LINUX Unplugged - 387: Tumbling Into the New Year!

Episode Date: January 6, 2021

We have some strong opinions about the state of openSUSE Tumbleweed. We've secretly been running it for the past week, and share our experience. Plus Microsoft's path to dominating the Linux desktop becomes clear.

So you hear all the hype about mechanical keyboards, but they're really expensive, let's be honest. So that's where the Buckle Spring Project comes in. It's nostalgic keyboard sounds that emulate the sound of the old faithful IBM Model M keyboard. The author of the project has painstakingly captured the sound of each key. This is a real conversation that I typed up to Wes Payne, and this is what it sounds like. And it runs on your machine in the background, and as you type, it makes these sounds come out your speakers.

Finally an answer for that problem where I'm at the coffee shop, but no one knows how nerdy I am. And now I can bother all of them. It's on-demand Model M typing. You're welcome. We'll have a link to that in the show notes. It's a lot of fun. Hello, friends. Welcome into your weekly Linux talk show. My name is Chris. My name is Chris. My name is Wes.

And this episode is brought to you by a cloud guru, the leader in hands-on learning. The only way to learn a new skill is by doing. That's why ACG provides hands-on labs for cloud Linux servers and much more. Get your hands cloudy at cloudguru.com. Coming up on this week's episode, we're finally doing it. We have gotten so many requests into the show, and it seemed like now, at the beginning of the year, when there's so much left to happen,

it was the time to try out OpenSUSE. We gave it a go. Essentially, we wrapped up last week's episode, and Wes and I loaded it up. Been running it on some metal directly. And I think we both opted for the tumbleweed route. Wes and I have not been telling been running it on some metal directly, and I think we both opted for the Tumbleweed route. Wes and I have not been telling each other much about our experience to save it for the show.

Secret, secret. You went Tumbleweed though, right? I sure did. Yeah, I mean, our kind of thought here was, we feel like there's a lot in the air for OpenSUSE, like some stuff's going to happen this year. And there's also a lot of momentum behind rolling distributions, and Arch is always the go-to example that's cited. But SUSE has been at this for a while now. And is it a viable Arch alternative?

Is it a workstation rolling OS? So we wanted to give it a go and relay our experiences to you with OpenSUSE at Tumbleweed. And we'll do that in a little bit. But of course, we also have some really great community news. We have a pick that's fantastic. It's going to make you way more productive and a lot more. So to help us get through all of this, we have to turn to our crack team. Time-appropriate greetings, Mamoru.

Hello. Greetings. Howdy. Buenos tardes. Happy New Year, everybody. Here's a New Year's taco for you. And thanks for being here. We have some community news to start with.

And I don't know what's going on here. So I just wanted to have an open conversation in the show because I bet a lot of you have seen the recent news about Steam data. It appears that Linux market share has slid badly over the holiday period. And one of the reasons I wanted to talk about this is because the numbers are so dramatic, it's going to kind of make its rounds.

And I think we can explain what's happening. But Steam published their annual, or no, monthly, whatever it is, their frequent survey that I never seem to get ever. Ever on my Linux box. Have you ever gotten a Steam hardware survey? or no, monthly, whatever it is, their frequent survey that I never seem to get ever, ever on my Linux box. Have you ever gotten a Steam hardware survey, Wes?

Actually, for the first time ever this year. I mean, I've gotten it a ton of times on my Windows gaming box that I seldom mention, but on all the Linux ones, no, never, except this one. So I was glad, and of course, I immediately filled it out. Yeah, well, so I think maybe once ever in all of the years. But according to the published numbers from Valve, this came out on January 1st. It was like a real kick in the stomach at the beginning of the year. The Linux market share on Steam had a pretty significant drop.

It was just 0.5%. We were hoping to get to 1% soon. It was a drop of 0.33%, which is pretty major in these numbers. And Windows picked up a pretty decent market share, just over 1.6% additional market share. macOS didn't do so well either. But the first round of numbers out of Valve looked kind of devastating just because, you know, over the holiday period, you hope more people are playing video games. Yeah, and there's been some updates there.

Valve's updated those numbers, but they still point to Linux regressing. It really doesn't look great. After being at 0.9% for the month of November, Valve's revised December 2020 numbers put the Linux gaming market share at 0.74%, or a drop of 0.16%. Yikes. Yeah, so not as significant, but still not good. And kind of hard to understand when Proton is making it possible to play more games than ever.

The library of available Linux games is greater than it's ever been. And the compatibility is greater than it's ever been. But I think you have to consider. I could be wrong because who really knows here. But you have to consider Cyberpunk 2077 came out around the time that the survey was conducted. That is true, yeah. And I

think, although people could confirm in the chat room, but I'm pretty sure that you're more likely to get the survey on a new install. So perhaps people were setting up Windows 10 machines or reloading Steam after not having it for a while. Oh, that is interesting. You know, I did get mine on a fresh Pop!OS setup for gaming that I recently set up.

Hmm. Mm-hmm. Mm-hmm. I guess there's also been an increase in Chinese users that are running pirated versions of Windows 10 on Steam. So that makes a difference as well. I'm not kidding. That's actually one of the things that's been attributed in here.

And then you have to kind of just think that just increased usage in general. So people that maybe weren't normally online but are on Windows came online to play Cyberpunk 2077. I mean, it was the most hyped game of the year. Well, and even just over the holiday periods, I know I've seen some friends I don't always or even frequently see on Steam suddenly popping back up with, you know,

a little time over the holidays to boot their Windows box back up. 100%, same here, Wes. I'm on Steam a couple of times a week at the most, really, just to play games with my kids. And I see more people on there than ever playing Cyberpunk and other games right now. I've been trying it out on Stadia.

Figured I'd put my Stadia subscription to use. And I've been playing it on the Coder Radio pre-show for about three weeks straight, and it's been fine. You know, just fine, really. Stadia's actually done a pretty decent job. You're almost making me want to check Stadia out again, just because

I mean, I'm probably not going to build a rig just right now for Cyberpunk, but I could play it on Stadia. No problem. Yeah, especially if you can get a Stadia credit like I had, or something, so you're not paying full price for the game. Because remember, you can't use it off on Stadia. No problem. Yeah, especially if you can get like a Stadia credit like I had or something so you're not paying full price for the game. Because remember, you can't use it off of Stadia. Right.

Yeah, yeah. But yeah, so as things now point, it looks like the Linux market percentage on Steam is around 0.74%. So we were at, we were really close to 1%. We've now slid back to 0.74%. And macOS is at 2.74%. And Windows, 96.51%. We're coming for you, Windows.

We are. And it does make me wonder, how do the folks over at Valve actually view this? You know, because we're looking at it from one direction. Surely there's a bunch of folks over in the, you know, Windows gaming camp laughing at us as usual. But it doesn't seem like Valve has really slacked off on continuing to build and work on Proton and continue those things. I just hope that, you know, some of the folks there internally

get the sense that I think we've had on the show and in the community that it's a better time than ever to be a Linux gamer. I'm gaming way more than I have in like the past five years. And I just hope that momentum continues. Great points all around. I'm gaming more on Linux than I ever have in a while too. My kids are as well. And I think your point is well taken. The reason why we care about this number, even though it is, I mean, let's be honest, it's rather, it's a rather insignificant number. I think maybe that's why we care about it so much is it clearly has made such a difference that Valve has invested as much as it has into gaming on Linux. And we are trying to understand the logic, justification, and likelihood that it will continue. That's what we're doing

here with these numbers. That's the conversation that's not being had is we are just trying to piece together what is Valve's motivation here. And my God, will they please keep doing it? And if these numbers go the wrong way, I think the unspoken worry here is that Valve will just give up and say, Jesus, look at this. 96.5% of our user base is on Windows. We're kidding ourselves here. They never built that store that put us out of business. Right, you got to hope that, you know, Gabe's involved and that there's not some bean counter somewhere that looks at that and then looks at the amount they're spending on engineers

or contracts working on it and say, this doesn't add up. And Eric, you're pointing out that they could do a better job at just answering some of these questions around how the survey works. Yeah, I've always been kind of frustrated when you see these surveys out there, but I did like Canonical's approach back when Will Cook was on the desktop team, and they did that nice blog post on kind of the metrics that they got at a glance.

But as a numbers guy, I do wish there was a bit more information on how they exactly identify which, say, Linux user they send the survey to. Like you said, Chris, you didn't get one forever, it seems like. I mean, what's their policy or what's their procedures around that? It just would be nice to know. Yeah, and it's definitely not just if it's a new install or not, because I install Steam on just about every laptop or machine I review. And, you know, that's about one a month or somewhere around there, right? And I'm doing that 12 times a year, and I'm still not getting the survey. So, yeah, I agree, Eric.

Some transparency there. We at least have to answer some questions about how that all works. All right. We can probably cover the rest of that if you guys have any other comments from the Mumbler in the post show because I think – I don't really know how – I think people probably care more about this next story to tell you the truth because I got fired up when I read this. I agree with Linus. ECC RAM matters. Error correcting code memory.

And he says it, quote, absolutely matters in a fun holiday mailing thread. As ever, Linus was happy to share a little more detail around his opinions and said that Intel has been instrumental in killing the whole ECC industry with its horribly bad market segmentation. Intel's been detrimental to the whole industry and to users because of their bad and misguided policies with respect to ECC. Seriously, the arguments against ECC were always complete and utter garbage. Now, even the memory manufacturers are starting to do ECC internally because they finally owned up to the fact that they absolutely have to. And the memory manufacturers claim it's because of economics and the lower power. But they are lying bastards. Let me once again point out Rowhammer and how these problems have existed for several generations already. But these effers

happily sold broken hardware to consumers and claimed it was an attack when it was always just we're cutting corners. Man, this is a classic brutal Linus takedown. He went on to say in this rather lengthy post, the modern DRAM is so reliable that it doesn't need ECC was always a bedtime story for children that have been dropped on their heads a bit too many times. Yeah, I'm pissed about it. You can find me complaining about this literally for decades now. I don't want to say I was right. I just want this fixed and I want ECC. AMD did it. Intel didn't. Wow, man. Like I thought, okay, here's what I really liked about why this happened, is I thought I was the jerk that was just going on about ECC all the time and that it really was just like in my head that my systems with ECC were way better

because obviously if ECC RAM actually made your system more stable, this is what I thought, obviously if that actually worked, it would be everywhere. It'd be in our phones. It'd be in our desktops. Right. Here's this better way to do it. Why are we doing it?

But I always thought, gosh, you know, I've had desktop workstation machines and servers with ECC RAM, and I've had lots of desktops and laptops and even servers that don't have ECC RAM. And unquestionably, in my personal hands-on experience, the systems with error-correcting code are more reliable. I like this reminder of Linus's deep hardware background and interest here. And the whole history is kind of interesting. I know there was a lot of stuff being discussed around Google's rise

as suddenly the data center was taken over by, you know, just cheap Intel boxes. And some of these bigger companies just sort of trying to handle all these corrections in software and sort of preaching like, look, you can save on your hardware budgets. Don't bother with the fancy ECC stuff. You know, just make sure you've got redundant calculations and that you're handling these sorts of sometimes failures in software. But especially these days, it just seems like if we need our computers to work right, security is more important than ever. Everything's interconnected and online. And we've already solved this, right? I mean, why can't you just buy it for the machine that

you want to run your software on? Yeah. And it also, I think, really gives us kind of insight into some of the blame that Linus puts on Intel for harming the industry. You and I were just doing a review of a lot of the stories for 2020 on the latest Linux Action News. And when we were going through those, you see several times where Linus is pinning something directly on Intel. And I think you might agree, although I'd be curious if you don't, when you kind of go through what Linus has said about Intel over 2020 chronologically, the more the year goes on, the more hostile towards Intel he seems to be getting. And then you start to see even Greg K8 starting to take shots at Intel as the year wraps up. Oh, yeah. No, I definitely think that's true. And it's interesting because obviously they've got, you know, a long history both in the industry at large and, of course, just working together because Intel, you know, contributes a whole bunch of great open source work to the kernel and to the Linux and open source community generally.

generally. But I think we're all really starting to feel some of their missteps that they really aren't on a solid course, or at least haven't been. And that's starting to hurt. I'm glad we have AMD, but it's interesting times ahead for Intel. It's clear that the pressure's on like never before. So you have Linus laying this pressure down. He even goes on later in the post to compliment and commend AMD for their more broad support of ECC RAM, even on consumer desktop hardware. You have Apple, of course, with the M1 chip. Even RISC-V applies pressure. You have all of these pressures that are getting applied, and then you bring in AMD, who is compatible with the x86 platform. It's a tremendous amount of pressure. You have to figure something's going to give.

It'll be interesting to watch anyway. I mean, there's even more pressure from investment advisors and capital groups who maybe even are pushing for Intel to sort of abandon their own system of having their own fabs. Now, I don't know that that's going to happen anytime soon, but probably the next, you know,

two, three years will be very instructive for what's next. I have thoughts about where Intel's going right now with their current processors. So, but we'll get to that later. I'll mention that maybe in a bit in the show. There's, there's so much really, boy, you look at it from a new year's perspective. It's kind of exciting how much it's going to, you can tell there's a lot that's going to shift this year. It's going to be a big year for this kind of stuff. You can tell there's a lot that's going to shift this year. It's going to be a big year for this kind of stuff. Linode.com slash unplugged. Go there to receive a $100 60-day credit towards your new account,

and you support the show. $100 in Linode terms is a lot. You can really build out some significant, cool projects, try things out, do a test migration of NextCloud. I mean, you name it. We, right now, are using Linode for everything. The show notes now that we're doing this on, we've recently spun up, thanks to Alex's help, a new instance of HedgeDoc. If you're not familiar with that, it used to be called CodeEMD. It's a real-time markdown collaboration tool. It just dominates anything like Google Docs if you're doing real-time markdown collaboration tool. It just dominates anything like Google Docs

if you're doing real-time markdown collaboration. And you can self-host it on Linode in seconds. Just get started by going to linode.com slash unplugged. You know, I was just looking at our Matrix server that we have on Linode. And Matrix is an interesting use case because there's some scenarios, like when a user connects that has a lot of rooms and stuff,

that initial login can put a significant load on the matrix server, uh, synapse. Yeah. There's, there's some work to do over there. Yeah. I'm just reviewing the CPU usage from, uh, from, you know, the last few days. And you can even see like, as the shows are live, the things spin up, the disk IO rate spin up, spins up. The way Linode makes it possible to view all of this in their dashboard is really, really smart. It's laid out in a way where you get all the information you need right away. So it's kind of in these sections. The first section up on the screen, and by the way, it's all beautifully designed. But the first section up is like it's your details you need to know.

How many cores? How much RAM? How much storage? What's your IP? And what's your details you need to know. How many cores? How much RAM? How much storage? What's your IP? And what's your SSH access? Like, you hit this button, copy it to your pasteboard, copy it to your clipboard, and log right in using this. And then you scroll down and you get all of the CPU and disk and network information. And we actually get a fair amount of IPv6 traffic, too.

But nothing like we do on the IPv4 side. But you get all of this and I can see, you know what, we're doing really well. We've proportioned this when this system bursts during like the start of a live stream. We're getting up there in usage, 85%, maybe 90% on a really busy day, but we're not maxing this box out. And I can tell that, yep, we have specced this one just right for what we're doing. And it makes it so simple to see that or add storage or tweak things or do backups, which again, because this is a community server as a responsibility to our community, we want to make sure that we have a backup of this thing if something goes wrong. So I can go over to the backup tab and I can see,

yeah, three minutes and 17 seconds ago there was a backup. It's really great. You can also take manual snapshots, like we might do that before a big matrix upgrade. There's all kinds of things you can do through the dashboard that make managing a cloud server really simple, even if you're not an expert. And if you are an expert, things are just a click away

to really take it to the next level. And Linode costs 30% to 50% less than major cloud providers like AWS, Google Cloud, or Azure. You get a nice balance of technology for the price with the optimal machines. SSD storage, 40 gigabit connections, 11 data centers all over the world. And they're true Linux users. Like as a company, like they're fans of Linux. And God, I just love that about them too. And they're independently owned.

They've been around forever. They started three years before AWS. They're independent, and they're the largest independent cloud provider. So go check them out at linode.com slash unplugged. Get that $100 credit, support the show, and it lets Linode know that, yeah, we're getting some investment.

When we support open source media, when we support independent show, and it lets Linode know that, yeah, we're getting some investment. When we support open source media, when we support independent media, and they can then turn around and make that for free, we get a return on that investment, and everybody wins. So thanks for supporting our sponsors. Linode.com slash unplugged. So I have a story in here that I think at first pass,

it's like, why is this on Linux Unplugged? Because the story headline is, Microsoft is building a new Outlook app for Windows and the Mac, powered by web technologies. Chris, this is a Linux show. You know that, right? Right. And you know what's also funny is how the tech press doesn't know what this stuff is. Like, they don't just say Electron, right? This is WindowsCentral.com. Microsoft owns Electron. Why not just say they're going to use Electron? Why say web technologies? But the idea is that they're going to build a universal Outlook

client for Windows and Mac, as far as the press calls it, that also replaces the default mail and calendar apps on Windows 10 when it's ready. That's how far they're going to go. The new project is codenamed Monarch, and it's based already on the Outlook web app. So if you're familiar with Outlook web, with the Outlook web app, which I am roughly familiar with, that's essentially what it's going to be. Yeah, and I mean, this kind of makes sense. Project Monarch is basically the end goal for Microsoft's One Outlook Vision, which aims to build a single Outlook client that works across the PC, Mac, and the web.

They really just want to replace existing desktop clients with just one app built with the same technologies that they already have to use to leverage those web versions. And I don't know about you, but even on the platforms I have that have that native Outlook app, I end up just using the web one anyway. Yeah, it is a lot better than the battle days of Exchange. It's gotten a lot, a lot better. And it works just fine on Linux in a browser. Microsoft told Zach at Windows Central

that they will do native integrations where possible. They're going to try to make it feel like a native application, look like a native application, offer offline storage, notifications, and more. But you have to figure, like they are doing now with some of their other applications, VS Code, which is also an Electron app, they're probably going to release this sucker for Linux too. And I think it will totally frickin' dominate on Linux when they do. I think it will destroy the Thunderbird market share for anyone who's in a corporate environment that needs solid calendar integration with the rest of the company, be it Google Apps, Microsoft Exchange, Office 365, whatever it might be. If you are in the corporate world, and, you know, I think this even applies to you, Wes. I mean, imagine if you could have an Outlook client that ties in perfectly with the day job. It does good with the reminders.

You can accept meeting invites. You can create meeting invites. You get the directory. Everything works as a full-fledged client, like your coworkers on Windows and Mac. You'd probably be compelled to use that. Yeah, yeah, I definitely would. Especially if you've got some of those nicer features.

I mean, the web version works well enough, but yeah, there's some nice desktop integrations you get if you go the full Electron route. Yes, and I have a ton of old school nostalgia love for Thunderbird. I still think it's a great personal email client. I have it installed. I have it. Totally agree.

I like that. Like all the family, like, happy holidays and Merry Christmas emails that I sent out, did them all from Thunderbird, you know, connected to my Gmail account. It's great. But for the two years that I was an employee at a company that had lots of meetings and, you know, did everything by calendar, it was a big part of the company culture. I spent that entire two years struggling on a daily basis trying to make things work. I tried MailSpring. I tried applications that house web apps inside a giant local application. I tried KMail.

I had Thunderbird, Evolution, Geary, everything. MailSpring got abandoned, which was already a mail electron app that sort of had the best integration and made it really easy to work with calendar invites. But MailSpring was abandoned. And to this day, I've just kind of resigned to using the Google Gmail web app whenever I am working on any kind of serious correspondence even. Even just serious correspondence and where I want my address book and I want calendar abilities and all of that, I just use the web app. Yeah, same here, right? I mean, the web email, basically.

Yeah, and I pine for a local email client. I prefer local email. I prefer a local window. I like actually having as few applications in my browser as possible. And I would absolutely give this a try. I'm at that point. I wonder too, if this might be, you know, just one more bolster, sort of to reassure corporate IT departments, you know, sure, like you can have your Linux desktop maybe, and you've gotten some special permission, or you've just set it up without asking. But suddenly one weird thing doesn't work, and they're like, well, you know, we're tested just fine in the desktop app on Windows and Mac, and I don't know what your problem is. But if suddenly there is this, you know, single blessed web-powered version, and it runs on

Linux, I mean, I remember Wes of four years ago, and I had to run a Windows VM just for Outlook integration because there was some stuff that OWA still couldn't do. That future, I mean, that looks like we're way beyond this. I bought every version of Crossover Office. I bought every release of Zandros OS, which was a Debian-based KDE

desktop environment that shipped Crossover integrated, so you could even put the Office 98 or whatever it was, Office 2003 probably into the CD-ROM drive. And it would even recognize the auto start file and just give you the full install experience. But think about this, Wes.

For better or for worse, think about the complete picture once Outlook is released. Fedora and Ubuntu out of the the box, during the installation, offer Active Directory integration. You log in. You've got Microsoft Edge to browse your corporate intranet. You've got Microsoft Teams to do all of your team collaboration Slack-style chat. You've got Microsoft Outlook to do all of your corporate email correspondence. You see where I'm going with this?

VS Code, et cetera. Don't forget.NET Core, right? That's available too. And let's be honest, they're not going to stop with Outlook. Microsoft didn't add Electron to their arsenal of technologies that they own so that way they could just let it piss off in the corner.

They're starting with Outlook because Outlook is the high bar. They're starting high and they're going to knock out Outlook and then they're starting with Outlook because Outlook is the high bar. They're starting high, and they're going to knock out Outlook, and they're coming for Word, they're coming for Excel, they're coming for PowerPoint, they're coming for the entire Office 365 suite, which are already web apps. And then they're going to ship them on Linux.

It's interesting, too. It does sort of seem like it's a more reinforcement that the Windows side of the house really is no longer the dominant tastemaker internally because they've clearly aligned to the goal of like we just want to get our stuff in the hands of as many people as possible, regardless of if they have to use our other platform. It's pretty wild to see this coming. And it may not be until the end of the year until we really see anything materialize. And then it may be a while after that before we see the Linux support. Just like with Edge, I wouldn't be surprised that if Microsoft takes the approach of they focus on the Mac and

Windows versions first because it's a narrower problem scope. And then once they have those shipping, they begin working on a Linux alpha and beta. Yes, and I mean, the same thing happened with Teams. It was quite some time before the Linux version came out. But hey, now it's here and seems to be working pretty darn well. Oh, live update from Popey in the IRC. He just checked around on their GitHub, and it looks like the developer popped up 20 days ago for MailSpring. Oh, thank goodness. So I'll be looking into that.

So maybe MailSpring's still alive. It may save me yet from an Outlook future. But yeah, they're not going to have anything until like maybe even 2022 is when you may actually see it start replacing the apps. They may have like an early version build client that you can get on Windows or Mac by the end of this year. But replacing the default application on the Windows desktop, that's a 2022 thing. So it's going to be a bit. And the Linux beta will be somewhere in there, right? We shall see.

We shall see. There's also just a couple of community things we want to touch on before we move on. Number one is FOSDEM, which is a pretty significant conference every single year. Europe's largest free and open source conference. Thousands of people, I think last year it was nearly 9,000 people, take over a campus in Belgium for a weekend and turn it into quite the event.

800 talks, 50 tracks, hundreds of exhibitors. It's really intense. It's like, I guess, the most intense of all of them. I've always wanted to go. Maybe this year I can, because it looks like the Matrix Project reached out to them and said, hey, we know you want to be virtual for February of 2021. I think we can facilitate that. Isn't that interesting? I mean, Matrix has all kinds of great technology under the hood to make this happen. And

well, you know, thankfully, FOSDEM 2020 snuck in just a few weeks before this whole pandemic really got into full swing. Of course, with FOSDEM 2021 happening just in February, this upcoming February, you've got to start figuring this out right now. I don't think we're going to be in any state to, you know, do an in-person one yet. So they just have to start planning for it. But Matrix, Wes, I mean, it's brilliant, right? It's an open protocol. There's something great about that for an open source conference like FOSDEM.

But the video stuff and all of that is really still early on Matrix and really Elements is the only client that has the most complete implementation right now. Yeah, I suppose in one sense, maybe that's risky or we'll see what parts all get adopted, how far the integration goes. the most complete implementation right now. Yeah, I suppose in one sense, maybe that's risky, or we'll see what parts all get adopted, how far the integration goes. But on the other side, that might be a really good sort of dogfooding stress test for some of those technologies.

Yeah. The other part that I like about this is it seems like more and more of the Matrix team, they're reaching out. They're pushing Matrix. They're trying to get it into places that make sense, whether that's, you know, various government things that are looking for protocols or open source conferences. And I both like that as a way to build Matrix.

But I think it also speaks well to just the project structure and internal organization. And they've got time for this. This is something they can take on and actively continue to develop Matrix. So it's just a good sign to me. Yeah, it'll push forward certain things in Matrix. It looks like they're going to use a good helping of Jitsi in here to handle some of the video heavy lifting.

Of course. That's brilliant. So what you're going to see from this is a notch up in the network effect of Matrix. And that I'm particularly excited about because I think you're going to see several projects over this next year switch their chat and collaboration infrastructure over to Matrix. There's some that are in conversations about doing it right now.

That's going to increase the amount of Matrix users. An event like FOSDEM taking place on Matrix, that's going to increase the number of Matrix users. And we're going to start to see this network effect where something that's based on an open standard that has open source implementations that is really kind of this next generation protocol for the web or its own layer on the internet,

it's going to have more users now. And that's going to be a net win for free software just in general. So by FOSDEM doing this, I actually think the contribution is more significant than just their event. It's actually fairly significant. And even if the effect just has the matrix team and the element team work a little bit harder in more of a sprint style to get ready for February to have some of these features more polished, even that is a great net benefit for all of us. And then last,

rounding out the community news, just a brief mention that KDE Plasma 5.20.5 shipped today. And it's just a real nice polishing of the 5.20 series. And this is, I think, a great indicator of the stage

that the Plasma Project is in right now. They announced today as we're recording this release, I think, a great indicator of the stage that the Plasma project is in right now. They announced today as we're recording this release, and it's just fixes for like visual glitches that maybe impacted GTK3 apps or system tray had an arrow problem that was hiding some things. And the one that is the most significant that drove me freaking batty, and I'm very happy to see it patched, is there was an issue with the Plasma Network Manager applet that would jump to a different network. So as you're searching and it's finding the networks of Wi-Fi networks, you click connect

and start entering the password, and then it jumps, and the connect registers for a different wireless network. And then you go through it over and over again. It was infuriating. And it actually struck me while using OpenSUSE at Tumbleweed, and at first I thought it was some weird, like, bug, but I quickly figured out it was a Plasma thing. So nice to see that fixed,

because if you've been using recent Plasma and you've run into that, you know how painful that was. It was really bad. I was also pretty pleased to see better support for WebRTC-powered meetings and screencasts and stuff like that if you're using the Plasma Wayland session, which I know not a whole bunch of people are. But it's exactly the kind of thing that needs to get solved if we're going to have more Plasma Wayland adoption. Yeah, if you're using Plasma for your day job and you're on Wayland,

you sometimes need to participate in meetings where you're sharing your screen. And so to see that kind of stuff get plumbed in is really nice. 521, so the next big release of Plasma, which is scheduled for February, has a new system monitor app in it. And I gave it a go because it's available already for some distributions. It's just Plasma System Monitor, and it's going to replace KSysGuard. Now, these are the tools that let you look at memory usage, process usage, CPU usage on the Plasma desktop, sort of like akin to Task Manager on Windows. And it's getting a really nice upgrade in Plasma.

I'll leave the link in the show notes for you guys to go check it out. But it's a completely rewritten UI, and you can now also easily create and edit pages where you can put your own items on this dashboard, sort of akin to Perfmon on Windows, just tons of categories of stuff in there. And you can put a nice heads-up display. It looks good. And this, in my opinion, Wes,

these types of system monitor apps where you get an insight into what your box is doing, they can never be good enough. Like, there's always more work to be done, and this thing's sharp-looking. Yes, it is. And actually, as noted by Nate Graham, who we just had on the show in episode 385,

notes that it's been in development for almost two years, and hopefully it will eventually replace K-Assist Guard once it's, you know, a little more tested out in the wild. But for now, it stays independent. But you're totally right. I mean, even Windows, you know, they've improved a lot of their resources. They've got pretty graphs now, if you want to go just pull up your task manager and check out what your system is doing. And even though we're going to promote all the next latest top replacement tools in the shell, of course, or just talk about net data, I think for users who just, you know, get into Linux, are starting things up,

or just want to have a native app that shows what their system is doing, it's important that this works well because the kernel in our user space, we have all the tools there. It just needs the plumbing and a little bit of prettiness. All right, we have a spot of housekeeping this week. A few things to mention.

Number one, I just kind of referred to it earlier, but if you have not yet checked out Linux Action News 170, there was a lot of open source development in a very crazy 2020. And we recapped some of the standout moments that you should know about. And there's a few items in there you may have missed. So it's a great episode to get caught up on some of the things that happened in 2020. That's linuxactionnews.com slash 170. Also, I'm going to encourage you to check out this week's Coder Radio episode 395, coder.show slash 395. It's not out as we record yet, but I've already recorded it with Mike. And I was sent the XPS 13 developer edition

that comes pre-installed with Ubuntu. And it has the 11th gen Intel CPU with Xe graphics. And I think you're going to be really surprised by my review of that laptop. And we're going to talk about more specifically the Xe graphics aspect in next week's episode. Wendell from Level One Techs is scheduled to join us and geek out on the Xe graphics under Linux. So it would be good context if you check out Coder Radio 395 to get an idea of what I was able to do with that laptop and sort of the situation I find myself in now.

So that will be Coder.show slash 395 later this week. And then that brings us to Sunday, where you can join the LUP plug every single Sunday at noon Pacific, 3 p.m. Eastern. You can get the mumble information at linuxunplugged.com. And Minimek, you wanted to put a word out to an upcoming special LUP plug. Yes, indeed. So first, I hope you all had a good start in 2021. And yay, we had our first Loveblog meeting for this year last Sunday, and we scheduled a new recording session. So the subject will be accessibility features in Linux. So we are talking about assistive tools like screen magnifier, screen reader and so on. And the recording will most probably be the 24th of January.

We have a community member that would like to share his experience with us. And he also contacted a special guest whose name I will not yet reveal. All I can say, he is a developer for a well-known desktop environment project. All I can say, he is a developer for a well-known desktop environment project. Now, what I would like to do is a call for participation for that talk. If you have some experience with accessibility features in Linux and want to be a speaker during that talk, please contact me either on IRC, Telegram or Matrix. So just send a message to Minimac, or even simpler, just join the next LUBLOG session next Sunday,

and we can talk about that. So that's all for me. Thanks, Chris. Yeah, great. That sounds like a fantastic topic. Again, that's going to be held on the 24th. And there is a LUBLOG Matrix channel.

If you'd like to get involved and chat there, you can find our Matrix info on the website as well. That's all at linuxunplugged.com. Okay, so after the show last week, Wes and I looked at each other and said, what are we going to do about OpenSUSE? Because we get a lot of requests into the show to try it out because it's not a distribution we cover very often because neither one of us run it on the regular. But we know there's a lot in the works for OpenSUSE over 2021, and we feel like it was time to check in. And I am intrigued by the idea that I could throw Tumbleweed, the rolling release, I am intrigued by the idea that I could throw Tumbleweed, the rolling release, on my laptop,

and we could do something like Leap on the server, which is the more LTS release. Isn't that a nice thought? Because there's so much, you know, I know we both love Arch running that, but even though we do have a server with Arch, we acknowledge that probably you wouldn't want to have a whole bunch of servers that are Arch, right? Why would you do it that way? But it's nice to stay in the same ecosystem. We don't have to suddenly switch things out or get used to all the little differences between things. If you could have just one system that could run everywhere, including all those Pis that you have,

I mean, wouldn't that be the dream? Yeah. And it does run. I'm told, although I have yet to test it, although I think I will put it on the Pi 400 just to give that a go. I was running it on a Dell laptop that I had in for testing, and I went with the DVD release. So they still kind of like old-school style, I love it, offer various sizes, and then they even offer like an i586 image if you want, along with the ARM and x64.

And then you pick like DVD or net install. As the name implies, net install is just what you need to get the live environment up, and then the packages are pulled down from the repo. And if you have a fast internet connection, I think that's probably the way to go. You tried the live version though, Wes, and that was kind of different. Well, I was confused from the start because, I mean, you go over to the, you know, the Get Tumbleweed page and they've got some tabs. If you want to go get, you know, you can get their cubic stuff, you can get the installation. And they do have a, they have a live tab with live

images. But right at the top, they say, please be aware of the following limitations. They should not be used to install or upgrade Tumbleweed. Please use the Tumbleweed media, which I think they mean the installation media, for that instead. And they also note that these have limited package and driver selection, so they can't be considered an accurate reflection as to whether Tumbleweed will work on your hardware or not. And, you know, that's a fair point and can probably be said of all live CDs, but it was a little strange to me that although I could kind of try it out, I had to then go get a totally separate installation media just to get things installed. So I ended up just downloading the net install and doing it that way.

Well, see, that's just as confusing as it was for me, because I think about 97.5% of the reason I would get a live disc would be to test how that distribution works with my hardware. And then if it works, I run the installer. Now, once it's a distro that I know works great and I'm just going to deploy it all the time, I prefer having something that just boots right to the installer like their primary media does. But the wording around limited package and driver selection, that could just be implying, you know, there's no NVIDIA driver here. Or it could be implying that, like, it's some sort of super slimmed-down kernel.

It was very confusing, and I'm like, but that's why I use a live CD. I've also kind of just been spoiled over the years of, you know, okay, maybe it doesn't matter right now when you've got, like, a test machine that you're using, you've still got your desktop. But if you're someone who just has the one Linux machine, it's really a luxury that unlike in the Windows world, I can still browse the internet with Firefox from the Live CD while I'm installing these new OS on my machine. And it kind of sucks to not have that option. Yeah, yeah.

Yeah, I remember how cool that used to be. Like Linux is so powerful. I can play Solitaire while I'm installing the OS. how cool that used to be. Like, Linux is so powerful. I can play Solitaire while I'm installing the OS. But so, you know, you had to hang around. While I'm sitting here downloading the 4.6 gigabyte or 4.28 gigabyte DVD image,

which is kind of quaint. I remember DVDs. But I wanted the full offline experience because I was going to take my laptop home after the show and just do everything from home where I'm on a pretty limited connection. And I actually, I want the DVD stuff. So I sit down and I, you know, after the download completes, I write to my media and, you know, you and I go our separate ways. I sit down at home and I boot off the live media. And I've always

really liked the Sousa Grub boot screens. I've always thought they just have always looked the best beyond like any other distro. And I think that still holds true. They're just, they really get that first power on experience nice. No kidding. I definitely have that written down in my notes of just, this feels slick and professional and wow. Things kind of then took a bit of a tumble, if you will,

when the installer loaded up and I'm on a 4K screen system and the UI for the installer was so tiny that the entire EULA fit on my screen. There was no scroll bar. The entire EULA fit. It was absolutely teensy weensy. And I couldn't see a way to magnify, you know, to double it or whatever during the installation, but I powered through.

And the installer got to the point where it seemed like this was the opportunity for me to put it on Wi-Fi if I wanted to do that. And boy, is that an experience. It's really old school because you set like, you first have to set the encryption type and it's like default to WEP. And that's on one tab. But then you on a different screen tell it to go scan for SSIDs. And then on a different tab, you set the IP settings and so on and so forth. I mean, it was a very strange experience. And I wasn't even sure if I was connected to Wi-Fi after all was said and done.

Did you experience this? No, you know, thankfully I was installing from an Ethernet cable old school, so I didn't run into that. Although, I did actually install a few other times in a virtual machine and had some similar problems getting things connected, which surprised me. I obviously did have it connected because later on in the installer, I got a prompt that came up and it said the installer has detected a network connection. If anybody's installed SUSE, you've seen this. Would you like to enable online repos? And I thought, well, what are they asking

me here? Like the, you know, Ubuntu will phrase it, would you like to pull down the latest packages? Are that, is that what they're asking me here? Or are they asking me that, do I just want to make sure I add additional repositories? Which I did. So I said yes, and then I saw which ones were checked, and I said okay. But then during the install, it seemed like it was clearly pulling down packages from the repository,

even though I had implicitly decided to use the offline media. And it just seems by opting to have online repositories, I then started getting packages from the repos during installation, which is fine, except for I was, again, on a limited LTE connection and would prefer to save the bandwidth. You had planned to not do that. You know, interestingly, on my, you know, the Netver install version, it asked me the exact same question. And I didn't try this. Maybe I will after the show. I don't know what happens if I would have said no,

because there weren't any other package sources to use. Yeah, that's a good question. Right, yeah, especially on the net install. And then the other thing that seemed particularly challenging with the large Dell trackpad is it was registering clicks weird. I had to get really far down in the corner of the trackpad in a way I don't normally have to do on any other Linux distribution and to kind of make the situation a little more frustrating

although yet helpful at the same time every time I would try to click next and I wasn't in the precise bottom of the far left corner of the trackpad I would get a dialog box that came up by Yas that says you seem to have misclicked would you like to swap the buttons but the way the dialogue is stated is not actually really clear. I wasn't sure what I'd be swapping and how,

and I didn't actually mean to do it. So then I had like a dialogue I had to clear, and then I had to go try to click my next button again. And I have never had this with this trackpad happen before, and I'm not quite sure what was going on there. It felt like it was acting like a PS2 mouse, but again, not really a deal breaker. It just made it not quite as smooth as maybe setting up Fedora or Ubuntu, but you know, at the end of the day, it was easy to get through. I think that's another version where you're not in the same environment, right? You're not on just a regular desktop that has a program that runs and installs. You're in a dedicated installation environment, which is simpler and might work really nicely, especially at scale.

But you don't know how it works or have any of the tools to go fiddle with, like, display settings. I do feel like you get a lot of nice user choice. And if that's your thing, and it is my thing, that's something I love about the OpenSUSE install. I mean, you can get access to something like six different desktops depending on what options you choose. But you get presented with Plasma or GNOME or other desktop, and that's where you get access to more or you get things like transactional server which is really neat and they give you a little brief descriptions of all of that but once you make your final decisions i really like that it's just it's just it goes man and then i i stepped away to go get a snack and came back and i was at my desktop it was just done

because i chose to use auto login for the test. And it was kind of nice to be honest. Like the computer just took care of itself and just finished the installation and brought me to my desktop. And I thought that was a slick experience. Yeah, none of that waiting for you to remove the USB drive or any of that other stuff that you figure out. And, you know, I was kind of impressed with the installation in general. It did feel kind of enterprise-y. I mean, like one of the first things it said to me was, searching for available products.

Probing the source type. And I was like, okay, what's going on here? And of course, as you sort of skipped over a little bit, there's also a license agreement that you're prompted with right at the start, which is not something I'm always used to when I'm installing Linux. But the flip side of that is that it felt mature. There's a ton of translations available. And as someone who loves to complain about Anaconda more than I should, I was really impressed with the disk setup tooling, the partitioning, especially how I could

go into expert mode. And it gave me a choice of, do you want to start fresh or do you want to keep things how we've set them up and then just tweak them a little in expert mode? And I thought that was really well done. And I liked its suggestions. a little in expert mode. And I thought that was really well done. And I liked its suggestions. I thought they were solid. And I wasn't sure what it would do since I had an existing distro on there. And it took care of just tidying it all up for me and doing a nice, you know, ButterFS setup, which is part of what I wanted to play with. I was disappointed to see that the Wi-Fi setup that I did in the installer

didn't hold over to the installed OS. So then I had to go through setting up Wi-Fi again, which is where I got bit by that current plasma bug of the menu jumping around. So I was like, oh, I always think that's nice. You know, when you get like on Ubuntu or something, you get the live environment on the Wi-Fi network, and then you reboot into the installed OS,

and it still has all the Wi-Fi stuff. Like, I just think that's slick, and I'd love to see them be able to do that if they could. think that's slick and i'd love to see them be able to do that if they could uh but i uh i i was plagued immediately with performance issues on the plasma desktop on uh open zeus a tumbleweed on this dell with an nvidia graphics card and the easiest way you'd see it manifest was i'd fire open k runner which is the plasma application launcher it's such like a muscle memory thing that I do alt space, and I start just typing the command, and I hit enter. But on this installation, I hit alt space,

and I'd start typing, and I'd hit enter, and then I'd see like my web browser or whatever I'm using freak out because all of the input had gone into my active window, and then even after I'd come to that realization of what had just happened then the k-runner menu came down and it was consistently laggy like this throughout the system so i thought well this has to be a video acceleration issue because this is a ginormous 4k display and it's doing it all in software right so or whatever whatever the whatever it does i better install the NVIDIA driver.

Do you have any systems where you tried to do the video driver, Wes? You know, actually, not yet. It's on my list to try. I think I'm going to be leaving Tumbleweed around on at least one system for a while. But on the Intel systems I was testing it on, no, no problem. I was, yeah, I was really pleased with performance. Yeah, works great. Now, I don really pleased with performance. Yeah. Works great. Now, I don't mean any disrespect to the project, but holy shit, this is bad.

Oh, my God. I cannot believe it's still in this state. So I found some great documentation that even as a 20-plus Linux user, I can kind of understand what they're saying, but I essentially have to use a series of tools to figure out what type of NVIDIA card I have. And then I have one of two options to install for an NVIDIA driver, but none of their documentation covers the use case of using a Quadro video card. So I was completely left on my own to assume what I'm supposed to do. And then once I even get the process of adding the NVIDIA repo and then going through YAST to find different NVIDIA package options

for a 50-50 shot of which one might apply to my video card, YAST would hang during the installation. The first time around 64%. Legitimately long enough that I stepped away, came back, sat with it for a bit, and pondered, well, I guess, do I reboot my system even though it's 64% through installing the NVIDIA driver? And got up, came back again, still there, sat down, thought, hmm, yeah, I think I'm going to have to reboot it, and then it finished. So I reboot my computer, and after I installed apparently what was the wrong of a 50-50 shot driver,

I tried the newest driver. There seems to be like a version 400 series and a version 300 series. I installed the 400 series because it's a fairly modern Quadro video card. Apparently that was the wrong one. So I reboot. No video. Everything's on the Intel still, still running real choppy, real laggy. And Yast helped me uninstall it, but I wasn't sure if it cleared up everything or not. And so then I installed the older driver.

And I mean, at this point, I'm kind of desperate to get the NVIDIA GPU working because the performance is basically unusable. The input lag was brutal. I was screwing stuff up. I just found it to be kind of an awful experience. Then you combine that with just the newness of having to set up a desktop environment and it's not working right. It's a frustrating experience. Then you combine that with just the newness of having to set up a desktop environment and it's not working right. It's a frustrating experience. And so I thought, okay, I'll clear out what I installed and I'll install the older driver. I install that, I reboot. I mean, I'm following the instructions that are on the wiki and I reboot and it still does not work. I've got no real errors that I can follow up on.

And to this day, I just kind of gave up on it because I figured I'll come back to it. And I wanted to test out other things on the system. But that experience of the process I had to go through feels like it's barely evolved from what it was like seven or eight years ago on desktop Linux to get proprietary graphics working. When it was a legit exercise in system administration to get a driver enabled on your box.

And now we live in an era where it's a checkbox on Fedora and Ubuntu and every other desktop environment or on Arch, it's a single package you install or a series of packages, a meta package. Some folks even, you know, some distributions and projects even have dedicated media for that. If you already know that's what you want, you know, it comes with it right there. And it was painful.

It was, I remember thinking how this used to be and how it just doesn't need to be like that anymore. And I was shocked that the experience was that bad and that that's not something people talk about a lot more. But there's other things that just seem like off that nobody talks about. something people talk about a lot more. But there's other things that just seem like off that nobody talks about. Both Wes and I on different systems couldn't get the ISOs to download using Chrome that particular day we were doing it.

I don't know why. You should go try it. See if it downloads on Chrome for you. We had to switch to Firefox. I actually switched to just a dedicated downloader to get the ISO to download. Nobody's talking about that. It was weird because it was clearly a link, right?

I could right-click and copy the link, and I did the same. I just pasted it and, you know, curled it down to my file system. But it was strange. And it's surprising to me. Maybe it's a new bug. I haven't downloaded OpenSUSE, especially Tumbleweed, for quite some time, I will admit, but not ideal. So then things really went wrong when I tried to get Slack working.

really went wrong when I tried to get Slack working. So there's obviously a pretty significant lead that the Snap store has in SEO when you search for getting Slack on OpenSUSE because the first like three or four results for me were all like, do it with Snaps. But when you go beyond there, you discover, and maybe you had a different route on this, I don't know if you tried this, but when you Google around, you can find that there are different community repos that are hosting Slack for you. And this is kind of like PPAs to the next level, because instead of it just being a simple, straightforward process to add a personal package repository, you have to go through this process through YAST where it asks you all of these questions. And I chose the community repo that looked the most legitimate

based on absolutely nothing and then went through the ask process where it had to add five additional repos to make it work. And of course, it aired out on adding the third out of the five repos and the process never completed and I never got Slack installed on the machine.

Oof, no, that is interesting. I kind of cheated there, and I went the easy route, and I installed it via Flatpak. Yeah, that's what I'm going to do. That's what I'll have to do. And honestly, I did that because that's kind of been my go-to for a while now. There's just that sort of set of applications that I just rely on having in, you know, these sort of containerized universal package formats now just because I know that they're there and I'm pretty sure they'll work wherever I'm going to run them.

But you're right, you know, software availability was one of the things we were wondering as a can I run Tumbleweed in place of Arch sort of meta question because with Arch, you know, that it's probably just in the AUR. And the AUR might not be simple. We should acknowledge that, right, you're learning more about how the ecosystem works here, whereas you're already familiar with Arch and the AUR. But this sounded a bit more complex.

I think it just kind of, it really feels like the experience is downgraded when you begin managing software via YAST. And I know that's probably the opposite opinion that every user that's listening to the show that loves to use SUSE probably has the opposite opinion of me. And I know that's probably the opposite opinion that every user that's listening to the show that loves to use SUSE probably has the opposite opinion of me. And I'm sure the development community for OpenSUSE

has the opposite opinion for me. But in my using it, and mind you, I've had SUSE Enterprise Server in production. I've used YAST extensively in production years ago. And I think one of the reasons I was first drawn to SUSE when I was new to Linux and I was deploying in production years ago. And I think one of the reasons I was first drawn to SUSE when I was new to Linux and I was deployed in production was because I felt like YAST was sort of an insurance policy for me. I knew there were certain things I was always going to be able to configure

on these Linux boxes because YAST gave me that option and I could do it in YAST. And that as a configuration tool is powerful. And it also breeds a lot of loyalty by the user base. But now having been away from SUSE for a long time, and coming back to it, my experience dies every time I go inside Yast. It is arduous, tedious, slow. It feels bloated. It never feels native. And the software management has to be the absolute slowest and most clunky and require the absolute most user input of any package management system

of any popular Linux distribution. It's an absolutely paper cuts experience at every point. And I think people that use SUSE and that have used this for so long are just so used to it being this way or perhaps like the benefits of what Yass brings that they just don't even see it. But for me, coming from the outside, it's a deal breaker for me with this distribution. I kept debating how to feel about it because on one hand, I'm kind of really impressed. Like it's almost a different style of how you interface and manage like a Linux system and there's a lot in

there that I like. I mean they've thought about a lot of things. It's been a project that's been around for a long time. There's clearly areas where it's very mature sophisticated and I could really see if I was deploying this in a corporate environment or some sort of you know I was trying to administer all these boxes and needed sameness. Yes it would be killer.

But as someone more used to assembling Arch systems, yeah, it was kind of a weird black box and I didn't feel, it felt like a rogue agent on my machine that left me more confused than enlightened. And I don't think it's just the pro user like coming from Arch angle. I also thought about it from a skill set that's applicable to the wider market. And when you learn how to do things in Yast, and you learn how to do it the SUSE way, the OpenSUSE way, you're not really learning a skill that's directly translatable to CentOS, RHEL, or Ubuntu. And early on in my career, I think I would have really have disserviced myself had I only learned how to do things the Yast way. I initially started with Red Hat servers, and I just did everything on

Red Hat. And then when Ubuntu started coming along with their LTS releases, I heavily advocated that for my clients because I really liked Apt back then. I remember this was a long time ago, and Apt was a significant competitive advantage, in my opinion, for managing a distribution. And I was able to take my four or five years of only using RHEL to Ubuntu because I understood what I was doing in RHEL, and I could translate that to essentially almost everything that was the same on Ubuntu with maybe a few things in different places or a config file that's in a folder instead of directly in Etsy or, you know, small differences. And so it made the skill set that I learned on RHEL applicable to other platforms, which made me more employable. And it was a net plus that way. But I don't feel like you would get that with YAS. So not only is it not appealing for people who are used to managing their systems from the ground up like an Arch user might be, but I think that people that want to participate

professionally in the industry, unless they're in a niche that serves applications that run on SUSE servers, which in that case, this is an obvious non-problem, you kind of are limiting yourself because you're learning how to do it the YAST way and not the way that every other single distribution might be doing it, even if YAST way and not the way that every other single distribution might be doing it, even if YAST is just pulling those same levers for you under the hood. And so I think it's a bigger problem

than just the power user perspective. And I know this is going to sound crazy to anybody who might be listening from the project because YAST is kind of like their, you know, competitive differentiator. But from where I'm sitting at, it feels like something that when Linux needed everything set up for you, when everything out of the box needed configured,

like your X config file and all of your devices and your printers and your USB scanners wouldn't just work or your SCSI devices, like it made so much sense back then. But today, outside of maybe managing things like boot managers and snapshots and stuff like that, I just feel like it's a brutal experience that really kind of turns me off the distribution completely. Like I'm never going to want to be dependent on that tool. I'm never going to want to just isolate myself to one distribution and one way of doing things. And I guess I find it really kind of disappointing that I couldn't get over that because I can see how appealing it would be to people

to have Tumbleweed on your daily driver and Leap on your server and Leap on your Raspberry Pis. Like I can really see the appeal of having it all with one distro and one community. And I really wish that could have worked for me, but at the end of the day, it just doesn't. I'm wondering, did you have any fun or not notice at all ButterFS by the default, you know? I didn't really get a chance to dig into it.

Great question. I feel like that's another thing I need to spend a little more time with. So I'm going to keep it on the box. I'm going to try to get the NVIDIA card working. I'm sure I'll get some tips on how to do that. And I want to try it on the Raspberry Pi 400 as well. I think that's an important note we should add, really, just to be clear about in this. You know, this is not an assessment of fact around Hope & Sue's Tumbleweed,

who it does or doesn't work for, anything like that. Really, this is just us getting our toes wet again in this ecosystem. And, yeah, like you, Chris, I'm going to keep it around for a while, keep playing with it, keep it on one of my machines, and just see what it feels like to live the lizard lifestyle for a bit. Live the lizard lifestyle. I like it. I want to make kind of an important announcement. Moving on, our unplugged core contributors,

we had a bit of a snafu with the founder's promo code. A learning moment, you could call it, that I think may have run afoul of a platform bug. I'm not sure. So I had to remove the Founders promo today as we're recording. However, the platform support folks at Memberful are trying to see if they can figure out who was in the Founders discount and just get it reapplied to you. That may happen. But if that doesn't happen, if you notice like you got a price bump or you just want to get in and become a core contributor at the beginning of 2021 and get the whole year in, I have a new promo code that will take $2 off the price.

I'm never really doing this anymore because honestly, you know, it's a big discount. But with this oops and the founder's promo code getting removed, I wanted to make it right. And so the new promo code that anyone is welcome to use if you want to get this locked in at the beginning of 2021 is just 2021. So if you go to unpluggedcore.com and you apply the promo code 2021, you'll get $2 off the membership. And that should remain on your account for the entire time you're a member. I apologize for the snafu with the founder stuff. I think I'm just going to probably get to the bottom of it in the next couple of days. I'm working actively with the memberful support

who has been awesome and super responsive about the situation. But it got kind of gnarly, and this really sucks. It resulted in a lot of our OG core contributors not renewing by mistake. So 52% of you have been accidentally unsubscribed. It really stinks. So if you could double check on your membership, and if you've been unsubscribed or it's not the right price, you can put that promo code 2021 on there. I think I'm going to leave it open for a couple of weeks, and you can support the show. That's unpluggedcore.com. And then you get access to a limited ad feed

if you'd like the show with the same full production, but just limited ads. We make that available as a perk. Additionally, we have a bootleg feed. It's the full live stream, all of our screw-ups, all of our resets, the stuff that never makes it into the show, with a much larger pre- and post-show that's essentially a whole other show.

That's also available to you as a member. And, I mean, $2 off the membership price is pretty great. So unpluggedcore.com. essentially a whole other show that's also available to you as a member and um i mean two dollars off the membership price is pretty pretty great so unpluggedcore.com and my bad i'm actually not 100 sure it was my fault but i'm taking responsibility for it i'm making it right so use the promo code 2021 all right westpain we have a really great pick this week. And I believe you and I are both Todoist users for a little over a year or so. Maybe you even longer. Yes, we are. And Todoist has released like an electron version of their to-do tracking application.

But you know me, I like something local. So if you want to keep track of your projects, your tasks, your goals in a local app that looks really, really nice. I do. I do. That sounds like you. Well, then check out Planner at useplanner.com. It's an example of one of the beautiful apps that's been designed for elementary OS that, I mean, because it's based on native local Linux technology, it's possible to get on other distros. I'm running it on a Plasma box. And if you're on Arch, you want the version of the package that has elementary in the name. It's elementary-planner.

There's two planners in the Arch repo or in the AUR. You want elementary-planner if you're on Arch. And the beautiful thing about it is if you just want something super simple, local, it can do that. You don't need any integration with anything. It'll just be a really nice local app. It's a beautiful GTK app, but it looks just fine on Plasma 2. Then, if you are a Todoist user, you can connect it to your Todoist account, and it will sync and import everything from Todoist, and then you don't have to use their Electron app. That I was pretty impressed with. It just worked right away, no problems. Just had to, you know, log in like usual.

I like that. Now, I installed the Flatpak, which there is a Flatpak available, which made it really easy, also on a Plasma desktop. I've got some sort of strange black borders around mine, but otherwise it seems to be working very nicely, and I think I'm going to keep it. You can go into the preferences and change some of the appearance options. So you might want to play around with that, you know, like put it in dark mode, OVS. And then the other thing you might want to check on is what your GTK theme settings are right now for Plasma. There might be something Wonka there because it is a GTK application, but hot damn, is it great. And Todoist is admittedly not free software. It's not an open source

hosted solution, but it has a lot of collaborative features. It has integration with freaking everything. It's got an API for days. And so I just sort of eventually fell into using it

because they have mobile apps, they have web apps, they now have third party client apps. And that's that I like a lot. And over the years, I've tried a lot of things, but I keep coming back to Todoist because of this kind of flexibility.

And the idea that I can have this nice, well-designed Todoist client that's native on my Linux desktop and then have a native app on my phone kind of puts it in the sweet spot for me. So check that one out. It's worth it. It's totally worth it. So we have a bunch of feedback this week, Mr. Westpain. And it's hard to say if we could get to all of it, but I did want to get to the Tuxes

feedback because we've heard a lot from that. I'm really happy that people seem to have really enjoyed the Tuxes. It's been really well received. People who won the tuxes are excited. And it was a nice way to say thank you at the end of the year. So Roy writes in regarding the award show. He said, you hope for more people to vote. I think you should have mentioned the voting on other JB shows as well. And I think he's absolutely right. I will attempt, when my goal is for next year's tuxes, is by around Thanksgiving to have a url that is just you go to a url directly that we maybe even launch a week earlier than we did this time around and that we do

promote it on other shows and if you're out there and interested in maybe helping me build like a static website that we could use to also show who won. Because that was a request that came in by some of the Tuxies winners was, hey, we'd sure like to be able to share something to tell the team. And that makes a lot of sense. So a site that could hold the winners for every year, and then maybe you could click on it and get like their individual award page that says, you know, Fedora project for this or whatever, who was a winner,

and then maybe even an embed of the YouTube video with the time code. And just something that's easy, simple, fast, that we could update every year with the tuxes I think would be great. And so if somebody's out there that wants to help me with that, get a hold of me. You can use the contact form or you can email me directly, chris at jupiterbroadcasting.com or

probably Telegram or whatever too. I don't know if that's a good idea, but you can give it a shot. Probably the show contact is the simplest for everyone. Yeah, yeah. So anyways, that is feedback we are taking in, and we'll try to get like a visual representation of that as well. We also had a suggestion from Sandro about the tuxes. You want to take that one, Wes? Yeah, Sandro wrote in and had actually a lot of advice for us, but thought that it might be overkill for the kind of contest we're doing, but we could consider having a nomination phase

first and then a voting phase, you know, sort of accept proposals for each category, make a big list, and then in the voting phase, only allow entries from that list. And, you know, that might make some of the tallying a little bit simpler, but wouldn't mean we need to get things started even earlier. So like almost like a bracketing system where you kind of like have a phase one that we take the submissions and then we have a debate and discussion about that one and then move a subset of those to the final round. Yeah, exactly. That's an interesting idea.

I don't know how we would do that in a way that wouldn't drag it out forever though. Yes, Andrew, even linked to some voting theory and some more technical resources that we can take a look at to see how we might actually implement such a system. Thank you everyone for bearing with us on this. You know, I think it was a lot of fun and interesting, but it was definitely a rough first year for the tuxes. But I'm already looking forward to the next year. And I think there's lots of stuff that we can do to improve it. So if you do have further ideas, suggestions, tools, things we just shouldn't try again, please do let us know. Yeah, it was definitely a minimum viable tuxes. We weren't sure if the idea

was sound. We wanted to try it because it's something that we've kicked around for years. And I think it showed that it was worth investing more into. And we're interested in doing that. And it also could become a yearly tradition that we get to look forward to, kind of like the predictions and reviews that as a team is kind of like it's almost the biggest indication that the holiday season is here is that we start working on those shows. So it has some sentimental value. It's like a milestone in the production year of the show. Like when we get to those shows, we know we have done a solid year of shows by that point. And it's kind of a moment. So it's something we want to do

and we would love to get your suggestions. We'll take one last email that comes in from Five on GNOME 40's impending changes. It says, happy new years. I am a GNOME user and GNOME has a killer workflow for me that no other desktop offers without a ton of customization. I use the AutoHide top bar as my only extension to create a clean desktop. I'm slightly nervous of the new changes proposed to my bread and butter activities overview,

which if you're not familiar listener, there is some significant work going into how the overview section of GNOME will work. I'm sure the changes will be welcome, but the activities overflow is what I use most to move quickly to other applications. It's much quicker than sequential alt-tabbing and doesn't require crazy key combos or memorizations. I also like how I can easily manage my work and home workspace within the overview. Before the pandemic, I did a magic trick

and replaced Windows 10 with Ubuntu at work. My coworker quickly noticed how changing from Windows 10, the company standard to a GNOME desktop, was really key for my workflow and productivity. That's interesting, Five, that you had other people notice that as well. Yeah, I think GNOME users are sitting here a bit nervous at this point with GNOME 40 looming out there. It's one of the things we're watching for. It's sort of the upside and the downside, right? If you click with GNOME and the workflow and the way it's set up, it can be really, really nice.

But because it is so specific, it also means that any changes to it are going to be way more noticeable than, say, on Plasma. I want to believe that they've received and kind of internalized some of the feedback over the years. And I think really since the GNOME 3.10, 3.11, and 3.12 series, we've seen a more compromised approach to some of the design choices and some of the changes.

And we've seen a real increase in performance and stability. So it seems like the last solid bit of GNOME history has shown us that the developers are working in the right direction. And so it is possible that GNOME 40 lands and people have a lot of strong feedback about it. But so far, history has showed us that over time, they'll work some of that back. They'll modify some of it to meet user demand.

It may be jarring at first. And as a project, they have to figure out, okay, what are people freaking out about because change is hard and it's j, okay, what are people freaking out about because change is hard and it's jarring? And what are people freaking out about that is legitimate and we have to improve? And they'll just have to sit with some of these changes for a bit to be able to pick out the signal from the noise. And as a community, I think we kind of have to give them some of that grace for a bit.

And we'll try it. We'll be honest about what works and doesn't work. But at the same time, we're not getting the pitchforks try it. We'll be honest about what works and doesn't work. But at the same time, we're not getting the pitchforks over it. You know, they'll refine the process. We hope. Just have to wait and see. We'd love to have you join the show live. We do it Tuesday at noon Pacific,

3 p.m. Eastern. See you next week. Same bad time, same bad station. We'd love to have you join us because, of course, we got the chat room, the mumble room. It just gives us that live vibe, and we like that too. But if you can't get the show live, well, then you're like 99.9% of the audience.

You can get it every single week by going to linuxunplugged.com slash subscribe with all the links there. Our contact form is linked there. Our matrix and mumble info and show notes for this week's episode are all at linuxunplugged.com. Thanks so much for joining us on this week's episode of the Unplugged program. Also, I should mention we're at Linux Unplugged on Twitter. I never do that. I'm bad about that, but you can get show announcements and stuff if Twitter's your thing. Either way, whatever you do, we'll see you back here next Tuesday! Okay, everybody go vote, jbtitles.com. One of the meta themes about this episode that Carl just started to touch on before the show kicked off is how distributions sometimes do things differently.

You know, can you have an Ubuntuism? Can you have a Seussism? And there is different that's maybe a little more local, and there's a different kind. And, Carl, you were just starting to touch on it, and I said, wait, wait, wait, save it for the post show. Do you want to pick that back up? You remember what you were saying? Yeah. I think the gist of it was, is that every package in a distribution is sort of like a soft fork in a way, unless it's just a direct build of the upstream software, exactly how they intended, which a lot of them are close to that, but you know, there'll be compile time options and other

things like that. But most distributions will diverge each package slightly with patch files. And, you know, some of those are just to adapt it to the build system. Other ones are to fix other bugs or fix features. Maybe it's something that's already upstream that isn't in a release yet that they just want to backport to the most recent release. But the key thing is, is that every patch that's applied, it has a potential to cause problems, a non-zero percentage chance. And if a distribution is adding patches that the distribution came up with that aren't sent upstream, if those patches cause problems, it's up to that distribution by themselves to fix it. They don't have any help.

They don't have any help. But if a distribution practices what we call upstream first, and they are working with the upstream developers, they won't apply patches that aren't at least proposed upstream. And that way, if one of those patches does cause an issue, it's not just on the distro maintainers to fix those problems. They'll actually be able to collaborate with the upstream developers. Right. That's kind of where the upstream first mentality comes from. I totally, totally track. But at the same time, I have to do the well, but thing. Couldn't you? I mean, doesn't that seem like a little ideal scenario? Because, for example,

this is a really small example, but I could totally see this hitting an outfit like Canonical, for example. You hire a developer to make some changes to an open source project. They create the code, they package it all up, and you submit it as a patch. And then the upstream project says, no, we're not interested in going that direction. No, thanks. And then you're like, yeah, but I have use cases or end users that do want this. And you're kind of left, well, I can make the changes as open source software. I have the ability to fix it for my users. I mean, isn't that kind of, in a way,

what kind of drives some of this forking to begin with is just a difference in directions for projects so people just have to do it themselves? Yeah, absolutely. I mean, and that's the thing is whenever a distribution is evaluating whether or not to apply the patch in the first place, they need to weigh that risk. So if there's something proposed upstream that the distribution doesn't have any idea if it would be fixed or merged, then maybe that's not a good candidate to apply as a patch in the package yet. But if it's something that, you know, the upstream has already said like, yes, this is an obvious fix. We, we, you know, thank you for the contribution where we plan on merging this, or we've already merged it into the master branch

and it'll be in the next release. Those are those, those are the things that are a slam dunk to go ahead and add as a patch to the package, because you know that it's, it's going to, it's at least likely to be, you know, you'll be able to collaborate with the upstream developers on it. So it's a risk evaluation whenever you're first applying that patch in the package. Yeah, they got to do that math of like, oh boy, do we want to take this on and be responsible for all the bugs and never really being accepted by the upstream? Yeah, I agree.

Every patch has that risk and has to be evaluated. It's complicated, you know, assembling all these different pieces from all kinds of different people, organizations, and groups into what we view as one cohesive

operating system.