LINUX Unplugged - 375: Wrong About Pop!

Episode Date: October 14, 2020

We're reminded that you can't judge a distro by its screenshots. We use Pop!_OS for a few weeks and share our embarrassing discovery. Plus our thoughts on the new Plasma release, a super handy pick, and more. Chapters: 0:00 Pre-Show 0:44 Intro 0:50 SPONSOR: A Cloud Guru 2:39 Plasma 5.20 7:50 Kernel 5.9 8:05 VMware Flirts with Arm 15:28 SPONSOR: Linode 18:54 Big News for Nebula 22:10 Code-Shaming the Kernel 27:40 Housekeeping 29:31 Pop!OS Exit Interview 31:44 Pop!OS Full-Time Staff 34:49 Pop!OS: The Last Ten Percent 37:46 Pop!OS: A Very Unique Distribution 43:13 Pop!OS: Driving Hardware Sales 47:40 Pop!OS: Strengthening the System76 Brand 49:51 Manjaro Arm 20.10 Released 50:48 SPONSOR: A Cloud Guru 51:48 Feedback: TLP Magic 53:23 Feedback: Chromebooks and Education 56:16 Pick: Autotier 59:09 Pick: Antennapod 2.0.1 1:00:30 SPONSOR: Core Contributors 1:01:10 Outro 1:03:18 Post-Show Special Guest: Neal Gompa.

have you seen mario 35 have you seen this no so the it's a switch game that takes like the the original mario game and then versus 35 other players and as they like kill koopas and stuff like that or goombas whatever there they've come to your screen and sort of a last man standing kind of game fun it's an overlay on top of a Super Mario Bros. emulator. It's really kind of fun. It is. You know what? They can just get me over and over and over again with Mario.

I tell you what. It's classic. Hello, friends, and welcome to episode 375 of your weekly Linux talk show. My name is Chris. My name is Wes. Hello, Wes. This episode is brought to you by a cloud guru. So many career possibilities in the cloud, so little time.

ACG's learning paths help you take the right courses to prepare for architect, developer, security, and many other high-paying cloud jobs. Get hired, get certified, get learning. A cloud guru dot com. Mr. Payne 375 is packed with lots of good community news. And then once we get through all of that, we're going to try something new. It's an idea that I'm kind of stealing from another tech podcast accidentally, if you will. And it's the idea of an exit interview. So we're going to do an exit interview with Pop! OS today. I've been running it for a few weeks on a Dell Precision test laptop, and I wanted to try different distros. And at one point, it crossed my silly little head that it'd be fun to run Pop! OS on a Dell laptop. And then I found myself sticking with it for a bit.

So I've been using it. And last night I sat down with Wes and I said, Wes, I got to confess something. And it was kind of embarrassing, if you know my history with Pop! OS. So I brought Wes over because I didn't want anybody to find out. And I said, Wes, come here, come here. And then we got in the cone of silence. The cone of silence. The cone of silence.

And I think, Wes, I think I, after using Papa Wes, have discovered something that makes it stand out and different than any other Linux distribution. That's a big claim. I'm going to try to prove it later in the episode, but don't tell anybody, okay? I don't want anybody to know. It's our little secret. Yeah, okay. The cone of silence. All right, well, now that we're out of the cone, let's bring in the mumble room time appropriate greetings mumble room hello hello hello they've really got that down now i just love that that was a good one that was a really good one

well let's talk about some news there's a brand new fancy plasmaasma. Plasma 5.20 is out, and it is an absolutely massive release. Running it right here on the machine I'm reading this news story about. And it's one of these releases where everyday things have gotten better, like the little utilities you use, the task manager, notification, system settings. It's just gotten some nice improvements and been overhauled just to make it more useful and usable. And you can really say that about all of Plasma from the way on-screen displays are handled, some new tricks that KWIN has. And for me, I'm always super appreciative of improvements to the way notifications work. And there's a lot of changes on the notification system in Plasma 520.

For one, you now get notified when your system is about to run out of space, even if your home directory is like on a different partition. Yeah, that's really nice. Yeah. There's just a lot of little things in there about like, you can make certain things sticky now in there. It's continued to get a lot of improvement. And I think Plasma remains out of every desktop environment, not just Linux desktop environments, but in the world. I think it's got some of the best notifications management now. Oh, yeah, it really does. It's just full featured. They've clearly thought about it.

They've iterated over a few releases now, so it feels kind of polished. Yeah, it's really solid. And I'm on Neon right here. So I just went ahead and did the old update and was kind of prepared for things to go sideways since it is such a huge update. But it seems like everything's nice. There is a change they've made, and the way Plasma works, that if you've already modified the default and you upgrade, it won't change it.

But if you're installing it fresh, there's a pretty big change to the way the Task Manager works now. Yeah, there's lots of improvements over in the Task Manager world. I mean, they did change the looks, too. It's icon- only by default. Yeah. But there's also big changes in how it behaves. When you open several windows with the same application, like maybe you might do if you're opening a bunch of LibreOffice documents, say,

the task manager will group them together. Clicking on the grouped windows will cycle through them, bringing each to the forefront until you reach the document you want. And yeah, that's kind of a different way to do it, but I'll give it a try. Yeah, I'm not a big fan of grouping it together. You know, one of the things I kind of do on a Plasma desktop is I really often, I'll just keep the task manager on the bottom. And I know it's hip today to put it on the right side of your screen or the left side

of your screen because, you know, you got more vertical or whatever. But I like the long taskbar. And I feel like if I got a long widescreen monitor, I might as well stretch that task manager out. I don't do anything for you there. I don't need it. I like having all the windows square. I'm more of an alt tab kind of guy anyways, though. So not really going to impact me. Really kind of one of these releases where, you know, maybe you're going to base a distro like Kubuntu

off of a future version of One. This would be a really good one because there's also some really good Wayland improvements that have landed. You know, back in 2019, the KDE community decided to set off as an official goal to prioritize building for Wayland, essentially. And I think the effort's really starting to pay off big. There's lots of features that have landed now. The Clipper clipboard utility and the middle click paste stuff are now fully functional in Wayland.

Love me middle click, copy and paste. And the multi-purpose launcher calculator search K-Runner app that's part of like my favorite thing of Plasma will stay in the correct places and works in the right place on Wayland.

It just behaves a lot better. Mouse and touchpad support are nearly on par now with the X counterparts and screencasting is now supported in Wayland. It just behaves a lot better. Mouse and touchpad support are nearly on par now with the X counterparts. And screencasting is now supported in Wayland on Plasma 520. It's awesome. Task Manager got better over there too.

It's got window thumbnails. And really, in general, the whole desktop is more stable. And you'll like this, Chris. It no longer crashes when X Wayland does. That is an improvement, isn't it? Yeah, and so the task manager is showing thumbnails. You know what that really tells us is that they've come up with a way now

for one Wayland component to get information from another application. One of the big things that's been the problem here is that one Wayland application couldn't necessarily read the other contents of the other application. Right, it's got to be able to go, you know, take a little tiny screenshot and show it to you.

It's a good time to be a Plasma user. Well, really, it's a good time to be a desktop Linux user because our desktops are really dialing this stuff in now. I got a couple of just nice quality of life improvements when I went to 5.20. I had some icon rendering issues in my system tray area. Some icons were essentially overlapping each other. Fixed now.

You know, it's funny when I go to other platforms we don't normally mention on this show. There's other platforms? You know, especially with like, you know, Mac's got a terminal, Windows, WSL2, the Linux story is way better there now. The thing I'm finding I miss the most, honestly, sometimes it's just the desktop because our desktops are so good. We've got all the different kinds you could want

and often they're very configurable. Yeah, and we're getting to the point where these releases are really just whittling off some of the niggly, kind of not quite right or needed another go kind of stuff, or they're positioning us for future technologies. We're getting both of that kind of at the same time. We're in a good maturity cycle for these desktops. Amen.

And 5.20 of Plasma looks like a really, really good one. I'm actually really grateful for all of the hard work the team does over there because it just has been such a consistent performer recently. Of course, everybody's freaking out this week because NFS, after being around for 100 years, got extended attribute support. So, Colonel 5'9", baby.

We're going to have ourselves a little party. But yeah, otherwise, 5'9 release came out recently and looks to be a pretty good release. but that got overshadowed by another announcement that i don't know how many people have caught and i think this could be huge vmware has announced vmware's they have like this fling have you seen this vmware fling have you heard of this no what is a fling okay yeah so i didn't know either so so I'm glad I feel better. Should I be ashamed I'm having one? I know, right?

So a Fling is a VMware program sponsored through the office of the CTO at VMware designed to offer early-stage software to the VMware community. It's not generally available. It's not necessarily production code. It's not even necessarily going to be a product, but it's something that you can submit bugs against. You know, it's something they have an kind of an opener process. So they're just having a fling, not guaranteeing it stays around or goes anywhere, but testing this stuff out, letting users play with it. And now their latest fling

is with ARM hardware. And while they announced support for a bunch of different ARM platforms, And while they announced support for a bunch of different ARM platforms, the most notable is that VMware ESXi is now available on the Raspberry Pi. You can run VMware ESXi on a Raspberry Pi. Now, that's only the 4B, mind you, but still, that's interesting. I happen to know someone you might be familiar with who has a lot of Raspberry Pis and has a lot of workloads on them. Yeah. Funny, because the number one thing that I kind of don't like about my Raspberry Pi empire of servers is that it's all on the metal. I don't have it virtualized. If I was going to have one layer of protection and redundancy and all of that, it would be to have everything running in a VM that I could snapshot or that I could move around.

And I just didn't know if it was possible in the Raspberry Pi. I played with KVM, and boy, was that a rough, rough, rough experience. I'm curious to see how this does. You know, if it's any better, maybe give KVM a shot again to compare. ESXi does have a lot of nice interfaces, APIs, just a familiar world for people already familiar with that from the server space. They do recommend you go with the 8 gig model if you can, though, just to get more features of vSphere. Which I happen to have

one right here. So I think I'll be doing that. I don't think this is something people are going to really use in production. I think it's more of experimentation. It's about learning. So this is a problem that VMware had, and this solves this problem pretty easily. In the past, if somebody wanted to learn VMware, you kind of had to tell them to go buy a $5,000 server to go learn how to do this, right? I mean... Yeah, you had to have stuff to run it, you had to actually have access to the software.

Yeah, even if you had a demo, you know, it would last for 30 days, and you got to license it, you got to run it on a machine that's an Intel processor with a certain base minimum set of features, you got to have some storage, there got to run it on a machine that's an Intel processor with a certain base minimum set of features. You've got to have some storage. There's some minimum. Really designed around those big data center work cases.

And it is very popular in the data center. So this now solves that problem because now you can say, well, go buy this $75 Raspberry Pi. You can learn some of the fundamentals about the interface. You can learn our language. You can learn how to build these systems. Yeah. It's just one more thing that the Pi's good at

if you're trying to learn and set up things and play with all this technology. Huge, but also, also think about this from a data recovery standpoint. So if you have a VMware server that has a big old set of VMware FS disks,

you know, you maybe got a big VMware FS device, something like that, and the host computer dies, the only way you're going to get access to that is to buy another $5,000 server, install VMware, and then mount the disk. Now you can buy a $75 Pi and mount that same VMFS device and get access to that one or two, three or four files that you needed to recover with essentially a Raspberry Pi acting as a data recovery node.

But additionally, these stupid things, I believe, are going to support vMotion, which is the live migration of a VM host. Very cool. So you could live migrate

a host to a Pi and then migrate it off again. And you could do it with load rules, right? So you could say the VM runs on this Pi until the load gets to a certain requirement.

Once we have a certain amount of processing and memory requirements or CPU requirements or disk or whatever it is, whatever the rules are, I haven't used in a long time, automatically migrate that workload over to a powerful x86 server and then really burn, you know, burn the coal and address that load. And then after it settles down, migrate back to the Pi. I can tell you're already going to abuse this. It is interesting. You know, they say this evaluation program is for the enterprise architects who are considering the viability of virtualizing ARM workloads. So there's also this sort of, you know, get people who are used to running it just on Intel machines. Yeah, all right, this can work on ARM too.

Yeah, and I think it's mostly going to be for testing. Mostly for people that are already VMware shops. Yeah, already familiar with it. mostly going to be for testing. Mostly for people that are already VMware shops. Yeah, already familiar with it. And the Zen project recently announced that they're also working on virtualization on the Raspberry Pi and other ARM CPUs. And in fact

to get Zen working, we covered a story recently where they had to solve some really particular problems. And so people are multiple groups actually, are taking virtualization on the Raspberry Pi really seriously. And I say good on them because look where this thing's gone and where's it going to go in another five years? And I'm sure it's going to just be even more powerful.

We're running into more places. It's eating the world. I think you and I should give it a go. You know, we've got some other things coming up for the next couple of weeks, but in a bit we should give it a go. I checked in on've got some other things coming up for the next couple of weeks, but in a bit, we should give it a go. I checked in on it. People are already submitting bugs. Great.

Stuff's already getting fixed. So it seems to be a pretty active project out of the gate already. And I think they're on to something here. I think they're on to something. So anyways, we'll have a link to it if you want to download the ISO in the show notes. You can try the Fling version without a license. It's free for anyone to use. I don't think you're supposed to use it in production.

I don't know if it has any other limitations. I'm not very familiar with the program. But you or I or whatever, whoever, that doesn't even have a license can go download this and try it. So if you try it out there, go to linuxunplugged.com slash contact. Let us know how it goes.

Tell us how it goes. It's just, I mean, you know, like four years ago, I would never have thought that this would happen, that you would ever take the Raspberry Pi this seriously. Right us how it goes. I mean, you know, like four years ago, I would never have thought that this would happen. You would never take the Raspberry Pi this seriously. Right. Yeah, yeah. You know, you're going to be limited, right, in how much the Pi is going to be able to run. You're going to be really limited. And I guess you're going to end up

with just ARM virtual machines. So you might not be able to, unless you could virtualize ARM on the X86. You still need stuff that runs on your architecture, you still need, yeah. Yeah, yeah, exactly. So I don't think it's going to emulate x86. Right. But if you could run it on a Raspberry Pi, you're probably building it, I would imagine, for a much larger production ARM server.

Like they do support a set of ARM servers. And that's probably where you'd see that workload kind of making sense is you build it on a Pi in your office or something like that. And then you deploy it on like an Amazon ARM box or something like that. And you've got some sort of edge devices out there. Which I wonder if that won't be a use case for the Apple Mac ARM machines, if people might be buying those for ARM development for ARM in the cloud. It's a whole new world.

It's a whole new world, Wes. It's a whole new world. We'll give it a go. We'll report back. I don't think there's much. We don't really have much to say on it other than that because it's still new and we haven't tried it yet. So when we try it and we have formed, I think, an educated opinion on it, we'll do an episode dedicated to it.

Maybe see how it stacks up. Sound good? Yeah. Virtualization on the pie. Take two. Congratulations to the folks at VMware, though. I know they've been working on this for a couple of years. They teased it a little while ago and it took a ton of R&D.

You've got to hope and figure they're doing the math that this is going to turn to something big if they put all this effort into it. Linode.com slash unplugged. Go there to get a $100 60-day credit for a new account. That's awesome. Linode started in 2003 as one of the first companies in cloud computing, three years before AWS. And now they remain as one of the largest independent cloud computing companies in the world. And you know, Wes, something that we haven't mentioned on this show before is Linode has been a longtime supporter and sponsor of our beloved LinuxFest Northwest. Sure has. I mean, I can't count the number of conferences I've seen them

with their, you know, green booth setup. I know. And actually, there's been a lot of times when maybe we had a different sponsor on here that I'd walk by their booth longingly and go, I wish there was a way. They look so cool. When we sold JB and I wanted to keep things separate from what was JB infrastructure, what was my personal infrastructure, that was my excuse to go check out Linode. That's when I was like, let's go see what Linode has to offer. And been using them since then. So that was about two years ago, a little more than two years ago.

And I've watched them really respond to a very fast market. So you'd think a company that's been around like three years before AWS got started would be slow to adapt. But that is not the case at all. Like they're clearly following this stuff. They're just always modernizing their infrastructure. They started using user mode Linux and, you know, one data center. And now they've totally modernized the infrastructure. They have 11 data centers worldwide. They have a cloud manager that's super easy to use. Wes was just looking at their GPU compute options. I think

you said you saw a machine that you could have up to four dedicated GPUs in. Oh, yeah. So they got everything from like a $5 a month rig up to like four GPU rig. And for the most part, when we're building a system these days, we've been doing the dedicated CPU rigs. Didn't we do a dedicated CPU rig for the Matrix server? Yeah, we did. Just in that way, it kind of has the performance we need. And it's really nice because everything's backed by native SSD storage, 40 gigabit connections, and they're super crazy fast. Really, the great thing about Linode is you can do the math like this. If it runs on Linux,

it runs on Linode. That's really just it. I mean, they started this whole thing with a love for Linux to begin with. So they've always been a fan of giving you access to everything you want. Every plan comes with their amazing human-powered customer support 24-7, 365. And you know, not just LinuxFest Northwest, they're also a sponsor of the Kubuntu project, which is a favorite here on the show. And that's just really great to see that too. So you can support the show. You can get a $100 60-day credit, and you can spend something up on Linode that matches your needs. You start by going to linode.com slash unplugged. That's the URL that tells them, yep, I heard about it here on unplugged. And that's the one that unlocks the $100 60-day credit for a new account.

They have a 99% uptime. They have a really easy to use DNS manager that allows you to easily switch your domains to your hosts that are running on Linode, something we've had to do a couple of times. It's really pretty simple and straightforward though, so don't worry about it. You'll be able to handle it. And if you want to use Linode to learn, they've got a lot of great options for that. But if you just want to get something started, you can do a one-click deployment and get going. They provide virtual servers that make it easy and affordable for you to host anything in the cloud. So check them out. Linode.com slash unplugged. Thanks to Linode for sponsoring this show. And thanks to everybody who supports the show by visiting Linode.com slash unplugged. Thanks to Linode for sponsoring this show. And thanks to everybody who supports the show by visiting Linode.com slash unplugged and

getting that $100 60-day credit on your new account. A little more community news. I think this is pretty great. There's a project that we followed on the show before called Nebula, which, if you recall, is a software-based overlay that Wes and I just went bonkers over. It's really simple to operate. It's resilient to devices coming and going. It's a software-defined network like the one you've always wanted. Lightweight, modern crypto, straightforward to use. And it just focuses on letting you build the network you want,

completely independent of where your devices are. So they could be across multiple providers. Slack uses this to just essentially bridge different public cloud providers and to create one flat network. Yeah, that's how they connect their production infrastructure. We've talked about it before here on the show. Yes, episode 329, Flat Network Truthers.

Go check that out if you want more info, if you missed it. And we had a chance there to talk to one of the founding developers about it and do the gushing that we did because we loved the project so much. And now it appears not only do they have apps for mobile that are official. We have some links in the show notes. You can run Nebula on Android or iOS to be part of that flat mesh network.

But more importantly, the two founding developers are going full time with the project. They left Slack back in February and are now doing this as a full time gig. They've launched a new company. Yeah, Defined Networking Incorporated, and they're focusing on Nebula full time. Man, they're still like working with their former colleagues at Slack. And I think really grateful just that, you know, Slack did this in an awesome way where they got to build it, test it, have it work internally, open source it, and now they can shift to just doing this all the time. Wow. I mean, yeah, it really seems like it's one of those ideal situations where they talk to the people at Slack and they said, we have this thing, it's taking off.

This is what we need to focus on. And Slack said, yeah, but we're using this in production. We need to keep working with you. And they came up with a way to amicably split. And they've done it well and continue to test new releases on Slack's production network. That's crazy. So good for us. And it just seems like such a neat, like this is possible partly, I think, because it's open source, right? You don't have to worry about this stuff. Slack can contribute to it internally from their own engineers still.

This new company can focus on it, and Nebula just exists as an open source project. Yeah, I need to play more with it. Especially now that there's upgraded mobile support. Yes, they've been toying with that for a while, and they actually are tying in on the iOS level with the kernel VPN

module, so it's like full-fledged. It'll work with every app. It's really well done, like the WireGuard support is. Yeah, exactly. It's funny that these two open source, free software, VPN-like solutions that we keep talking about, Nebula and WireGuard, they are really first-class mobile experiences too. The developers were keenly aware that they needed to be on mobile. Doesn't that show you the times? Yeah. But it's true. I mean, like that's a huge benefit. And the whole thing about Nebula is having this one flat network of all your devices. But if the device I use most commonly can't be part of it, it's not maybe as useful.

Yeah. Like, I want to get to my notes or I want to back up some photos from my phone. Or stuff running at your homeland. If you haven't checked out Nebula, go listen to Linux Home Plug slash 329 and get a taste of what it could offer. It could solve some problems that you may have been kicking around. We found it very useful. Congrats to Brian and Nate. Totally.

Very excited for them. All right. So last story of the week. And this is one that's been getting some grief. And the headline reads, the AMD Radeon graphics driver makes up roughly 10.5% of the entire Linux kernel. It's a bit of code shaming through size is what's happening here at scale. And looking at the 5.9 kernel, there's about 20.49 million lines of code and more, right, when you include comments. Yeah, you know, like 3.58 lines of code comments, 3.72 million lines, just blank lines,

something like 27.81 total lines of code in 5.9. I mean, for a while, the AMD GPU kernel driver has been huge, making it the largest entry kernel driver easily. But with Linux 5.9 comes that additional 2.16 million lines of code, plus, you know, even more than that when you include comments and line spaces and all of that. And a lot of it is auto-generated headers for GPU registers. That's what 1.79 million lines of the Linux 5.9 code are. It's just auto-generated header files. But, I had an actual, like, insightful comment I came

across on Hacker News. Like, when does that happen, right? It happens occasionally. Well, it happens when it's a comment that's made by a kernel developer. So, this is a comment that we'll link to on Hacker News from Broodbucket, and it goes, My two cents as a kernel developer. The driver is pretty abominable compared to the code quality of the rest of the Linux kernel. However, and I think this is the point we should all take here, having a GPU driver not just

be open source, but upstream in the Linux kernel is a gigantic deal. That's why I don't like the code shaming. I don't like it when we look at the lines of code and we, oh, look at all these lines of code, and then we kind of code shame. When it's like, from AMD's position, it's like, well, you wanted us to open source the code and contribute it upstream. What did you expect?

It's got to be what they're thinking, right? Like, so we finally do what you ask, and now you're bitching that it's too much code. Well, and, you know, there's some cultural differences. I think a lot of this hardware or the embedded world, you know, you just have this duplication. You've got similar boards that are slightly different. You copy the code base. You make the changes you need.

It's not this, like, highly polished Linux kernel type open source atmosphere. So there's a little mix match. And to kind of bounce off your point a little more, they're revving the GPU one or two times a year. There's a lot of change. Right. Much more than what you see in the x86 or even in the ARM world. This commenter goes on to say, kernel development takes a long time. We have millions of lines of code in the AMD GPU driver.

And if every one of those dealt with the lengthy review process, it would have never made it into the tree. So it's a necessary evil, they say. I do wish they would clean it up. I've even sent some fixes upstream for them to fix things up where there is massively, largely duplicated files. And that kind of stuff just doesn't fly anywhere else in the kernel. But it's kind of like the compromise you make to get something as complicated as a GPU driver that covers such a wide span of not just architectures,

but different vendor implementations of those boards, it's a massive chunk of code. And perhaps gives us a little bit of insight into why NVIDIA keeps its cards so close to its chest. Right. This is something that the company released after they knew it was going to be public, and they probably went through a process to...

Oh, yeah, right. Yeah. To vet, go through, maybe clean up. Right. right and so can in nvidia having not done that work uh probably is in a lot less publicly polished state yeah well and you know the other the other part here too sort of unsaid is you know this commenter submitted a change in a fix so like having it in the kernel while it doesn't mean it's where we might want it to the extent possible, like fixes can be applied. The kernel community is in control. It would be funny though, in a alternate reality where NVIDIA has just all of a sudden released their code and it was also about this size. And then like 20% of the kernel is just GPU.

That's ridiculous. I mean, it's already a huge chunk of drivers. Yeah. Really? Seriously? I feel like people don't understand the complexity of like what you have to do to deal with another processor on your computer that has to just do stuff. And unlike normal, you know, CPU processors, which if you look at the kernel code is not an insignificant chunk in its own right, based on all the different architectures and whatnot. But GPUs have to deal with multiple different instruction interfaces. You've got GL, you've got GLS, you've got Vulkan, you have Direct3D.

Like all of these things happen in the GPU. And I'm not even talking about the compute stuff like OpenCL, you know, and SpirV and all that other stuff. Yeah, fair point. And they are essentially like another computer in the computer. And they are a constantly moving target. Plus, then you have all of the management code that is necessary to just get those frames onto the screen for the different operating systems. I mean, there's a lot to it for sure. So I just I'd say, yes, it does seem like a lot of code.

Clearly, it does seem like there's some room for improvement. But I don't know if I like the idea of campaigning for years to get code released and then code shaming when they do it. It just doesn't feel quite right. But I guess there is probably room for improvement, too. Mr. Payne, there's some room for improvement around here. It's a mess. Let's do a little housekeeping around here. Minimech, rumor has it that there is something special going on for the Luplug this Sunday.

Indeed it is. In fact, we would like to talk about browser experience, how you harden your browser, what browser do you use, and how do you handle cookies and stuff like that. So we try to get a group together that's talking about their experiences with browsers and with extensions they have. And it would be cool if you joined us. And even if you have some input, we would be really happy to hear from you. Yeah, you can find the LUP plug on Sundays at noon Pacific. It's on the Jupyter Broadcasting calendar.

And we have the Mumble server info at LinuxUnplugged.com, link there at the top. You get that set up, you join in the lobby, and you hang out with a bunch of other Linux-loving users. And this week, talk about browsers. I think browsers are actually once again going from boring to interesting all over again, so it's a good time to talk about that. I also want to suggest you consider grabbing the Jupyter Broadcasting All Shows feed. You never know what we have in the works or what we might put out in that feed and get every show.

Place to find it. Yeah, everything. It's like the ultimate JB experience. The Jupiter Broadcasting All Shows feed is in your podcatcher of choice or linked on our website at jupiterbroadcasting.com. Let's check it out. I also, we should just put a link in the show notes. Let's do that. We'll put a link in the show notes too. And do that. We'll put a link in the show notes too. And we'd love to have you subscribe and get everything that we put out. Because we have learned many things over the years. And we've learned kind of what Jupiter Broadcasting's focus is.

And we're really proud of everything we produce. Like if you're not listening to self-hosted or Coda Radio or Linux Action News, you're missing out on some great content. That's all in the All Shows feed over at jupiterbroadcasting.com. All right, so we wanted to try something new. The exit interview. After we've used a distribution for a while, instead of giving you the review when we first try it,

we're going to give you our thoughts after we've used it for a little while. But I wanted to do it for Pop! OS, which is the current version as we record is based on 2004. God, this is really embarrassing. This is embarrassing for two reasons, actually, for me to talk about. Number one, I shit on it when it first came out. Just going to say it. I did.

I did. I didn't think it was a good idea. You're skeptical. And then I ran my mouth like multiple times on multiple shows about it. So I got it. I got to eat my hat on that. I don't normally get get it wrong so badly, but that's when I got it wrong on. And I got to admit that it's also embarrassing because System76 is a sponsor of Linux Action News and Coder Radio. That may cause people to think that is influenced my thoughts.

But those two things have no bearing on what my opinions are of the Pop! OS desktop itself and the experience that I walked away on at Dell. So I loaded it on this Precision that I am going to give you guys a review, but I loaded it on this Precision because it has a badass Quadro NVIDIA card in it. It's got an NVMe SSD.

It's got a gorgeous edge-to-edge 17-inch 4K display. Beautiful, right? I mean, it is the creme de la creme of Dell Precision laptops. And I wanted to see what the System76 Pop! OS experience was like because they have focused a lot on high DPI. And the general, I would say, kind of, you know, word on the street about Pop!OS is it's great if you want a game, which is exactly what I wanted to do for a little bit. You know, part of the review is to put it through gaming paces. And so I spent a couple of weekends trying

different games on it. It just seemed like Pop!OS would be a good way to go. As I do, as I start using it, I run into things or I start having questions. And I have the privilege of being able to call up the people at System76. And so I did just that and I asked them some questions about it and got a few clarifications on strategy and stuff. And I thought I'd share some of those experiences as I'm now taking it off that Dell. I'm going to remove it. So I wanted to just share my thoughts as I transition off of Pop! OS. Yeah, you spent some time with Pop! Really lived in it. Yeah. This, I swear, is not hyperbole, but it sure sounds like it when I'm about to say is

I think with Pop! OS, they have something no other distribution has. And if I were to boil it down into a couple of things, then I'll expand on that. It is limited scope of what they have to address.

And it is sustainability and staffing. Sustainability and staffing. Let's go with those two. Honestly, not a lot of distros even have that. They don't necessarily have full-time people. It's the exception of the exception. Like, we can count on our one hand what distributions have full-time desktop staff. Yeah, right. I mean, like many open-source projects, there's a lot of volunteer time or off-hours time. Or a lot of integration of upstream, right? That's essentially what the volunteers are integrating upstream, and it's the good work of upstream

that is making these distributions possible, and then they, these volunteers, will theme it, or they will modify it, or they'll bring some of their own special sauce to it. Pick your chews, put things together, yeah. But they're not working a nine-to-five. They're not commuting into an office to go work on that thing. They're not building software that is at the scope of, like, say, the System76 firmware tool, right?

Even if System76, say, has three people working on Pop! OS, that's three more people than just about any other distribution has when you look at all the distributions out there. But additionally, because they've decided to ship Ubuntu, well, the desktop team at Canonical is doing some of the heavy lifting there, and they're shipping the heavy lifting that the GNOME desktop team is doing, right? System76 hasn't come up with their own desktop environment. There's no, you know, there's no System76 desktop. They have taken the hard work of the giants underneath them, and what's left is something that their small team can easily address. Yeah, right, and they're working directly with customers.

They've got an interesting sort of vantage point hearing people using their desktops and figuring out where maybe there are some rough edges, and they're in a position to actually not just hope Upstream fixes it, but fix that themselves. and they're in a position to actually not just hope Upstream fixes it, but fix that themselves. Right, the support department acts as sort of a unique feedback cycle to the development features that go back into Pop! OS. Again, that's unique. But additionally, when we talk about sustainability, System76 has come up with a way to make it viable

to release a Linux distribution for free to anyone, and they distribute it via DigitalOcean Spaces. So even downloading it costs them. And it's fast, but they're paying for it. So they have come up with a scheme in which they can give the distribution away for free but still have a sustainable funding model because of the benefits and features it brings their laptops. And while it's a small team, they can do things like work on auto-tiling. And because they have time to do the research, they'll base it on something like TypeScript, which is much more sustainable and portable. Well, yeah. I mean, if you've ever used

the GNOME flavor of JavaScript, it works just fine, but you don't have all the niceties of modern JavaScript by choosing TypeScript. Not only do you have the nice TypeScript compiler to help you out, but you've also got all the abilities of modern JavaScript tools, access to library ecosystems if you need that, and a much friendlier development experience. If they were maybe choosing other features, maybe I would have a different opinion, but what they focused on is auto-tiling that seems to work for mortals.

They've brought fractional scaling to standard GNOME shell. Yeah. They have enabled just recently hybrid graphics for external displays, which is kind of a big deal for a lot of customers who are buying nice GPUs. It's workstation features. Yeah, right. I mean, they realized that the firmware updates were a thing that needed to happen, that customers were going to want addressed in a good way,

and they were early on that game. They can work on these 10% gap closure features because they're building on all of the work of the other projects, and then they can focus in on that kind of stuff. If I could wave a magic wand and make Pop!OS anything I want, I might prefer to have it based on something like Arch. Oh, now that's an idea. But a pretty good second desktop,

it would be Ubuntu LTS or an Ubuntu base. Right. I mean, I also, I feel very comfortable with it. I can have all the benefits of Pop, but I still feel like under the system, you know, I can do whatever server side things on my desktop I need. And so that actually makes it a pretty reasonable choice too. If I'm not going to run Arch on the desktop, I'm probably going to run an Ubuntu flavor

or something like Pop OS because it just, well, it just makes running a lot of applications that I want to run simpler. It just, it's one step removed. It's a well-supported environment, yeah. It's one step removed from being on one of the commercial platforms, essentially. And so you get those benefits in Pop as well. And then you look at some of the other choices they've made, like the installer. They worked with Elementary OS. They didn't go out and create their own proprietary

installer. Instead, they funded development for Elementary OS to create a free installer that they collaborated with Elementary OS to modify it for Pop! OS. And now Pop! OS has a better installer than Ubuntu does. It's real nice. The Pop! Shop is better than GNOME software. And it's just simpler and cleaner and faster.

And it looks better too. Yeah. Everything feels very integrated. It feels like a holistic, you know, complete experience. So you've got something unique here in that you've got dedicated development time by people where it's actually their job

and they're held accountable and they have features they need to ship to match product releases. So there have features they need to ship to match product releases. So there's actually that drive to ship. You have a sustainable system where they can give the distribution away for free to people that want to run it on their own hardware, but yet it still returns enough value for them on their hardware that they can justify the cost to make it. And you have them developing features that close the gap, that add usability and pro features like keyboard navigation of GNOME Shell is a lot better on Pop! OS. You almost don't even need the mouse now.

And the reality of that is that it's while developed for Pop! OS, it is being upstreamed. It's available. It's open source code. So they're also doing that aspect of the good open source citizen aspect. When you take any one of those slices, like auto tiling or independent development staff or any of that, you slice it on its own. I mean, it's neat. It's good. But when you total it up in totality, I think it makes Pop! OS a very unique distribution that I didn't fully appreciate until I saw a few iterations of it and then spent half a month with it or three weeks with it and realized, oh, oh. It does feel like it's very broad appeal in a way that sort of surprised me as well. Like I'd feel comfortable letting my mom use it, but I'd also be entirely comfortable having it as my, you know, workstation desktop in the office. Yeah. That's what I'm thinking about. You know, eventually when I do

get my own laptop, I kind of assumed I would just go Manjaro, Plasma, and watching Gnome Shell, we, you know, what it started with, it started when we started beta testing Fedora 33. And I was like, oh, oh, hello, Gnome, it's so nice to be home. It really felt like going home. And I thought, okay. So then I actually put Fedora 33 first on the Dell. I can't remember. Do you remember why? I think I was troubleshooting some sound card issues or something completely unrelated. And so I tried out Ubuntu 2010 and then I tried out Pop!OS 2004 and stayed. It's funny because I just recently did a snapshot of 33, 2010, Pop then I tried out Pop!OS 2004 and stayed. It's funny because so I just recently did a snapshot of 33, 2010, Pop!OS and Pop!OS is the one I stayed home with, Gnome Shell

at the dark mode. And you guys, if you've listened to these shows for a while, know that we've gone on about like trying out Regolith and we've done different discussions about tiling window managers and I always end up just going back to wanting a windowed environment. But having the auto-tiling built in, where I turn it on, and then about once a day for an hour, I turn it off for that particular work thing. And then all the other times, I just click a button in my toolbar and it's on.

And it's so much nicer. It's so much nicer. Plus, they have an option where you can highlight the window that's active with like a border. And that's just a little bit quicker for when I come back to my machine, I can just look immediately at the screen. I know what is the active windows. I don't start typing in the wrong chat box. It's very practical.

I really like it. I really, it's like the first auto-tiling that's ever stuck for me. It just makes it better. It doesn't get in your way. It's easy to turn off if you don't want it. It just enhances, you know. I can't believe they got me. Easy to turn off if you don't want it.

It just enhances, you know. I can't believe they got me. Because, you know, the thing is, if I could have anything, I would just have all of this be in mainline Ubuntu. You know, that would be the mainline Ubuntu just has all this stuff. But that's not the world we live in. It's just not the reality. I also, I have to be completely frank, I prefer that PopShop defaults to DEBs. You know, it'll go for like when I install Telegram,

it installs the Deb of Telegram. And the reason why I like that is because when I'm sending and receiving attachments, I just want to be on my actual file system. And I like the faster startup time of the Deb file, and I launch Telegram every day. So I prefer that about PopShop too. There's just little things. The installer is a little bit better. Auto-tiling, the theming is better. The software install, they install by default a very practical tool to just be able to double-click and install a dev file without having to launch the entire software center. It's like these little tiny things that all really add up, plus the development focus, I think make it kind of a special contender. focus, I think make it kind of a special contender. I mean, I'm not switching anything to it today, but I think when I do get my next laptop, I think I'm going to go pop. I think that's going to be the way of the future for me, especially if it's one that has a video cut in it. If I'm going to do a little gaming, I really went well, super solid. I know you gave it a little bit of a go

too. How did your Steam experience go? Yeah, that was actually really pleasant. You know, I just installed it right from the Pop shop, loaded everything up, and it played very nicely, much better than the previous gaming setup I had. Mm-hmm. Yeah, yeah. I was just thinking about it

for my son. If I was going to reload my son's machine, I think elementary has worked so well, but I could see elementary for some of my family and then Pop

for the more technically inclined members of my family. That'll probably be the distro I start recommending to people when they ask me what distro should they switch to from Windows. This is a big change for you. Yeah. I'm pretty impressed.

And I think part of what my calculus is is assuming that System76 takes this trajectory and runs with it, right? And kind of builds on where they're going. Yeah. If they fade on it, then my opinion of this is going to change. You know, but right now, if they continue their current trajectory or even add to it, I mean, then we're really talking about something special. And I was an idiot. I was an idiot for getting it wrong. Well, I think it was just, you know, we some of the things that have worked out in their advantage of finding the right levels to contribute.

I think that's where we were skeptical. We're kind of like, well, do we just need this additional theme or this customization? But with the work they've put in, it really made a bigger difference than I would have guessed. Yeah. So I really recommend you check it out. So I guess our exit interview is you are a better performer than we expected in just about every single category. They have also some slightly different software selections available in the Pop shop that I think are kind of nice. And all of that just made it really simple to get up and going and just keep it all up to date really easily.

That's it. It feels tasteful to me. I think you can really tell that, you know, the folks at System76 use and love desktop Linux, are in all about that and are using Pop as they're working on Pop. And so it's just feels already well lived in and makes me comfortable. You and I were speculating, like, will this long term pay off for them? Will it create like the iPod style halo effect where today you buy an iPod, tomorrow you buy a Mac? where today you buy an iPod, tomorrow you buy a Mac? Will Pop! OS users like me who are running it on Dell hardware,

would we be more inclined to buy a System76 machine? I think I've had a great time here on Pop! Maybe I'll just get the real deal. I could see it, but I'm a little more skeptical because it's not like the brand is smacking you in the face anywhere. I don't even see it really. So maybe. I mean, I could't even see it really. So maybe. I mean, I could see the argument that it keeps the System76 brand

forefront in your mind. So maybe when the time comes and you're thinking, hey, I want to buy a computer. Well, they do a good job with their operating system. Yeah, maybe. I don't know.

I'd be curious to punt that one to the mumble room. Does anybody have any thoughts? Does creating a distribution that you end up running on HP hardware or something does it make you more inclined what well i don't know i don't know toshiba maybe you have an old toshiba satellite i don't know what you're rocking uh but does it make someone more inclined you think to maybe want to purchase the hardware later when time comes i don't know so like um if the aspect of it being made by System76 might be a little bit more prominent in Pop! OS, it would it may introduce the other hand, I don't think that's the point of Pop! OS.

And it kind of shows in how they think about it and how they release it and how they brand it. I think they make Pop! OS, you know, just to so that they have an area in which they can drive their own opinions about how a desktop experience should evolve and how they feel it should support their users. And that itself is shipped on their hardware because, of course, they make hardware. But they're totally okay with supporting other hardware as long as it's supported by their upstream distribution because the point of Pop! OS isn't to drive people to System76 hardware. The point of Pop! OS is to drive people to their opinion of how a Linux desktop should function. Sure, and maybe in the future, drive them to a Custom76 keyboard because I can see the way this is going. So if anybody tries out this auto-tiling extension, one of the things I recommend you do is bring up the keyboard shortcuts

and look at what you can do in GNOME Shell now from the keyboard and tell me that that doesn't get extracted out into the future when they ship this System76 custom design keyboard that happens to work great with Pop! OS. I mean, it's going to be a perfect combo. I mean, if I'm running Pop! OS at that time, I might buy the keyboard, you know? Back in the day when we had the computer, it was like, oh, cool, I can install, for example, Ubuntu on it. You put it on and there was pretty much always one thing that didn't work. And that annoyed me. And it also was one of the things where I couldn't promote Linux to other users that were like, yeah, but there's always something, little things that don't work. And you always have to tinker with it. And what PubOS is doing is standardizing hardware where they can provide support.

So this way you can promote it to a family like, okay, this is going to work and you don't have any problems with it. If you've got some problem with it, you can report it and they will fix it. And this is also where companies get more and more comfortable by purchasing hardware from, for example, Pop!OS with the service they can get with it. And this way there is more and more adopting of Linux because there's standardization of hardware with software. Right. It kind of means it's like a safe bet in a way, because you know that the vendors integrated it, but also they have a support system for it too. A long time ago, there was also on the show where someone tried to sell the software and they were like, yeah, it's free.

And companies didn't buy for it. And when they sold it with the support and the software was free with the support, companies started buying it. Yeah, that's an old lesson there. So there's two things that this makes me kind of come to, like if we were to shoot this way out. Number one, So there's two things that this makes me kind of come to, like, if we were to shoot this way out. Number one, is System76 more compelled to brand the UI in a way, perhaps, to bring their branding more forward to the user? Could that show up? You know, and we've seen this, we've seen a mistake here in the past, like, Canonical never really got over the, putting the Amazon links in the search results. I mean, but, you know, is there something they could do that would make System76 more front of mind? You could see going that direction perhaps one day.

Maybe not. And if Neil's analysis is correct, then there would really be no motivation for that. But my other thought on this is, is this a play we start to see become much more common? We're starting to see it now with announcements of distribution A with hardware Y and a partnership between the distribution and the laptop maker or whatever. We're seeing a lot more of that. And I wonder if that isn't sort of a validation of hardware integration with software does provide extra value to the customer. I think we're seeing that when you start to see all these projects that have announced laptops. Just this morning, Slimbook, which is a Spanish laptop maker or

PC maker, because they actually do make desktops as well. They announced they're now shipping the Slimbook Essential, an affordable sub 500 euro laptop. They're shipping it with Fedora and they do the work and they work with the distributions to kind of, you know, make sure that their hardware and software work together. And I think what this kind of goes to for what you're saying, Chris, is that the key issue that has always been the case, and I think even Linus Torvalds has said it like back in the days of when Linux World was a conference, you know, he's always said that the reason that Linux doesn't win in the desktop is because there are no preloaded Linux on computers.

That's changing. And it's happening at big vendors and small vendors. You've got Tuxedo, Slimbook, System76, Dell, Lenovo, Hewlett Packard and others. Like it's just starting to happen everywhere. It's happening. This is CNN Breaking News. Dan Johnson in the mumble room has some breaking news for us as we wrap up the show here.

Manjaro Arm 2010 has been released. Dan, do you have anything you want to share with the class on this fresh new release? Well, not much. We've updated the kernels and mesostacks on most of the devices. Hey, that seems great. I'm looking at it right now. It looks like a pretty nice plasma image is available. So are you spinning these images yourself? Yeah. Oh, thank you for all of the hard work. This is a lot of images you're spinning. You have Sway. Of course, you have them for the Pinebook and Pinebook Rock Pro 6.4.

You have it for the Raspberry Pi. This is a big matrix of download options. Yeah. How long does it take you to do all of that? Well, it's clocking in at about 12 hours now. Oh, well, thank you for doing the good work. We'll have a link to that in the show notes so people can go grab that. Before we go into the feedback, though, I want to thank a cloud group sponsoring this episode of the Unplugged program. They've got

a course that might be good for you, especially if you are administering a system with SystemD. They have SystemD management for Linux. We'll have a link in the show notes. This course has 5.8 hours of content, 40 total lessons, and eight hands-on labs. Yes, it is a course designed to help you demystify SystemD. You can get things done in SystemD perhaps today already, but did you know there's more and more added to SystemD all the time? There's a lot to learn.

So you can take this course and learn how components work together and how to configure, manage, monitor, and troubleshoot SystemD. Check out the link. We'll have it in the show notes at Linux Unplugged slash 375. And thanks to a Cloud Guru for sponsoring this episode of the Unplugged program. Go check out SystemD Management for Linux from a Cloud Guru. All right, Mr. Payne. from a cloud guru all right mr pain after the show last week you and i had to think about the feedback and picks segment and i like

what we came up with we got two emails this week and one wicked awesome pick like so so cool uh but i'll save that first we get to the emails This one came in from Eric, and he was listening to the back catalog. So I guess he just recently discovered the network and thought, I'll start from the back catalog and listen up. Hey, that's a great idea. That is kind of a fun idea. That would be interesting. By the time he gets here, I wonder what he's going to think about. Where he'll catch up.

About all this stuff. But anyways, he's in 293 right now. So this is 375. Some work to do, but good work. And he said that in that episode, we were mentioning PowerTop and TLP, which is the power management solution for Linux. He said, I'd never tried TLP before. I installed it thinking it would give me more battery life. Well, not only did it do that, but I also fixed a really annoying issue with my Wi-Fi where it would randomly disconnect from the network.

Nothing I ever found on the Internet and research had fixed it, but somehow after I installed TLP, my Wi-Fi was fixed. Well, how about that? Yeah, we were scratching our heads about that. It must have been some sort of power management rule that was knocking his Wi-Fi off, like maybe putting his card to sleep or something. And TLP actually changed that to be less frequent, perhaps. I mean, that could be, right? Yeah. I mean, that sounds plausible.

It does, but it's not the kind of thing you would expect. And it really shows you how interlinked all of this stuff is. And you never know. Trying one of our random suggestions on the show might just improve your system. You never know. Or the reverse. Or the reverse. Wetzel writes in to us about Chromebooks and education, something I think you know a little bit about, Chris. Yeah, all three of

my kids are using Chromebooks right now. Wow, okay. Well, here's how Wetzel handled that. Hello there, long-time listener, first-time emailer. I was writing to talk about kids using laptops, specifically Chromebooks, in school. As we all know, kids all over the country are going back to school. Now, whether it be in person or online, my oldest just started kindergarten this year. We all have computers at home, including him, all of which run various flavors of Linux. We've got a Linux household here. He's never used anything else.

So when he was getting ready to start school and they announced that they were going to be giving all students Chromebooks, I, of course, did not agree with this at all. I know, right? I mean, talk about, I know this sounds weird, but I would have almost preferred they would have had them use iPads. I just, because like the Chromebook means the Google account and the Google account, it's like you get them in. You're really starting your whole online identity. And it works. My kids think everything web first and Chrome first, like even video editing and picture editing, like they just browser first everything. I mean, it really works. So Google's very clever in pushing this.

So yeah, when I heard that, I was so disappointed. But, you know, as a parent, there's so little you can do because this is how the school does everything now. Like they turn their papers in this way through the Google stuff. They do the classroom meet tools, like everything is powered by this. And that's how they're doing the entire Rona homeschooling. Oh, yeah. Okay. So when Wetzel ran into that, he writes that I called the school, I spoke with his teacher, the principal, and even the superintendent and explained my concerns. I also told them that I would be more than willing to provide an alternative for them to use.

And that's what it sounds like he's been able to do. So it might still have the Google account integrated, but at least, you know, instead. It's on OS that he has some oversight on at least. Yeah, his son's using a ThinkPad T420 with Kubuntu installed. So how about that? Perfect. That's great. So I could have done that, I suppose.

I could have petitioned the principal, but I did not. I relented. Yeah, and I think that's really the heart of Wessel's message here is, you know, if you care about it, if that's the right thing for your family and your setup, hey, it doesn't hurt to ask. Yeah, I guess that's, he says, for anybody that has a kid in school that got hunted at Chromebook, call the teacher administration and talk to them like I did, you know, and maybe it will work. You know, you can ask for a special accommodation there.

You might be surprised once you hear in response. I wonder, though, if they still end up having to have a Google account. I bet there's got to be some integration, right, if there's online management. Because they're using Docs and everything. Meet, I mean, they're in Google Meet for the school sessions. So I don't know what you really gain other than having full control over the os which has its advantages but my solution there is they just use the chromebooks for school work only only yeah yeah and you know their school property too they're not bad little devices but just not what i what i would have

picked all right speaking of pics how about this how about a system that auto magically manages different tiers of storage for you so, imagine you got yourself some fast storage. And if you access a file a lot, you want it on the fast storage. But then maybe you got your files that sit around. Like, for me, I have movies that will sit around that maybe get watched once every five years, right? You still want them, but you don't need them much. I don't need them on the most precious storage either. And this is where our pick comes in.

It's called AutoTier. It's a pass-through fuse file system that intelligently moves files between storage tiers based on frequency of use, file age, and tier fullness. And of course, anything like this, you're going to want something that has a very easy to understand and read config file. And I think that's where AutoTier really nails this.

You just define two lines plus the definition, tier one. And then you just define another tier and you just give it the mount point to that. And then another tier and you give it the mount point to that. And you can tell it what your watermark is, which is essentially your waterline in a tank. How high do you want the storage to get?

And you can say, I want this water line at like 70%. And then once that tier reaches 70%, it's taken out of consideration. And then it just sort of acts merging them together. So you get like a transparent view. You don't have to see how all the tiers work when you're actually interacting with it through Fuse. I've not extensively tried this, but I'm definitely going to give it a go. I mean, it's pretty neat.

And I could really see you could have online disks that are, you could even, I mean, you could even do something like where you mount some sort of like backblaze or glacier storage, right? And that's another tier in this whole thing. Combine this with like our clone or similar, yeah. Or even a USB, like a large USB disk. Alex sent me a link this morning for a 12 terabyte external USB disk for $175. I mean, that's not bad for 12 terabytes.

Yeah. And you wouldn't want to run, you wouldn't want to be streaming all of your Plex videos or MB videos or Jellyfin, but you would definitely maybe be okay with things you haven't accessed in two years or something being migrated off to USB storage.

That seems totally fine to me, especially if it's just doing it all in the back end when my system isn't super busy or on scheduled times. If you're on Fedora, you can basically just DNF install this sucker. Otherwise, you're going to have to do some building. But when it comes installed,

when you install it, it includes a systemd unit file so you can set it up to run automatically. Or you can just run the autotier command and do what, like a one-shot, spread things out. So you could just do it from time to time if you prefer to do it more hands-on and not have it automated. Or you could schedule it to run automatically and just take advantage of the systemd service. That's so cool.

It is really cool, Wes. I kind of wish it was built into the OS. I kind of, you know, like using Fuse and whatnot. That would be the next layer, yeah. Wouldn't it? Wouldn't it? Also, a bonus pick, because we love the folks at AntennaPod, the free Android podcasting app has released a brand new version, version 2.01 with over 20 volunteers.

And I think the team's pretty proud of this release. I think so, yeah. They got a new logo, refreshed user interface. Hey, here's a nerdy one for us. Support for chapter images. Yes. There's only a couple of podcast players that have that. Right. That's nice. They also have included a new set of app shortcuts, allow users to go directly to their QR episode from the home screen. So tap it there and it goes right to that. So you can have a Linux unplugged icon on your desk or on your Android screen and just go right to it. It's also available in 17 languages.

Stop it. I just love that about AntennaPod. You know, it's so cool that we have such a nice mobile podcast client that's open source. Oh, that is so neat. You know, this new logo came after a call to the open source design community and they received 12 proposals of which two were selected and then put to a public vote. Wow, over 6,000 users participated, so hey.

That's so cool. Good for them, good for them. Yeah, and we don't talk about AntennaPod enough because it's just one of those apps that just works and you don't really think about it, but when you use some other podcast players, you realize it's got features the other ones don't. It's really, yeah, if you're doing something slightly off the beaten path, it's the way to go.

And as podcasters, we really appreciate that. Also, thank you to our core contributors, the members of this show that keep us independent forever and also get two options, a limited ad feed or, if you're so crazy, an extended live mix that is pre-edit. It is every flub, every mistake, but it's also what you get before and after the show that never makes it into the final recording. Honestly, sometimes we cut topics too because we try to keep the show to a certain run length. We're not intentionally excluding that from the public release. It's just a production call and that stuff is also in the live release.

It's all there. You can become a member at unpluggedcore.com and we really appreciate it. You're making this show possible. We'd love your feedback at linuxunplugged.com contact. Reading your emails and whatnot is a big part of the show.

You can get links to everything we kind of talked about or maybe implied we talked about or thought we were going to talk about at linuxunplugged.com 375. And last but not least, we'd love to have you join us live. Oh, please do. It's a great experience. You get to participate either in the chat room or in the mumble room if you so are inclined. And you can find a link at linuxunplugged.com for how to set up the mumble, how to get in there. And then don't

forget, once you have it set up, you can also participate in the LUP blog every single Sunday. It's like just a great hangout. And I haven't been able to make it as much now. I've been feeling sad because I've been doing Linux Action News. But I can tell you, there has been some great conversations. You get great project ideas if you're stuck on something or if you're chewing on a piece of hardware. And now, this Sunday, you can also talk browsers. And we will be back next Tuesday, noon Pacific, 3 p.m. Eastern, over at jblive.tv.

Get that converted at jupiterbroadcasting.com slash calendar. Thank you to our mumble room. Thank you to our chat room. Thank you to you, Wes. Well, thank you to you. Also, thanks to everyone out there for listening and sharing the show with somebody. Word of mouth is the number one way podcasts get marketed. Pretty much nothing else works. So if you know somebody who would like this show, share show with somebody. Word of mouth is the number one way podcasts get marketed. Pretty much nothing else works.

So if you know somebody who would like this show, share it with them. Thanks so much for tuning in to this week's episode of the Unplugged program. We'll see you right back here next Tuesday's Tuesday's

Tuesday! We have forever on Coda Radio discussed the Oracle and Google lawsuit. And I want to play this clip that I found for Coda Radio discussed the Oracle and Google lawsuit. And I want to play this clip that I found for Coda Radio. It's Sun Microsystems from 1995, really just top of the world. You know, ZFS perhaps was a twinkle in their eye, and Java was going to be this thing that took over the world. And they're actually very forward-looking, considering how old this clip is. When you think about where the rest of the industry was, it's great.

So I thought this would be a fun little post-show thing to listen to, so check this out. See how retro it is already? Oh, yeah! The first 25 years at Sun have been just huge amounts of innovation. We popularized TCP IP. We stunned the world with NFS, we delivered Spark and Solaris and Java and all the other new technologies that you've seen

Sun deliver and we're on six billion devices. Sun started with a very simple idea which is to take open standards on the software side and combine them with open standards on the hardware side, which at the time was apparently quite revolutionary and everybody liked it. So this whole notion of open systems really started with Sun Microsystems 25 years ago. Early on in Sun's career, long before I got here, was a fundamental idea that if you propagated a standard across the world, literally shared it with the world, you would create for yourself more opportunity for yourself and for your customers and all the developers who worked with both of you.

Our whole strategy of sharing open interfaces, open source, even open sourcing the hardware is going to create just an explosion of technology. I think what really sets Sun apart is the idea that we live so fully by the idea that you should share your innovation with the world. Sun was really the first company to define open standards and open systems in the computer area. And I have every reason to believe that that strategy will be very successful going forward. I'm not sure any company has ever had a tagline that has been as enduring, as right on the button, as the network is the computer. on the button as the network is the computer. We tried to identify one phrase that would really show what we were trying to do as a

company, which was to make the computer and the network indistinguishable from each other. It is all about giving people access from anywhere, any device, any time, to open and share technologies and innovation.