LINUX Unplugged - 337: Mystical Users

Episode Date: January 22, 2020

We make an appeal to keep Linux powerful and avoid the Macification of the desktop, and review the latest developer-focused XPS 13. Plus some community news that's getting missed, picks, and more. Special Guests: Alex Kretzschmar and Brent Gervais.

Got an email saying that my Pine phone has shipped. What about you, Cheesy? Oh, yeah, I got that email. Super excited. Pine noticed you weren't on the show last week, and they're like, what can we do to get Cheesy back on the air? What they said is we'll just ship the Pine phone. That's pretty amazing, man.

Yeah, I mean, that's a good call. They got me back here. It's the Braveheart edition, so this is going to be the rougher one. But they're shipping out, and I don't know. I guess maybe I'll have mine in the next week or so. We should come up with some fun projects in-house. We've got to put it through its paces. Right? We've got to demand the utmost performance. This thing has to be completely production- oh, wait. It's our responsibility to the community. Actually,

we just want to play around. Yeah, I'm a little jealous. I'm a little jealous. Hello, friends. Welcome in to Lytics Unplugged. My name is Chris. My name is Wes. Wes, we have quite the fun episode today because for more than a few weeks, we've had some really great hardware in-house. We've been kicking it around, benchmarking it, and just recently put it through some

upgrade paces. So we'll tell you about our thoughts on the new Dell XPS 13. Well, new-house. We've been kicking it around, benchmarking it, and just recently put it through some upgrade paces. So I'll tell you about our thoughts on the new Dell XPS 13. Well, new-ish. More on that, too. And I'm not alone in saying that we have so much to get to. So to help me say it, I'm going to let Alex and Chi say it. Hey, guys, say something great about the show today. It's a big, big, big, big show. Yeah, that's what I usually go with. Something great about the show today.

Oh, Alex. Dang it. I mean, I usually go with. Something great about the show today. Oh, Alex. Dang it. I mean, I thought that was funny. You're wrecking this. I'm sorry if it wasn't, but I don't care. Okay, good enough. Accepted. You know, I mean, I'll give you a checkmark for effort,

but I'm not giving it much more than that. And, of course, really, the thing that we can rely on today is the Mumble Room. We have a huge virtual lug. Time-appropriate greetings, Mumble Room. We have a huge virtual lug. Time-appropriate greetings, Mumble Room. Hello, hello, hello. Good day. The deafening cries

of the Mumble Room. That's massive in there. I mean, I don't know, it doesn't really come across as like super massive, but there's 14 on air, 21 people total just hanging out

in the Mumble streaming. It's the place to be. You know, these Tuesdays are really kind of a lot of fun because we stream live. We're doing a double recording. we won't be live next week. And so we get to hang out for a couple of hours, talk Linux, or sometimes the conversations go in weird directions.

I'll admit that's true for the pre-show. And that's why you want to show up live. But as always, we have some community news to get into, some really interesting stuff, really kind of important community news this week. There's two things that I think are worth chewing on this week, and one of them is when a community gets left out. It's both good news and bad news in this case. Fedora Core OS has come out of preview,

and they've announced that it is available for general use. That's great. But it also means there's a community of users, the ones that ran Core OS Container Linux, that are left out. They were told they should migrate to the new Fedora CoreOS. However, documentation is not yet provided to actually make that migration possible. Sometimes these things happen. And as all great things in open source, there's a fork.

And a community has spun up around flat car Container Linux, which is a direct compatible continuation of container Linux, a new community forms. But what are the admins that are running this in production left to do? They have to decide to either go with a non-compatible conversion to Fedora Core OS. Right, which will require a bunch of manual interventions. Or use this community offshoot fork

of a product that's no longer around. It's a lot to maintain, and these are production-grade systems. If you'll recall, one of the great advantages of container Linux was centralized configuration across all your different machines using etcd and automatic OS updates.

Isn't the irony here that the distribution that provided seamless automatic OS updates. Isn't the irony here that the distribution that provided seamless automatic OS updates has gone away and now you are faced with a completely incompatible upgrade? Isn't there some rich irony to that? Oh, yeah. And, you know, I think the team at Fedora understands that too.

They've got some sections here in the announcements like how do I migrate? What do I have to do? And what are the stability guarantees? And unfortunately, kind of just due to the way Fedora works, they're not in the same ballpark. And there's something else that's embedded in this story. This has all really been pitched as a positive thing,

because it's all been really about what's happening with CoreOS when it goes over to Red Hat after the acquisition. Then the story became what happens with Fedora and their atomic version versus Core OS. And oh, isn't it funny that now it's Fedora Core again? Like all of this has been talked about a lot. And then going from a preview to a general availability release and mastering images for AWS and whatnot has been all talked about fairly positively, but it sort of whitewashes over the reality of users that are left behind in a transition like this. And it particularly is stinging when it's enterprise

grade software. Right, you've been paying for it and you're using container Linux, they get bought and now you have to change how you're, you know, what infrastructure is running your business. Does this just sort of happen? Is this the way you roll the dice in the tech industry? Sometimes you bet on something that goes away and that's just how it rolls? Or is there something different we could be doing here? Well, I don't know about the, is there something different we could be doing?

I think we need to put a bunch of smart people in the same room and solve that one. But it feels like this is just the reality of the tech industry. Like if you look at open source projects, that's certainly true. But even if you look at the proprietary stuff and corporations, you know, they change all the time. So it's almost like you have to expect that the landscape is just always unsure or always changing. Maybe that's the thing you can count on. And so, I don't know. I wish I had a better answer, I guess. In all the history of software, computers, and so forth, have we ever really had any true stability? Really? I mean, so this is kind of expected in this industry entirely. I mean, I mean, I'm not an IT specialist, but dealing with the different product lifecycle management databases that I've used in my time at my employer, things just don't stay static.

So there's always going to be that pain of an upgrade. There's always going to be that pain of a change of a vendor. So this is just par for the course. This is just the business we live in. Essentially comes with the territory, in other words. And I can't really disagree with that. And you could even make the argument that the platforms and areas that have remained the most stable in the terms of changes in development also come with some of the highest risk profiles. Like I'm thinking of x86 and Intel's architecture.

It's been very consistent for a long time, but it's also now strapped with a legacy of baggage and vulnerabilities. Same with Windows. So there's downsides to it as well, to not doing this change. Yeah, that's true. And I mean, all these workloads are containerized, right? So you still have system and infrastructure configurations

you need to change, but hopefully your applications can migrate a little bit easier. Absolutely, Wes. I totally agree. And I think if you look at Red Hat's strategy right now, OpenShift is their primary future kind of direction, right? And obviously Relate's very important.

But if you look at it from the enterprise adoption perspective of what can we start getting more people excited in, what can we start selling to people, all that kind of stuff that Red Hat unfortunately do to pay the bills, you know, selling stuff. That's got to come from somewhere, and Fedora Core OS is a key part of that. You know, without that kind of upstream route to feed into the Core OS

that is, you know, Red Hat Core OS that ships with OpenShift, into the core OS that is, you know, Red Hat core OS that ships with OpenShift, you know, where are you going to get that kind of upstream feed from? So I think it's a necessary evil. You can't really have progress without change, as Tiger Wolf says in the IRC. Thank you, Alex. I appreciate that perspective. By a bit, and I wanted to give you a chance to jump in on this. Yeah, so we have a lot of changing technologies. And also, if you look at Google, they create a

technology, fuse it with others, and some technologies fall away. And that kind of forces us to innovate and move forward. So it is a kind of process and progress. Good point. Thank you. All right. I don't really know. I think I feel better. It was a good conversation. I'm sure the people that are running Container Linux from CoreOS, there's a lot of different names now,

I'm sure they'll find a path forward. It's still a good open source moment. You're right, this can happen to proprietary software. Things always change. You don't get free support for your products, but you do get the code, and that means we can continue on. Lots of stuff can happen to proprietary software. Things always change. You don't get free support for your products, but you do get the code, and that means we can continue on. Lots of stuff can happen, and you can still keep running yourself. And there's that whole, the way we communicate about it.

This whole aspect of the fact that there's all these users getting left behind gets left out of the coverage today because of the way coverage is engineered and done in this space. And as somebody who's in the space life, that's the aspect of the story that I find extremely fascinating. But we move on because there's a thought-provoking post over on Tobias Bernard's GNOME blog. And he posts over there semi-frequently talking about different aspects related to the future and present of Gnome Shell development. And he wrote a post titled Doing Things That Scale. And it starts with an area that kind of rings really true to me, and I wanted to share it with you and kick it back and forth

with Wes. He starts, there was a point in my life when I ran Arch. I had an elaborate personalized terminal prompt and my own custom icon theme. I stopped doing all of these things at various points for various different reasons, but underlying them all is a general feeling that it's taken me some time to figure out how to particularly articulate. I no longer want to invest time in things that don't scale. We will link to the full blog post because we're only going to cut through some of the highlights here. But he continues, Not only is it highly wasteful for me to come up with a custom solution to every problem, but in most cases those solutions would be worse than the ones developed in collaboration with others.

It also means nobody will help maintain the solutions in the long run, so I'll be stuck with extra work forever. Conversely, things that scale. A few examples that he cites is keeping things fairly simple. He says, until recently, I always set a custom monospace font in my editor and terminal when setting up a new machine. At some point, though, I realized I wouldn't have to do that if the default was just nicer.

So I just opened an issue. A discussion ensued, and a better default was agreed upon. And voila, my problem was solved. So far, this really rings true for me. It resonates. It ringsonates for me, Wes. Does it ringsonate for you? Yeah, you know, I certainly agree that sometimes the defaults are the reason that you need to go spend some more time and customize.

Now, I think you can do both in some of this, right? Like you can go try to get a better default, but even if the font is better, you might still prefer your own custom font. And fonts seem like the least applicable part of this. Like, yes, we should do that. But I do think the idea of if you want your tools to be better, if you can see how those could be better, that's valuable. That's worth contributing

because you can be a developer working on a project that might not, you might not know what the users think is best for them. Well, and to that point, you remember recently we did that Plasma live stream

where we went through how I customized the Plasma desktop. And Jonathan Riddle and a few others from the Plasma team watched it. They're like, you know, we're taking some notes. These might just be some defaults we tweak for you. We didn't really think that. You know, that's a good idea. That is, first of all, rare and incredible.

But it is truly, to your point, like, they're different use cases. They have different uses. They use it differently. They can't necessarily articulate in their mind all the different various ways that most people might want to use it. And so seeing different use cases and, oh, that does make sense. It gives them an opportunity. it really, for me, touches on that nerve of I push back on a few fundamental assumptions of the rest of the Linux community. And I feel like I am on an island onto myself in which I say something and people just think I'm a nutter.

And number one is I don't think Linux is for everyone. Full front, I don't think it's necessarily great for everyone to use Linux. Not right now, not yet. Number two, and quite frankly, I don't think what's hurting and holding Linux back is a marketing issue. It's so much bigger than that. It's so much bigger than that. It comes down to corporate structures. It comes down to agreements.

It comes down to ecosystems. It comes down to market demand and consumer willingness to switch. It's a massive issue. And when you take it all in, when you really think about it, when you go have yourself a hike and you think about it, you realize there is a narrow niche. Thankfully and amazingly, that niche, and I've witnessed it and you've witnessed it, has expanded over the last few years. Incredibly now, just at an even more accelerated rate than ever.

But there's a narrow niche of people that are going to have a truly successful transition to Linux. We're joking ourselves if we think Windows 7 users are all switching to Linux. Maybe some of them, some in the engineering and scientific fields, I bet. So when I see that we should fundamentally limit the technical capabilities of our systems so that way they're, quote, scalable

and easier for average people to maintain, I kind of put my hand up and go, well, can we talk about this for a moment? Because some of these make a lot of sense, but some of them don't. Like dropping separate home partitions, dropping dot files, dropping multiple Firefox profiles, dropping encryption, dropping email backups. These kinds of things I appreciate are hard for users to manage, but I think my fundamental argument

would be the niche that Linux is actually really applicable to. Those are nice features to have. That user base does care about those things. Because I picture it to be software developers, engineers, geeks, hobbyists, creatives, people who tend to really care about the state of their tools, because that's a tool to do a job. And I just don't agree with limiting the system to make it appealing to a fantasy user base

that you can't even articulate. It's a nebulous concept. The new user to Linux is a nebulous concept that each one of us has a version of in our heads. I think you are hitting here. I mean, there's a lot that's going unspoken. One, the definition of scale that we're using for this conversation. And you're right, I think

it's easy to take this as an intent to limit. And that's where it gets really tricky because as existing Linux users, we love that configuration, right? We love the ability to automate, to interface, to customize our solutions to fit our need because it's a tool we're heavily invested in. Yeah, how many times have you heard the mantra,

it's so flexible? Right. And I think what this article is trying to point out is that There's too much of that. There's too much of that. And where it's tricky is you try to get both, right? We've all had tools where you're like, I wish that there was an escape hatch in this tool

because it works 80% of the time, but I need to make one little change. And the reverse is also true where you're like, I could use Handbrake or I could use FFmpeg. One, you know, maybe I don't have all the options. And the other, I have to go cobble together my whole bespoke custom solution. It's hard to figure out what's in the middle of there that works for both of those use cases. I'll tell you what fits in the middle, Wes. And that's, I know this is a Linux podcast and I'm almost afraid to say it, but macOS. You know, that fits in the middle for a lot of people.

It has a decent terminal. It can run the Adobe Suite. You've heard all these arguments before. It's not Windows. Yeah, there's nothing revolutionary in what I'm saying here. It just gets the job done. You're actually kind of making what my next point was going to be, is I feel like if we go too far in this direction,

we're macifying the Linux desktop to a degree that just makes us a also brand to Mac OS. I wouldn't necessarily object to that, but then I suppose you could argue that that's what elementary OS is. Or at least attempting to replicate some of the successful aspects of what makes the Mac work. But I think if you limit some of this, what you really are left with is idealism, people who want to use the software for moral reasons, and people who would be running their own custom desktop environments that don't have anything to do with this whole stack.

But from like a market competitive analysis, say I'm, I was just watching, I can't remember his name right now, maybe somebody in the mumble or the chat remembers, but a YouTuber just posted, oh no, surprise, a very good video where he built a Pop! OS workstation to run DaVinci Resolve. And he compared it to a Hackintosh and Windows and the new Mac Pro, because he has one of those as well. So I actually found it right at my alley,

because this is my head-to-head kind of battle. And he clearly was new to Linux, clearly new to the whole system, but figured out his way, after looking at different options, of getting to Pop! OS and then getting a way to convert the RPM to a DEB so he could install DaVinci and get it all working. He even got the hardware proprietary dongle stuff set up and

had very positive things to say about it. Just after watching that, I looked at it and thought, there's a real market here for somebody who wants to build a custom system with a set amount of hardware that they need. Maybe it's disk, maybe it's GPU, and they have a budget, and it's less than $35,000. It really worked for him. And he liked some of the power. Like, Gnome Shell was really hitting that sweet spot for him

because it wasn't too complicated, but he could tell it gave him more than what he got from the Mac. Right, that's it. That's kind of the best case that we could hope for, and I think that's maybe some of the spirit that this article is intending, that better defaults, maybe not focusing on the limiting part, but better defaults can help new users and existing users. Because sometimes I

just need to test the latest Fedora out. And if GNOME looks great, that means I don't have to change anything. Would you agree that it's probably never actually going to be possible to target the quote, new users, unquote? Because like, if you you think about it everyone kind of has different expectations coming into something so whether that's someone who is fairly technologically savvy or you know grandma down the street i feel like it's really hard to define what the perfect situation is for the quote-unquote new user I could see two potential types of new users. I should probably address this, so thank you for asking, because I don't want to make it sound like I'm saying,

blah, don't make Linux for new users, because we have to keep expanding that niche. We have to keep marching forward, and the only way to get there is by people working on it. It's just we don't want to sacrifice what makes it so great for the thousands and millions of people it's pulling in right now. You know, we're about to talk about the XPS 13. Dell is very intentional in their targeting of the developer market with the Sputnik.

And I think that's a clear strategy that's working for them. And they didn't just go for like new users. They didn't go for a best buy blitz. They just focused on developers with a low-key marketing campaign. And I think that's telling. And maybe you could see in 10 years, they do have something that's on a best buy shelf because it has, that niche has expanded over time. So that's one type of user, just as the niche expands, it'll pull them in. I think the other type is, and I'm sure a lot of people listening have done this, family members, friends, associates, whatever, where you set them up a Linux system

and you just kind of check in on it for them from time to time. Like my son, famously, I talk about it all the time on the show. My son has been running Linux for like five years now, and it's no big deal at all. In fact, I very rarely even, he mostly just does the updates and all that. And I've done it with family members in the past for a few years or more. And typically I'll just pick something that is reliable and I just can do updates every three to six months when I see them. Or most distros now just say, hey, you need to do an update. And I've just told them, go ahead and do

them. And it works. And with a little bit of help, with a little bit of management, and of course, there's the corporate user, where they are just given a workstation, and they can be new users to it. So we still need to make it usable. I'm not advocating for not making it usable. I'm just advocating for being realistic about who we're appealing to as just a direct product. And to be, you know, conscious as we change, as we must, to keep some respect for the culture that we've had. Yeah.

And we have to, those that are advocating that, have to also be willing to compromise on those positions as things change. Very true. And I think that's something to keep in mind, too. It's not a call to arms. It's a call to further conversation. How about that? They should take that. too. It's not a call to arms. It's a call to further conversation. How about that? They should take that. They should just take that and run with it.

So I will put a link to that YouTube video because while we were talking, I did find it. And I just think it was super great. I had been watching this guy for a little bit because I liked his style. And then to see that video drop, I was like, wow, that's awesome. And it's so cool because he didn't even know

to position it as Linuxy clickbait because he's not a Linux guy. Of course. And so I didn't even really know for sure it was a Linux video. And I'm like, oh, man, he could have totally titled this differently if he knew what he was doing. Like the clickbait voice that's sometimes in the back of my head. Of course. You know, I think whenever we're talking about Linux and the user space and making it easier for entry level users or maybe that person that's an IT professional who's always been in a Microsoft Windows environment, switching over to Linux and things like that. I think one of the things that we really have to consider, too, is that Linux is really the kernel. And then the user space below that is ours to do with what we want, right? So different distros can target that different ways.

They can make, you know, a very user-friendly, entry-level kind of user. Like Chrome OS, for example. Exactly. Or you can make it as deep as you want. You can, you know, arch it up. You can Linux from scratch. You can, you know, so there's, I don't think there's one particular, like you said, I think there's this fantasy user that they're, that they're kind of perceiving is going to move into the Linux space.

And unlike all of these other available operating systems like macOS or Windows, you know, there's no, it's not like anyone has a stand up and says, hey, UI UX developers, let's target this for Linux, right? Like you have distributions that do that. Elementary has done a great job with it. Fedora has done a fantastic job with it, you know, but that same conversation is going to be completely different if that stand-up was with the Arch developers, right, or the Manjaro developers. They're going to look at that completely different than the way the normal layman would. I think your point's well made, too, that the eventual new user product will be something that is assembled using Linux, but is not likely just a raw ISO

that has been flashed to a thumb drive and then thrown on an x86 compatible or ARM compatible system. It's going to be your Chromebook. It's going to be something that's packaged up and ready to go. Or, I mean, maybe it'll be, who knows, I'd love to see this.

I'd love to see just vanilla Linux systems on good hardware, especially with some of the improvements we see coming to the Ubuntu desktop. The next update makes it look very pro. We have Ubuntu 20.04 installed on the XPS 13. We wanted to spend the last couple of days on the latest Ubuntu to see what it's like on this XPS 13. And one of the things that has come out of a recent get-together

by the canonical folk, like our buddies Wimpy and Popey, is a really nice new theme. So you remember Yaru, right? Of course. I think it's been the default since 1810. Right. And now it's getting a big update in 2004, and we got it installed on this XPS 13, and it's very gorgeous. I think it might be my favorite theme. Specifically the purple touches.

I don't know. Dark theme with purple. It's one of those themes where I've set it, and I've been using Yaru even on my Arch systems now for a while. Once I set it, I'm good. Yeah, it's really nice. Checkboxes, radio buttons, and switches all have little hints of purple. I think actually, really, it's eggplant.

Oh, my gosh. Thanks, Wes. Thanks. I think you're right. I think, yeah. Oh, it's it's poppy and wimpy, so it's probably actually taken from the emoji. You know, these guys and their emojis. You know what else is exciting

about this, though? Do you remember like 10 years ago, back in like, won't you, 1004? I don't even remember last week. With like radiance and ambiance, those themes? Yeah. So, those had light and dark variations of the same theme, right? Way back when, you could like, you could sort of customize it, have the dark version, have the light version.

I mean, I believe you. Yeah, right? Well, that's coming back now, too. So, now you're going to have Yari Dark, Light, and the regular Yari. Oh, if it wasn't clear, we are dark mode on that XPS. Oh, okay, I see, I see. You just weren't even considering the other options. I mean, I tried it this morning, because Wes was playing around with it yesterday on 2004, so I grabbed it this morning to play around with it on 2004.

And first thing I did was to check to see if yet they had baked something into GNOME settings, so I could just go select it there. Good news, that's coming. Ah, it's not there yet. Yeah, you still got to use GNOME tweaks. But not a big deal. It was installed, so.

Oh, you know what that means? It's time for a little housekeeping here on the old podcast keys. I already mentioned it, but I have a note to mention it again. No Linux Unplugged Live next week. We'll have an episode for you. So this episode's all about hardware. Well, it's about to be. Next

episode is going to be about software. Really looking forward to it. So bring your time machine. Yeah. And maybe this is a good time to go make sure you're subscribed in all the places you want to easily get the podcast. Linuxunplugged.com slash subscribe for those. You might remember we mentioned we were doing a live stream on using the terminal like your desktop. A day in the life of cheese.

And it turned out great. So it's up on YouTube now, youtube.com slash jupiterbroadcasting. You can go watch it over there. Cheese and myself just doing our thing on the live stream. Yeah, you know, and we got some really solid feedback too in the comments. And I would suggest anyone that's listening

that ventures over there to watch the video. If you have any terminal applications that you really swear by and use every day, drop them in the comments because there's been some really excellent suggestions in there. I love that. That's great. Did you see that Alex was on the Home Assistant podcast? Go check out Alex on the Home Assistant podcast. He did his thing over there representing the self-hosted program. One of the reasons, well, the reason we're not live next week is I'm taking Lady Joops down

for a major project off-grid update doing the solar, doing the lithium, redoing the entire power system, talking a little bit about that on self-hosted. So check out selfhosted.show if you're not listening to that. Also, rumor has it that

I'm going to be going on the show again in a couple of weeks to do like a deep dive into my setup. Even deeper than we do in self-hosted. So even more home assistant. Watch out for that. Nice. Good for you. And speaking of great content you can catch, our man Brent sits down with Mr. Jim Salter of the TechSnap program and Ars Technica.

He was on our show here recently in the latest episode of Brunch with Brent over on extras.show. I love brunch. I want chats with Chris, dinner with Drew, and we need something for you, Wes. Something about drinks with Wes? No, that's not it. Wine with Wes? Oh, there we go. That'll work.

And then a special note, special programming note, everybody. Texas Linux Fest has extended their call for papers until this Sunday, my birthday, January 26th at 5 p.m. Their time, which is central time. So you got a couple more days. Remember, you don't have to have the perfect fully done talk. You just need a great pitch, a good summary, and an idea of where you're going to take it, and then you use now until the fest to make it just right.

And as long as you get it finished by the fest, no one will know. We're going to be there. Are you going to be there? I can't remember. Yeah, of course. Oh, my gosh. Great. And Alex is going to be there, and Brent's going to be there, and Cheesy.

Absolutely. I'm sure L, of course. Carl's going to be there and Brent's going to be there and cheesy. Absolutely. I'm sure L, L, L, of course. Carl's going to be there, of course. Yes, sir. And it's going to be in Austin again this year. So barbecue what? Yes. Carl, we've got to make that happen because Hedia has been giving me two years of crap

for not taking it. The marriage is on the line. It's important. Hey, Carl, is it in the venue across the road from Terry Black's? Did that happen? Yes, sir. We're going to be smelling it on the line. It's important. Hey, Carl, is it in the venue across the road from Terry Black's? Did that happen? Yes, sir. We're going to be smelling it from the fest. You know that because they pump all that.

Right. Oh, my gosh, you guys. So there you have it. Texas Linux Fest Call for Papers extended until this here Sunday as we record. So go check that out. And then Linux Fest, Scale, and Texas Linux Fest all coming up. Scale first, right? Then Linux Fest. Then Texas Linux Fest, Scale, and Texas Linux Fest, all coming up. Scale first, right? Then Linux Fest, then Texas Linux Fest.

Right now, we're planning to go to all of them if we can, so we hope to see you there. Who knows? Will plans change? I don't know. I don't got a magic 8-ball. You lost the magic 8-ball? Come on, I trusted you. I let my kids play with it.

They asked all kinds of questions, mostly about farts. So Wes, do we have the hardware in here? Yeah, we did. It's actually at your feet. Oh yeah, there it is. So we're talking about the XPS 13 7390. This is no longer the absolute latest.

Can you order the brand, brand new one that was announced at CES? I don't know if it's shipping. It's on Dell's website. I'm not sure when you'll actually receive it. So we wondered, is there still a point to this laptop, number one? And number two, where has the XPS line gotten to? Because something that is extremely important to me when I'm investing in a hardware platform is I really like to have insurance that that platform will get better over time. And let's say after I've used a laptop for three or four years, I can buy another one of that same generation,

and it's iteratively gotten better. Right, it's no fun having to jump brands each time you need a new laptop. I mean, we saw how upset the Mac users got when the MacBook suffered there for a few years, and they've really only gotten one good one. That's right. Let's be clear, there's only one that's got a fixed keyboard, for heck's sake. Anyways, I really think you

can say that is true for the XPS 13 line. Yeah, we've both been using them over the years, and there hasn't been a huge regression. You and I both have been and are XPS owners. Not our daily drivers, but still really great. My wife uses an XPS 13 as her main computer.

She loves the portability of it. Great screen. That's always been true. So where are they at now? And where is it at performance-wise? Well, like we mentioned, we recently launched this. We came with, I believe, 1804. And we messed around with a couple of different distributions. We finally just put Ubuntu 2004 on here for fun.

But something kind of hit me right in the face when I first started playing around with this thing, as I had some trouble booting other Linux distributions on there. It's this newfangled hardware, Wes. What exactly was going on with that? Well, there was fast boot enabled. Basically, the default BIOS settings made it very difficult to get a USB drive recognized. You kind of, first of all, it's only USB-C ports on this thing.

The default BIOS, you say? Yeah, that's right. No, it's UEFI now, Chris. The default BIOS, you say? Yeah, that's right. No, it's UEFI now, Chris. UEFI? Is that a worm? So you're going to live the dongle lifestyle, right? So you got your little USB-C to USB-A adapter. You've plugged in your Linux distribution.

You're hoping it's going to boot, and you just get Ubuntu again. Yeah, you get kind of weird stuff, actually, depending on the distribution. In elementary, I got the elementary OS splash screen, and it started booting elementary OS, and then when the splash screen disappeared, I was in an Ubuntu desktop. It was the strangest thing.

And then when I booted Manjaro on it, I had really kind of strange mouse glitchy issues. But all of that kind of, I just attributed to the thing being so dang new. But after you solved it, it was really kind of no big deal at all. Yeah, I spent most of my time with it running Arch, and it was performing great. I love that you put Arch on it.

You're such a maniac. So like we said, this is now no longer the most latest and greatest edition. They have one that has been just announced at CES. that has been just announced at CES. So we called up Dell and we said, hey guys, what's the difference between this here laptop and the new one? What's the big difference? And Damon from Dell obliged.

The big difference is the focus on the 9300. So first, it is a completely redesigned machine. Even though they look very similar side by side, and again, that is because the design tenants have not changed radically, but the materials and the technology that we've got have let us continue making it smaller, smaller, smaller, right? And you can only eek so much out, but in this one, we do, you know, have managed to get it a little bit smaller and maximize the screen. So we're maximizing the things that matter, the screen, the keyboard, the touchpad, right? So the touchpad is bigger. The keyboard, if you look at it from the two models, the keyboard now goes edge to edge.

Yeah. And of course, the screen is probably the biggest point of difference between the one we have in studio and the one announced at CES. The display, we've gone from a FHD and UHD, which is, you know, your 1920 by 1080 and your 3840 by 2160. We've moved up to a 10, excuse me, the 16 by 10 aspect ratio instead of the 16 by 9. So you get that extra vertical space. And the way we did that is by almost eliminating the bottom bezel. There we go. A little bit larger screen ratio, a little bit nicer keyboard to body ratio.

And those sound like small changes, but, you know, to give credit to what he said, to make those kind of changes on a laptop that's already this small with such a nice screen, that's a lot of work. Yeah, but we, so we had this one in-house since before Christmas, and Wes and I traded it back and forth. Wes got to spend quite a bit of time on it, and then we decided to just put it in studio and run benchmarks on it. And I'll tell you, if you're thinking about a laptop like this, here's a couple of things that I really like about this. Number one is it's all USB-C now, even the power, we've all gone USB-C. And because I kind of just over the last couple of years have been making the transition to getting a dongle for this or changing from an HDMI cable to a USB-C cable,

just kind of making these changes. Pretty easy to slot this in now to my existing workflow. Right, once you've got USB-C docks hanging around, then okay, not a big deal. The keyboard remains pretty good. It's not my absolute favorite keyboard. No, I mean, I'm using the ThinkPad most of the time, so it wasn't quite as nice of that, but it's so much more portable, it kind of makes up for it. And unlike our ThinkPads,

this has a nice backlight keyboard too. Backlight keyboard and that screen. Yeah. Okay. So the screen is really great. Even on ours, the non-CES edition, the screen is really great. And so this is what I wanted to say, is I think this laptop now that they have released the CES edition is sitting in a very sweet spot for those of you who are okay having something that's just about as good as the latest and greatest edition. And if your workload is disk, IO heavy, or CPU heavy, or those two things together, you would be very impressed with what this laptop can do. Again, we'll have comparison benchmarks linked in the show notes. You can run the exact same benchmark on your hardware and see how it compares to this laptop. But in short, this thing was

benchmarking at ranges of high-end Ryzen desktops with high-speed disks. Like, it's really crunching numbers in the CPU as fast as possible, really lightening up the cores, and it can suck data on and off of that disk. Yeah, we did some tests doing a whole bunch of video encoding, and it performed very admirably. Sure, the fan does kick on, but it doesn't get crazy hot because it actually has very good thermals. So I couldn't tell you, well, it needs more performance.

I couldn't say to you the screen isn't good enough. The trackpad, I think, is the best trackpad you can get on a computer pre-built with Linux. The build is, it's like all the previous XPSs, it's very good. The question is, can you actually buy it now that the new version's been released? And could you maybe get it at a slight discount? Maybe you want to do a little budget shopping. Well, I asked Damon just that. So I'm glad you asked that because interestingly enough, a lot of times you'll see manufacturers discontinue a model

after they bring in the new one. And they're going to actually let the older generation model live on for a while next to this one. I think it's an experiment to see how, you know, folks feel about the changes that we made and whatnot. But so you'll be able to go to Dell.com, and you'll be able to click on XPS 13, and it'll give you an option. You know, XPS 13, 7390, or it'll say new XPS 13, and that'll be the 9300. So they will live on side by side. Right now, I think both of them are starting at 999, but things do go on sale from time to time.

That I think is the hint there. And depending on how you price them, the prices will change. But I think if you watch for a sale, the version we have here in studio, the 7,300 version, it will be very competitively priced soon. Last time I looked, there was $100 off going on right now over at Dell.com. Oh, really? I mean, mileage may vary, but yeah, it's definitely worth looking at. Yeah, because I could see them keeping this around and doing special discounts, Black Friday sales, all those kinds of things. Because there's just a scale of economies just since they've been building this for so long and they have the newer version now.

And there's really nothing I could ask. Would I like a slightly wider screen? Sure. The other thing that's pretty great is the new edition has been changed to 1610. This is a 16 by 9 ratio. But again, I don't know if it's really necessary. So I thought, well, where is this going? Because in talking with them, it's clear that Dell is feeling pretty good about the success of the Sputnik program.

It started as this crazy idea of let's just ask them what they want from a computer and then let's take our best rig that we sell and give them that, which was a weird concept. Right, it didn't start with like a netbook line, something cheap and easy.

Which is what everybody else was doing. They had to actually take in a crack, I don't know if you remember, like at a really cheap laptop. Oh, right, right. That was from a different area of the company. And this was a risk because it involved spending a lot of time ahead

upstreaming drivers. Before they even shipped a product, they had to pay developers to start upstreaming, right? And work with their partners to do licensing and like this. Wait, how do we Linux again? Yeah. So it was considered sort of Barton going out on the edge with this idea and using a fund that had been established in the company for crazy ideas to just seed this thing.

Now you fast forward and it appears this has been looked at as one of the more innovative successes inside the company in a while. Like that's where we're at now. Like we could be seeing big things coming from this. Right. I mean, with a big corporation like that, if it's not going to perform well for the company, at some point they're going to exit. And it's been going on for a long time. And I think we as a community need to recognize not only the importance of that, but how great it is that we have a really rich hardware ecosystem

where each of them have their advantages. There's clear advantages from each vendor right now. And so you have to wonder, where are things going in the future? Who's going to take the risk on the next big thing, like an ARM laptop or maybe an all Ryzen system? I know a lot of people listening are asking, did he ask them if they're going to do Ryzen? Well, I did. And it sounds like maybe it's not outside the realm of possibility.

You know, I think if you just look at the product lines, Dell already carries AMD product lines, AMD components in our product lines. There's no reason why in the future, if that makes sense, and we hear that feedback from our customers, why we couldn't incorporate something like that if it makes sense in the bigger picture, right? Arm, I'm not really the right person to comment on that. You know, I can tell you that we are always looking at the technologies that are out there because we don't ever want to look out there and just say, oh, well, this is off the table because we just, you know, we're always looking at the technologies, seeing where they're going, where the pros, where the cons, strengths, weaknesses. And does it make sense for our customers? Is it a good strategic move?

So it sounds like if enough people ask over time, they could go that direction. And that'd be pretty neat to see. Something like that XPS 13, but all Ryzen, would really be cool. I would love to play with that. And Ryzen's getting more and more applicable for something that size. So Wes, I want to give you a chance

to talk about it a little bit since you spent a considerable amount of time with the laptop. Yeah, I pretty much use it as my day-to-day work machine for at least a week, I would say. Yeah, and then plus all of the horrible torturing. I just basically say to Wes, like, let's run these benchmarks. And then Wes really has to do all of the setting it up.

And I was impressed. I mean, we had like net data going. I actually set up a little data going into a Prometheus cluster so we could watch as it was encoding. It's pretty geeky. It was pretty geeky. Yeah. I was impressed. cluster so we could watch as it was encoding.

It's pretty geeky. It was pretty geeky. I was impressed. Now, there's a couple of things. Dell still hasn't fixed the EFI on their XBS line. It's not like it's terribly broken. EFI, you say? Yeah, but if you want to do fancy stuff like that, you can boot the kernel with what's known as EFI stub, where due to some clever

hackery on the kernel's part, you can make the kernel act like an EFI executable. So you can just boot the kernel, no bootload or nothing, but you obviously have to give it some commands, right? Like tell it where the init ramfs is and the root and all that. EFI supports all of this. Unfortunately,

Dell's EFI does not pass those variables. The upside is I discovered a neat little project which is an alternative. Maybe you don't want grub and you don't even want systemd boot. I don't. Nope. Super minimal. It's just a little C program. All you need is like the GNU EFI development libraries and you just build

yourself a custom bootloader to just boot your kernel. Nothing else. That's all it does. That's pretty slick. Yeah, so if you're interested in that kind of booting hackery, we'll have links to that. I actually kind of appreciate you mentioning that kind of feature specifically for this class of workstation.

It's meant for developers that are building things you might need to boot into another OS. I mean, ideally I should think Dell would love the idea that mainstream open source and free software developers, large, maybe Linus or Greg KH would be using one of their

devices. But supporting things like that are kind of necessary. It's the kind of details that that market is going to find in your product. But again, if this is what we're getting down to in terms of our criticism, it's also a measure of how good they're doing. Right. So the flip side is once I got that working, boy, this thing boots fast too. It's just performing all around. Yeah.

I can't stress how fast that thing can get bits off the disk and shove it into the CPU. It's so glorious. It's just performing all around. Yeah. I can't stress how fast that thing can get bits off the disk and shove it into the CPU. It's so glorious. It's amazing. Now, the other issue I ran into, I mean, out of the box, it was working great, but as we mentioned, even if this isn't the newest laptop, it's still pretty darn new and it's actually got a Wi-Fi

6 chipset inside of it. Yeah. Tell me about that. Now, it has already been upstreamed, but if you're not running one of those distributions that gives you a nice new modern kernel, you're going to run into some problems and probably won't have Wi-Fi out of the box. If you install, say, an older elementary OS. How was it under Arch? Oh yeah, Arch worked

just fine. And did you have to do anything with 2004? 2004 worked right out of the box. And thankfully, it is Intel Wi-Fi, so there's actually just driver firmware you can go download and add to your system even if you don't have

the newest kernel. You know, and if I'm not thinking through this, when we did the 2004 install, we opted to use ZFS. Sure did. So we're talking about

how fast this dang thing is with using ZFS on it too. Just as a note, it doesn't seem to have impacted our perception of the performance. No, not at all. Huh.

Or really, the benchmarks. I mean, we did those benchmarks all using ZFS for this last round. Yeah. It screamed. Pleased to report to you that in our early 2004 testing, the ZFS option is still performing great.

Yeah. It's been a lot of fun. My only kind of criticisms that I really have for this thing is the fan noise is a little significant. Once it's up, if it's really doing work, you're going to hear it. And the kind of workloads that I would throw at it would push the noise levels that I am comfortable with because I work on mic a lot. And so it is an issue in that regard. And so if you're sensitive to noise, it's not also a very, it's not a white noise. It's a little bit of a harsher kind of a pink noise. You can definitely tell those fans are getting aggressive.

Yeah, but it's also, you trade for that, you get something that's so thin you don't even notice it in your bag. What was so great about it is both Wes and I could just slip this in our bag that's already full of stuff and it was no big deal. Right, it would also be perfect, maybe you have an existing work laptop and you also want to bring your personal to keep that separate. Yeah. Beautiful. And I think if you remember, they're targeting the developer where that workload is often, it's bursty in terms of CPU, maybe you're building,

but otherwise it's terminal or it's your IDE and a browser. And in those conditions, like when I'm, so what I, when my famous, like just sort of sitting there working mode that I did is YouTube video. Then I pop that out into picture in picture, then I use the browser, I'm using the terminal, I'm doing stuff but nothing massive, chatting, looking things up, putting together a doc. Active multitasking, trying to get stuff done. In that scenario, you don't even have to hear it.

I mean, it's not. I mean, we're talking like when you're really pushing with that multi-person view. The thing to know is that the curve settings, it's quiet almost all the time and then it gets really loud. And as you would expect, but it's just one of those things that I'll also mention, is in that, I saw battery life ranges from five hours if you're not pushing it very hard to three hours if you're pushing it pretty hard. Remember that thing's got a very bright screen, too, so you can

back that down a little bit, and that'll also vary your mileage. I think if you do some tuning, you'll probably see, you know, seven or eight hours if you're not doing too much with it, but you're right. There are some trade-offs for the portability. We both really like holding it too, which is a weird thing, but we both, I think, really enjoy the feel in the hand. Where they've put the bumps on the bottom, it makes it easier to hold onto so it doesn't slip out.

It's got some traction. It's just a pleasant machine to interface with. Yeah. If you are in this area of considering a MacBook Air class machine or a MacBook 13, that range of device or a really nice high-end thin ThinkPad, I would definitely add the XPS to your line, if that's your category. The thing about this XPS 13 is it packs more of a punch than you expect in that small packaging. Yeah, I think the only complaint I still have, including with the new model, is that 16 gigs of RAM. Yeah, it needs more RAM. It needs a lot more RAM. But they're using a very special low-power everything in that thing, so I can understand that.

But it's pretty great. We didn't play around much with the battery testing, so there's probably others that will give you more details on that because we use it. First, a lot of times our process is use it as a work machine, then run it through all of its paces as much as possible. And then something we try to do in all of our reviews is post at least one comparable benchmark

so you can see where it would fall in your range, see if it's worth the money for you. And then the other thing we try to do if we have time, and thankfully Dell was very generous with our time on this one, is we try to throw a couple of different distros on there. Because I like to do that. If I buy a piece of hardware, I want to know that I can run more than just what ships on there. I think that's what's so funny about this. We're really celebrating the fact that you can just buy this Linux laptop and it comes with Linux, but all of us want to go customize it. Isn't that the thing, though?

Well, for one, I like to explore. Two, I want to vote with my wallet, and I want to show that I'm buying a Linux machine. Absolutely. And to me, if it's running a mainstream distro, what it says is it runs a mainstream kernel, likely. And so even if when it's brand, brand, brand new, it's got some patches,

the great thing about these Dells is they're upstream in that stuff. So you give it a little bit of time, and it's just in the mainline kernel. I love that. And it's pretty nice. So it's a pretty good machine if you're in the market. And we should put some links to it in the show notes.

That would be good. And thanks to Damon for stopping, well, for letting us call him up and ask questions. I was going to say stopping by, but you could probably tell we called him on the phone. We do that every now and then. We call them so you don't have to, right? That's right. Yeah, before we get out of here, we have another unplug to record,

but this is nearly the end for the download audience. We have a pick, and this one's pretty cool. It's called Glow. And I actually don't remember which one of us found Glow because I don't think it was me. I was going to say it was you, but maybe it was Cheesy. Cheesy, did you find Glow?

Maybe. I don't remember. I love it. The thing is, we've been pick shopping. We spent the holiday, a lot of us, making lists of our favorite picks and stuff. And so we've all got ones that we're just throwing in the mix right now. This is a really nice-looking way to render Markdown on the command line. It must have been from Cheese's video.

It's super easy to install. You just brew install glow. Anybody? Get out of here, Mac guy. Actually, no, there's brew for Linux now too. It's fine. It's fine. I know. I know. I know. But I was proof for Linux now, too. It's fine, it's fine. I know, I know, I know, but I was just trying to trigger you.

I just was seeing if you were listening. Of course, there's other package managers that have it, too. Yeah, it looks like there's binaries for Windows, Linux, and FreeBSD. Yeah, it's in the AUR, too, by the way, Arch. Anyways, it's just really nice because we often will be talking about these massive Electron applications. Well, instead, you can do it all on the command line. It might all be NPM on the back end, but at least it's on the command line. That doesn't matter, right?

Web apps on the command line don't matter. What you do is you run this in Hyperterm or Hyperterm or whatever it is, that Windows terminal that's based on Electron, and you're really set. You've got to make sure you keep Electron around. So we'll have a link to Glow, which does look really nice. I kid around, but it actually looks really fantastic. I love that their description is render markdown on the CLI with pizzazz. And pizzazz is in italics.

Which made me think of pizza. I want a Linux powered pizza oven. That's what I want. I want to do a Linux unplugged pizza party like we did that I want. I want to do a Linux Unplugged pizza party like we did that barbecue. I also want to do a barbecue, though. There's going to be barbecue pizza, my friend. You're reading my mind, Wes. You're reading my mind.

All right. We'll go over to linuxunplugged.com slash 337 for links to everything we talked about today, as well as linuxunplugged.com slash contact if you want to correct us or, more importantly, share a pic or an insight or something you've been thinking about. We love hearing from you. And last but not least, go get more Westpain over at techsnap.systems and get daily weekday news linuxheadlines.show every weekday in three minutes or less. We'll see you back here next Tuesday. It's for people who like to mess with computers.

And you know who you are. It's for people who like to mess with computers. It almost sounds threatening. And you know who you are. Does it? Leo, don't come for me. It's for people who like to mess with computers. And you know who you are. Does it? Leo, don't come for me. It's for people who like to mess with computers. And you know who you are.

If you're somebody who doesn't want to mess with it, I just want to surf the... I just want to buy something on Amazon, send an email to my kids, look at some websites. If that's you, you don't want to mess with it, probably not a good choice. That's not what I was saying today in the episode.

Just want to make it clear. That's not what I was saying. That's not what you were saying? Don't, don't want to make it clear. That's not what I was saying. That's not what you were saying? Don't, don't. Ooh. All right, let's go pick a title over there. JBTitles.com. It makes me think, you know, in the U.S. we've got this whole impeachment thing,

but it's always very popular to go find the clips of the politician from like 10 years ago saying the opposite. It makes me think we could find some last clips of you. That might be pretty entertaining. Oh, no kidding. That's where some of this comes from is reflection on I was one of the loudest mouthpieces for advocating everyone use Linux. And I mean, you remember

the Switch competition? Not only was that manifested in the podcast, but it came out in my career. I made a consulting career out of going in and converting people's Windows installs to Linux in the last part of my my career. I made a consulting career out of going in and converting people's Windows installs to Linux in the last part of my IT career. And sometimes looking back at it, I think in most cases for file and print and basics,

I was always right. But I never even really considered Windows. Right, to stop and think of like, this is really what I should be doing? And then if I had to use Windows, it'd be in a VM. I'm like, this is really what I should be doing? And then if I had to use Windows, it'd be in a VM. And now I've gone from a diehard, what would you call that, evangelist to, I don't know if advocate's the right word. But now I'm much more of this is a tool that works really well for me and I think works really well for a lot of uses. But if there's a tool that works really well for you, and I think works really well for a lot of uses.

But if there's a tool that works really well for you, that's fine. It seems like it's kind of the stage of any sort of relationship, right? You fell in love with Linux at first. It rescued you from the depths of Windows. So it's natural to be infatuated with it. Yes. But now it's just, it's another tool in your tool belt

and you can think about it a little more critically. Yeah, initially it was solving really significant problems for me. And it was doing it for free. And I was blown away. And it was just, yeah, this is something I can even do. And now it's transitioned to,

well, I can understand why using Photoshop's important for you. Or I can appreciate that you like the Mac better. That's fine. I think this is great. You know, I really, and I think part of it too

has been my distro hopping and my desktop environment hopping where I get really into something for a bit, and then I move to something else. I'm like, well, this is great. And then I realized, no, it's all kind of great. And whatever I want to use is fine.

Right. It doesn't have to be the only solution for everybody. There's a lot of good options. Yeah. And I think the other aspect to it that we sort of failed to appreciate in 2020 is it doesn't really need the same kind of evangelizing. We won in terms of market penetration in server and mobile

and as a default way for most software development that happens at scale. Like that's just now a given. That wasn't a given in the late 90s and the early 2000s. Especially after Microsoft survived the antitrust trial. It was like, oh crap!

They're still here? Yeah, that's where I think a lot of the anger towards Microsoft came from is it was a competition for how software was going to be developed for the next decade and beyond. 20 years now, right? 20 years. Right. This is when we were dealing with Y2K. I mean, put this in context. This is when we were, this was the time. And in that time, things were not so great. There was no Linux in everyone's pocket. There was no every

router running Linux. There was no cloud. There was other people's computers, and they ran everything from Unix to NT to Linux. Right, I mean, we'd had the whole Unix wars era and Windows and... Yes, SCO. Oof. There was a lot that just was uncertain. And so it felt necessary to evangelize Linux

in a very direct way. Now here we are where most people's information's in a cloud provider that they're getting access to through a web browser. We have Electron. We have Snaps and Flatpaks. You have just about every vendor doing their best to appeal to every type of user with Microsoft with the subsystem and with Apple's new Pro hardware and with the various different Linux distribution attempts and Del Sputnik program.

It's like there's so many options that all can do the job now. So it's kind of like whatever works for you, I can't really pass judgment on anymore. Who the hell am I? I just know it works for me. I know it works for our group. And beyond that, I'm good. Have at it.

Whatever you want to use. And so I think that's where I look at this and go, don't make my tool that I use for my job less powerful. Don't take sudo away. I mean, imagine. And you laugh, but they're talking about no home partitions. Like, that's better. No drive encryption.

That's better. It's like, oh, yeah, for average users. But that user doesn't actually exist where this user base is here and expanding at a very rapid clip. Right. And it'd be interesting to have some more examples because I think home partitions are a good solution to, or are a solution to a lot of things, but probably at least a subset of those solutions could be better handled or more approachably, scalably handled in

other ways, right? So with encryption. It's a technology problem, not a human problem. The software could be managing the encryption and managing the partitioning, and the user could be blissfully unaware of it. Or, you know, maybe in modern parlance, they should be sub-volumes or whatever. Right. But I think disk encryption is that example, right?

Because five years ago, you had to go learn a whole bunch about Lux and do it all yourself and customize it to get it at all. And these days, it's just a checkbox when you install Fedora.