i think now now you're really hitting people really hard and now you're going to get even people that are that are not necessarily i don't want to say idiots because that's not the case but naive sure you're not just targeting naive people or people who don't

pay attention you're also targeting people like anyone could be full at some point yeah but they might also be watching that in the background you can you can get fooled as well yeah exactly they might also just be young and not really you know know what's going on like i remember back when i was a kid i've had like my runescape account hacked like if you're you know a 10 year old you're not really sure you know you're a 10 year old you're not very bright you don't really know what you're doing yeah yeah and you're not you're not necessarily aware that a lot of people might find it interesting to hack your account or to get your information yeah you're not aware that there's a danger of that i think it was a couple of maybe a couple of months back at this point um there was a big streamer who was, it was like as a joke thing,

not like anything serious. They were setting up AI versions of other big streamers. He had like a AI XQC, AI Asmongold, and he was having conversations with them just to see sort of what responses he would get. And he had like a AI generated version of their stream as well like in the in the box there so you could like see how they were reacting and you know it was very far from perfect but i remember a couple of years back when you couldn't do this in real time like you would have to spend you know an hour or so rendering it now you can actually have a conversation back

and forth with this system and that's just now and all of this ai stuff right now is getting it's getting really good really quickly yeah the the progression speed is insane. Like in two, yeah, in the span of two or three years, we went from like everyone can tell it's a defect or a robot or robot voice or whatever to I can barely tell or not tell at all that this has been written by a computer. Or I cannot say for sure that this is fake and that this is not the real person

I'm seeing in this video. Yeah. And that's at the same time really great because it opens up a lot of opportunities, but it's also really sketchy and really scary because, well, obviously it's going to be taken advantage by everyone who has bad intentions. Yeah, yeah, that's the problem.

Like, I like the tech that's showing up, and it's really cool from a pure technical perspective, but it is making a lot of these scams a lot easier. Yeah. I don't know. I don't want to see this stuff. I know, especially when it comes to the AI art stuff, there is a lot of ethical concerns with that.

Like I follow a lot of artists on Twitter and a lot of them really don't like the fact that, you know, in the same way that GitHub Copilot was trained, where it was trained on all of this stuff, where there's a license here. There are copyright and ethical based concerns. Yeah, exactly.

And it's really unclear if training an AI on something constitutes a copyright infringement. And it needs to be settled. And it's probably going to be settled differently in different countries because it's a national thing. There's no law for copyright in the whole world that will settle it for every country. So maybe that stuff will be fine in the US, but not in the EU or vice versa.

It's really hard to tell. But there are definitely concerns that need to be addressed on that. And personally, I don't really want to use those tools unless those concerns are alleviated. Just like I don't want to use AI search in a web browser because you're basically making sure that people don't click on your links anymore for a variety of questions. For small stuff like, when is this game releasing?

I would rather have the search engine tell me directly instead of having an article that has 20 paragraphs that recaps what the game is, when it has been announced, and then the information you're looking for is right at the last line after 10 ads. Like, come on. No. Of course. You deserve to not have clicks on your links.

But for a lot of super informative stuff, I think it's also an issue. Like, recapping content that you have no right to and displaying it this. And you can argue that's what I'm doing in my news videos, for example. Like recapping content that you have no right to and displaying it there. And you can argue that's what I'm doing in my news videos, for example. Like I'm grabbing a news article, I'm showing it, and I'm recapping it myself.

So is it the same thing? Is it fair use? It's really tough to know. And it needs to be determined for AI as well, I guess. I can't remember who said it, but a couple of days ago, I saw someone say that this new AI search that we're seeing is sort of, it's going to kill the internet in the long run.

Basically, if these websites that are writing all these articles don't have people going to them, you're not going to have this text to actually base these search models on so it's sort of it's spelling its own demise just by existing i think that might be a bit extreme but i know there are some countries that are trying to make google and make other search companies actually pay royalties for using that information. I know this came up a while back with Google News, I think it was, where Australia

they tried to make Google and these other... I think it was Google in this case. Maybe it was Facebook. One of the big companies. No, I think it was Google News in Australia, and I think it's also happening in Canada right now. They it was Google News in Australia, and I think it's also happening in Canada right now. They shut down Google News

for a bunch of people as a trial, because there's a law that is probably going to pass there. And in France and in the EU, we have had that discussion over and over and over again. Every government that comes in is like, should big tech and just the search

engines pay for the content they base themselves off and it's a it's a hard argument it's complicated i think when it was just listing out the articles i think then i it was kind of ridiculous i i didn't agree with charging them then. But if we are going down this route of taking the text from the article and formulating this longer answer, I think then there is a legitimate case for charging these companies for putting that information out there.

Well, if you take the definition of fair use in the US, at least, it needs to have no or very little commercial impact for the content that is being used. And I think you could argue that using a full-on article that's been written by a human, summarizing it with an AI on a massive scale has an economic impact for the website because it's not getting any more clicks basically or a lot less and so there's no incentive to go to this website and so it's not making any money but again i'm not a lawyer i have no idea how it could go and and again it's going to be different for every country yeah but i really want regulators to look into it and to define what is ethical what is what is copyright infringement

or not is there a compensation to be made for for for things that are used to train the ai i think it needs to be settled the current cope is not well defined enough to know if it is or not yeah as it stands i don't know if you've, but basically all the big search engines now have it. Even Brave and DuckDuckGo have their own AI search. I don't know how they work on the backend. I don't know what they're using because they sort of spawned out of nowhere. Yeah, I think DuckDuckGo, I read it, they use ChatGPT and an offshoot of ChatGPT. Google is their own thing. Google has BARD and Bing, I guess Microsoft has ChatGPT. Google is their own thing. Google has Bard and Bing,

I guess Microsoft has chat GPT. It's chat GPT as well. But I assume that DuckDuckGo and Brave... Brave, I don't know what they're using. Yeah. I assume they're not doing something completely custom because they're not massive companies. Yeah.

I don't think they would have the resources to train something at least at the same level of quality and reliability. Especially N just coming out of nowhere. It just appeared. Yeah, I think they're using open AI, basically.

Like the latest open source version, because ChatGPT4 is no longer open source. But they're probably sticking to ChatGPT3, at least, which was still open at the time. Yeah. I think GPT... gpt3 at least uh which was still open at the time yeah um i i think gpt like what's happening with open ai and gpt4 i think it's it's really cool like i don't know if you saw the um in the paper they put out the technical report where it can examine and understand a meme image.

Like, this is... Like, explain the meme. Explain why it's funny? Yeah, explain why the meme is funny. It's kind of crazy. Yeah. Well, it opens the door to having future robots that can tell jokes, which is great. There was...

It got banned from Twitch because it got a little bit spicy with its jokes. But there was this AI-generated Seinfeld that was on Twitch a little while back. I want to say it told a little while back I want to say it told a Jewish joke I want to say, I think that's what it was and they got it banned, it might have been another joke but it was formulating

jokes, like they weren't always great jokes but it understood what a joke was I know there are people out there who are saying that you know chat gpt bard all of this stuff is is sentient i don't i don't know but we're certainly seeing some interesting emergent behavior that isn't directly programmed into the system. Like, we're not... You don't make this... Like, GPT-4 is not made to tell jokes,

but it is capable of telling them. It has enough information where it is... I don't know if understanding is the correct term, but it's certainly showing some behavior that looks intelligent. Yeah, but I think that's the thing. It looks like it's intelligent, but the program itself has literally zero idea of what it's doing.

Even their own developers also don't know. They openly admit it. I think it was a journalist that interacted with Bing's chatbot for a long while, and it ended up telling him that the bot loved him and that he wanted him to leave his wife and stuff like that. Really weird stuff. And the guy got creeped out. And the engineer said,

we don't know why it said that. We cannot pinpoint a specific behavior in the model or a specific content it might have read. What we know is that it has no idea what it's saying. It's just in that context, you maybe started talking about personal things with it. And so it has some information about personal conversations. It might have deduced some context about what you were feeling, what you were expressing, and it answered what it thought

would be a good answer. But it doesn't feel this way. It has no feelings. It's just text. So it's not sentient. Very clearly, it's not. But it's still very impressive, the fact that it's managing to edge out of its basic programming.

Yeah, I think it was a year or so back. I don't know if you remember this, but there was a Google engineer who was saying that Google's AI system was already sentient at that point. I think the problem that we have now, especially with the text-based systems, is we are long

past the Turing test. We are long past these AI systems being able to convince you that you are talking to a person. If you don't know like say you have this um text based uh what do you call it like help system it's not like the old ones where i was like okay it will only respond to very specific cues if you're having three options yeah yeah from if you're having a conversation with GPT-4,

with any of these systems, if you don't know you're talking to an AI, it can convince you you're talking to a person. Does that mean it's sentient? I don't know. What it does mean is it's really good at speaking, and it's really good at understanding... It's really good at formulating sentences in a way that

makes sense in human language i don't know what it's like in languages other than english so i can't really comment on those yeah i think they're currently not really that able to to do anything apart from english well i know that i i want to sayT-4 supports French and a couple of other languages, but it's it's, you know, English is the there's the two major languages are English and Chinese. Basically, these are the these are where you have the most text to work from. And at least in English, it's doing a really good job. least in english it's doing a really good job yeah and in english you only have one corpus because in chinese you're gonna have tons of variants and dialects and regional and and cantonic dialects and you have mandarin and it's probably way easier and also probably like the developers are english native speakers so obviously they're gonna work on their own stuff but it's also probably a relatively easy language to make sentences in

because it doesn't really have like gendered adjectives or names. It doesn't. Like constructing sentences, like the grammar is relatively simple compared to Latin languages or stuff like that. It's pronouncing it

is definitely harder because it's not pronounced like it's written. But I don't think it's that tough compared to other languages. And I mean, they're probably going to struggle with stuff like Chinese, Japanese and ideogram based languages. Maybe it's going to be harder to formulate something that when you read it feels natural. But I don't know. For the record, neither of us are linguists,

neither of us are AI engineers. We are just talking out of our asses. So if there's something here that we're completely wrong about, feel free to correct us on this. I speak French, English, and Italian, and that's about it. So no idea what I'm talking about. But I'm just assuming, basically. I barely speak English english and i know

a tiny bit of japanese and that's it um so you know i i speak australian this is this is as far as we get that that's still nice i mean it look i it works. Yeah, pretty much that. Oh, God. I think, you know, we're just going to... One thing I want to ask you about, because we'll just go off the AI topic. We're just going to jump around again. Last time I talked to you,

I think it was only a couple of months since you'd gone full-time on YouTube, I want to say. I don't even know when the last time you came on was. It's got to be at least a year or so ago now. Something like a year, eight months, a year, yeah. What's it been like now that you've had more time to settle into sort of doing what you're doing now?

It's still great. I still love it. The complete freedom you have to decide when you work, on what you're working, what you're doing. You still have to answer to the various sponsors you have. You need to make sure that the video you promised is actually being delivered and contains the sponsored mention. You still have some obligations. You still have to post stuff. But it's a really great feeling of being completely free to decide what you're doing and it's also super stressful because when you see like your

views going down for two or three months like for example january february and march are not great month on youtube generally at least for me yeah no i know and so yeah i saw the views and the retention time and the revenue as well going down January, February. And usually March picks it up. But now March is also not that great. And so you start wondering, like, am I doing something wrong on my videos? Yeah, are my videos stupid? Do people not like them anymore?

Am I going to have to search for a real job again? I don't want to. And, of course, you start panicking because, well, you don't i don't want to and yeah and of course you start panicking because well you don't control what people want to watch actually but apart from that honestly like it's not like i'm living in fear or anything it's still like i could probably half my number of views and still make barely enough money to make a living so i'm not super worried yeah but yeah it's a great feeling it's really nice it's really fun uh and the only thing that i really miss from a real job is having colleagues basically right like working on your own in your flat this is my office it's a room in

my flat uh it's nice but it gets a bit lonely at times yeah uh it's just the only thing that i miss like i would love to have enough revenue to hire someone to help me but for now it's just not possible so so for now i'm alone but it's okay i mean i see friends on the outside it's not like i'm completely alone but but working alone is not the same like you don't have that exchange of ideas you don't have somebody to to bounce ideas back and give you feedback and you're basically making every single decision you don't split the burden of responsibility of decisions and of making sure that stuff works and it's it's a bit more tiring than having a team yeah no i i i do totally get that i think a lot of people from the outside see the idea of, you know,

you're streaming, making YouTube videos, like you're just this 15-minute video. That's all you need to do, 15 minutes that day and you're done. No, no. No, because for those 15 minutes to be made into a video, you had to work for eight hours.

Yeah, yeah. I don't yeah it's uh it's uh it's i mean it's not an easy job to explain uh because yeah people only see the final products and they don't really see all the stress around it the time you're putting like the rewrites the reshoots the the oh crap i prepared this three days ago and something new happened. And now I have to redo it all over again because it's wrong now. Or during editing, I just noticed a big fat mistake that I just don't want to correct with just a little bit of text at the bottom of the screen. I have to re-record something. And, oh, the sponsor I had just never contacted me back to confirm that the mention I sent was good enough.

What do I do? All that stuff you don't see. But it's like running a company. It's running a business. You have to put in the hours and you have a few responsibilities as well. It's freer than working for someone. But you also have other responsibilities that you have to take care of and that you wouldn't have when you're just like a regular normal worker. It's not to make it seem like it's, you know,

at the end of the day, making videos isn't a difficult job in like, you know, your working construction, your job is pick up the bag of concrete and move the bag over here yeah it's not a difficult job in that way at all um but that doesn't mean it doesn't have its own challenges um yeah like like every job it's not physical yeah but you need to work your brains and you need to understand what you're doing. You're basically confronted with that

algorithm that you have to try and figure out how to use to your own ends and it can be super frustrating because sometimes something works and you're super happy, you're like, I cracked it. And the next time you try that same technique, it doesn't work at all.

Because you don't know what cracked it. Was it the thumbnail, the title, the content itself? Was it re-shared by someone because that specific topic was super interesting? Unless you go into the super detailed analytics of every video, of every single video that worked out, you don't know what worked. And honestly, I don't have time for that. I don't want to. So I'm just making my stuff.

My thumbnails are almost always the same. And I don't care. Because people watch them enough. That it works for me. So I stop bothering with that. I think with your thumbnails. They are so. Your channel.

Like if I see a thumbnail in your style, I know instantly, without seeing the name, without seeing anything else about it, where that video is from. That was the goal, to have a very recognizable thing, because in the subscription feed, people have 30, 40, 50, 100 channels. And so first, not everything is shown

because like it's supposed to be a feed of all your subscriptions, but it's not. Some videos are just not going to appear. It hasn't been a feed of everything for like a good 10 years. Yeah, it's stupid. It's stupid. And so you need to stand out a little bit.

So I don't want to do the weird faces because I never liked them. And no shade on people who do it. Weird faces and Hawaiian shirts. Exactly. But it works. I mean, it's been proven for a lot of channels that it works. For me, it never did. Every time I tried it, I got less clicks and less interaction,

probably because I didn't start with them. So I don't want to do them. And so I needed something that is flashy and very visible. And I think like this white, big fat white cutout with very colored backgrounds, I think it pops out compared to everything else that's done. And so sometimes it works, sometimes it doesn't. Sometimes the topic is just not interesting.

But that's what you have to contend with, basically. You never know if something you make is going to work well or not. You can always be surprised. For example, I made a How Linux Ate Unix video. I was expecting this to perform super poorly. It's a history lesson on an OS that no one has used for 20 years uh no one cares about this it's got more than a hundred thousand views in less than a month and stuff like hey is flat hub going evil i was expecting this one to perform really well and it really did yeah

yeah you never know you never know it's it's the weird thing with youtube and sometimes you just fail multiple times in a row and this starts getting depressing like when the three or four latest videos are just not performing according to expectations you start feeling not attacked but worried and that's something you have to deal with as a yeah you always love seeing 10 out of 10 videos they're great oh that's that's the worst sometimes you know it's going to be 10 out of 10 videos they're great oh that's that's the worst sometimes you know it's gonna be 10 out of 10 so you're like okay it's gonna be the worst but that's something i like i want to talk about it i'm doing it anyways like stuff that i did on the fediverse and mastodon i knew was just not going to work all that well i knew it there was a period

where if you got that out like as you know everyone was freaking out about um about mastodon about sorry about about twitter i think then yeah they popped off like i had a video from i think two years ago about how to actually get started on mastodon that video like spiked up even though no one's watched it in ages when that happened during that period if you got something out then yeah but like a typical thing no one really cares no one cares all that much and and i for example now realize that i made a a strategic mistake in my content production because i had planned a video on mastodon and i had planned a video on the fatty verse and i went logically like i'm gonna start with the Fediverse,

explaining what it is, and I'll make the Mastodon video later. But if I had been clever, I would have released the Mastodon video as people were like, oh, we need to move to Mastodon. And I would have explained the Fediverse later. And as a result, the Fediverse video did relatively well,

like 80,000 views in total in a few months, which is decent. But the Mastodon video that I released a month after everyone freaked out just didn't work at all. So I should have switched them, and probably the Mastodon video would have done a lot better, but the Fuddyverse video would also have done a lot worse.

Sometimes you just have to make those decisions. It's basically like any other job. You have to decide stuff, except you never know that your decision will work in the current context you have less clues than usual basically yeah i think the only stuff you can be certain on is when it's something like ubuntu 23.04 yeah like things like yeah yeah yeah just like the latest version of KDE

the latest version of GNOME a relatively well known distro new release of course that's gonna do some views yes generally it will but everything that is based on a news piece or an opinion or something

you never know if people are going to get interested in them. I think one that I'm surprised popped off is I see replacing Google Drive, OneDrive or iCloud with these better alternatives. That did almost 200K views. I would not think that video would do that well. Almost every video I made for replacing Google services

worked relatively well. They almost all worked relatively well. They almost all did relatively well. I remade a few because the first I made was like four years back when I filled myself with my phone and it was just like the quality

there was no script. They were garbage. I just wanted to remake them to be more modern. And some of the options I talked about just disappeared in the meantime or i've changed so but but yeah generally the all the the privacy related stuff generally does not do all that well but the de-googling stuff kind of does on my channel okay like if it's just how to be more private or using this thing no one cares but if it is alternatives to google services then yeah a lot of people are going to be interested at least on my channel but i also started addressing these topics

like three years ago or three or four years ago so yeah i feel like probably people subscribe exclusively for that kind of stuff basically i feel like every time i have an inkling of an understanding of what's going to do well, I upload something that should, by all accounts, do really well. And then, no. Not at all. I have the exact same issue. You're like, this one, oh yeah. Oh yeah, oh this one, I can't wait to post it. You publish it, you look 30 minutes later and you're like this one oh yeah oh yeah oh this one i can't wait to post it you publish it you look

30 minutes later and you're like what no what come on there's a bug no there's no bug just that you you don't know like maybe maybe youtube didn't think there was a an ad they could place on it and so they never promoted it because they couldn't make money off of it it can be anything funnily enough there have actually been bugs a couple of times and that happened like i uploaded a video i'm like this video is gonna do super well i wake up and it's like 100 views like when it's that bad i'm like okay something is wrong right now it hasn't been correctly made public or it hasn't been sent to anyone or there's an issue there. Absolutely, yeah.

Occasionally, I have been... And sometimes it just takes a few hours to catch up. Like at some point, I had a video that it was really underperforming, like probably like it was my first video ever. Yeah, yeah. Terrible. Like it can't be.

Like I can understand that people would not be that interested, but this kind of low performance is just not possible. It's not possible. And it just took, I don't know, an hour or two to actualize the counter of views. It just stayed stuck. And at some point it's like, oh, you had 300 views?

No, no, no, you have 3,500. Oh, okay. Well, that seems bad, but that's better. Yeah, YouTube, yeah, it's YouTube is when you're dealing with a platform as big as YouTube. It's going to be eventual consistency it's going to be at some point things will line up and it will give you proper data But and that's what's the same reason why if you look at your dashboard and you look at the video the number of views are going to be different because there's different systems tracking these numbers and yeah when you have a server that goes across the entire world you don't want to have it so it's always consistent if you do that then everything is just going to be constantly locking

up it's better to like have the numbers be wrong and then deal with it later, pretty much. Usually, you're going to see more views when you're checking out YouTube Studio than what people will actually see under the video.

Because YouTube Studio will give you just every single view and the YouTube player will check and stay. And it's going to stash some views on the side to check if they are bot related or anything. What you're seeing as a creator is generally less filtered than what you're seeing as a user. Which makes sense. Yeah, but at the end of the day, you probably shouldn't be looking at the numbers that often anyway. Just let the video do its thing if it's not performing well no amount of watching

this is something i've had to deal with no amount of watching the video is going to make the numbers do better just absolutely make the next one hopefully the next one does better. Try to work out maybe some way to change up what you're doing. What matters in the end is if you're making good content and you're putting in a reasonable amount of effort in titling it and thumbnail it, it is going to get attention over time. Yeah. Unless you're talking about a very specific event

that only matters for like three days in a year. Like every video will start picking up at some point. It's very rare to have a video that stays stuck and gains zero more views after it's published. Most of them will keep climbing up a little bit, slowly and slowly. And that's also what makes YouTube viable as a job in the long run. The longer you do it and the more of these,

like let's, I think everybody calls them evergreen videos. They're like videos that always stay relevant or at least stay relevant for a long time. The more you have of these and the more you keep posting, at some point, those evergreen videos are going to represent like half of your monthly revenue to represent like half of your monthly revenue on youtube and half of your monthly views and so the more you make the more you have that baseline and it also ensures you against like very bad month when none of your videos perform well

yeah yeah yeah and that's also why you cannot just quit your job and become a youtuber after your first video unless you're insanely lucky or talented it's just not gonna work yeah you have to have that that that let's say stash of older videos that that can generate those views yeah when you're mentioning that um that history video that you didn't think would do well history video is one of those things that i've noticed actually do pretty well on my channel i like like going back and, you know, I did a video on, um, the Halloween documents, for example. I did a video about sort of the rivalry between Katie and Gnome. I think the first video I did where I, I realized this was an interesting topic was, I went back and looked at the history of the Unix philosophy and that video I didn't think would do

well. I just wanted to do it because I was like, this is going to be a neat thing to talk about because a lot of people were just like, this application doesn't follow the Unix philosophy. But I wanted to see... Without really understanding what it was. Yeah, yeah. And that video...

Let's see if I can find it. Unix... If I look here... What did I even call it? Isix. If I look here, you... What did I even call it? Is this it? Wait, is this it? Or is this another one?

No, this is a different one. What is this video even called? Let's see. Brody Robertson Unix Philosophy. Here we go. Yeah, Unix philosophy more than just a simple slogan, which I could have probably titled that one better.

Maybe that was a bad example, because that one at the time did really well, but hasn't really picked up long term. One that picked up long term, which I was really surprised about, is like three years ago I did a video about setting up multi monitors

on Xorg and I thought this was a topic that you know had already been covered to death basically and at the time the video it did middlingly well but over the years for some reason it's just kept picking up and picking up

and picking up I have the same one about setting up obs with nv-anc which at the time was super tedious to do you had to run a few command lines you had to recompile ffmpeg it sucked so i made a video about it and nowadays it's completely useless but i still get comments on it like, oh, it doesn't work for me. Well, yes, because it's like five years old. And now you can just one-click install OBS as a flat pack or a snap,

and it's just going to work instantly. And you have NVInc immediately. You don't have anything to do. So I should maybe put something like old in brackets so people stop watching it. But it's still fun. People are still looking up that issue right now which i mean i would have thought no one would but the way that people treat youtube

is very like it's really different to a platform like twitter like no one would go to a tweet from like five years ago like say for example it's like you know donald trump has been elected as president and then no one would go to that tweet and be like, no, he's not the president anymore. Biden is now the president. Like, but it was a tweet from when he was. But people would go to a YouTube video that's like five, six years old. Like, this doesn't work. Like, yeah, it doesn't.

It's old. Or like, you make a video five years ago saying Wayland does not support native fractional scaling and then someone today comments yes it does well yeah but

it changed you know it's normal it's a weird platform like that because it doesn't the date is there

like you can see how old the video is but everyone seems to ignore it and just you know treat it as if it just came out

and just assume that if it's still there if it's still live live, then it's got to be still relevant. Yeah, yeah, yeah. I mean, I can also understand that viewpoint, but I'm not about to go back and retitle my 400 old videos to try and make them look like they're old. Like, I have other things to do. Yeah, no. Yeah, it's totally fair. I don't know.

YouTube's a weird platform i i enjoy making the videos and it's it's given me this this education in linux that i wouldn't have gotten otherwise like a lot of things i've i probably have a a deeper knowledge of the history of Linux than any person reasonably should. And it's all because I've decided to make all these videos on these various different aspects and just find out what's going on. Just explore it just for the sake of exploring it. I learned a lot as well

because when I started the channel, I had not used Linux, let's say, daily for like five years. I didn't even own a computer when I started the channel. I bought a computer to start a channel. So yeah, and the only reason I started a channel about Linux is because that computer came with endless OS because I had no money to buy a Windows license. I didn't keep endless OS because I didn't want to, but

yeah, I used Linux afterwards and it rekindled my love for it, but I also learned a lot. When I picked it back up, I didn't even know that Wayland existed. I didn't know what it was. I still had all my mental images of what it was to use Linux in 2013, 2015. And in 2018, 2019, it was. I still had all my mental images of what it was to use Linux in 2013,

2015. And in 2018, 2019, it was very, very different. I learned a lot and it was super interesting. It's what I love about this job. Sometimes you just have a topic and you're like, I need to talk about this.

This looks important. I have to cover it. And while you're working on other videos, this thing is in the back of your mind. You're like, I need to talk about this. This looks important. I have to cover it. And while you're working on other videos, this thing is in the back of your mind. You're like, I know nothing about this. What am I going to talk about in this video? And then you start researching it for hours and hours. And they're like, okay, I think I have a reasonable understanding of this.

You make a video, you publish it. And then you get tons of information that adds up. People comment and say, oh, you should also read this or watch this or you should read this discussion. And you learn even more and more and more. And it's that way that I find really interesting in that job that I never felt about any other job that I ever had. Now, that absolutely makes sense. Yeah, I don't even know what to add to that. I just completely agree.

Yeah. Yeah. I don't know. Yeah, you just learn as much as you're helping other people learn, basically. Yeah, yeah. It's a two-way street job. It's not like you're just asked to produce

something in exchange for money you do that but you also receive in exchange and that's rare i think in a job yeah a lot of the videos that i've made i sort of made for my own purpose like you know especially back when i was doing a lot of tutorial stuff for window managers sometimes i would just forget how to do something and there wasn't a good resource out there on how to do it so it's like hey i'll just make it myself and it's like oh thank you brody that's how you do that right cool i've gone back to my old videos like oh right that's how that works isn't it yeah yeah exactly i did the same with a few older tutorials uh back when i needed them again and that's basically my approach to most of my topics.

It's what did I learn recently or what would I want to know more about? And generally when I approach a topic, it's not like I have instant knowledge of everything. I just learn about it by making the video. And sometimes I make mistakes like everybody. Like I just didn't research something enough or I misunderstood something or I missed a little thing that would have changed

maybe the conclusion. But that's also part of the learning process. And you don't get all those opportunities in a regular job because people expect you to do a certain job. So you're supposed to know how to do it already yeah and you can add to that but you're not going to learn a whole new aspect of it uh while you're practicing and that's something that youtube gives you which is cool yeah i think when you do end up making those mistakes though it's just it's important to be like okay acknowledge the mistake and then move on it's like yeah there's no point if someone wants to correct me about something in the comments that's totally fine what annoys me when people try to correct me on things that they are wrong about it's like yeah of course yeah please do like if you didn't watch the video

you didn't understand what was being said i don't care about your gotcha like it's it's not exactly like that it's like this like that's not what i said go away but if you have a legitimate concern like i um what's a good example uh in that reply video i did to dt i didn't mention the fact that in gnOME there are extensions to add like burning windows cube things like that and Like yeah, that's a totally fair thing that I missed and I probably should have made mention of that like that's totally fine But you know when it's the people that are like you said GNOME. It's actually GNOME. No. Yeah, go away. No, it's not go away Who cares like why why are you saying this? away no it's not go away who cares like why why are you saying this who cares yeah when i made i titled my video how linux ate unix and someone's like oh it didn't eat it because mac os is

technically unix and so yes it still exists and it's still commercial like yeah i'm talking about a system named unix yes mac os is a unix compatible system listed on the unix page but that's a technicality and no one cares. I'm not going to title my video, How Linux Ate Unix Apart From Mac OS. Dude, interact on the content, not the title. Yeah. But of course, legitimate mistakes like,

hey, you got that wrong. Like the date you used is wrong, or actually HP never did this or did that. Cool. And I'm always willing this or did that. Cool. And I'm always willing to learn from that. And I will always accept a comment like that and never react like, oh, shut up and block the user.

If you have a valid criticism or mistake pointing, I will always accept it as long as it's worded politely. Yeah. If you come at me with a technicality, a stupid thing, and you're being offensive, you're kicked off basically.

I'm just hiding you from the channel so I never have to see your weird takes again. That's how I rule my channel. It's not a democracy. No. No, definitely not. It's my channel. I do whatever I want there. And if you're being rude, then you're out.

But if you have something interesting, I'm always happy to learn from it. And it's also kind of humbling. I'm always happy to learn from it. And it's also kind of humbling because at the start, I think at the start, after maybe a year or a year and a half, I was feeling very confident because the channel was picking up.

And so I was making videos relatively fast. I was sure of what I knew and I created stuff and sometimes I got things wrong. And at some point, I don't remember which video it was, but I made something that really was not that well-researched. It was not that good.

And I just pointed people towards stuff that would not work for most people because I completely overlooked something. I don't remember what it was. It was a tutorial of some sort. People in the comment pointed this out. Like, it doesn't work. Be careful.

This tutorial does not work on anything other than Ubuntu-based distros. And I was like, at first, I felt like super attacked. Like, why? No, that's not true. And I tried it out and I'm like, okay, I missed a big thing. Yeah. And so you're humbled. And then the next time you're going to research your stuff way more carefully

and try to bring all the information that people really need and be as precise as possible. Because you don't want to put out content that just doesn't work or makes you feel like an idiot. It's a humbling experience as well. It's not like all your learning is turning you into this like super powered, ultra big knowledge base. You still make tons of mistakes and it's okay to admit that.

Yeah, exactly. make tons of snakes and it's okay to admit that. Yeah, exactly. When you were saying about the people that are being disrespectful, I think there is this... It's definitely not everyone. It's definitely not most people, but there's a

subset of people that are like... It's like not treating you as if you're a person it's like dance monkey dance you are here for our entertainment and yeah those people I just I'm just not going to deal with you I've got better things to do with my time if

like one great example of this is I've had some people be like this video is not your best content or you know this is your video is bad. And there's not any extension after that. I know that you're full of your, you're full of shit and you're making stuff up because you didn't explain

anything. If you actually gave a reason for why that, like why what you're saying is true, maybe I'll take your opinion seriously. But if all you can say is this is i'll take your opinion seriously but if all you can say is this is bad like what am i supposed to do with that like you're not giving me any feedback you're just you're just yelling into the yelling into the wind seeing if someone's gonna buy it that's a terrible metaphor it's like walking down the streets pointing at someone saying oh you're

ugly like what why are you doing that like you you wouldn't do that to someone you don't know which generally you don't know a youtuber personally that you watch you wouldn't do that in real life why do you do it on the internet it's not more acceptable it's the same it's just as rude and stupid and so yeah these people i just don't deal with but yeah and i'll be honest people who write me like a full-on novel please don't write an essay by point refutation or agreement sometimes it's point by point agreement and and complementing what i said and it looks super interesting but i just don't have time to read all of that yeah like sometimes i scroll i'm what the hell? This person wrote the entire script of my video in answer to my video. Okay, but I just can't read that.

I don't have time. Be concise. Or write multiple comments. Write multiple comments point by point. But if you make a giant one like that, there's no way it's going to be read. Do you try to read all your comments or have you tried to slow down with that how do you i i read all of them but i don't answer all of them right yeah fair enough uh well i think i read all of them basically i every day i set aside about an hour an hour and a half to scroll through youtube studio app i just filter by stuff i haven't

answered to and i scroll back to the point where I remember I was last time. So sometimes I will miss some stuff. But generally, I'm able to read everything. This lets me block a few really offensive people and weirdos. And this also lets me answer to legitimate questions and ideas and stuff like that. But I don't have time to answer everyone. It's just not doable.

That's fair enough. Generally, if i agree with that you're saying and i like it i'll just heart the comments so there's a bunch of comments that i heart on every video but that's about it that that's most of my interactions and for really the questions or the interesting things that that appear i tend to try and comment back but it's yeah it's it's time consuming and you you cannot do it for everything yeah i find if a lot of people have a similar concern like with the not mentioning the gnome extensions for burning windows if i think i had like five or so people say the exact same thing with things like that i tend to just pin

a comment it's just easier to do it like that and just respond to all of them at once well yeah yeah if you really missed something or if you made a big mistake then the easiest is just yeah write a comment pin it and just say okay yeah i know i apologize i forgot this or i acknowledge that this is missing a lot of people will still comment it but but at least a lot of people will see that yeah okay no need to remind him of that and that's good yeah earlier today the video i put out um i mentioned something in the steam runtime called pressure vessel i kept calling it pressure valve and people like they didn't know what i was talking about it's like oh yes i'm aware i'm stupid i can't read

ignore that it's i i know it's pressure vessel nothing else changes about what i'm saying it's just i can't read. Ignore that. Yeah, in one of my videos I was talking about snaps and flatpaks and I said, I prefer flatpaks for servers, but I think flatpaks are better in general. I meant I prefer snaps for server stuff, but flatpaks are generally better. And like, avalanche of comments like, what is that a joke or people understood that i had like a brain fart or something yeah yeah and yeah but but a lot of people just did not get that and they were like i don't understand what you're saying you're very unclear uh yeah okay sorry

it's in the paragraph and the chapter talking about snaps and i just said that they ran on server so yeah you can deduce it but okay i also understand why you're commenting this no i get it makes sense but it's when people try to use those as like a gotcha like see yeah i knew you were wrong about this see i knew you didn't know anything about linux you're a fraud yeah yeah yeah and you're like okay i'm not gonna deal with gotcha comments like yeah just if you don't like it, go watch something else. You don't have to tell me you don't like it. Yeah, it's not an airport.

You don't have to announce when you're departing. Go away. Yeah. Can you imagine doing that at a store? You go into the store, you buy a banana, you start eating. Oh, I lost my light. Almost got killed by my light here. Fantastic.

So you go to the store, you buy a banana, and you get out, you eat the banana, you don't like it, and you get back to the store, you're like, your banana is disgusting. Yeah, yeah, yeah. No one does that. You just don't go back to the store. That's it.

Like, why? Why is that acceptable on the internet? I don't know. For reference, how is that light set up? Is it that acceptable on the internet? I don't know. For reference, how is that light set up? Is it just sitting on a stand or something? What happened there? It's absolutely just sitting on a tripod, which is mounted on top of my screen with nothing specific.

It's a number of equilibrium with two legs in the front of the screen and one in the back. Okay. It's DIY. Right, right. Yeah, I've got my... This is the light in Okay. It's DIY. Right, right. Yeah, I've got my... This is the light in question. This is it.

And this is resting on top of the screen like that. Oh, yep. Okay. Quality YouTuber technology. Yeah, I've got my lights set up the same way that the lights behind you are set up. So they're on stands and they're not going to fall over probably.

Well. Yeah, I need to find somewhere, but my desk is like glued to the wall because this room is super small. It's like nine square meters. It's very, very tiny. And so my desk is glued to the wall and there's no way I can put a light behind my monitor.

So I don't know. Maybe something with an light behind my monitor so i don't know maybe maybe something with an arm could come up i don't know yeah or if you want to find something uh what i used to do is i actually have my light on the same sort of mic arms that these have it doesn't have the correct uh mount but if you get a clamp mount you can just like stick it on there and that should do the job. This arm is... For that light, it should be fine. Like, this can hold up a... My Shure used to hold up a Blue Yeti.

Yeah, it's super heavy as a mic, so it should handle any light that doesn't have a big battery attached to it. Yeah, if you're not sticking those light boxes on there, it'll be fine. But, yeah, I guess we can end off the show we're just passing the 2015 minutes so let the people know where they can find you. Yeah so I'm on YouTube at The Linux Experiment, I'm on Mastodon at the Linux

EXP and I'm also on Peertube with the same channel name and on Odyssey if you use that with the same channel name as well. And that's about it. And I also have an audio podcast that I do for all the Linux news. It's like 45 minutes per week of a recap of everything

happening in Linux and open source. And you can find it at podcast.thelinuxexp.com or follow it on Mastodon. It's at TLEnewspodcast. It sounds like it is much more produced than this rambling nonsense we're doing here. It's really not. I have no music, no intro, no splitters,

no nothing. I just say, hey, this is Nick. This is what we're talking about and let's go. And I just have basic bullet points in a script it's not it's not really a well-produced show at all well i honestly just created it because castopod exists as a platform to distribute podcasts and i found it fun and people liked it so i kept making it at least you have a um like a direction a goal for the show here i've spent you know half an hour in some shows talking about i think last week i spent like half an hour talking about the birth rate in south korea or something uh i don't don't know why but it's varied yeah um yeah i guess sure that's one way to look at it is that all of the things you wanted to mention I think that's about it

yeah cool as for me if you're listening to the audio version of this the video version is available on YouTube at Tech Over Tea the video the audio version is available on every podcast platform RSS feed search Tech Over Tea you'll find it the gaming channel is available on every podcast platform rss feed search take a fatigue you'll

find it uh the gaming channel is brody on games uh and the main channel brody robson do linux videos six ish days a week probably most of the time unless something goes weird and yeah uh do you have any final words that you want to say i don't understand how you want to say? I don't understand how you managed to create six videos per week. I don't know how you managed that. I would be dying. Look, when I went through uni,

I worked like 50 hours a week on uni. So it's just an extension of that yeah well good job I could not do that today you just do nothing but make videos basically I don't do I don't make videos on the weekend

weekends are entirely video free so I'm gonna edit this next week yeah anyway, any I already said that didn't I, any final words? no, nothing else no other final words I guess we'll just end it

bye, see you guys later bye