Hey everyone, this is Nick and welcome to your Linux and open source news podcast. Now this is the first episode ever of that podcast, so brace for a little bit of an introduction. If you don't know, this thing is based on the Linux and open source news videos that I make on my YouTube channel, which is called the Linux experiment.
And this podcast will cover the same topics as in the video, but it will add some more that I found interesting, but I just could not fit in the length and the format of those Linux news videos. Now all the links to the articles I used and cited to make this podcast, they're all available in the show notes for attribution, but also if you want to read up more on each topic and each topic will have its own timestamp as well.
I'll try to make chapters work, apparently I need to create a JSON file, depending on your podcast client, you might see those chapters or not.
Now since this is the first episode, don't hesitate to send me some feedback on the episodes page on the podcast website at podcast.thelinuxexp.com I'll try to upload that podcast to various services like iTunes and Spotify, but you will also have a ready-made RSS feed right there on the podcast's page, so you can just copy that and use that in your own client as well.
The podcast isn't sponsored and probably never will be, but it still takes some time to make, so if you want to support this new venture of mine, don't hesitate to check out my Patreon page, it's in the show notes or at patreon.com slash the Linux experiment. Even the cheapest tier at $1 per month will make me more money than if you watched my entire video catalog on YouTube back-to-back with ads enabled, so if you like the podcast and you wanted to continue, please consider that.
So with that long intro out of the way, let's begin with the actual news. So to begin with, we have a website claiming that Arch is the most popular Linux distro. It's called Linuxiac as a website. It's one that I generally trust and respect and that I use as a source for some of those news videos and podcasts, and they've used Reddit, of all things, they've used Reddit to try and see what the most popular Linux distro was, and apparently it's Arch Linux. That's kind of surprising.
Basically what they've done is count the number of members in each subreddit for various Linux distributions, and they took that as an indication of popularity, which means that Arch Linux apparently is the most popular, with 226,000 members, followed by Ubuntu with 209,000, then Tails with 91,000, and then Fedora, Mint, Kali, Linux, Manjaro, Debian, Pop!
OS, and Red Hat, OpenSUSE, and Elementary OS that follow way, way far behind, and the poor Solus, Kubuntu, MX, Linux, or Slackware are just like at the bottom of the barrel.
Now, of course, this might seem surprising because Arch definitely is not the most used Linux distribution out there, it's clearly Ubuntu or Linux Mint, there is no question about that, but the methodology here is completely flawed, because Reddit is not a good indicator of popularity, except if you're trying to measure popularity on Reddit. Reddit is not used by all categories of users, whether they're technical or not.
I'm a technical user, technically inclined at least, I never used Reddit, I don't like it, I don't understand why people like it, I don't understand the structure, it's just not for me, and I'm pretty sure that a lot of regular average Joes also don't use Reddit even though they could use Linux.
So basically, someone being a member of a subreddit also doesn't mean that they actually like the distribution, you could be part of the Arch subreddit because you used it two years ago, or you could be part of it because you like watching what's happening, or you like trolling. It's not an indication of popularity.
So there's a huge selection bias with this article, which is also why Arch has most members, Reddit isn't used by average people, that might be more at ease on more simple distros than Arch, and since Arch is more complex, it probably also has more members in its subreddit because they have more issues that they need help to solve, and they turn to Reddit for that, while users of more simple distros use them quietly without needing to go on Reddit to talk about the problems that they have.
And also the fact that Arch users can't help but communicate the fact that they use Arch is probably a big indicator as well. So yeah, probably yes, Arch is a very popular distro, but it's definitely not the most popular or at least the most used. It's just very popular on Reddit, but there's an obvious bias in the selection of who contributed to the popularity of a distro used by the Linux Yak website. Now let's talk about X11.
And apparently its development is more than tapering off, it's basically dying out. In 2022, there were only 156 commits made to the whole code base, to the X server master branch on Git. That's half of what had been done in 2021. That only amounts to 3,600 lines of code added and 900 removed. And as a comparison, in 2021, there were 31,000 new lines of code and 180,000 lines had been removed.
So basically the level of contributions on X11, on X.org in 2022, is as low as it was in 2003 when they switched from their old development model. And all these changes were made by 32 people only, down from 48 last year. And that's basically the lowest number of contributors the X server has seen since 2003 as well. And I mean, it's not hard to imagine why. Companies just don't contribute to X11 or X.org anymore. They don't care.
They all moved on to contributing to various implementations of Wayland. Because that's now the default on the biggest distros out there that have the biggest base of contributors, whether they work for a company or not. Ubuntu, Fedora, OpenSUSE, they're all big contributors. Well, at least Red Hat, Canonical, and OpenSUSE are big contributors to open source projects. And they're not contributing to X.org anymore because their distros are using Wayland nowadays.
The code bases for Wayland are leaner, they're easier to learn, they're easier to handle, and they let them implement new features, which is basically impossible on the enormous old bloated spaghetti code code base of X.org. They will be able to implement HDR support on Wayland, which they would never be able to do on X.org. Which means that this big old X.org is now in maintenance mode. There will be no new features added, or I would be very surprised.
And if there were new features added, they probably wouldn't get any support from most desktop environments. They would not be implemented by the window managers or the various desktops. Because they're all focused on Wayland right now, and they would not go back and try and implement new X.org features. So I think it's time to say goodbye to X.org. It served us faithfully and loyally for a long time, but it's really dying now.
And yes, it will endure for decades as something that you can still install and use if you prefer it, if you have a niche use case that is not covered by Wayland. And there are some of these use cases. But it probably won't be maintained at all in a few years. It's an enormous code base, and basic maintenance just cannot be done by just a few people. Now personally, everything I own has moved on to Wayland, including all my Nvidia GPUs bearing PCs, because it now works flawlessly for me.
Okay, now a big complex topic that has more to do with privacy than with Linux and open source. And I talked about it in a previous Linux News video, Meta was going to have issues with their targeted ads, because they basically forced users to consent to have their data used for targeted ads in the terms and conditions. And the EU was not having it. Well, the Irish Data Protection Commission confirmed that decision this week.
So Facebook will be fined 210 million euros, and Instagram will be fined 180 million euros, because those are two parts of Meta that infringed on GDPR by just bundling user consent inside of the terms and conditions, instead of letting the users opt out of this data collection and use. Now, the EU Data Protection Board confirmed basically that access to personal data for targeted ads is not a necessary part of providing the core elements of Facebook or Instagram.
You don't need to have targeted ads to provide the rest of the service, which means that you should not have to bundle that in the contract that users sign when they decide to use your service. That's basically the gist of it.
So Meta has been ordered to bring their products into compliance with the GDPR, they have three months to do it, and they will have to ask their users for consent, which will absolutely hurt their business, at least in the EU, because if I was Meta, I would not implement this for every other country, I would do it for the EU, and everywhere else keep doing what I was doing because they have more lax privacy laws. Of course, Meta will appeal this decision, and they absolutely disagree with it.
They say it would be weird to provide a social network service without any kind of personalization, any kind of tailoring, but they basically conflate tailoring what the user sees and ads. It's not the same thing. You can provide a Facebook service, and you can provide an Instagram service that is tailored to the user without tailoring the ads to what the user has seen or watched.
The two are completely separate, but in Facebook's mind, that's what they're going to use as their legal defense, basically. The two are one and the same, but they're clearly not, and I don't see how any court could rule in Facebook's favor right now. They also said that even if they lose on that front, they could avoid using an opt-in form for users.
They could still force consent, but with other ways, as they say that other companies also do it, but in other ways than just bundling consent inside of the terms and conditions. Now, what's interesting here is that the Irish Data Protection Committee didn't seem to agree with the EU as a whole here. They've been basically forced to push out a decision that they don't necessarily agree with.
For them, yes, Facebook and Instagram had breached GDPR because it was not clear enough that when you sign up for these services, you also sign up for a targeted ad contract. It's not made clear enough, but they also think that it's not illegal. It just should be explained better, and they would have dealt out fines that were 10 times lower than what the EU decided to do. That's because GDPR enforcement in the EU is basically crowdsourced.
One data protection commission or one board or organism for a country can say, hey, there's a problem with that, but then the EU basically puts that information and just networks it around with the other data protection commissions in other member countries, which then rule together. Basically, you can say, hey, I think they're doing something shady here, and I think we should punish them in this way. But then the EU can say, oh, no, buddy, you're way too lax. This is unacceptable.
And as a whole, we decide that, no, it should be way more punitive than what you were talking about. And that's what happened here. Basically, the EU dealt out way bigger fines and a way bigger sentence than what the Irish commission would have done. And that's probably because Facebook's business in the EU transit entirely through Ireland.
And so it's not entirely impossible that they lobbied the Irish government, which then enforced a decision on the Irish Data Protection Commission to protect one of their big employers, one of their big tax providers. Maybe they are scared that Facebook will leave or fire people or something. And so they probably just told the Irish Data Protection Committee, please lower the fines and lower the sentence that you're going to rule, because we don't want to scare Facebook off.
That's probably what happened. So we'll have to see how it goes. I think Facebook, well, Meta will have a hard time arguing that providing tailored ads is a necessary part of providing a tailored social network, because that's absolutely not linked neither technologically or ethically. But I'm not a lawyer. So maybe they'll find a way to wiggle their way out of this one. Now, a lot of people always ask me, where are the AMD computers running Linux? Why are there always NVIDIA GPUs?
Well, you have a chance to have a fully pre-built AMD computer now. It's from Slimbook and it's a desktop. It's the Chimera Ventus AMD in its black limited edition. Now it has a nice aluminum metal case, which is all black. It has a nice stylized wireframe Tux logo on the front and it runs exclusively AMD hardware, which is to say a Ryzen 7000 series CPU. And you also have a choice between a Radeon 6750 XT, a 7900 XT or 7900 XTX as in the GPU choices.
They are basically the latest AMD cards, the last two. It supports DDR5 RAM. It goes up to 64 gigs of RAM and it goes up to four terabytes of storage. And since it's a limited edition, the inventory is finite and it starts at 2000 euros, taxes included for 16 gigs of RAM, one terabyte of hard drive spinning rust, not SSD. And you get the Ryzen 7 7700X CPU and the Radeon 6750 XT as a GPU. And you can trick it out with all the tacky RGB you want.
You can ruin this fantastic black look with shining lights and spinning things and yeah, you can do it if you like. And you can also opt for a non-black edition that offers a choice between Intel and AMD CPUs and also Nvidia GPUs instead of AMD ones if you want. They have a mini ATX variant that could make for a nice Steam console with SteamOS. Here you have choices.
So yeah, it's not necessary that I think it's a great value, even though it doesn't seem too bad compared to the current market prices, but it's just that for once we have a fully AMD computer with AMD GPUs in and people are always clamoring that they are tired of computers running Linux shipping with Nvidia GPUs out of the box. So finally, there's an option there.
Now personally, I've been running a Slimbook Chimera as my main desktop for the better part of two years now, and it never let me down. It's seeing really, really heavy daily use, basically, I don't know, six or seven hours of usage for video editing, for gaming, and yeah, it's running well. It never crashed. It never had any issues. None of the components have died on me or anything. Very solid, and it came in a really great state.
So I can definitely recommend it if you're looking for a prebuilt, the Slimbook Chimera is a good choice. I have firsthand experience and I like it. Now all we need is a laptop with an AMD GPU instead of a Nvidia GPU, but I don't know where or who will send something like that. Maybe you'll have a review on the main YouTube channel soon, because I think System76 has a pangolin which fits that description, and so I will be reviewing that as soon as I can receive it.
Now going back to Google as a topic. Google is working on a free moderation tool to let small websites identify and remove terrorist material. Now this was prompted after there were some new laws introduced in the EU and in the UK that compel companies to remove illegal content, including this terrorist material. Now Google says that most websites do not have the manpower to basically enforce this decision.
They say that building the algorithm to remove that is extremely time-consuming, doing it by hand with hand-picked moderators, with human moderators, is also extremely time-consuming and hard, and basically even with an algorithm you can't do away with human reviews.
And since websites who don't remove that kind of terrorist material will be fined or can have legal pursuits leveled against them, it's understandable why they would want to have access to an easy and simple solution to help them remove, automatically remove that kind of material and then just deal with the various appeals from users saying hey it wasn't terrorist material, and they could look at that and say okay yes it wasn't, or yes buddy it was, you should not have posted that.
So yeah, basically they're also saying that since Google, Facebook and the big platforms have learned how to deal with that kind of terrorist material better and better, this kind of stuff doesn't get posted as much on their platforms, but it gets posted on way smaller websites and smaller social networks that do not have the manpower to remove it.
So that's their reasoning, that's Google's reasoning for saying we're building a tool that small websites will have access to, to help them deal with that kind of moderation. Now I have multiple issues with that. First Google is not good with moderation, they suck at it, as you can clearly see in YouTube comments, where YouTube and Google will remove stuff like hey great video, but let something like a super hateful racist insult go through.
It's weird, they're not good at moderation, they're not good at it on the main Google search engine, they're not good at it on YouTube, they suck at moderation, so I would not trust them to build an algorithm to remove that kind of crap. And also, why would you want to trust Google with that kind of stuff?
They're going to treat that as a Google analytics sort of thing, which means that they will have access, I'm speculating here, but I'm pretty sure that's how they're going to do it, they will have access to what your users post on your website, even if that website is not part of Google's family of services, which means that they're going to be able to collect more data about your users, even though they said nothing about wanting to
use Google or opting in, so you're going to have to implement some kind of cookie system that users will have to accept to use your website, and then Google will use all your data, all your users' data, to forcibly remove that kind of content, or at least give you the tools to remove terrorist content, but on the other hand, they will probably also keep that data and use it for themselves.
That's exactly what they're doing with Google analytics, for example, all your traffic data, when you're using Google analytics, is used for other Google products, and to help them fingerprint users and build cross-website profiles to sell more ads. They're going to do the exact same thing with that moderation tool.
If it was an open source tool that they opened up to everyone and that could be worked on by everyone, I think yes, this would be for the greater good of the internet, but as it stands, it's probably going to be a closed source, they never said it was open source, so definitely closed source, it's going to be a closed source tool that only Google has access to, and that's going to be a black box with what it does with the data you transfer it. Do not use that, basically.
I would say do not use that if you value your users, your users' data, or your users' privacy, do not implement that moderation service. Now, a little bit more positive Google news. It seems that they're interested in the RISC-V architecture, which is an open source CPU architecture project.
It's super interesting because it lets manufacturers create designs that use the same instruction set, so basically you have the x86 instruction set, the ARM instruction set, and you have the RISC-V instruction set. Using these instruction sets, manufacturers can build super small, super efficient, super low power CPUs, or they can build super powerful CPUs for super computers.
They have that whole range of stuff, and since it's open source, they can share the designs with everybody else and have improvements and feedback done. It's a very interesting project, and it looks like Google has recognized that it is because at the RISC-V Summit conference, Google said that they wanted the architecture to be seen as a tier one platform on Android, so basically on par with ARM, which is the only, let's be honest, the only CPU architecture that is really used for Android.
They support x86, but I don't know of any large-scale device that uses Android on an x86 CPU, like an Intel or an AMD CPU. It just doesn't exist.
Google says that there's a lot of work to do to optimize Android for RISC-V. They outlined a multi-year roadmap, so they're not there yet at all, but still, they already have started to merge patches into the Android open source project to support RISC-V, which is cool, and you can already grab an Android build for RISC-V, even though it probably doesn't run anything right now. They promised initial Android emulator support for 2023 and support for Java applications in the first quarter.
Now the question is, why would they invest so much time into a yet unproven CPU architecture? It's a very cool project on paper, and I think it's definitely going to have a big impact in the future as consumers and manufacturers are tired of paying licenses to have access to other instruction sets like ARM, for example, but it's still completely unproven and hasn't been marketed or sold to many consumers yet. There haven't been RISC-V CPUs mass-produced that consumers have used.
I think it's because Google is afraid of ARM right now. They think ARM is starting to be more of a RISC than it was before, because the company that handles this architecture and licenses the architecture to manufacturers like Qualcomm, like NVIDIA, like even Apple, this company was almost sold to NVIDIA before they backed out. NVIDIA backed out because there were some regulatory problems and antitrust problems. Now then ARM decided to basically have an IPO.
So they're currently working their ways towards having an IPO, which feels, if you combine the two, it feels like the investors want to pull out. The investors are like, okay, we're going to sell to NVIDIA. Oh, we can't? Well, we're going to be publicly traded and we're going to sell our stock as soon as the IPO has gone through with a high price. We're going to sell everything because they want to pull out.
It's the only reason why they will want to absolutely sell either publicly to anyone or to a big company like NVIDIA. ARM also is suing one of their biggest consumers. They're suing Qualcomm because Qualcomm bought a chip design firm also using ARM patents and ARM would basically like Qualcomm to have that acquisition, sure, but to start over.
They don't want Qualcomm to be able to reuse the designs that the other firm has, because that would mean that they would not have to rent another license from ARM. Basically ARM goes from having two licenses, one for Nuvia, the company Qualcomm bought, and one for Qualcomm to only having one, which would combine both designs. And ARM is saying, no, you can't do that. If you buy the other company, then their license expires and doesn't exist. And so you have to delete all their designs.
You just don't inherit them. That's not in the contract, which is completely insane if you think about it. But so basically ARM feels like a company that doesn't really know what it's doing or where it's going.
And since Android is completely dependent on ARM to have any hardware to ship on, I think Google is seeing an open source alternative like RISC-V as a boon that they could even use themselves to design their own chips, their Tensor chips, which currently are ARM-based chips, but could be RISC-V in the future. And also an open source project like RISC-V could not be controlled by a potential competitor. They could use open source designs for other people.
They could open source their own designs so people can improve on them and help them do this chip design and have extensions to the architecture and the instruction set. That's all a benefit for them. And also they could not, they could avoid having that risk of a major company buying the license holder. So they would not be beholden to any other company for their chip design, which is starting to be very, very important in the Android space.
They're not developing their Tensor chips for nothing. They're doing it because they're seeing an advantage, a benefit for it. So personally, I have no issues with the fact that Google would move on to RISC-V or at least heavily try and adapt Android for RISC-V. I would rather have a giant like Google contributing to an open source architecture like RISC-V instead than having them fighting it tooth and nail and trying to level patents against it or anything.
And since the project already exists as a governance model, Google will not take control of it anytime soon. So that's reassuring as well. Okay. Now we're going to talk about desktop environments because we have updates for a lot of them. So first we have System76, which shared more details about their upcoming Cosmic desktop environment, which is developed in Rust.
Now, they had already said that they would use the IST toolkit, which is a graphical toolkit made to be used with the Rust language, the one they used to develop Cosmic. And they will base something off of IST, which will be their equivalent of LibetVata on Gnome, but for their Cosmic environment. But they will also be integrating the Slint toolkit, which is not made specifically for Rust programs. It would open the desktop up to other programming languages, like for example, JavaScript.
So people could write applets and applications in JavaScript that would still look like Cosmic applications. They're also looking at AccessKit, which is a project written in Rust as well. They're really going all in on Rust here that wants to combine basically all the accessibility related code from various desktops, from various toolkits, sorry, into a single cross-platform framework. So stuff like screen readers or font size scaling would be easier to integrate in Cosmic.
They would not have to redevelop their own implementation of everything that makes a desktop accessible, and that would be a boon for people with disabilities. They're also reworking the notification indicators, which they call applets in Cosmic.
These are now all integrated into the IST toolkit, and they already have applets for audio, for battery, for graphics, which probably means hybrid graphics switching, for the network, for notifications, for power, probably power profiles, for time, and for workspaces. And they're apparently also expanding the feature set that people might be used in the GNOME version of Cosmic.
For example, they're adding the option to limit the battery's charge to 80% on a laptop to save up on battery cycles and avoid damaging the battery if you use it plugged in a lot. They're also only sharing tidbits of information, really. Basically every month in their newsletter, you will get a few tidbits of information like what I just gave.
But it still gives a nice perspective on how well they're advancing, how well they're moving on that desktop, and it's nice seeing the desktop take form bit by bit. So it might look like they're just re-implementing the basic stuff, and that's basically exactly what they're doing, but it at least looks like there's a clear direction for the project, and that's a pretty cool thing to see.
Now on the GNOME side of things, they haven't announced many big things, but they have a new simple app joining GNOME Circle, which is called Lorem. And as its name implies, it is used to generate placeholder text like the Lorem Ipsum text, although why you would want to use that instead of using Bacon Ipsum, that's much funnier. That's up to you.
They also have the Fragments torrent client, which lets you change a torrent's location, which now lets you resume all torrents at the same time, and that handles errors better. There's also GAFOR, which is a UML modeling tool. It had a big new release with a new welcome screen, with access to keyboard shortcuts, access to tabs in the application to work on multiple files at the same time. It has better toolbars.
Authenticator, which is a two-factor authentication tool, now supports multiple cameras if you want to scan QR codes, and it will use OpenGL to scan those things to accelerate the webcam's view, which is always nice. It will also make use of newer libadvita widgets. Converter, it's a small app that I had never heard about before, but it looks amazing.
It lets you convert images into any other format, and now it supports HEIF, HEIC, which are high-efficiency image formats that Apple uses for their photos, for example. It supports bitmap, AVIF, JXL, and TIFF formats, and it can now export PDF into images as well, as well as converting GIFs, or GIFs if you prefer, into WebP. You can even split GIFs frames into separate images. And finally, there's Moni, which is a personal finance manager.
They have rewritten that whole thing in C-sharp, and it now has a Windows version, both versions being compatible with each other. It's currently in beta on Flathub. So basically, as always with Gnome, they don't really talk about what they're working on on the desktop side of things. They're not really letting anything transpire on what will happen in Gnome 44, I guess. We're at 43, yeah.
On Gnome 44, they're not really telling anybody what's happening there, but they're talking a lot about their applications, which is cool because it's basically turning into the best and biggest application ecosystem there is that follows all the guidelines, that looks good, and basically that does one thing, but does it well. Now if you're more of a KDE user, we also have some very, very small updates.
The Info Center can now display OpenCL information, which might be useful if you rely on that for various applications, just as a debug way of seeing if your OpenCL implementation actually works. On Wayland, the Window Manager can now send tilt and pan rotation information to various apps, so that should be really, really cool for people using, for example, Krita to draw some stuff. And KRunner has improved how it displays information about time zones. You can look for time zones in KRunner.
For example, if you type GMT minus five, it's going to tell you the time there, but it's also going to tell you how many hours before or after your time zones they are. Some system settings pages are also getting a new look, which I guess will progressively be brought to all other settings pages, which makes use of big, bold headers that are centered and that split the page into more legible categories. I really like the look.
You can't see it on the podcast, obviously, but if you click on the relevant link in the show notes, you'll see it. And yeah, I think it looks really good. I hope they generalize that to the whole settings page because it makes the page way easier to parse at a glance and to find where you want to be instead of just scrolling endlessly trying to look at every option. You can just now look at headers to triage in your mind the various options. It's better.
They also disabled tooltips in the sidebar of the system settings. You could have tooltips when you hovered over categories, but they just didn't add any information. They just added clutter. So they removed that. And now Discover will also be a little bit clearer about the updates that have been applied or that you will have applied when you have offline updates turned on. Basically, you had a small recap, a list of packages, but it wasn't clear. It wasn't well presented. Now it's much better.
And they also disabled the notifications when you're in the overview, present windows or desktop grid effect before critical notifications could appear in that view. And they kind of messed up how you could type in the general search bar or how you could click on windows. So now they just won't appear here. They will appear when you exit these present windows effects. Now, of course, all these changes should be available in Plasma 5.27 in February.
And after that, they will be working on Plasma 6, which means that the release calendar probably will not stick to a new release every four months. That's just not doable, I think, for Plasma 6. I would be very surprised if they could whip out a first release in four months. We probably will have to wait a little bit more than that. And KDE Plasma 5.27 will keep getting updates and bug fixes in the meantime.
I don't think they will be getting new features, but at least they'll be getting bug fixes, which is still good, I guess. Now we're still on the desktop environment side of things. We have the MAUI project, which you might have heard of in the mobile space, but also on the desktop space, because they're working on MAUI Shell, which is a Qt-based desktop environment.
The MAUI Shell now better handles client-side decoration and window shadows, and it better supports full screen, which are like two interesting things, because client-side decoration is what basically all GNOME apps use for their header bars, and so supporting that makes GNOME apps basically work on your desktop when they would have displayed very weirdly before or not at all. They also cleaned up the code, but that's something else.
PIX, the image viewer, now can perform searches using the text inside of images using OCR, which is a very nice functionality for an image viewer, something that I wish every other image viewer on Linux had. And you can also extract that text from the image. They have a new clock app, which they call Era, which has stopwatch functionality, timer, world clocks, anything you would expect.
And they worked on their MAUI Kit frameworks, which is what you can use to develop apps that fit into the MAUI desktop and the MAUI look and feel. So they improved the backend for every single app with better Android integration, because most of these apps can also be used on Android. And they're also apparently planning to port the whole project to Qt 6, just like what Plasma will be doing with KDE 6. And actually, these mobile apps, well, they're mobile, but they're also desktop.
You can scale them. They're responsive. That's the whole goal of the project, MAUI Shell being a shell for mobile phones and desktops and applications being responsive or adaptive. That's what they want to do. And since they are on Android, a lot of them, it could be a nice first step. Like you could try those applications on Android before and see if they're enough for your needs. And if they are, you could then try and move to a Linux phone running MAUI when it's ready, of course.
OK, last desktop environment will be Budgie. Budgie will get an official Fedora spin with Fedora 38, planned for the end of April 2023. Budgie is already available in Fedora's repos. You can install it right now if you run Fedora. But they had no official variant of Fedora, so they had applied for that. And now they have, which is a way for Budgie to ship a completely as developers intended version of their desktop environment. So that request has been approved.
And so Budgie will join the official Fedora spins that Fedora already has. They have one for KD Plasma. They have one for XFCE, for LXQt, for Mate and Compass, for Cinnamon, LXDE, i3, and something called Sugar on a Stick, or SOAS, that I had never heard about and that is apparently designed to be super accessible to children.
Now on top of that Budgie spin, Fedora will also have an official spin using Sway, the tiling window manager for Wayland, which is basically exactly like i3, but for Wayland instead of X.org. So I think it's good.
While I don't personally think that Budgie is all that crucial compared to other desktop environments, it's basically GNOME plus add-ons in my opinion, it's still cool that the developers have a platform to express their original vision, because the biggest distro that they have that ships Budgie is Ubuntu Budgie, but they really tweak that thing. They change the theme, they add applets, they basically do not ship Budgie as intended, and even Solus also tweaks it quite a bit.
So having it out Fedora where basically no one will touch the vision that the developers have is I think a very good thing to let users experience that desktop as it was intended. Okay, and now we have some gaming news. Well we have the gaming news, I think you're supposed to say the in front of news in English, but I'm French, what do I know? So it looks like the growth of Linux gamers has stalled a little bit after seeing that Linux market share on Steam grow month by month by month.
Now it's been dipping a little bit in December, down to 1.38% compared to almost 1.5% in November. It's not concerning at all, we'll have to see how the trend evolves. It's probably because a lot of gamers got new games or new hardware on Christmas and so they wanted to play that, and so they played that, and they played that on Windows probably because well there are more gamers on Windows than on Linux if you can believe it.
Now SteamOS is still the number one distribution used with 24% of Linux Steam users using it, which means that basically the Steam Deck has increased the market share of Linux users by a ton, at least on Steam. And since it's now shipping to Asia, it's probably going to grow up a little bit more as well. Now something more exciting, we also have Valve teasing HDR support for Linux games, at least on SteamOS and the Steam Deck.
It is super early work, but developers have reported that thing working on Death Stranding and Deep Rock Galactic. Nothing is available to the general public yet, it's not in Proton or anything, or in GameScope or in SteamOS, and that work will probably only be useful for SteamOS at first because it's implemented at the compositor level, and the compositor they're working on is GameScope and is only used on SteamOS, it's tailor-made to play games in full screen.
So if other desktop environments want to benefit from that, they will probably have to replicate a big part of that code or to re-implement it in some way that works with for example KWin or Mutter on GNOME. But we also have AMD ray tracing getting closer to being usable. Mesa is going to enable support for it in two games for now, only two games, Quake 2 RTX and Doom Eternal. And why these two games?
It's just because the work isn't done yet, it needs more debugging, but they want to enable it for specific games at first. So they can't just wait for the implementation to be feature complete and then launch it for everybody because it will not work well for everybody. So they're going to test it out with these two games first, get some feedback from players of these games that will test it out, and then they will have more feedback and more stuff to fix and to work on for future games.
So this should land in Mesa 23, which will arrive later this year, probably before the end of the first quarter. So you can try that out, well you will be able to try that out pretty soon. And that's it for this week! That's 40, 42 minutes of podcast, probably a little bit too long, but it's the first one, so we're trying things out. If you managed to go through all the things and end up here at less than an hour, let me know what you think about this podcast, let me know what I could improve.
You can either tell me on any YouTube video by leaving a comment, but if you want to make sure that I will see that comment, you should probably head to the podcast.thelinuxexp.com page and leave a comment under the podcast itself so you can be sure that I can see it and interact with it, because that's where I'm going to check for those kind of comments.
Now thank you all for listening, I hope you enjoyed, and if you did, please consider maybe looking at my Patreon supporter page at patreon.com slash TheLinuxExperiment, it will really help me keep making this podcast and this show, because it's taking way longer to make than I expected. So thank you all for listening, and I guess you will hear me in the next one. Bye.