LINUX Unplugged - Episode 275: Year of the Relevant Desktop

Episode Date: November 14, 2018

Christian F.K. Schaller from Red Hat joins us to discuss seamless Linux upgrades, replacing PulseAudio, some of the recent desktop Projects Red Hat’s been working on... And the value they get from them. Plus a big batch of important community news, Wimpy’s Thunderbolt Dock experiments, and way to run pacman on any Linux distribution. Special Guests: Alan Pope, Christian F.K. Schaller, and Martin Wimpress.

Maybe before we start the show, let's talk about something that we can't really include in the show because it's not only extremely visual, but it's one of those things you have to just kind of poke and explore yourself. Radio.garden. And I could only describe this to you as a way the internet may have once looked in a retro future. You remember when in like the late 90s, we would visualize the information superhighway as like this 3D plane that you would drive down and websites and company sites would be like these 3D objects that you would approach and go inside of. And that's how we thought the web was going to be, Wes.

That's a Unix system. Yes, yes. That's a perfect example in Jurassic Park where they're flying through all the 3D objects. Well, now it's real. Radio.garden is an overlay on top of a Google globe of the Earth, and it's hotspots all over the Earth of radio stations. They've mapped online radio streams to their physical geolocation

and then represent that on a globe as points of lights. And so you can move about the globe, and it moves a cursor over the area, and then as soon as the globe stops, it begins playing the stream from that region of the Earth. And it's a super cool way to, like, move the planet around and get a different feel for different music cultures all over the globe. It it's a super cool way to like move the planet around and get a different feel for different music cultures all over the globe. It's so neat. Yeah. Right. Instead of just like scanning your radio band, do it visually. And you might actually discover like, you know, you oftentimes don't know where a random radio station that you can hear it's coming from.

Yeah. I had, I had not ever seen this. Pobo, you've seen this thing before? No, because I get punched in the face with a this experience requires WebGL, please try again on a WebGL supported browser dialogue. What are you using? What are you using? The latest version of Firefox. Oh, I got a Firefox here in the studio. Interesting. I'm getting a retry for a live version cloud for your prompt and I try. Uh-oh, we've DDoSed it.

Well, all right, that'll be an exercise for after the show then. This is Linux Unplugged, episode 275. Welcome to Linux Unplugged, your weekly Linux talk show that's just trying to stay warm. My name is Chris. My name is Wes. Hello, Wes. It's good to be connected with you. We have a great show, some super great community news, a bunch of it. We've kind of defined and distilled the best news stories of the week from the community.

We'll have those in there. And then Christian from Red Hat is joining the show. He'll discuss seamless Linux upgrades, replacing Paul's audio, some of the recent desktop projects that Red Hat's been working on, and the value that Red Hat gets from those desktop projects. Why are they doing it? And then a little Thunderbolt 3 dock follow-up. Really looking forward to chatting about that with Mr. Wimpy

because I'm looking for a new Thunderbolt dock from Alanovo. And then we have the poll to name the automation system, and then we'll wrap it up with a couple of app picks. It's a solid show, Wes. Oh, it's huge. There was a lot going on these past couple of weeks, so it took a lot of effort to distill this. I hope everyone appreciates.

You know, to make sure we get it right, we're going to bring in our backup group, and that's our virtual lug. Time-appropriate greetings, Mumble Room. Beep, beep. Hello. Hello. Hello, Brent and Cubicle Nate and Mac and Minimac and Popey and Sean, the silent drifter, and Mr. Wimpy.

It is good to see you gentlemen in the show. I am remote again, so it's not really different for any of you, but it always feels different for me. But it's probably the same for most of you out there but I had quite the experience that I just wanted to share at the top of the show won't take a lot of time I met a Zune guy, Wes you remember the Microsoft Zune?

Yeah, of course, that MP3 player that didn't go anywhere? I actually owned one I got it on a Black Friday sale I went down with Brian Lunduk to I think a Kmart or, you know, like one of those type stores, like a Fred Meyers, a Kmart, a Toys R Us. I think it was, maybe it was a Toys R Us. And we waited in line. This was years ago. Yes, to get a brown Zune,

which I wanted, I think, for a honeymoon flight. I was going on a flight for my honeymoon and I wanted something to, to like watch videos. And it's interesting though, seeing the Zune again today, because it is so huge. These things were discontinued in October of 2011. And they all have their own Zune service with its own Zune software. They are so thick. They're like two decks of cards stacked on top of each other thick. And I'm looking over at this guy. And I noticed, you know, he's got sort of a look about him, like a mid-level management look. And he sits down and he busts out the Zune player. And the first thing that strikes me was how much of that UI is in use today,

where you have a very sort of like what became sort of a famous Microsoft mobile UI really started on the Zune. So that was really fascinating to see that. But towards the end of the flight, he starts to bust out his laptop. And I can't help it. I'm sitting next to this guy. He's got a Zune player and this convertible Fujitsu

or Sony Windows PC laptop. And then the guy, I'm sitting right in the middle in coach, in the cheapo seating, and the guy that's got the Windows seat has got an iPhone XS Max, an iPad Pro, an Apple Watch. And so I've got a guy with a Zune and a Windows foldable PC next to me, and then a guy with an iPad Pro and the phone and all that stuff next to me. And I'm sitting there watching both of these two, how they use their technology. And I don't mean to pick, but the guy with the Windows PC was too big for the seat on the plane.

And if you picture Kilroy or the home improvement neighbor where they're always just like their nose up is over the fence and they're peeking, This guy had the laptop in his lap with the screen flat on his lap and then his nose peeking over the ass of the laptop looking at his screen down into his lap because he couldn't fit this thing. And then he keeps getting all of these Windows 10 error messages and prompts for firmware upgrades while he's watching the movie.

And then the browser edge crashes on him. He has to reload it. he has to reauthorize flash and the movie loses his place and i mean i'm trying not to stare but the just watching him have to interact and like haphazardly mash the tap screen that the touch screen to do all of this it was a real interesting use case and as we landed i lean over to him and i say you know i couldn't help but notice you're using the Zune. Like that's, that's pretty retro, man. Like what's the story there? He's like, not only do I use the Zune, but I have two Zunes. I said, what? You have two? Yeah, I carry two Zunes with me because he says, as you know, because I told him I was a former Zune owner, the power bricks for

these things are huge. And so instead of packing a power brick, he packs two Zunes and he runs one Zune down and then he runs the next Zune down and he hasn't synced them for like two years. But he tells me there's still an active community out there of people that love their Zunes. And I'm looking at this guy like, you've got to be kidding me. You're still rocking this thing. And as we are deplaning, he picks up a bag with Microsoft logos all over it. And I say, oh, so do you work at Microsoft? Yeah, I work at Microsoft. Ah, okay. Everything makes sense now.

Like my whole world was collapsing. Like here's this guy. He's using this touch convertible PC. He's got a Zoom player. Like he must be this big Microsoft fan. That makes a lot of sense. And like my whole world's like, it just doesn't compute, right?

But no, he's an employee. So it makes a little more sense. There's also some people out there that just don't, you and I, a lot of the people who listen to the show, probably we're upgraders, right? We want technology. We like technology. There is a subset of the populace

that they find a solution. They put in work finding it, right? They're like, I spent all this money. I bought two Zunes. And they don't want to have to reinvest. They don't want to learn a new ecosystem. Yeah, I suppose. I suppose.

I mean, I understand not upgrading and getting money. And that was my first impression. Like, my first impression was kind of one of respect. Like, man, that guy is getting a lot out of that purchase. Like, I respect when you can push a piece of technology that far. There's something about that I can respect. But I have a sense it was more of an unwilling to change kind of thing. Ah, yes. Combination of loyalty and just stubbornness. Yeah. And so what he did is he just doubled down and bought two instead of replacing the Zune. He just bought

another one. That's a lot to manage, boy. Okay. Then my last travel story is I saw a side of Texas I've never seen before this morning and I've been sold, I'm going to say it right now, a pack of lies about Texas. And I am angry, Wes. I have been told that Texas was warm and temperate even in the winters. It is below freezing out there right now, Wes. Wait, what? It's below freezing outside, and there is a damn wind chill that makes it feel like it's 10 below freezing out there. It's an aggressive, mean Texas winter right now.

So you flew hours south of here, leaving the beautiful Pacific Northwest just to be uncomfortable. Yeah, and I get the sense that this doesn't happen a lot because everyone lost their damn minds this morning on the road. It was a commuter's battlefield. I've never seen anything like this before. It was intense.

So I guess the sort of chain of events that happened is a few fire hydrants exploded because of freezing. Yeah, I mean, imagine they're not super well insulated, right, when it's normally warm. Yeah, right. So a few fire hydrants exploded because of freezing. Yeah, I mean, imagine they're not super well insulated, right, when it's normally warm. Yeah, right. So a few fire hydrants explode,

which then spews water all over the road in below-freezing conditions, which just started a chain of effects that really screwed traffic up. But what really blew it open was all of the train track crossing bars went down with the red lights, ding, ding, ding, ding, ding, but there's no train. All of the intersections in this entire Keller town that I'm in right now, they all had the train tracks come down. So here's Chris bopping into work this morning because he's here to do a live stream in about a half hour or so. I mean, I had a little bit of fudge time, but not much,

cutting a little close because I thought I had a 15 minute drive. And I end up getting rerouted by cops. I end up circling around for 40 minutes. Long story short, I eventually found a patch of train track intersection that had recently been shooed away by their local cops. And then they had moved on. So there was no other cars. And I, I was able, let's just keep it short and say, had to go into oncoming traffic, had to put my rental vehicle into four wheel drive and had to drive over train tracks. And, and me and about 30 other Texans had to drive over these tracks through the navigation, uh, guide bars, just all this crazy stuff. Then through a multiple lane highway intersection, it was the most insane

thing I'd ever seen. Texans driving in the wrong lanes, going against traffic, people doubling up in single lanes, people like me driving over train tracks. It was every Texan for themselves. It was total car warfare this morning. And that's my drive as I'm coming in to do the live stream for their big 200 announcements. I tell you what, nothing wakes you up like that kind of adrenaline rush. Wow, we are lucky to have you today. You could have died on those dangerous Texas roads. It was a side of Texas I'd never seen, Wes. It was a side of Texas.

You know, it's a balmy 55 here, so come back soon. You bastard. That's my travel update for you. Man, oh, man. Let's talk about a side of Sony we've never seen before. You know, Nintendo has the Super Nintendo Classic and the NES Classic. They're these remakes of these old consoles, and they come preloaded with a whole bunch of essentially ROMs on them.

Yeah, right. Turns out these old games are better than anything we come up with now. Yeah. Apparently. And it looks like following in those footsteps, these old games are better than anything we come up with now. Yeah. Apparently. And it looks like following in those footsteps, Sony's going to be releasing a similar take on the PS1 next month. The PlayStation Classic, you may have heard of this,

it's loaded with 20 of the different ROMs. But here's something that's different than the Nintendo versions. The PlayStation Classic is going to use an open-source emulator on the hardware to run Sony's games. Yeah, I guess Kotaku got some early hands-on time with this PlayStation Classic and discovered that it uses the PCSX Rearmed Emulator, which is a modern version of the open-source PCSX emulator that was originally developed way back in like 2003. Wow, I can't believe they've resurrected this and it's still working.

I guess it actually still does a decent job of being accurate to those old games. Yeah, and it's in a way, it's sort of Sony saying, this emulator is better than what we could build ourselves. Nintendo went their own route. But here they're saying, this is probably the closest we're going to get

to really recapturing that original experience. Now, that's kind of, feels like a bit of hypocrisy to me because Sony's a bit famous for going after this type of community, the ROM community and emulators. And they've been criticized in the past as being very aggressive. And so now that very community that's in some ways been chased by Sony for years is calling them lazy.

Like, you give us so much crap and then you use our software? You lazy bastards. There was that whole, like, running Linux and then not running Linux anymore. I mean, there's been a lot of Sony-related scandals.

Yeah. It's also, I mean, it might just be a good testament to the quality of open source because you would assume in-house, you know, paid engineers, access to all the proprietary stuff, they could develop a really good emulator, specs for the chips,-house, you know, paid engineers, access to all the proprietary stuff,

they could develop a really good emulator, specs for the chips, all that. But no, I mean, it turns out that open source is the way to go. Yeah, it sort of goes back to what we talked about last week, open source by default. Also, didn't know this, but the Super Nintendo emulator, SNEX9X, one of my favorites,

still alive and well. In fact, it's seen some really nice improvements. On the Linux side, the GTK version is now GTK3 ready. That means you also get Wayland compatibility fully implemented. I know, right? Yeah, that's for sure. So GTK3 and Wayland support in a Super Nintendo emulator. I am impressed. Also, yes,

I mean, that's just a great emulator. Yep, recently got my son into that. And don't tell him this, because it's going to be a Christmas present. But there are some really solid Super Nintendo controller knockoffs that just show up as USB input devices to Linux.

Oh, really? Yeah. They're not very expensive. So I got him one and I got me one so I could try it and test it. Of course. Of course. That's for the show. Right.

Totally professional. But yeah, it does work. And so I can't wait to give him one. And he'll plug it into his elementary OS laptop, which I just updated to Juno recently too. Oh. So he's getting all the goodies in there. Did that go pretty smoothly? It went super smooth. and he really likes the

update. He feels like he got a new computer out of it. He even feels like it's a little bit faster. He was running two releases behind. Hey, that's how it should go. Yeah, I was like, that's a best-case scenario. That is great. He's really thrilled. Now, let's talk about Canonical for a moment.

Mark Shuttleworth had a chat with Frederik at TechCrunch at the recent OpenStack Summit. And that's in Berlin, I believe, because we have a couple of people from Linux Academy there right now as well that are giving talks and doing like project meetups and just all kinds of really cool stuff. So really well represented, actually. You know, we have there's a lot of experts there. I didn't even think about going. I mean, it's in Berlin. That'd be awesome. But anyways, Mark's also there,

and he had a chat with Frederic, and he says, I value my independence. He told Frederic that during a brief chat on the outskirts of the OpenStack Summit in Berlin, the outskirts in Berlin, in part, that's because he simply doesn't personally need the money,

but also because he'd like to see all the way through his vision for Canonical and Ubuntu. Now, of course, it's no secret that Shuttleworth would rather like to see Canonical, you know, get an IPO. But apparently, it still needs some work

to get the numbers right. I would also just say that after this whole Red Hat news, the IPO, the public company, and the Linux backing, I feel like there's a little bit of questions there, right? A private company is a little more free to, you know, make their own risks, be a little less profitable, or at least sort of choose philosophy over profit, less so in the public space. Yeah. I mean, that's obviously why this is even coming up. You know, that's why they're asking about this because of the Red Hat sale. And I saw a story recently that more and more tech companies are sort of bailing on the idea of IPOing. They're just kind of, I don't know, like maybe it not coming to an end, but the hype cycle around it, like everybody rushing towards the IPO, I think has faded.

And now people are taking a calculated, really a much more practical and realistic approach to doing an initial public offering. Right. We had this world of unicorns and other things. You get a bunch of VC money. You try to go for the IPO. You try to build this really big, successful company. But for niche players, people that aren't going to be an industrial giant, maybe that doesn't make sense. Yeah, yeah, I think that's well said, Wes, yeah.

The Open Invention Network has a bit of news. This is after Microsoft joined them. And then we all started talking about that Linux system definition. It's all about the Linux system definition. Well, the Open Invention Network, which is now the largest patent non-aggression community in history, with more than 2,750 community members,

has announced today that it's increased its patent non-aggression coverage through an update to its definition of the Linux system. This expansion is part of the Open Invention Network's program to regularly revise its Linux system coverage to keep pace with innovation, as they say. Right. Now, this expansion includes 151 new packages, bringing the total number of, quote-unquote,

protected packages to 2,873. As they say, with this update to the Linux system definition, the OIN continues with its well-established process of carefully maintaining a balance between stability and innovative core open source technology. Oh, of course. Of course, Wes. Of course.

And while the majority of the new additions are widely used and found in most devices, that update also includes a number of key open source innovations such as Kubernetes, Cassandra, and packages for automotive-grade Linux. That last one, kind of surprising. Yeah, I mean, it's not bad, right? And this is directly addressing the major complaint that a lot of people had about Microsoft joining the OIN was that, well, it's only covering the stuff in that definition. Well, I think the message they're trying to get across here with this press release is,

we really evaluate that definition on an ongoing basis and are open to expanding it. That must take quite the process, though, because you've got to get – I mean, don't you have to get buy-off from 2,000 members now that we're going to just all of a sudden expand this definition? Like, that's got to be a process now. I mean, anything surrounding all of these patents, right, that's going to be a lot of lawyers, a lot of rooms full of lawyers negotiating. And that just takes time. It also kind of strikes me that we only need this, you know, the open invention network because a lot of the patent system, especially in the U.S., it's just broken. Yeah.

Yeah. And like architects pointed out in the chat room, it's not a perfect solution. It still has limitations. It's kind of a bone, right? It's a bone thrown to the open source from the big companies. We have to have this because the lawyers and the executive suites say we've got to protect our intellectual property, but we also rely on all this open source technology. Yes, and you also want to give some insurance to companies that are considering building products around open source. to give some insurance to companies that are considering building products around open source,

that they're not going to get sued because that was a major topic for a long, long time, especially in the bad old days of SCO. It was a major concern. I personally had several conversations where I was trying to pitch companies on Linux and was shot down because they were concerned about being sued for using free software. And it was super hard to hear that every time because it felt like they were buying into FUD. Like companies were generating intentionally to just create doubt. And it was, you know, I wouldn't label this heartbreaking, but it was really hard for me to hear people say that. You're like, here we, I have this technology. You can use it for free. And then the cynicism of lawyers just shuts it down. Yes, exactly. And it was better. It was better. You can use it for free. And then the cynicism of lawyers just shuts it down.

Yes, exactly. And it was better. It was better. You can use it. It'll make my work cheaper. It'll be more stable. It is a better solution I can give you. And the software is free.

It's a business move. Yes. Yes. So it was really, it was hard. And so I do appreciate the need for just setting people's concerns down, just turning down the notch on the concern. I do see the need for that. I've got also kind of a concern about LastPass these days. I mean, I know we've talked about it before.

I know the general take is just try to always use free software, but I wanted to give them a solid period of time after LogMeIn bought them just to kind of see how things went. And as time has gone on, and I just, I wanted it to go another way, but I've gotten less and less happy with LastPass. So it's with kind of a wanting eye that I've been looking at Bitwarden. And this week, they may have announced something that's going to tip me over. Bitwarden has completed a third-party security audit of their software. And if you're not familiar with Bitwarden, it's a 100% open source password manager, identity manager is probably the term they would use. And the nice thing about it

is it also has a public bug bounty program as well. And they write on their blog that they're pleased to announce that Bitwarden has completed a thorough security audit and a cryptographic analysis from the security experts at Cure53. You know, actually, it is a good step. It's not perfect. There were several issues that were identified by the audit, some with immediate impact. But the good news, all of those have been resolved in a recent Bitwarden update. Yeah, and it wasn't anything disastrous, really.

I looked at it in the PDF. There's about five Yeah, and it wasn't anything disastrous, really. I looked at it, the PDF, there's about five vulnerabilities and they range, like one of the vulnerabilities was you are allowing people to use two simple of passwords for the master password. Yeah, fair enough.

I mean, that's not exactly a security vulnerability per se, but it is something they needed to change. Right. Other things in there, like there was like a vulnerability with one of the browser extensions.

But the core stuff all looked really good. And, I mean, it's just a big moment for an open source project, right? Because that takes money, that takes community, it takes funding and commitment to development. Yeah. It is a big milestone. And it takes a lot of work. So, that's awesome.

I will probably be talking more about Binward and I have a sense, Wes, don't you? Oh, I think so. And then one last little bit of just sort of updates here, and then we just got a couple other community things, is the LVFS project has a nice improvement coming down the pipe. They've just added an optional feature to FW update and LVS that some people may find useful. The firmware update process can now tell the user how long in seconds your update is going to take. So you'll actually get a bit of feedback on that.

That's pretty cool. Yeah, right. I mean, this means that users can now know that a dock update might take, you know, five minutes or so so they can start the process before they go to lunch. A UEFI update might require multiple reboots, could take up to an hour, 45 minutes, so the user can know, like, don't do this right before your important meeting. That is really cool. And LVFS

is one of those projects that I'm going to talk with Christian about when he comes on the show. It's, I think, one of those that has checked one of those fundamental boxes on Linux. Now, only users running the very latest versions of FWUpdate and GNOME software will be seeing the duration countdown.

Older versions, it's unchanged. Nothing really doesn't work for you. It's just... That makes sense. You're not going to see it. And it's just ignored. Now, and of course, for people that aren't familiar, LVFS is the Linux Vendor Firmware Service,

which is a great way that you can get firmware updates right on Linux. Yeah, especially these days when those chips are essentially running little micro operating systems packed full of vulnerabilities. It's pretty important that you update that stuff. It matters more than it used to. And so it's really good to see LVFS out there shipping that stuff. But Arthur writes into the show, and this is so awesome. He's looking for help with Super Tux Kart networking. He says, Hi, first of all, I want to thank you for the suite of great podcasts. I've personally been a listener to several of them for years now, and I enjoy them a lot. I am

representing the Super Tux Kart project, which as you may know, is a free software and open source racing game. So I don't know if he's actually a developer, but he's representing them. And focused on fun rather than realism, which probably means it would be a hit with my kids. We are nearing a new stable release with the headline feature of bowl, local, and internet multiplayer availability. We're looking for people willing to test and compile the game from the latest sources and help test gameplay over the internet in particular. Later on, we'll announce a beta release, which will provide binaries. But at the moment, we want testers to be able to compile for themselves in case we need to debug the output or we need them to test a fix later on.

He's got more information linked at blog.supertuxcart.net. And they say they'd be grateful if the listeners could help spread the word about this. You'll see it posted a few other places. And if you say if you do read this, it'd probably be easier if you just call everyone by my nickname, Arthur. Oh, okay. Very good, Arthur. We will. He says, I look forward to hearing more on Linux Unplugged.

So I went over and checked out the Super Tux Cart blog, as he talked about there in that email. And holy crap, Wes, they got a lot going on with this multiplayer stuff. This looks good, huh? Oh, it looks great. I mean, it has been about a year since the last release. And I would also like to say, adding network play, especially non-LAN network play, I mean, it works on the LAN too, but like internet play with real latency, that is a significant undertaking, especially on top of a code base not necessarily designed for it.

Yeah. I mean, you know, Wimpy is really into the racing games. So this isn't like the type of racing games that Wimpy's probably into. But if you like the Mario Kart kind of racing games that are more like,

it's like bowling when you have the bumpers in the lanes. Yeah, right. You don't have to focus on realism or accuracy. You just get to go wild. Yeah, exactly. You get real have to focus on realism or accuracy. You just get to go wild. Yeah, exactly. You get real drunk. You go on your go-kart course, and you have a good time.

It's a great party game. It so is. Or you play it with your children, which is where Super Tux Kart fits in my gaming agenda these days. Absolutely. That's absolutely what I'll be doing. Also, what's the game called where Tux slides down on his belly? You know, that old game that we've all been playing for years now

that I've just completely blanked on the name where he's in the snow? Tux Racer. Yeah. Okay. Is that just Tux Racer? Kids love that one too. I think so. Yeah. So Super Tux Cart, pretty awesome. And we may chat more about that in a bit. It's interesting they're asking for just people that are willing to build it themselves. What do you think of that, Wes? I mean, I think it's just, mostly it's early days. And they want to be able to switch things into debug mode if you run into a problem. So they can get all the necessary information to go actually investigate that.

Things that you would leave out in a real production build. It looks like when everything is all said and done, you'll be able to run your own SuperTuxCart server. You know, you can host it yourself if you want, if you just want to have a nice little LAN or something like that. And they've got an IRC channel set up. It's SuperTuxCart on Freenode if you want to get in and chat with people about it.

I'll also say I checked out over in the AUR, one of our favorite little user-curated repositories, and there is a supertuxcart-git package already. And if you go look at the package build, there's a line to use the network branch, which currently has all these changes. Nice. All right. So you need the network branch. Okay. That is good to know. That's so cool that they reached out to us to talk about it.

I love hearing that from them. And I hope they do get some people to help them test it out because that's such a fun one. And it's great. Even if we can't have like AAA open source titles right now, if it's just something you get a lot of fun and entertainment out of, it doesn't really matter. Well, the other thing too, right, is if you're an up-and-coming person

interested in both open source and game development, while it might not be the cream of the crop, here's an open source game you can go look at. You're curious how to make a networked game? Soon, you can learn from them. Absolutely. So I wanted to chat with Christian from Red Hat here because they have a lot of desktop projects that they're working on. And I started to notice a theme across all of them. But nominally, the thing I really wanted to get them on the show about was they've just completed another Pipewire Hackfest.

And now the conversation is just beginning to change to how are we going to manage the pulse audio replacement rollout? That's a big milestone. And so I thought, let's get Christian on the show. He just landed back in his hometown a little bit ago. And I wanted to make sure I got the name right, too. Christian, welcome back to the Unplugged program. It's been a while, and you deserve a proper pronunciation of the last name, so help me there so I don't sound like a doof.

How do you pronounce the last name? Yes, I pronounce my surname Shalir. Technically speaking, I'm not sure if it's correct in the sense that it's a German surname, and I probably pronounce it in a very Norwegian way, but that's how I pronounce it. But you are located in the States? Yes, I am. I live in Westford, north of Boston. Ah, and your day job is technically working for Red Hat, correct? Yes, that's correct. I'm the senior engineering manager for the desktop team at Red Hat.

Ah, okay. Well, then you probably are the perfect guy to talk to. We've been emailing back and forth for a while. It's been ages since you've been on the show, But the thing that has always kind of kept you on my radar is you're often one of the forward facing people talking about Pipewire developments. We just recently had Wim on the show. And you also were with Wim at a hack fest even more recently. And so I thought, gosh, let's get Christian on talk a little more about Pipewire, but then kind of get more in towards your your general day job, the Linux desktop itself and that position and all those things.

So why don't we just start with a quick recap of the Pipewire Hackfest? How did that go, man? Yeah, it all went really well. We got a bigger group than I actually expected together. Cool. Because, I mean, the background, of course, was that Wim had been working on Pipewire for quite a while, and he felt it was time now to see if he could try to help pull in more people and sort of get a larger group working on it,

because the more people helping out, the sooner we can actually have it in production. Sure. So, yeah, it was really good. We had more or less the whole core Pulse Audio team attending, and we had people from the Ulster community and we had multiple people representing various kind of embedded use cases. So was that conversation being had about replacing Pulse Audio

exactly kind of how that might look? Is that going over? Is that too awkward? What about that? Yeah, no, the conversation went really well. I mean, the thing is we had been talking to them a long time before. I mean, Wim is actually a contributor to Pulse Audio over time. He knew the people well. It wasn't like, you know, he was like this person from outside who's suddenly jumping in saying, hey, I have this new thing.

Ah, okay. So when I started to put together the hackfest, I actually reached out to Aaron Ragaban, who's one of the co-maintainers of Pulse Audio, and said, hey, do you want to help co-host this thing? Because he and I had been talking about eventually wanting to move towards PulseWord, sorry, Pipewire anyway.

So he was more happy to do that, and then we got the group together and, of course, started discussing, well, first, where Pipewire is today and where it needs to go to be ready to be that drop-in replacement for both Pulse Audio and Jack. That seems like a major milestone when that happens.

And in a way, it's sort of, to me, it seems like the Pipewire model is more of a Jack model than it is a Pulse Audio model. So when we replace Pulse Audio one day, people are going to see something that might be more familiar to them as Jack than Pulse Audio. That's a kind of a transition in itself too. that might be more familiar to them as Jack than Pulse Audio. That's kind of a transition in itself, too. Yeah, although I think for, like, you know,

if you're an end user and all you use the sound server for is basically, like, playing back music or, you know, getting audio from a browser, I don't think you actually would even notice the differences. I mean, of course, the interesting thing is that this combination of things makes some interesting combinations. I mean, Tal Newkaskinen, who's another commentator for Pulse Audio, actually brought a

post on Patreon recently where he talked about Hackfest. And the one thing he mentioned he thought was interesting was the fact that, you know, Jack has his tools so he can see all the different applications that Jack's application is running and how they're linked together and so on. And the fact that with Pipe Over and Over

you can certainly get Pulse Audio applications into that same graph. Oh, okay. That's personally exciting to me. Yeah, and so we were basically running unaltered jack applications and seeing Totem or Rhythmbox or whatever appear in that graph so that you can basically connect your Pulse Audio applications directly to your jack applications.

I mean, I'm sure there's sort of practical concerns where you might not necessarily want to do that just because they are not written to work in the kind of way a Jack application is meant to work. But in theory, you could connect these things together now. Hmm. This is sort of one of those bigger, under-the-covers projects that is in the works or has been in the works semi-recently

at Red Hat that is kind of what I want to talk to you about today. Because, okay, so you have Pipewire, which I think is just a fantastic initiative that Linux really needs for professional audio production. So I'm very excited about it. But there's been other kind of like lower level improvements that are just kind of important and needed. Like the one that comes to my mind almost immediately that I don't think people think about very often is Bolt, which is crucial for Thunderbolt security settings and having a way for the user to interact with that on the desktop. So why don't we take a second here in the context of Pipewire and some of the other initiatives that Red Hat's doing. and some of the other initiatives that Red Hat's doing. How do we talk about the desktop in a whole as something that's just as relevant or almost as relevant as all of the other major desktops?

I'm not talking about millions and millions and millions and millions of users. I'm not talking about it being the year of the Linux desktop. I'm simply saying, how much work is it to get it to be a relevant desktop, one that's at the discussion table with all of the other desktops? What are your thoughts there?

I think the way I've been thinking about it for a long time now is that we don't need to be at 50 or 80 or 90% to be, call it a player here. I think if you can get the Linux desktop up to, let's say, 5% of the total marketer, which I guess is similar to where Mac is today,

I think that's enough to make us relevant and make sure we stay relevant. And I mean, that's not necessarily too far away from where we are today. I mean, I think, you know, depending on the statistics you look at, the Linux desktop often ends up maybe around 2%. So if you are able to climb another 2%, I think that sort of brings the number where it should be. Right. Or where it needs to be.

That actually seems achievable to me. Yeah, to me too. So actually when I came in about six years ago and took over at this, we sort of sat down and said, hey, let's have a serious discussion about, A, where are we putting our resources today? And are we addressing the things that need to be addressed to actually get users in?

And in some sense, we said, to achieve growth, you basically need to have two things happen, right? A, you need to have new people come in and you need to have existing users not leave. So we said, like, let's try to look at these two. And we said, like, OK, in terms of leaving, we end up doing some surveys and trying to figure out why were people leaving their Linux desktop behind. And I think what we found was the number one reason, at least, we could see for people not wanting to use Linux desktop anymore was hardware support. That after they maybe started Linux during their university years, and then as the time went by, they got more and more frustrated with the fact that, you know, every time we got a new laptop,

it was, you know, half a year or more where things maybe were a little bit hazy whether how well things worked or there was like always these two functions on the laptop that never worked. So we felt that, you know, A, that needed to be addressed. And then of course there were other things too

in terms of like, you know, application availability and so on. But the hardware issue was something that we sort of said, OK, that's definitely where people feel the pain and get frustrated with Linux. So we basically started saying, OK, how can we reallocate all the resources we have available to try to look more on the hardware side and also try to be a bit more proactive? Because I think what often has happened also with Linux is that a new hardware feature comes along and eventually we do get support on Linux, but it's sort of like maybe a year or two after it becomes mainstream available in new laptops. So I think one of the first things we started looking at where we wanted to be a little bit more ahead of the curve

was that we spent a lot of time working on high DPI support to ensure that, you know, as we saw that these laptops were already on the way into the market, that it wouldn't be once again the case that, okay, you got your brand new laptop with a high DPI screen, but you had to run it in, you know, 1080p mode because Linux didn't support anything else. Right. So we've been sort of trying to, you know,

build relationships with harder hardware vendors and start planning things up front. And in Bolt, I guess it's another area where we're seeing that, hey, there are going to be a lot of peripherals, especially docking stations now moving to Thunderbolt. As you said, let's put some on it and work together with hardware vendors to make sure we have an infrastructure there to handle

these kind of things. Or another one that crosses my mind, it does serve a purpose in the data center, but the Linux vendor firmware service, which has been huge for those of us on laptops now that can get security firmware updates via our software update mechanism. I mean, to me, that's a massive, huge, first-class desktop experience that that project is bringing to desktop Linux users. And it seems like it's kind of checking off

one of those fundamental boxes. Like, this is a fundamental thing we need to have. Yeah, no, I fully agree. And it was also like a product that came up, I wouldn't call it randomly, but one of the people on my team is a guy called Peter Jones, who's actually a Red Hat representative on the UFI Standards Forum.

And he actually just approached me one day saying, hey, you know, it'd be really good if you can do something here to provide UFI firmware updates because, A, that's what's coming. I mean, people are moving away from the old BIOS model and switching to UFI firmware. And the thing we hadn't known for years, right,

was that people, well, first of all, it was extra work for our kernel team because often they end up doing workarounds in a kernel driver to work around for firmware bugs. And in some cases, people would often have a laptop and say, oh, Linux is broken, this thing sucks, and my battery life sucks, or my hard drive is slow, whatever. And the problem was that they didn't have an up-to-date firmware, Of course, if they had, they would have gotten a fix, you know, years ago that actually proved their situation. So, yeah, so we kicked off that LVFS effort and, I mean, had to give a big call out there to Dell, actually. It was one of the first major companies to come on board and said,

okay, hey, we want to work with you and make sure this happens. And, of course, today it's,, today, we also have Lenovo signed up and, of course, a host of peripheral vendors. Yeah, it's so great, especially as we get more and more software in these chips that are full-fledged machines that are doing a lot and have some security issues, potentially. And so that kind of first-party update service is just super important, maybe more than it ever has been. And in a way, I think maybe a lot of desktop users can't even

really wrap their heads around because it's all just a black box to us these days. Well, you know, Christian, I feel like I'm kind of seeing a theme here. So we talk about Bolt and we talk about the firmware service and getting those firmware updates is super critical as these devices do more and more. And this is like another one of those fundamentals. But the one that really seems like the most practical compromise coming out of the Red Hat and Fedora group recently is the support for binary drivers like the NVIDIA driver. And this is something that I think I've always kind of criticized a little bit. They made it for a bad end user experience.

You know, you buy a machine, it's got a $400 part in there and you get about $80 of value out of it with the free driver. And that was always sort of my proposition to the Fedora camp. And the response generally was, we really want to focus on free software and building a free desktop. But as time has gone on, there have been, I don't know what else to call them, maybe compromises to make it a little more practical, to sort of check another one of those fundamentals. I'd love to know the backstory around getting those binary drivers into Fedora. Yeah, no, it was, it came from the same thinking here, right?

It's like, we know a lot of people, of course, are getting laptops or desktops, for this matter, with NVIDIA chips, and they needed a binary driver. And it was always this pain because you had the conflict between Mesa and the binary driver. And, of course, the binary driver was available only from a third-party site and so on. And I sort of said, like, hey, we are a major backer of Nouveau. I mean, we are the only company who actually are paying people to work full-time on Nouveau.

But at the same time, we need to realize that, you know, as you said, when people buy expensive hardware, they want to be able to get max use of it. And at the same time, we should then use this as a way to develop an even better relationship with NVIDIA to improve the overall level. So we started

up having conversations with NVIDIA saying that this is what we want to do. We want to step one, figure out a way to avoid this SO file conflict between Mesa and the binary driver because it makes life hell for packagers. And users, of course, we have to deal with the fact that

they might have a package depending on Mesa that gets uninstalled to deal with NV driver and so on. So we set off a lot of discussions and projects and of course we did the whole GLV&D project that had this vendor neutral dispatch system so you can have multiple implementations of

your GL drivers and talked about how can we eventually get to a point where we can support discrete graphics better. And of course, at the same time, you know, Fingal and Hawk, we also collaborate better on making sure Novo works better. So, I mean, all in all, I mean, I don't feel we're there yet. I mean, this has been an ongoing process, but I think we're miles ahead of where we used to be. And I believe that, you know, within a year, I think you will suddenly feel that, you know,

all these challenges around NVDA hardware will, to some degree, be a thing of the past. Interesting. I like that prediction. What I see here is a multi-team effort to solve this problem. And when you add it all up, it's a pretty sizable investment. And I'm curious, in your opinion, being inside Red Hat now for six years, what do you sense the value is to Red Hat to invest in these particular areas? You know, the desktop for Red Hat isn't almost a special product.

I mean, you know, Red Hat was actually started as a desktop Linux company. I mean, I think people tend to forget that these days, but you know, the original starting of Red Hat was being a desktop operating system. to this day, we see the desktop as a valuable asset, not necessarily

as a revenue generating one, but it's important in terms of just training people to use Linux. I mean, a lot of people who today work as sysadmins and or developers around a lot of companies around the globe, I mean, they got their start with Linux through running a Linux desktop. Absolutely. So for us, it's more like the more people have using Linux on the desktop, then the more people are available for both us and our customers and partners to hire who has, you know, experience and knowledge around Linux. That is an interesting answer. I hadn't considered the education angle.

That is playing the long game, but it is a smart one. I guess my only other question to you would be, is there anything else you want to mention or send the audience to, something going on? Is there any resources you'd like to inform people about? Because you guys are working on a ton of good stuff, and some of it deserves a lot more attention than it's getting.

One of the projects we're currently doing that I'm pretty excited about is called Silverblue. I think you have covered it before on your show, which is basically this idea of trying to make an image-based operating system out of Linux or for the Linux desktop. Because I think what we are concerned, going back to when we looked at all the problems people have, right,

is these recurring problems that upgrades are not seamless. So, of course, I think we've done a lot of work over the last five years for Dora to make it a lot more seamless, and I think it mostly works now. But the problem still often happens, right, is that time goes by and you install that, you know, random package that you found on some website that, you know, provides application or feature you wanted. And that package starts creating problems when you try to go to the next version because it, you know, depends on an SO file that has changed versioning or whatever. So, of course, I guess we have two efforts trying to address that problem, right? One is, of course, Flatpak,

which is this core effort of trying to make application bundles that, of course, keep most of their dependencies with them and just avoid this problem of migrating. And, of course, for Red Hat, Flatpak brings another valuable feature, which means that we can have one package that targets both Fedora and RHEL. We don't need to rebuild Firefox, for instance, for each.

Of course, yeah. So that's part of the story, but at the same point, we felt that we still want an upgrade setup where that's sort of rock solid, and that's sort of where Silverblue comes in, where we can say that, hey, we have an image of the core and all the packages that are part of that core

are upgraded as a group so that we can basically upgrade you know call it an oversight verify that everything works and that there's no scripts or anything that messes up and then just ship that as a as a an altered image to place on your computer my goal is that a more people to try linux because it works better and at the same point, less people leave because they have problems with it. That's sort of where we're coming from. And I guess to cover another thing that I know was sort of a little bit covered in some stories recently was that,

you know, it's been announced now that we're not doing KDE in RHEL 8. Ah, yes, right. In fact, I did want to chat with you about that. Well, there's a couple of things to make clear. One thing is that, yes, we're not supporting running KDE as a desktop, but we're still supporting people running KDE applications on RHEL. So that's an important distinction to make. But I mean, once again, it went back to when we start to look at,

like, why are people leaving Linux? And I guess you could say that, you know, if there's one area where Linux beats every other operating system, it's in the number of desktops available for it. Yet, of course, as we see, it doesn't mean that we have a sort of taken over the market based on that. So I think we've in some sense concluded that having a choice of desktops isn't really what drives adoption or retention for that matter

and and it's quite costly to maintain because for for instance for my team then who maintains all the desktops in red out it's sort of thinking it's like well first of all the most horrible message you can give any customer or user right it's like oh if you want to do that then you switch to this desktop environment but if you want to do that then you switch to that environment and if you want to do both, then you have a problem. So we said, no, we need to be able to say this is what you support, and this is when we get requests from users for features where we'll implement those. And at the same point, in order for me to be able to move quickly on these things and do products like Pipewire and Flatpak and Silverbeam and all those things, I need to be able to do them once as opposed to

sort of feel that we have to re-implement the feature two, three, four, five times, depending on how many desktop environments we are supposed to support. That seems like a tight line to walk, though, because at the same time, if you want something completely universally accepted, of course, you'll need the Plasma desktop to be on board with Pipewire as well. There's a line to walk there. There is definitely. And I mean, the thing is, of course, we are still actually, you know, despite the plan not to not ship with the KDE desktop in Relate, we are still contributing to KDE, especially, of course, in this area that we care about. So Jan Grulich, who's a member of my team, he attended a tech fest for Pipewire in Edinburgh, and he's the one who's been working on making sure that we have full Pipewire support in KDE. That's great. That's great.

I mean, but it really, I mean, to steal a business metaphor, which everybody loves, it really feels like it's about putting all of your wood behind a single arrow. You know, like it's about focus. Yeah, it's about focus. So it wasn't like, you know, an anti-anything. It was more like, hey, we need to focus our resources,

which are limited, on solving what we know we are costing as customers. And as opposed to, you know, trying to keep a small subset of people happy by offering them like this choice of desktops, which, as I said, doesn't seem to be actually what drives people to adopt an operating system. Yeah. Yeah, I could see it maybe keeping people around every now and then, you know, some desktop environment has just failed you.

And instead of switching back to Windows, you could switch to Plasma and be happy. That could be a potential. But I see what you're saying. People aren't coming over to Linux and going, oh, boy, after I spend a week picking my distribution, I get to spend another two weeks picking my desktop environment. Yay!

That's not a big advertising point. No, and of course, as I said, we had cases in the past, right, where people were dumping desktops basically to, like, you know, hey, Multimontor is working better here than there, but then I discovered that my Wacom tablet worked better there than here. And then, of course, it puts the customer

in an impossible situation, right, because it's sort of like, well, I have to choose that feature versus that. And then we are on the other side sort of struggling with, like, well, I don't have enough engineers available to fix both this and that. So what we're doing here. And people also might think that features haven't been fixed because they're, let's say, using the wrong desktop environment compared to one where we implemented the fix. And so it's like, it's a challenging problem to overcome.

Well, Christian, I really like the approach that you're taking here. It's sort of, you know, shave away at these rough edges and make the desktop just relevant in the marketplace. Don't necessarily chase the market share number. Just make the Linux desktop relevant. And I think we're close. And that seems like an actual attainable goal. And I think you guys are playing a big part of getting us there. So thanks for coming on.

Thanks for keeping everybody informed on what you do and traveling around to participate at those hack fests and whatnot. I'll have links to the blog and other things. Anywhere else you want to shoot people to check out? No, I mean, once again, I actually would like to say that,

you know, of course, yes, we are doing a lot of work there and trying to fix it. But I mean, there's other people, I think, doing critical work these days. I mean, once again, I would like to actually give a shout out to Dell for all their recent efforts.

I mean, I guess we're the first major maker who got serious about Linux on the laptops. And then, of course, in addition to those, we have smaller but thriving businesses like System76 and Purism and so on, who are also pushing Linux on laptops forward and making that a more reliable experience. And I mean, another thing that I often felt held us back, and we knew we got feedback from people saying that they ended up using or keeping Windows around for was gaming. And of course, Valve has been doing an incredible effort, in my opinion, getting Steam going on Linux and making sure we have a huge selection of games.

It really is a full-spectrum community effort. Well, Christian, thanks for coming on the show and chatting with me. Yeah, my pleasure. I really like talking to him because it's a reminder of all of the things that I use every day that they're working on. And that is more important than ever in the context of the news with the IBM

purchasing Red Hat. And, you know, that was sort of what I was getting at, too, with what's the value to Red Hat for these projects? And so I enjoyed that entire conversation with Christian. You can find links to his blog and other things in our show notes at Linux unplugged.com slash two 75,

including his most recent post about the pipe wire hack fest, which I know Wes and I are super excited about pipe wire. Oh boy. Yeah. Well, Wes, we should keep moving right along because on the Christmas list for me this year is a Thunderbolt 3 dock. I think I'm going to do it.

Ideally, I'd like to do one that even maybe has a GPU or something. Like I've got all of these. Like embedded right in there, just stock ready. Yeah, ready to go. Thunderbolt aspirations big time over here. But I don't know exactly what to expect, where the Linux compatibility is at. And so what I'd like to do is let Wimpy take all of the arrows on some of these hardware purchases and then check in with him. We'll learn from his mistakes.

Exactly. Or successes or successes. I know Mr. Wimpress has recently become the proud owner of a ThinkPad P1, a beautiful, beautiful new rig. And I believe he has tied that with a Thunderbolt dock. And I'd just love to get your take on how this is going, Wimpy. Yes, I have. And I've been using the dock for about a week now. So I've got a ThinkPad Thunderbolt 3 dock and it's the second gen unit. All right.

So you went with a Lenovo branded dock. I did. And I will explain why. So technically, there is a single umbilical cord that connects the laptop to the dock. But what Lenovo have done is made a connector on the laptop side which is a single connector but actually bridges the power uh port and the first thunderbolt port and then that connects to the thunderbolt port uh and connects the same way on the back. So it connects into a power port and a Thunderbolt port. So technically two cables, but one umbilical cable to connect the thing together.

Okay. All right. That's nice. And is that delivering a higher wattage? Yeah, right. Because you can only deliver a hundred watts over Thunderbolt 3. So insufficient for a laptop with a Xeon processor and a discrete graphics card. So that's why you have to provide power and Thunderbolt. Okay, and so this allows you to hook up Ethernet, USB devices. What's the port layout on this thing? And what's the capabilities that it brings you?

Because these things are, over here at least, nearly $400 US. Is it worth it yeah well it depends what you're going for in my opinion yes because it now makes for a very convenient way to connect and disconnect the laptop so this this laptop is now going to be my main workstation and laptop and i will just you know disconnect it from all of this gubbins via one cable now to take it out with me when i go out you know and travel and work yes um but the first the first thing you have to do is use bolt which of course you talked about um in your last interview um so bolt is the mechanism by which you actually authenticate your Thunderbolt device so that it is now trusted.

So you run that to connect that, and there's command line utilities, and if you're using GNOME, then that's all built into the UI. And then once you've done that, on the back of this Thunderbolt dock, there are two full-size DisplayPort connectors. There are two full-size DisplayPort connectors. There are two HDMI connectors. There's a plethora of USB 3A ports. And there is gigabit Ethernet. And on the front, it's got USB-C and headphone, and it's an audio interface.

And is that everything? I think that's most of the things. And Linux is pretty cool with all that stuff just appearing and disappearing. It's not really a big deal. Yeah. So I've just been, I mean, I think back to my days when I had a Commodore 64 and you used to have to sort of turn the computer off when you wrench the cartridge out and that sort of baked into my DNA. So i think can i can i really just you know yank this yank this out and it'll all carry on working and it turns out it does so yes i um i just pull it out and the screen has a little bit of a dance as everything sort of collapses down to to one screen on the laptop lid again um but yes it all works fine um the one the one area where i've had to do a little bit of um sort of hackery is i've got two external 1440p monitors and the internal panel on this laptop is

um uhd so you run into this uh mixed dpi environment where one of the dpis is going to be dominant and what i've done there is i've written a little script that manipulates x randa so um it enumerates what monitors are attached it lays out the monitors in the right order. So effectively, I've got the monitors in front of me, left and right. And then the laptop sits right in the middle between the two of them. And this X-Randa layout actually positions the laptop monitor in the center of the crease where the two monitors line up and uses the correct offset. So as you move down as you move down

you know you're you're directly you know moving into where you'd expect the mouse to appear on the laptop lid beneath you and then i use some scaling in x rander to basically um make that internal panel uh the effective resolution approximate to 1080p so that now means that i've got effectively the same dpi across all three of those machines now there isn't a ui that enables you to facilitate that in x at the moment but using x rander it is possible to create that so yeah i have this little script that just you know checks periodically sees which which panels are detected and um and adjusts itself accordingly i know it's sort of a small script but have you thought about tossing it up on on your github or something because that's something i might use yeah i will i was discussing this with poppy

earlier and the first thing he said is i want it um. But at the moment, it's very much sort of hardwired to my screens because it's my display IDs, it's my resolutions, the offset calculations are all hardwired. But I think I'll put it up as a gist and just sort of document what it's doing, and then other people can try and take advantage of that and figure out how to reuse it but ultimately what what we need to do is make the um the display arrangement uis in the desktop environments be smart enough to be able to accommodate these more exotic you know display layouts absolutely especially as this becomes more and more

possible there's going to be lots of laptops with Thunderbolt 3 ports. So if I recall correctly from a couple episodes ago, you got the Thunderbolt 3 workstation dock, which Lenovo sells for $356 here in the States. Now, this is where I'm torn, because you're right, it's a beautiful dock. It's got two HDMI, two display port, Ethernet, and it's got that bigger connector, which delivers more power. But Lenovo also sells a Thunderbolt 3 dock for $362 that has a few less ports, but has a GTX 1050 built into it. So I would go from Intel graphics to getting a 1050 when I plug in that Thunderbolt dock. Now I grant you, a 1050 is not going to rock my world, but it's better probably

than the Intel embedded. And that's really tempting. Do you think that would possibly even work? Yes, I know it will work because when I have this dock attached, i have a single additional thunderbolt port spare and i've been able to connect my razor core to that and use the 1080 ti in that you know to amplify things now as it happens this laptop also has quadro p2000 discrete graphics in it which is more or less equivalent to a 1050 and the 1050 isn't to be sneezed at. I was just using it to play Horizon Chase Turbo before we came on air this evening. It's not the most demanding game in the world, but it was flying along at plenty of frames a second

and was very enjoyable. That's what I was thinking. Like, I could play a game, and all I have to do is plug in my dock. Right. And it still has a couple of display ports. I think it has two display ports on it, but only one HDMI. Right, right.

Yeah, I mean, I'm only ever going to use two monitors. I don't have space for any more monitors than that here. And the ones I've got are plenty big enough. I might give it a go and report back on how that GPU integration works with Linux too, just after some testing. Yeah, and I can explain, you know, the way that that actually works is I've written an applet for Marte to make this easy. But the way you use it is you just run the NVIDIA settings.

There's even a switch to take you to the right portion of the settings where you can just flip-flop between NVIDIA and Intel. Ah, good to know. Thank you, sir. Well, I always like learning vicariously, as I say. It's the least expensive way to learn. It really is. Maybe we can pass it along. But, you know, I've got dozens of USB devices connected here.

You know, webcams, the mic, the USB audio device, multiple keyboards, mice, a bunch of other stuff, controllers, all sorts of things. And now I just have one cable to connect and the monitors and all of those devices just power up. That is the dream. One cord and it's powered data, everything. It's stuff and it's no longer just on the USB bus.

It's on the PCI bus now. So we're talking like real hardware too. It's on the PCI bus. Yeah, yeah. So yeah, this is not too shonky. So there's four, four PCI lanes available to that hub, to that dock rather, um, to deliver all of those, all of those pieces, obviously the more displays and things and the more bandwidth you were using, but, um, it runs really nicely. I've had no problems with it at all.

I'll put links to both the one that Wimpy got and the one that I'm going to grab and try out in the show notes if you guys want to check that out. LinuxUnplugged.com slash 275. How's that P1 treating you? Is it going good so far? Yes. Well, that doesn't sound like a super yes, though. Well, you know, I'm an early adopter,

so you can always expect problems, can't you? So there are two issues. These issues will be the same for ThinkPad P1 owners and ThinkPad X1 Extreme owners. The first one's minor, and that's that the, what do you call it, fingerprint sensor scanner thing is not currently supported. Oh, of course. So that's a minor issue um the other one that's a bit more interesting is um hybrid graphics don't work uh on this at

the moment and i've been uh i spent the weekend having a poke around um with the gpu manager and i've i've got a hacked up monkey monkey patch thing here where I've kind of got it half working. So, uh, where I say half working, um, it means that I don't have black screens and I can't do anything. I actually get the displays to initialize, but I'm not at the moment able to switch between a hybrid and discrete graphics. Um, but I'm a bit further forward in that, you know, when I first tried to use it, I just had black screens. And now I have screens to function. Well, yeah, that is a bit further forward. You're right. Okay. Yeah, that's good. Well, I wish you luck there on your journey, sir. I would love to know how it goes.

Yeah, it wouldn't be fun if it didn't have a few problems. You don't feel like you own a computer until you've solved some of these little gremlins. You know, I've been there where there's times where I buy sort of the ambitious system. And then there's times where I buy the really practical system that I know is going to be taken out of the box and get to work. And it just sort of depends on what you want to get out of that system. And the ones where you eventually get them working are sort of rewarding in themselves because it's like an accomplishment. Now, we have just a couple of bits of business to take care of before we get out of here for this week. We won't spend a lot of time on it

this week, but the official voting is now opening for our automation system. You can find the link in our show notes. Go give your vote. We'll go more into that next week. But if you want to get the early vote in, you can find a link to our new system. But I just want to take a second and talk about a really cool tool that Wes found this week that he's actually already managed to put into work. And I think it was from an email from a listener, Brandon, that tipped you off. Yes, that is right. Brandon wrote in to us and was asking if we'd ever heard of the Junest project. I don't know if I'm saying that right, but J-U-N-E-S-T, or Jailed Under Nest. It essentially lets you run Arch Linux with fake root access.

It really uses proot, but that basically means you get a whole little user space Arch install. You don't need root. All you need is a home dir, a little bit of modifications to your path, and, you know, git. But you probably already have all of those things, and now you have all of the Arch packages, the AUR, and that wonderful package manager,

Pac-Man. Right there on top of any distribution you want to use, and is that what you used to try out the SuperTuxCart build? Yes, it is. I mean, because it was already in the AUR, right? That's one of the nice things about Arch, even in the days of snap packages and flatback and all that. Arch just has a lot of packages. And if you're like me, I mean, I just spent so many years on Arch, even though a lot of my systems are Ubuntu

these days, I'm familiar with it. It's easy. And a lot of times you kind of just get that minimal baseline. You don't have to worry about of times you kind of just get that minimal baseline. You don't have to worry about a whole bunch of different extras you need to do. It's all right there. It's up to date. You get the latest. And now with Junist, you can have it all in a throwaway environment. It's pretty neat.

That's pretty neat. Thank you to Brandon for tipping us off to that. And thanks, Wes, for kicking the tires and giving that a go while I'm out and about. Junist, Arch on Linux without root. We'll have a link in the show notes for that as well. Go get more Popey and Wimpy. Mark as well at the Ubuntu podcast, ubuntupodcast.org.

A special shout out to producer Michael for helping us out with our mumble room disaster we had right before the show. Go get his great This Week in Linux on YouTube and go get more Wes Payne. He's over there at techsnap.systems for another podcast that he's on with us. And he's over on the Twitters at Wes Payne with a Y.

Oh, that's right. Payne with a Y. Yeah, I'm at ChrisLAS. The network's at Jupiter Signal. Thanks so much for joining us and we'll see you back here next Tuesday! you know what so uh producer michael in the chat room says Lass, could you not have given the option of Castablasta? I had that same thought. When I looked at that list, I thought we should just maybe call it Castablasta because that was the first podcast we ever did. It's too perfect.

Yeah, it does seem like it would be really perfect. But then I thought we're creating name collision with like a piece of art history maybe. Or it's an homage. I don't know. So I just sort of dropped it. It's an homage because it's like it's not going to be messed up because if you go to the site and search for it, there's only a few that actually are listed on the site

because most of it was like an archive. Yeah. Oh. Yeah. I feel like I wonder, can we edit that poll and add it? That's a good question, yeah. You could just change Cast Blaster to it if you wanted. Yeah, yeah, that I think would be the thing to do.

Yeah, I don't know if we can edit that poll, but we should. Because I had that same exact thought, and so I think we should definitely do it then. Yeah, that poll I think is going to be interesting because I am trying to picture saying that as a verb. You know, run it through the Cast Blaster or run it through whatever the names are. So when I read those names in that poll, that's sort of the scenario. I run it through my head. So some of them are pretty funny when you do that. We got a whole bunch of submissions.

We're going to have to do a giveaway or something forever wins. I think I think that'd be good. What are you linking me? Oh, this is the script. Oh. Nicely done, Mr. Wimpress. Yeah, definitely. Mr. Wimpress for the win.

I'm going to, I got to grab that. I'll put that in the show notes. So Wimpy just linked his gist that we were just talking about. That was quick work. Well done. I think he was multitasking.