LINUX Unplugged - Episode 204: Awkward Distro Puberty | LUP 204

Episode Date: July 5, 2017

TUXEDO Computers & System76 have announced their own Linux distributions, but both these new efforts betray a much larger issue that no one is discussion. We’ll have that awkward conversation. Plus OutlawCountry is getting a bunch of attention, BFQ scheduler finally gets its day, XDA Forum is going to give Phoronix some competition & some important info for Fedora users. The an update from the recent SNAP sprint, community news & a lot more!

Someone here at the barbecue though, he asked like, so do you guys submit bug reports during your reviews? Like if you come across a problem, do you submit a bug report? And I said often no. That's not always the case. We sometimes do submit bug reports. But most of the time when we review it, we're reviewing something that's been officially released. And so we review it as if it's a finished product and then give our thoughts. But we do submit a ton of bugs and work with the communities that build

a lot of the projects that we use internally to make the shows. So what's happened with this video? So it's a Python. If you search for Python Audio Visualizer in Google, you'll come across the project. It is amazing what's happened in the last couple of months because of the Jupyter Broadcasting community.

They have taken this thing from a pretty decent application that sits on top of FFmpeg. And I mean it almost chokes me up looking at it because they have taken this into something that is more competitive than anything that's available commercially, including some products I have looked at that require a monthly charge. And this thing is more competitive. That's crazy. And just because the community just got involved. And it's going to make our shows better.

And it's going to make late night Linux better. It's one of those great examples, right? Where clearly this was a need. People wanted this. Maybe not everyone had that clear conception yet. But once it was there, once it was started, there's just this natural momentum that's fascinating to watch. Yeah, I love open source.

You could see how media production on Linux can be more successful in this paradigm. I don't know if it's going to be the true, maybe it's not the same for video editing, but this is something you can't do on a Mac or Windows. This is Linux Unplugged, episode 203 for July 4th, 2017. Oh, welcome to Linux Unplugged, your weekly Linux talk show that's feeling the patriotic spirit or the bourbon.

My name is Chris. My name is Wes. Hello, Wes. Quite a show for us today. That's amazing. Not only do we have the big barbecue going on right now as we record this, but we have some breaking news applicable to us Linux users, and apparently it is the season of stupid. We'll be talking about some developments in the community that really have your host quite ponderous.

Then we'll give you a few updates on some stories we've talked about, discuss something big and bold the XDA developer forums is doing that us Linux users are going to like quite a bit. And then PackageKit. You know it, you love it. It makes cross-platform package management possible. Turns out, a little crashy. A little crashy.

A little unstable on Fedora. We'll give you the reasons why you might want to avoid PackageKit on Fedora, at least for this release. And something might change in the future. And then we'll discuss System76's new operating system, as well as a way for you to hide your Steam... Well, I won't quote RMS. We'll just say it's semi-negative in the Freedom Dimension games. We'll tell you how you can hide those and isolate them from your system. You found that one.

Oh, yeah, I did. You found a good one. Plus, it seems that Rust is taking over as the language of choice for the Linux desktop. I think we're officially Rust fanboys on this show. I don't even know how that happened. Well, I think because we both listen to Coder Radio. It's because you like Firefox so much and you're excited about Servo. That's totally what it is, dude.

No, you know, it's like Apple's got their Swift and we've got Rust. You know what I mean? And I think we— The branding is traditionally an open source. We're not quite on their game with that, but it is fun to say. And Rust looks like a really, really serious, really good technology. And we'll tell you about how it's making—potentially. I guess it hasn't actually come yet, how it could potentially make GNOME much better, a little more crash-proof, thanks to a Google Summer of Code project.

But, Mr. West, I hope that the audience out in the living room, we have a small group out in the audience in the living room, as I just said. You say small, but those are some heavy hitters out there. Jeez, and boy, was it some good food. So we're doing the barbecue right now, the 4th of July episode 200 barbecue. And all-star Wes Payne right here. I mean, knocked it out of the park with the bratwurst. Unbelievably good. And rumor has it you are up till 2 a.m. working on those Sons of Guns.

Oh, maybe. Just a little bit. Soaked in butter, beer, and onions. Nature's finest. Super good, Wes. Super good. And everybody ate those things up. The food here has been amazing today.

We're cooking ribeyes. We're cooking burgers. We're cooking the bratwurst. And we've also got some great veggie skewers. Don't forget that. Oh, you got it. No, go ahead. I was going to say that avocado.

The fried avocado. Just really a game changer. But I'm hoping that those people out in the living room, I'm talking directly to them, I hope right now that they will keep an ear out for a knock, knock, knock at the door because UPS is supposed to deliver my brewery today, which is a Linux-powered, one-touch Linux brewing machine of the beer. No way. Of the Beer, Wes. And that should be delivered as we do the show today. It was something that Blaster and I went into. So, buddy, I think it's almost here. If you hit me up on Telegram, we should definitely do a vlog or two about this thing.

I mean, you put all the ingredients in there, you push a button, and then you come back in a couple of weeks and you got beer. So you're saying in a couple weeks, I can come back. I won't have to bring beer. You'll have beer. I think that could be the case. In the meantime, I've got something better for you, though. You ready for this? Surprise me.

It's our virtual lug. Time-appropriate greetings, Mumble Room. Greetings, Mumble Room. Hello, gentlemen. It is good to see you. I hope everyone is well in there. We have a really good show for you this week, so I'm glad you're here.

Thank you very much. Now, Wes, before we go any further, we've got to stop and just mention something really quick. Because I know not everyone listening to this show listens to User Air, although they should be. I know they're not. They should be. I understand. And I just want to give them a heads up that we have a bit of a giveaway going on, a sweepstakes, if you will. Don't call it a giveaway, Wes.

It's a sweepstakes. It is a sweepstakes, 100% sweep. That's right. User Air is doing a sweepstakes. It is a sweepstakes. 100% sweep. That's right. User Air is doing a sweepstakes for the Dell XPS 13. Yeah, that's right. You have one of those, buddy. Oh, yeah, I do. Yeah, you do.

You have the original. And we're going to give one away. We're going to give a brand new one away. Give it to me. Give it to me. It's for me, guys. Sorry. You already have one.

Sorry the giveaway is gone. It's for me. It's a sweepstakes. Don't call it a giveaway, Dylan. It is for me. Sweepstakes. Now, here's the thing. It's for me. It's a sweepstakes. Don't call it a giveaway, Dylan. It is for me. Now, here's the thing. It's for U.S. residents. It's a sweepstakes, by the way.

Yeah, and it's for U.S. residents only. You have to be 18 years or older. I'm 19. No, you are not. You go to jupiterbroadcasting.com, click on shows, and go to user air. It's in the user air description right at the top of the user air show. We're going to be giving away the laptop. Oh, I'm sorry. It's not a giveaway. It's a sweepstakes. Yeah. We're going to be giving away the laptop. Oh, I'm sorry.

It's not a giveaway. It's a. Sweepstakes. Yeah, we're going to be doing the sweepstakes in user error to give away. I'm sorry. It's a. Sweepstakes. To stake an XPS 13 to a listener of the user error show.

And we want you to have a chance to get in there. Just go to Jupiter Broadcasting. So we sweep the stakes. Yeah. Yeah, we're going to sweep them. So go to jupiterbroadcasting.com. Click on shows and then go down to User Air. And right there in the description where we have what the show's about and our contact information,

we have a temporary link for the sweepstakes. I want you to get an XPS 13. If you're over in the EU, just come over and live in the U.S. for just a little while. It's worth it. These XPSs are great. Get the Dell XPS 13. We hear that the U.S. needs a roommate. That's true.

Yeah, you could just crash here in the garage for a little bit. Yeah, there you go. Sorry, you have to be 18 years. Sorry about the 18 years or older, Dylan. Sorry about that. Sorry about that. Let's start with that Rust thing I was just talking about because this is neat and it's summertime. Google's has the summer of codes. I don't really

need to explain this, do I? Because everybody knows what this is at this point. Google throws some money around during the summer to employ some students to build things and often they build open source things that we all use and love and sometimes have actually

turned into long-term projects that we use today and Philip is one of these blokes who is going to be working on the I think it's the gnome JavaScript server I think is it's GJS so

I'm kind of guessing at the acronym. And he's worked on it for quite a long time. And he's going to be mentoring someone that's paid by Google to essentially rewrite this JavaScript binding for GNOME and rewrite it using Rust to make it stable, essentially more crash-proof. And it's not just GNOME Shell that uses this.

GNOME Documents, GNOME Maps, Polari, Sushi, GNOME Photos, a lot of things that are modern GTK applications use this back-end, this GJS back-end. JavaScript really is everywhere, even not the hip places. I think right now the current version uses Mozilla's SpiderMonkey for the engine, and so this is going to be a rewrite of that in Rust. And the reason why I mention this, this was actually a highly submitted story to the show this week because a lot on JV, we've been talking about some of the gnome crashing issues we've had recently. And so somebody submitted this as a possible like, hey, look, things are improving. also heard some it's not rumors because they're blog posts but i i have heard some implication that there has been some work at canonical on mutter and i don't i'm not clear on what that

work has been but you combine different people working on this stuff and it could start shoring up some of these problems that we've experienced with gnome pretty quick and this is just the kind of thing that you need you know it won't be the bl end all it's not going to be the next thing maybe that you run on your system even. But this can lower that barrier to entry. Suddenly, if Rust becomes a thing that like, oh, yeah, you already have build support. If you want to add a module that's in Rust and not in C or C++, you can. I think that really opens the door.

So if this works out, hey, that'd be great. And you can find out more in the link in the show notes. Let's talk about the new kernel that came out. So I just want to do a couple of pieces of breaking news right now. Let's talk about the new kernel release and let's talk about the WikiLeaks document dump that shows the CIA was targeting Linux users. Because this is probably the one that's getting the most attention as we record the show right now. It's named Outlaw Country. You may have heard that name floating around. And

it's a Linux kernel module that gets loaded into the system and creates a new entry in IP tables. And then once it does that, it forwards all the traffic to whoever they want it forwarded to. And the attacker, in this case, the CIA or whoever has this code, could then reroute to whatever mining system they have or whatever. Now, what I'm not clear about from this exploit is it seems like you could just do an iptables list and you would see this entry. So not only would it be pretty easy to detect if you were looking for it, but you also would have to have executed this exploit with root privileges. I think some of the issues there are that, one, it adds a new table. And so unless you actually specify that table, by default, you won't know, right?

You'll be like, oh, well, it's not in my NAT rules, and it's not in the default table. Right, so it's not like you're going to look at your config and see it. Plus, it's most likely on, you know, here we're relying on admins in the middle, because it's not like they're going to be infecting you most likely with this kind of thing. It's going to be at some router or some NAT. Or some web server that wants to know who's visiting that website, and it's been running on a VPS for six months, and you're not checking the logs.

What's interesting, though, is this doesn't give itself access to the system, so you have to take advantage of another existing vulnerability in the Linux system to get access to run this. So this doesn't give you the access. You know, it does also say to me maybe that, hey, this is one of those things that people who are pushing UEFI maybe were thinking about, right?

Otherwise, if this module has to be signed, hey, that would be nice. Can I tell you the honest truth though? After the Snowden leaks, I was feeling pretty cocky because they had stuff on there about exploiting Macs through Thunderbolt and they had stuff in there about going after Windows and Windows servers. But Linux desktop users, if you go by the Snowden leaks, were just totally off the radar. Like, they weren't even worried about Linux users. And I'm like, look at me. I'm like James Bond just by running Linux. And then I see this WikiLeaks dump here from the CIA, and they're going after Linux users.

And here, I thought Linux users were upstanding moral citizens. Yeah, yeah. I guess I was wrong. Yeah, of course they're going after the users. And here I thought Linux users were upstanding moral citizens. Yeah, yeah. I guess I was wrong. Yeah, of course they're going after the servers. That's the problem. So yeah, if you hear the story going around and you hear them talking about outlaw country, just keep in mind that it's just,

all it is, it's a kernel module that maybe it does some other nefarious things, but mostly just injects rules into IP tables, which is just, the reason why it's a kernel module is because like Wes just said, it doesn't check your config. And so it has to be able to apply it to your config after it's been loaded. And that reason why it's a kernel module is because, like Wes just said, it doesn't check your config. And so it has to be able to apply it to your config after it's been loaded. And that's why it's a module.

And really, the worst part is that it makes you have to go look at IP tables. So sorry, everyone. You know how you bring down the entire spying industrial complex? You sue them for GPL violations, right? I bet you that's not GPL compliant. You follow up on that. And then just before we get into the good stuff, I mean, the stuff I really, I got a couple of more things I want to talk about with you guys. Kernel 4.12 came out today. That's the other big breaking news. That might have been yesterday, but it just came out. And the big thing in here is BFQ,

the BFQ scheduler and USB type C manager that I talked about in a recent episode has been integrated into the kernel. Also, actually, I should say Vegas support too, which is going to matter to a lot of you. Vegas support has been added as well. And the thing that I'm very excited about, although I've gotten mixed feedback on this and I would like to get more, and maybe if anybody in the mumble room knows, please feel free to interrupt me. BFQ, which is the Better, Faster, Quicker Scheduler, the scheduler that sets the priority for like all the things going on and access to IO. If you've ever had a Linux box, I can replicate this on Linux every Wednesday night. I replicate – I have this scenario. I open up Chrome and then I go to Pinboard and I load up all of the links I have for that night's unfilter and I go boom

boom boom boom boom and I'm using my middle click and I click click click click click along my pin board list and I open tab after tab after tab after tab after tab and some of them have flash most of them actually have flash because I'm linking to new sites and the whole system, the entire system locks up

for about 10 seconds, 15 seconds just long enough you can't do anything else. No. You can't get up. No. And BFQ promises to solve that problem. Now, a lot of people say there's ways that you can fix this without BFQ. But in a nutshell, it's an IO scheduler that balances how the system accesses data on a

drive. And supposedly it has improvements for SSDs. BFQ targets desktop users by ensuring that at no point a single application is using all of the bandwidth. Sometimes when copying large files in Linux, a user sees systems slow down or freeze completely. This will now be a thing of the past. I like the way the Arch Wiki, as always, describes it. Budget-fair queuing is a disk scheduler which allows each process slash thread to be assigned a portion of the disk throughput. So that kind of explains how you're, you know,

again, this may be the case where you wouldn't want it on a server where you might minimize your maximum throughput, but it should give, you know, better latencies for desktop users, or at least that's what people say. And by the way, the USB Type-C stuff is like the stuff that figures out charging based on what has what charge. Oh, my God, the future is here now. Yeah, if you're living the USB lifestyle,

that's a good kernel for you to get. You're not mandatory. Your USB-C devices will still work fine. But if you upgrade to that kernel, you'll get the smart intelligence where it decides, oh, that's a phone. I should charge it, not charge from the phone, which could happen right now

under Linux. Kind of adorable. We had the first USB-C support, so I'll give us credit for that. We were there first. We just do it our own way. Yeah. Okay, and then last kind of newsy thing before we get into the nuts and bolts of the community stuff this week.

Wes found this, and I just found it awesome because when you stopped by User Air a few weeks ago, you were talking about playing around with SystemD's Nspawn and container support. Yeah, that's right. And so here's somebody who self-admittedly is not a Linux expert. His name is Drew, I guess. I'm not sure.

Let me look. I don't know what his name is. And he's been messing around with confining Steam into a SystemD Nspawn container. And I don't – I've never set one of these up yet, but if I'm understanding correctly, well, he did it to play around with too, but it was a great way to sort of solve the 32-bit library problem because instead of spewing 32-bit libraries

all over your system, you contain it all to this container and you don't have to like, if none of your other applications need it, you don't have to bother with it. Yeah, exactly. It's this crossover now. Finally, like containers are big on that in the development side of things because exactly that.

You don't want all these – I have six different projects. I don't want the dependencies or the libraries from all of these different things on my system. I don't want to make a mess on this. I don't want to cause dependency hell or have some weird app thing go on. Containers are perfect for that. And now it's like, yeah, I don't want 32-bit libraries. I don't want to have to deal with that. containers are perfect for that.

And now it's like, yeah, I don't want 32-bit libraries. I don't want to have to deal with that. And, yeah, he points out some things like FireJail or other alternatives which can work for, you know, some of the, you know, getting more control over the process. But he's using systemd's NSPON, it seems like, mostly just because it's built in. Yeah, it's built in. It's right there.

It's really easy to go. And it makes it super simple to be, you know, you can pretend maybe you don't have, you know, you don't want to use a cloud provider or something like you need something something local like for Steam. Boom. Perfect. Throw away Linux distribution.

Yeah, exactly. And the thing that would have tripped me up, and this is why I like his post and we'll have it linked in the show notes, is he talks about getting the X applications out of that container. And actually, at first, this was a bit major hurdle. But these days, between Docker and systemdnspawn, there's like a thousand guides out there for how to do that. It's really pretty easy. X needs some stuff that's under slash temp. You can also do other things like you can share special devices, you can mount stuff systemd and spawn makes it super simple to just bind mount stuff. Networking is really easy. If you use systemd network D, they have a lot of primitive setup to just boom, that's already there, your containers automatically get you know, set up behind the NAT.

They've got their own private IP address. Super simple. And Systemd and Spawn can run Docker containers. So you can use all the Docker build stuff. You can use – yeah, exactly. Oh, I did not know that. But it makes sense. Because it's just a file system that you start a process on.

I'm going to chew on that. So I don't know. In the background, you might be able to hear some of the barbecue going on the barbecue continues while we do the show which is awesome and not only do we have a crowd out there but we have a big crowd in the mumble room which is pretty exciting and Mr. Popey's

back after a little bit of a stay Mr. Popey did you happen to go to the recent snap sprint held by Chronicle I did I was there Monday Tuesday Wednesday and then I had a conference to go to on Thursday recent Snap Sprint held by Canonical? I did. I was there Monday, Tuesday, Wednesday, and then I had a conference to go to on Thursday, Friday, to go and give a talk about Snaps.

So, yeah, it was great. It was nice to see a bunch of people from different projects, people from KDE, from Elementary, from GNOME. It was really nice to have everyone in the Canonical office and all working together to make Snaps better. Yeah. And I've seen some recent news about GNOME software. Is that as a result of some of the work that came out of that sprint?

Yeah, definitely. Robert Ansell, who maintains GNOME software from the Ubuntu side, was there. But also Richard Hughes, who's the upstream maintainer for GNOME software, was there as well. And so there were lots of conversations they could have about improving the experience for installing snaps on any distro, not just on Ubuntu, and improving things like how options are presented to users. And there was lots of discussion about how to improve that kind of stuff. So it was really, really nice to see all these working together in the office. Yeah, that is really cool.

All the different walks of stuff. So it was really, really nice to see all these working together in the office. Yeah, that is really cool. All the different walks of Linux. You know, I was listening to the Ubuntu podcast today, and I noticed that you sort of built a better, improved Mycroft. And I was wondering, are you going to launch an Indiegogo for that anytime soon? Because I might be willing to back that. You mean launch an Indiegogo and then not deliver for two years? Yeah, you got a long time yeah that's what i'm used to at least you could you could have a stretch goal for palmer violets yeah included okay hello mr wimpy how are you i'm very well yourself i'm good we've had a great we've i mean we've had a great unbelievable food here at the studio it turned out it was so so so good were you at the sprint yourself i was yes oh well welcome welcome i'm jealous it's nice both you guys here now if rumor has it are you

going to be afk soon are you going to be uh traveling are you gonna be on staycation like what's happening soon because i think uh if i'm not mistaken i didn't expect you to be here today yeah so we our schedule's a little bit off by one at the moment on the Ubuntu podcast. We just recorded one episode of Ubuntu podcast this evening. So we were let out of school early. Very good one. I'm glad you made it here. Now, we have some good stuff to get into today, including – I'm really excited. I know Foronix gets a lot of crap.

Some of it from you. Okay, fair enough. It does. I mean, what? No, you're right. I mean, I will gets a lot of crap. Some of it from you. Okay. Fair enough. It does. I mean, what? No, you're right. I mean, I will own that.

I have. But I do at the same time – I do respect Michael Larble's hustle to put it in sort of – Whatever you think of the output, there is a lot of output. There's a lot of output. There's a lot of hustle. But actually where a lot of my respect comes from for Michael is what he's done with OpenBenchmark. It's just the Pharonix test suite, if I were him, I would rename it. And I know that's really a hard thing to hear. But if I were him, I would honestly rename it to the OpenBenchmark test suite because it uploads to OpenBenchmark.org. And I have found it to be such a useful tool, not because I'm all into benchmarks and numbers, but because it gives me baselines to see how machines compare to each

other that I'm testing up against each other. Some heuristics you can use later when you're trying to evaluate them really quickly. Yeah, absolutely. And then as a reviewer, I can publish links to that and say, now run it against your own system and just see the delta. It's not about what, it's about the gap in the numbers, not what the numbers are. There's no absolute measure here. It's not about what – it's about the gap in the numbers, not what the numbers are. There's no absolute measure here. It doesn't matter. And that is such an amazing resource and that database is growing and growing and growing.

So Pharonix has been, I think, actually in total a pretty good resource. And when anybody is working that hard and publishing that often, you're going to get a couple of things wrong here or there, especially if you don't have an editor, if it's a one-person team. So I can understand some of the mistakes that people have criticized in the past. They're somewhat unfortunate and I think it's been a good resource but I think what we could really use in this space is some competition. And this week, XDA, the ones that have traditionally exceedingly well covered ROMs and that whole development community, announced that they were going to cover PC hardware with a specific focus on Linux and open source. And they've got some really good things that they're thinking about. And They capture and cover some of the feedback from their

readers. I've been talking to the author. I don't think he's going to make it. He was going to meet us up today in the show, but you know, mumble. Hashtag mumble problems. Wow, did you hear that? That's America. Did you guys hear that in the mumble room? Did you hear that in the background? Yep.

That's fireworks that are going off already right there. We're having a small war with the neighbors. I think we're going to win though. The charge is happening as we speak. We do have a larger group here. We do have a larger group. So I want to talk about what they're all doing and why I think it's a big deal and how they're doing it differently. And honestly, the brilliant point they made about why they're starting it. So we're going to talk about that next. But first, I want to talk about Ting. Go to linux.ting.com. That supports the show, but it also gives you $25 in service credit if you bring a device or $25 off your first device. Mobile plans used to drive me crazy. It's obviously a scam.

You can sense it. You can suss that out. You're smart enough. You know that they're scamming you. They're really only capable of doing this because they have a duopoly. It's AT&T and Verizon, and they have a duopoly. And so they're able to do all these kinds of crazy things, and then the competitors have to wear crazy leather jackets and pink shirts

and make all kinds of crazy claims and completely destroy net neutrality to sell phones. It's a horrible competitive environment. If you go to linux.ting.com, you can make a difference. Linux.ting.com is mobile the way it would have been done. If you hit the reset button on wireless, this is how you would do mobile, right? Oh, yeah. It's pay for what you use. It's $6 a month for your line.

Think about that for a second. So you want a cell phone line? It's $6 a month. Okay, now your kid gets a little bit older. You want to get them a cell phone? $6 a month. You got a small business. You're growing.

You want to give a couple employees some phones? $6 a month. $6 a month for a you're growing, you want to give a couple employees some phones, $6 a month. $6 a month for a phone and then your usage on top of that. Ting just takes your minutes, your messages, your megabytes, you add them up and that's what you pay. It's simple. It's straightforward. There's no contract. There's no early termination fee. And the best part, CDMA and GSM networks to work with. So whatever you got, you're likely going to bring over to Ting. Just check their BYOD page. It'll give you all the details you need. You bring it over to Ting, you get a $25 service credit. The reason why that's great, average Ting line, $23 a month.

Once you include like the line cost and the minutes, $23. So if you get a $25 service credit, it pays for more than your first month. Or maybe you don't have a phone that you're going to bring. Like you need a new phone. Then you get $25 off your first device, Wes. I know it. Linux.Tting.com. A big thank you to Ting for not only making wireless not suck so bad, but also for supporting the show for so damn long.

But really, the first thing is even more important, honestly, because wireless is the worst. Also, happy America birthday up to our friends in Canada. Linux.ting.com. You hear that party out there? Oh, yeah. They're having a party out there. We left, so they're now – Well, and there's booze and there's food.

I mean there is a party. Let's hurry up and finish. XDA has announced they're beginning regular PC hardware coverage designed for, quote-unquote, their audience, which I think means the Jupyter Broadcasting audience. And they go through some recent comments they've had about some of their articles that they were kind of preemptively planning this, and they're like, let's test out this material and see how it works here. And so they published a few articles, and the comments are like, do more of this. Yes, please, get more into computer hardware. And they say, at the time, we didn't mention that we were kind of considering this,

and we were monitoring the feedback. And I love what they say they're going to cover, because they're so brilliant about it, because they say XDA has traditionally covered mobile devices. But how do you make all these apps and OSs for these mobile devices? You have to use great PC hardware. And there's that link, right, between – I mean I think there's a lot of people who get into Linux because they want to mess with Android ROMs or whatever and it turns out,

hey, Linux and open source is the best way to get started with that. Yeah, yeah, yeah. Well, and so they say like – so here's the logic. They say after two months of working behind the scenes, we're ready to say it. XDA will start covering PC hardware on a regular basis. And if you dig down a little bit, they talk about the type of hardware they're going to be covering. They say reviews will largely focus on the Linux operating system and continue to have a developer focus in them. Whoa.

We decided to keep our reviews on Linux for two primary reasons. Oh, I love it. Okay, number one. First up was the lack of dedicated Linux coverage for new PC hardware in the global market. We hope that this will help introduce, what's that word right there, Wes? Cooperation? I don't know that. I don't know what that is.

What's that? I don't think that's something we do down here in the States. No. No, I'm not familiar with that. Is that something where you fight other people and then whoever wins? It's just another word for argument. Okay. Okay.

Okay. Anyways, they're going to take that and they're going to work with other people and do Linux reviews. I don't understand that part of it. But what they go on to say is there's not really many competitors, and they address this directly, beyond Foronix. That caught my attention there. The other half, by the way, is one that we have had... Here's what they say.

A lack of information on how to get most PC hardware working on the day of release. As part of our reviews, we'll be including notes on any issues using hardware with Linux, as well as any solutions and workarounds available. You know, I hadn't really thought of it until this article, but you're right, there kind of is that gap. Like we talk a lot about software releases, those kind of things,

but you're on in by and large, I mean, yeah, you might have some Pharonix articles, maybe there's an ArchWiki entry for your laptop, but hardware is kind of neglected. It's like, well, just go to the regular PC hardware news, you'll be fine for that. Yeah. Yeah. So I've been talking with Daniel Moran. And he'll I think that's how you say his last name. He. You'll be fine for that. Yeah. Yeah, so I've been talking with Daniel Moran, and I think that's how you say his last name. He'll be doing the writing over there, and he's available on Twitter. I have him linked in the show notes,

and he's pretty excited about it. I think he's got a good vision about what he wants to do with it, and it's not going to be exactly what Foronix does. It'll be somewhat competition in that space, but I think it's going to be also more like, here's first day release of this new piece of hardware that everybody's talking about, and here's

how it actually works today on Linux, and if you really got to use it, here's the couple of things you have to modify or tweak to make it work. Awesome. Yeah. Yeah. Yeah, that's the kind of stuff that Windows users typically get to enjoy. And they talk about that, too. They say, well, look, there's tons of other people that are doing Windows

coverage. Right. And, like, before, we've been kind of left waiting,. They say, well, look, there's tons of other people that are doing Windows coverage. And like before, we've been kind of left waiting like, OK, well, I hope someone buys that laptop and writes a blog post sometime in the next six months because otherwise, how will I know? Yeah. I hope that somebody who's on like Planet Gnome or something that blogs writes about buying this thing and they – yeah, they fix it. Yeah, that's exactly it. So I'm actually super glad that Wimpy is here because I was going to – I hadn't even planned on having the expertise of Mr. Martin Wimpers for this next story. But Joey over at OMG Ubuntu has an article on the new software boutique app that's going to ship in Ubuntu Mate 17.10. A couple of things I noticed that jump out at me, Wimpy, is it looks like some of the Mate branding has been stripped from this. More theming support has been added and snap support.

It kind of feels like this is starting to be positioned as any distro could use this thing. Hopefully, yeah. We're using all the right technologies for that to become a thing. We're obviously going to be developing it to hit the 1710 release deadlines, and it will be squarely aimed at doing what Ubuntu Marte needs. But because we're going to be using SnapDGLib, which is the library to do all of the Snap stuff, and then PackageKit now for the traditional repository installs, it could potentially work anywhere.

So decoupling Software Boutique has several advantages. It can be installed and used in other distros. Software Boutique and Ubuntu Mate welcome and use fewer system resources, and Software Boutique will dynamically adapt to the currently selected theme to better fit in with your preferred look and feel. So I didn't realize this before, Wimpypy because i've always used it after immediate installation but did the software boutique not follow my desktop theme in the past well it did in the usual places but then within the application itself um it has its own css to style it and it didn't take any cues from

or it didn't take many cues from the theme that you actually had running to uh you know blend in and it will do that now oh i see um so can you give me a little peek behind the curtain like are you doing the work on this are you paying somebody to do the work on this how are they getting paid and that kind of stuff so uh most of this work is bit well all the front end work is being done by luke horwell who is a contributor to ubuntu mate yeah we've talked about him before definitely yeah he's responsible for all the cool stuff you see in ubuntu mate welcome and the software boutique and if you use um razorals, you are probably using his other project, which is called Polychromatic, which is the tool to do all of the RGB configuration. Cool.

Okay. Well, gentlemen, because this is one of my favorite, favorite experiences on Ubuntu Mate. I honestly would – for internal uses, it has tipped the scale in the past to know that I could – you log in after your first login. You run this thing and now with the queued installations, you just go through and all of the great applications for Linux are in here. It's one of those things that makes me actually think, OK, I could install this on someone. I was setting them up on Linux for the first time and I wouldn't have to be there and walk them through it. I could just give them this and be like, call me when you have problems. Yeah, and even on systems where I've run the Unity desktop,

I've still gone out of my way to install it. Yeah, right? Why not? And we hear that a lot from people. We hear a lot that people install Welcome on other Ubuntu flavors just to get the boutique. So there are other advantages for us. It's easier to maintain the boutique when it's decoupled from welcome. So I don't know how much you can say, but if a great software –

Well, it's open source. I can tell you whatever you want to know. So if a great software project like this comes along that happens to hit things like snap packages, happens to have an apt get back in, happens to use PPAs. What could I as a user do to let Canonical know that I would be interested in actually having that ship in mainline Ubuntu GNOME, like the standard Ubuntu desktop, starting with 17.10 and beyond? Like, what could I do as a user to voice that opinion?

That's a very good question, and I don't know the answer to that. It seems like, I mean, I know maybe 1710, you know, that's too soon. But in the LTS, what a piece of software. You know, you give folks an entirely new desktop environment, but you give them this guide to get their whereabouts, to get their bearings and get to the applications that they use and know. It seems like it's that perfect kind of, I know this is new, but this will help. And even just ironing out, I mean, GNOME software is great

and actually that is getting pretty good. But just ironing out that extra little bit of friction would be really nice in a totally new desktop environment. I'd like to see we set the same person down in front of clean Windows laptop, clean Linux laptop. You got this, you got the Windows store.

Have at it. Where are you going to be in a day? We should bring Will on. I'm going to try to get a hold of Will Cook. I'm going to ask him because I really want this. It's so great that I install it even on mainline Ubuntu desktop. Oh, yeah. Why not? Yeah. So we did a lot of this work at the Snap Sprint last week.

And Will Cook was there, as were, as Popey said earlier, people from KDE and Fedora and Elementary. And there were a couple of breakout sessions in the week where we have coffee, and we all sat around and showed each other our wares. The KDE guys have added, well, they had Snap support in their store, but they'd written their own interface for Snaps, and they've now used SnapDGLib and its cute wrappers. And they were showing us all of the new stuff that they'd added as a result of making that transition.

So we were all doing a little bit of show and tell, and everyone was looking at what the new stuff in GNOME software is, what's new in KDE, and they were all looking over our shoulder at what is a very embryonic software boutique to see what we were up to. It was kind of fun comparing notes and seeing how we were going about stuff i love it it is i'm i didn't realize is this has it has canonical always ran this many sprints or is this something that's picked up in the last year well you might remember about this time last year i was a snappy sprint in germany yeah okay yeah i do remember that and that was quite a big and and this was like the continuation of that effort, but it was a much smaller focus group.

So I think there was about 30 people in total at this one. And this one was more about doing rather than making decisions and drawing up plans. So these were not that we've already decided we're going to do these things. Now, how do we go about doing it? And for some of us, we spent the week, you know, kicking out a lot of code the great unblocking festival i mean yeah right doesn't it really doesn't it just sort of kick everything through and like you make serious progress it's fascinating yeah yeah i mean um for luke and i this was something that we've wanted to work on for ages and and it gave us a week to actually and it's the first time we've met in person

so it actually gave us a week to like work on this stuff and actually make the progress we needed to do to sort of kick this in a new direction. Because Luke has decoupled Boutique months ago. We just haven't had the time to work on it. Oh, I didn't realize that. I did not. I thought that was a recent thing. Well, we announced it recently, but Luke started this work maybe six months ago sure wow wow so this is something you've been thinking about for a while yeah smart i and i hope it i hope i hope

it takes off because it's such a great user experience and i i really i don't know how much emphasis and underscore and exclamation marks i can put on this but as somebody who is feeling back through his years of trying to use Linux in the corporate space and trying to use Linux as a creator and trying to use Linux as a small business owner and trying to use Linux as a father with photos, like all of these things,

I come with the weight of saying this makes a difference. And it seems it's like one of those things is so obvious, but I want to, so I want to go to the rooftops and I want to shout it around and say, make everything better with this. Well, and it's like we – it seems like the time is ripe in many ways with the Windows Store and everything that Apple has been doing. Like consumers are now expecting this sort of model. But the traditional workflows we have for – we have the package management and those systems and all of that but but something that says this is really the best of the best that you can hit one button and have it installed at links and icons are here and it's just it's just great and and it does all the ppa math for you it does all of the snap math for you it just it just handles all of that and

gives it to you in a cute interface where you can go check check check check check check boom go make my system usable and one new feature is we're going to retain that you know curated list of the hundred or so applications and we've had a search facility in the boutique for a while but it's only ever searched its contents now when you run a search it will query the entire archive and the snap store so what you see in the boutique is just the stuff we're recommending. But when you search for stuff, it will interrogate everything it's got access to. Smart. So any Snap package that's been submitted since that thing's been released

will be available in the search. Yep. Wow. These kind of things are like the critical momentum pieces for Snaps and the whole ecosystem, it feels like that suddenly it's cohesive. Now it's, it's thought out. It's well-rounded.

So let's talk about kind of related to this is package kit. Now, if I say package kit, you probably think that thing I think Fedora uses. Well, it turns out a lot more than just Fedora uses it. Pretty much, pretty much all Linux.

I use it. It is. Think of it as an API that sits on top of your package manager. So at some point in time, people came together and said, I want to write something like GNOME software or Neon Discover or whatever, and I don't want to, or KDE Discover, and I don't want to have to write a version for DNF, well, I'm sorry, YUM, then DNF for Pac-Man, for Apt.

You want one API that you can write to. And so that was the idea of PackageKit. And you can actually use it as a command line tool as well, which is kind of nice because you can bounce around distros and use the same PackageKit commands to install software. Hey-oh. Yeah. So there is some real like worthwhile investment here in PackageKit. However, right now, and these things always change, but right now, package kit

on Fedora is not in a great space. And I think this isn't a big deal. DNF's great, and Fedora users aren't married to GNOME software. But it relies right now on a library, libdnf, pretty sexy name, really, that is pretty new and they've just recently ported it from Python to C so like when I say it's new it's like not only is it kind of new but they've also just ported it to C kind of new so it's all kinds of new

and they actually say in the readme they say libdnf is not currently I'm sorry is currently being reworked and is considered unstable once major users like packagekit and and DNF are fully ported, a new stable release will be considered. In other words, GNOME Software and Plasma Discover, two front-end center applications for end users of Fedora workstations,

especially GNOME Software, rely on unfinished work in progress in Fedora. It's worth noting that both Pac-Man and apt backends, like the one for Ubuntu, are in a much better shape. I think that highlights the whole thing here, right? Like the value proposition is that you're pushing this off to some middle layer. Yeah. And what you can do is only as good as the binding

between you and what the underlying system is. You know what? DNF is newer than apt or Pac-Man. And so that middleman layer hasn't been finished yet. And that's really the problem. And you wonder like, OK, well, how bad is it? Because that was my consideration. Like how bad is it?

Well, actually, it's pretty shit actually to tell you the truth. I was pretty surprised. Like I wasn't going to cover it on the show. But then I realized how shit it was. I was about to say. Is there a story here? Yeah, there is. So LibDNF doesn't support Delta RPM.

So if you're on a mobile connection, go screw yourself. Also, LibDNF just totally ignores EtsyDNF conf. EtsyDNF slash DNS. That's where I put the conf I don't want to use, Chris. Right. So like where you maybe turn on automatic fast mirror selection, package exclusions, controlling network utilization. Nope, sorry.

LibDNF ignores all of that. Also, it doesn't uninstall things properly and it can leave orphaned packages behind. When you uninstall software, it won't necessarily clean things up. That's less than desirable. Yeah, it also lacks safeguards to prevent users from uninstalling crucial software such as GLIB-C, DNF itself. Also, LibDNF itself can be uninstalled via package kit and other things that can cause complete devastation on your system. Now, these are not total likely scenarios, but it's a bit of a shit show right now.

It's kind of amateur hour and, you know, full disclosure, they say that in the readme for libdnf. They say it's not ready yet. So it's not like this is a surprise here. But if you're using GNOME software or other graphical managers, you may— You might just not have any expectation, right? If it works so well, it works just fine on apt or— Sure.

Well, here's what I—yeah, especially if you're coming from Ubuntu and you're trying out Fedora. You don't even consider the fact that I might be talking a totally different language. Here's Chris's recommendation and this is really going to be fine. If you want to discover software via GNOME software, if you want to discover something new and install it, you're going to be fine.

Maybe just do your updates from the command lane or using the Fedora update. That's not available anymore. Just use your... Just do DNF update and upgrade. Just do that because DNF will do... On the command line, it will do delta packages. It will also – and when you uninstall software, just use DNF from the command line. The command line syntax for DNF is exceptionally simple. So I normally wouldn't say like go to the terminal, but DNF is like simple human readable commands.

Here's the problem with DNF though. When you tell someone to use it, they're going to ask you what it stands for what does i don't remember anymore what does it stand for do you remember uh does anybody in the mumble room remember what dnf stands for i thought it was something kind of womp womp kind of meh like does not stand for much dandified yum oh is that oh okay that's yeah of course yum right right of course of course were Yum. Right. Right. Of course. Of course. Were you going to say that, WW? It's usually did not finish in other. Yeah, that too. No, dude, that's lib DNF.

That's which is bad. Yeah. Yeah. You know, it's just something to know about. I don't think anybody listening to this show is using package kit in the command line. If you're just adding new applications from the GUI, either Plasma Discoverer. It really seems like an API type, you know. Yeah, and probably by Fedora 28

this isn't going to be an issue anymore. This blog post over at control.blog, wow, what a great freaking domain name. Control.blog, which I have linked in the show notes, it talks about how 26 and 27 are going to have a problem, but by 28 this should not really be a problem anymore.

But it is interesting. Like you said, it is interesting because we kind of think – we talk a lot about – there's a common conception, and this is going to actually come into what we're about to talk about, that Linux is all kind of the same. You can use different distros. You can use Pop! OS. You can use Ubuntu GNOME. But it's all kind of the same thing. You can use Pop! OS, you can use Ubuntu GNOME, but it's all kind of the same thing. You can use Ubuntu Mate, you can use Ubuntu XFCE, you could use Ubuntu Studio.

It's all kind of the same. And then when you switch over to something else, you know, you get the GNOME desktop on Fedora, you're like, oh, this is kind of the same. And there actually can be huge, really, really, like, screw-your-distro-up differences in things like package kit and other things. And I actually think that's an interesting topic to consider

as we're about to go into our next thing. Yeah, you're absolutely right about that. So this week, I don't know. I almost feel like... You can take a sec if you need. Do you think so? Yeah, that's fine. Do you think so?

Because I almost feel like I can start to see – my heart rate just jumped up. Like I'm starting to get a little upset. Could you see that in me? Just like scoot back. Because there's no video. But like Wes Payne, like theater of the mind. Do you see me starting to get a little upset right now? Yeah, I'm getting a little upset.

There's the sign. All right. Okay. Let's take a second here. Let's take a deep breath. Let's talk about digital lotion. That's a good idea. You're right, Wes. Thank you. You're right. Because this is a deep breath. Let's talk about DigitalOcean. That's a good idea. You're right, Wes.

Thank you. You're right. Because this is going to help me. This is just something that makes us all happy. Focus on something that makes us happy. DigitalOcean is so great. DigitalOcean.com. Go over there, create an account, and then apply our promo code D-O-unplugged.

One word after you create the account. That's all you got to do. And then you get a $10 credit. You can spin up a rig. They have a – my favorite. It's not even the most expensive or the cheapest. but they have a three cents an hour rig. It's kind of like right there in the middle. And the things you can do for three cents an hour. It's crazy.

DigitalOcean is the simple, easiest way to spin up some great Linux machines on their amazing infrastructure. They have data centers all over the world, New York, San Francisco, Singapore, Amsterdam, London, Toronto, Germany, India. And I do prefer San Francisco, but if I'm going to spin up another droplet, I like to get close to Al and Jude, as close as possible. Oh, yeah. Yeah, so I go Toronto. That Toronto data center. Speak and span.

You know what else? Turns out, no NSA in Toronto. No. Turns out, DigitalOcean.com, use our promo code DEOUnplUGGED. Create a machine on their great infrastructure in seconds. It's simple, easy to spin up infrastructure. FreeBSD is also available. Tons of great Linux distributions and a dashboard that makes it easy to use. It's like butter on top of corn on the cob at a barbecue. DigitalOcean.com.

Use our promo code DEONPLUGGED. And they have a new feature too. I'm sorry. They have block storage. But object storage is a new feature. Oh my gosh. I'm really excited. And I'm just waiting. I signed up for the have block storage, but object storage is a new feature. I'm really excited. And I'm just waiting.

I signed up for the beta. I haven't heard back yet. Yeah. Literally every day I check my email. When you use, see, when you switch over to using something like object storage, it means that you're, even a small application, you're building it in a way that you're going to be able to use that code base for 10 years. Like just, and the storage aspect of, too, is the biggest bet when

you're writing something that could have images or media or anything with rich content. Object storage is such a game changer. And it feels like one of the last things to drop here in, like, they made a ton of improvements this year, last year. Lots. Yeah. And this is one of the last ones that seems to me like they're suddenly

not only is it simple, intuitive, super easy to use, but they now have basically all the features you could get from any of the major cloud providers whatsoever. I mean, they've got the same thing. With an interface and an API that none of them can even touch. And it makes such a difference. Clear, upfront prices.

And it's fast and easy to use. It means that in a moment before a show, before we go live, if we need a little additional infrastructure... No problem. How live, if we need a little additional infrastructure. No problem. How many times have we done that? A lot. A lot.

And in fact, we've even done it on the live stream as like a bit. Like how long? How many seconds can we get up new infrastructure? We spun up like a game server one time. Yeah. Whatever you need. It's so awesome. So DigitalOcean.com, go there, create something, and use our promo code DL1PLUG to get the credit.

Thank you to DigitalOcean for sponsoring code DL1PLUG to get the credit. Thank you to DigitalOcean for sponsoring. DL1PLUG. I love it. I love that they're in the background because there's no other show where we've ever had that. We've talked about going to a bar. Yes, a bar would be awesome. This is fun. You know, we have a ton of viewers in Utah.

I'll go to Utah. I know. No problem. I'm like, we should go to Utah if we're going to do it. So, okay. So there's a lot to cover here. Let's start with Tuxedo. Tuxedo is a manufacturer of Linux notebooks in – I believe they're based in Germany.

I've never had an opportunity to be a customer. I've never had an opportunity to interview because there's a bit of a language barrier and they have a different market. And today they have announced Tuxedo XFCE, which is their own version of Linux, a distro respin, if you will. And they've worked on something like this in the past, but now they're getting really, really serious about it. and they say it makes a big difference. Tuxedo Computers has announced the release of their own highly customized version of Linux. Tuxedo's Ubuntu will be its own distribution based on Linux kernel 4.1.1, even though it is based on Ubuntu 16.04, which does not ship with that kernel. The kernel is expected to significantly increase battery life compared to Linux Mint while the Infinity Book

13 will go from six and a half hours of battery life with their new magic sauce special sauce distribution to twelve hours twelve and a half hours of battery life. Say that again for me. Twelve and a half hours of battery life just by

installing their distribution. You install their distribution of Linux, not that piece of shit Ubuntu, but you install their special curated spin of XFCE's Ubuntu and you'll go from measly, pathetic 6 hours to 12.5

hours of battery life. Now that's something to brag about. Also, by the way, XFCE, they say, uses less memory than Unity. And the most important advantage for Linux beginners, all the necessary drivers and the latest firmware are already installed, and adjustments have been implemented.

Tuxedo's Ubuntu, now available. When you buy a Tuxedo notebook. Well, I'm looking at them right now. Are you? How do they look to you, Wes? Hey, I mean, not too bad. Here's one that looks kind of... I've never tried the Infinity Book, but Wimpy, you had a chance to review it on the Ubuntu podcast, didn't you? Yeah, we did.

We reviewed the Infinity Book Pro 13 a few weeks ago, and Marius, who's in the mumble room from ubuntu fun uh podcast has also recently reviewed the same unit great this is awesome awesome so what were your thoughts if you don't mind just sort of the elevator pitch and i i remember that episode but i wouldn't remind just a quick recap well i'm gonna skip anything to do with the hardware because obviously this announcement is about them making their own distro. And I think this is probably where Marius and I have most opinion. you receive an infinity book or or any one of their computers pre-installed with ubuntu or one of the ubuntu flavors it comes already with it with the customizations that they allude to in this um uh announcement okay that was my impression like they've been kind of doing this yeah and the

problem is is that well one i i can't stand by their claims for the battery improvements because i didn't see anywhere near uh that kind of battery endurance and my understanding is talking to them the modifications that they make specific specifically the kernel changes are to try and improve battery endurance but that comes at a cost in that it's effectively a mainline kernel so docker doesn't run properly um lex d doesn't run properly there's no live kernel patch for you and snaps don't work and a whole lot of other stuff so it's kind of you know a not a very useful Ubuntu experience at that point that's something I really hadn't thought of before but that totally makes sense is that

hey suddenly you've disconnected this kernel normally I think of having a kernel with more features that I'm adding onto from the base you know config profile for a kernel but here you have something where you have this user land that expects things in the kernel that you just won't have all of a sudden right right yeah like z Yeah, like ZFS modules, for example. User namespaces, probably. Oh, jeez. So, Marius, does my summary sound familiar? Pretty much, yeah.

I've been using the Infinity Book Pro for a review for like two months. And even back then when I was on the show to talk about Magic Device, I was actually using the Infinity Book Pro. And halfway during the show, the screen went off, and I couldn't get it to turn back on. I basically did the whole show blind, and I didn't know what to do. And it's basically one of the aforementioned kernel problems or problems because they were using the mainline kernel.

Okay. So this is not encouraging then. Kernel 4.1.1, that must be a typo, right? Is that a typo? It's 4.11 is what it's referring to. Okay, thank you. So are we just talking about some mindshare here that doesn't really matter?

Is this the market that we're going to? Would a developer be comfortable enough adding in a different kernel or making some of these changes? I mean, OK. So what if we take a second here and we use the announcement of Pop! OS as a lens to look in this? So Tuxedo has been doing this now for a little while, as Wimpy just implied there. They've been doing – this isn't their first time. But now they're like serious about it and they're calling it a name and all that.

first time. But now they're serious about it and they're calling it a name and all that. But does it change your opinion, Wes? Specifically, because I've talked about it on User Air, I've talked about it on Linux Action News, but I have not heard your thoughts about Pop! OS. Does it change your opinion of their announcement if you

think of Pop! OS or are we talking about sort of one and the same here? You know, there are a lot of similarities there. talking about sort of one and the same here. You know, there are a lot of similarities there. It just makes me think about what's the maintenance burden going to be and how much is this really the special sauce?

And that's what I've been trying to think about for Pop! OS as well. Is this the kind of investments that I want to see? I think System76 has a kind of clear idea as to what they're doing. I'm not sure yet. I haven't seen, I haven't really played with it or anything like that.

And so I don't know that much here about Tuxedo comparably. I can see why a consumer vendor would want that level of control. But from the Linux advocate side, the Linux power user side, there's a lot of compromises and there's a lot of, this isn't what I want you to focus on for me. I just want the hardware and that part of the experience to be the priority.

I'll tell you what I see similar here. And I don't think they are the same thing. But what I do see similar here is we want to provide something for our customers that is competitive. And I believe that is what this is rooted in. Battery life being an example of that. I kind of want to run through the Mumble Room really quick and see what everybody thinks.

I have definitely had my opportunity to share my thoughts. Bitten, would you mind sharing, what do you think about things like Tuxedo, Zubuntu, or Pop! OS? Are they one and the same? And do you have any strong feelings about their announcements? I haven't really tested them at all. No, not tested, but what is your impression just aside from that? Just new distros coming in.

So here's an idea that I've heard floated around, is maybe every OEM should be making their own distribution. Perhaps every OEM should be focused on the end user experience. And Bitten, they think, here's the theory, Bitten. If every OEM wants to satisfy their customer, they should focus on the experience and ship their own distribution that enables their hardware. What do you think of that? Well, it sounds good if it doesn't break any stuff.

I haven't read about the German distro thing, but I heard some stuff about the Popo and for now it's only been to Gnome and other teams and stuff. Yeah, I agree. For what I've heard.

Yeah, yeah. What about you, Marius? What do you think? Well, I think it's a great thing if more companies like Tuxedo or System76 focus on getting the hardware out of the box, working with the drivers and all this stuff. But if the experience is like it was for Martin and me with the Tuxedo Infinity Book Pro, when the experience is actually worse than if you put vanilla Ubuntu on it. Sure, sure. Then I don't really see the point. Right, right. So what you – I mean I think to boil down what you said is you just – if that's what they have to do to make sure that out of the box when the customer opens the machine,

the Wi-Fi works, the sleep works, the resume works, the video works, that's probably understandable. Yeah, definitely. Well, I can see that they want to ship it out of the box with with their with their logo and on the on the wallpaper and that's totally fine but that's not worth an whole os yeah i okay okay so it's not worth like a whole os okay yeah okay uh so jm did you want to chime in did you have a thought on this yeah um i had an interview with Ryan Sipes from System76, and he talked a lot about making those last mile improvements, which they had a hard time doing while waiting for Canonical to ship new versions of Unity. I believe his words were, we waited three years for features for our operating system, right? Yeah, and it makes sense for them to do the hardware and the software.

So if I said to you that people were not buying System76 hardware to get their OS, but they were buying System76 hardware to get the Ubuntu OS, then I think Ryan's comment in that context doesn't reflect the motivations of their customers. And far be it for me to say what their customers prefer. But if I was buying a System76 machine, I mean the point I always made sure to emphasize for 10 years in Linux Action Show was System76 machines were born to run Linux. Or I'm sorry, born to run Ubuntu. They run Linux. But they're born to run Ubuntu.

I.e., this company has made it their mission to make sure this thing runs Ubuntu. And that is why I thought they were a great vendor to sponsor because if you want an out-of-the-box great Ubuntu experience, then wouldn't it be wonderful if there's this one company and their sole purpose is making sure that these systems run Ubuntu really well? They're born to run Ubuntu. I'm not waiting for System76's features. I'm not buying a Bonobo. I'm not buying two Bonobos and a Retel. I don't buy Oryx because I want System76's features to Ubuntu. I buy them because I want them to run Ubuntu and mostly, actually, if I'm being honest, I want them to run the Linux kernel very well.

That's what I want from them. And that's why I'm buying them. And I really couldn't – to be honest with you, could not give two shits what features they want to add to the desktop. If they want to put a great theme on it, God bless. And you know what? Papa bless, have at it. I really – I think that's a great idea. But myself, I'm going to format that shit and I'm going to put Solus or Anagross on it.

And then I'm going to make it work for me as a workstation that gets actual work done. And I'm not going to care about any pop theme. I'm not going to care about anything that's a special distribution. I just want to make sure that this thing runs Linux really well. And if I know it runs mainline, good old, fashioned Ubuntu, then I know I can get my Arch Kakamani or Fedora shenanigans working on this thing. What does

it mean to go from changes, maybe even a large amount of changes on top of an OS or distribution, whatever verbiage you want, to do that to make it your own? It seems like we're okay we want

we recognize that like but when is it a distro versus a fork versus an os right like clearly there needs to be some customization there can be driver customization or whatever but what what does it mean to change that branding what does it mean what is that line i want to i want to i i had i had a really interesting realization when i visited Dell. And because I hadn't really ever gotten inside a company that was selling Linux at a massive scale. And what I realized is at a certain scale, and it's actually less than you would expect. But at a certain scale, what was explained to me was you are working with Canonical on a contractual level. Like you become like a – you get a partnership at some level where there's an agreement to work

together. And this is official, it's between two companies, and there's money exchanged. And in that process, get ready for this one. Because I think if you watched our Dell coverage, I think you would have the opinion and the impression that Dell did all of the modifications to Linux.

Dell made the changes to make high DPI work on Linux. Dell made the changes to standardize the Ubuntu installer. Dell made the changes to do an after first boot splash screen. You might walk away with the impression that Dell contributed all of that code upstream and Canonical just fell backwards into adopting it. While Dell did write some of that code, the reality is, and this is explained to me, and if anyone from our friends at Canonical wants to correct me, please, because this is just third hand that I got from Dell. My understanding is if you sell a certain amount of computers, you become an OEM partner. And by doing so, Canonical does all the heavy lifting for you. want to have high dpi support if you want to have special branding or if you want to have a special theme or if you want to have something that works with your wireless chip that isn't upstream yet

canonical actually does all of the grunt work to make that code actually work and then they turn over to you and say here's your repo here's your repos here's your image deploy this on the systems and so what dell has to do is they just have to figure out what hardware they want to ship. Then they go to a company that their entire job is to make sure that their operating system works on your hardware. And their entire staff is intimately familiar with Ubuntu. And then they make all of the modifications necessary and they deliver that end result to their business partner. In this case, in my experience, it was Dell. And then Dell ships that. And what's beautiful about that is instead of having Dell write software like they used to for Windows, you actually have the manufacturer of the operating system writing the modifications necessary to make it work on the OEM hardware. OEM, every dick and hairy OEM that gets the idea of shipping an ISO on an x86 box needs to write their own distribution is completely negated when you realize that if you just

if you work at a certain scale and you create a business opportunity with Canonical, they'll do the work for you and you can call it Ubuntu legally. Problem solved. And then your customers that want Ubuntu can buy Ubuntu. It's called Ubuntu and the software is developed by Canonical. And it is much more straightforward than Tuxedo as Ubuntu or whatever it's going to be called. What does that – what do the small-scale people do?

So I'll just chip in and say, yes, there is an OEM team, and that's how that arrangement works. My understanding is Ubuntu are not the only distro manufacturer that offer services like that. And, you know, I would suggest that the likes of Tuxedo and maybe System76 as well consider what they're going to spend on engineering resources in terms of dollars to actually pull this off and whether they do better to actually partner with, insert name of distribution here, who can offer those professional services. So I've not spoken. I think Chris needs to take a breather for a moment. I do.

I appreciate you guys filling in because, yeah, I want to calm down because this is really upsetting me. I'm not kidding. I've not spoken to System 76 anyone from system 76 since the announcement of bob os i don't doubt that i will at some point soon um it feels to me like they when canonical made the announcement back in april that we were downsizing slightly and refocusing our attention on different things and as a result the desktop was changing I think perhaps from an outsider's point of view it looked like system 76 utterly freaked out and did what they thought was best for them and their customers now to be fair system 76 know better what their

customers need than anyone else right because they talk to them every day and their customers give them dollars to make this stuff work and they're super connected too like they are they're not like a company that just sort of sells a piece of hardware and they just they like they really are good faith like they have employees who talk to their customers all day every day and they get good feedback from their customers about what they like and what they don't like and they're getting support calls and they're telling you know they're working with these people and okay yeah this is always something we get a call about yeah here's all you have to do to fix that i cannot totally understand that right but this is reminding me of something in the it world that's been happening for years.

Have we learned nothing from Android where every single manufacturer has their own strange, quirky version of Android that's bastardized from the original and is slightly different and not well-maintained, doesn't get updates, and isn't as consistent as all the others. It's a perfect thing. That's a dangerous road that they're going down. It sells iPhones. It's a road that sells iPhones. It's a road that leads to the damn Apple Store. What I see is that I just don't want, like, are these upstream chains, are these going to go upstream?

Any of these distros, are they going to be updating us? I suppose. But, you know, that's a, so upstream is something that gets tossed around a lot. But let's take a second and talk seriously. Let's be real about it. So if you're going to make a modification that works with your limited range of hardware, and then you submit it upstream, there is not a guaranteed chance that's going to be accepted. There's no – no one knows what the quality of that code is going to create something and toss it upstream is not the same as actually having upstream stuff you send upstream getting accepted and integrated into the wider ecosystem.

Right, and it raises the question too, right? It's upstream's job to guard against modifications or changes that are only – make sense for your base but may not make sense for the broader community. Right. Right. So say Tuxedo decides to – in a year from now, let's say that after 1804 ships, Tuxedo – I'm sorry. Tuxedo – I'm just – theory here. Tuxedo launches Unity 7 Legacy Edition because their customers have heard – they've heard from their customers. They got support tickets. They got phone calls. They got emails.

And they have tweets that all of these customers miss Unity 7. So Tuxedo launches the Unity 7 Tuxedo Linux. And it's great, Wes, because you log right in and you just get the great traditional Unity 7 desktop. My favorite. It just knocks it out of the park. It's really nice and standard, and it's easy. And so they, you know, but a couple of users have noticed that the online accounts don't really support

any of the modern services anymore because it's been a couple of years. And, you know, it'd be nice to sort of maybe get those updated. So Tuxedo writes a patch to add support to the online accounts to make sure that you can log into, you name it, service Mastodon so you can get your fucking Mastodon account logged into unity 7 and they submit that upstream and then guess what the canonical team looks at that and goes yeah well that's great that's code for a desktop we don't maintain anymore for a social platform that's dominated by social justice

warriors that we don't support anymore so thanks but no thanks we're not going to integrate that upstream but thank you for writing the code and then Tuxedo writes a blog post about how they've written upstream code and they submitted it, but nobody accepted it. You see, there's no winning here. No, there's not. I suppose there's not. That's the world of distributions. I guess that's what's been frustrating me here

and when I was trying to get it earlier is I just don't see the kind of the kind of tenets or axioms or clear thoughts behind what does it mean for this to be a distribution? What does it mean for it to be an OS? Why are we doing this? What are our changes actually be?

Because when it was just like modifications that include drivers or new firmware or some customizations that are the boot screen or whatever okay those all make sense. Those seem like a reasonable level of customization to be applied to a base operating system. But when I compare things, like when we've talked so much about Solus recently, et cetera, that feels much more like something that has goals, it has process, it has a philosophy behind it. I don't see these in that same light, and it doesn't feel like an organic space for Linux distribution to really thrive. I think the other thing that I find slightly irritating

is seeing a company who's in a difficult position like System76 or Tuxedo who clearly have customer demand or their own internal brainstorming that's decided they need to do this and then tell us that they've waited for us to do something for three years, they could have contributed to unity over those last three years. Like, it's the same argument that I keep seeing.

When a project gets shut down, a lot of people pop up out of the woodwork and then fork it. It's like, well, the reason this died is because for the last three years you didn't contribute but suddenly you want to contribute after the death of the thing has already been announced if you wanted the thing to be better three years ago you should have contributed three years ago not now okay i want to play devil's advocate because i watched uh ryan's interview on the lundukuk Hour and I at the end of it I felt like he didn't mention anything new like there was nothing that hasn't been in a blog post or replies

to comments that we didn't already know and I was a little disappointed I was hoping to hear something new because what he did toss out there was some something about flat packs and snaps that I didn't make any sense to me but I I actually did kind of understand a bit where they're coming from because he said something that really resonated with me. He said, we just want to make great things. We just want to do great work. And fuck if I don't understand that, man. I really understand. Like when he said that, I was like, yeah, OK.

I can see where they're coming from to some degree and I can almost actually understand why they had no other choice. I feel like they legitimately had no other choice than to do this. And so I'm sitting here like really super skeptical and at the same time like I'm going like, yeah, I kind of understand, guys, why you had to do this. And so I kind of want to share some of that with you. So let's talk first about Linux Academy. It's our last sponsor this week. And we've got to get this in before I forget. There's a lot of barbecue stuff going on.

And the summer is here. And Linux Academy is a great use of your time. You're sitting around wasting time right now. Use Linux Academy. If you're sitting around in your car driving to work right now and you're thinking, man, I wish I had a lifestyle that enabled me to live a little bit differently, Linux Academy. LinuxAcademy.com slash unplugged. And I say this because before they ever became a sponsor, I had a lot of conversations. In fact, I had conversations about producing content for them.

I had conversations about running shows on the JB network from them. We really kind of came eye to eye on one thing and that was spreading the adoption of Linux now this was several years ago now which almost sounds like a stupid thing to say because it's like so crazy I'm not sure

I can't say it post-Burman you are not that that's not your role here please just talk about Linux Academy I need to know Linux Academy Polifacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacacac SSH into a server and you're working on a real server. It's such a difference because that's exactly how I learn. And it means when I go to – not only does it mean I'm going to test better, which is obviously a benefit. But I think the thing that matters a lot more than that is when I actually go to do the work, I have the confidence that I've done this before because I actually did it in one of their hands-on labs. And they have instructor mentoring too, like real human beings to help you. They have

certs that, like course if you want to go for certs they have courses for that like just HyperCard. You're familiar with that? They don't have courses on that. They don't have. They also don't have courses on like barbecue. No. Because they're focused. They know what they're trying to do. They don't have

courses on Windows command line essentials. No. Or working with the IPX protocol. They don't have courses on Windows command line essentials or working with the IPX protocol. They don't have courses on that. But they do have flashcards that are forked by the community and made better and better to help you study. They have nuggets when you have just a little bit of time and you want to do a deep dive. They have a course scheduler when you've got a time frame and you need to just stick to it. Kind of like an exercise for your mind. LinuxAcademy.com slash unplugged.

So I was looking at both Pop!OS. Kind of ironically, I have it installed on a Dell Precision laptop. I have Pop!OS right now out there. I've invited everybody here at the barbecue, which is how many people do you think we have here?

I couldn't even... That's a great question. Yeah, we have a lot of people here. We have a good amount. They're the best people. And more than a dozen, not including us. And I've asked them all to kind of take a look at it. Here is the Dell Precision running Pop OS, and I always snicker. And one of the things I was really impressed with, and I'm showing the Tuxedo Zubuntu screencast right here.

Let me just mute that because that music is distracting me. I think I would like XFCE more if it came with that music. Yeah, right? And looking at both Pop! OS and Tuxedo Zubuntu, I get it. I do get it. And I'll talk more about System76 because I get it more than Tuxedo. Tuxedo is a joke because uh it looks it looks it looks like a house of cards in fact in fact if i'm not mistaken their previous attempts have been

pretty pretty bad i'm sure you must have had some experience from their ubuntu mate attempt in the past so the the review unit that i received had an ubuntu mate image on it but they they'd made a number of customizations beyond the the kernel stuff i i referenced earlier so they'd um they'd replace the greeter they replaced the gtk greeter with the unity greeter which as a distro maintainer as soon as i saw that when when the thing booted up i knew it was going to be bad because i know full well that breaks things in mate because it pulls in different input managers. So IBUS, for example. And it means that the keyboard layout is completely unconfigurable.

So that was one example. But also because of that change and the way that Ubuntu Marte 16.04 is put together, this is something that they've created. They have a script that runs on first login. Now it's in German language. So I didn't know what it said, but I had it translated since. But it basically says welcome and you click okay. But behind the scenes, that does an apt-get auto,

apt-get minus y auto remove. And that actually removed most of the Ubuntu mate meta packages and no software no yeah so that was the out-of-box experience now these are considerations so you know when where you can hear the full gruesome story on ubuntu podcast there's a full review of of the of the laptop it was good too um you were actually pretty polite if i if i'm gonna say you were actually fairly polite yeah i mean it's disappointing from my point of view because you know i i obviously make a bundumate and i've put a lot of effort into it over the years for it to work hopefully the right way when people install it for the first time so to think that

people might be buying computers and this is what they expect, that reflects badly on my work and it also reflects badly on Tuxedo. But it's an illustration of the kind of naivety that someone who isn't a distro maintainer has about how easy it is to maintain a distro. You know, all I'm doing is changing the greeter and changing the theme with not a full appreciation of what the knock-on effects of doing that might be. A cynical person, Martin, might think that the negative feedback that you gave to the people at Tux, whatever they call it. Tuxedo. Tuxedo. That negative feedback might reflect it badly on

the recent blog post where they talk about how their xfc edition uh is better memory footprint even a buntumate yeah i don't care what they write they also took a couple of shots at systems they took some shots at system 76 in that post it's in german so it's a little hard to translate but yeah they did yeah you know. I mean – That's fine. I mean they're welcome to make the comparisons that they want to make. That's fine by me. So here's where I could kind of see no other choice than doing this. And I have to make some assumptions and I'm not saying these assumptions are correct.

But I'm going to make some assumptions to just try to share what I perceive to be their side based on what I've read, talking in years of conversations with Carl and recently with Ryan. I know every person in the System76 office except for the couple of new people they've hired since I've been there last. But I actually met one of them at LinuxFest, two of them. So actually, I think I know everybody there. And that third one, he's coming for you. So I'm not saying this is not officially their position, and I'm not saying this is the position. I'm saying this is my insight, maybe as an 11-year customer too.

I think that they have reached a point where they have maximized the leverage that they can apply against the ODMs, and the ODMs being Clevo, perhaps, and others. And so they are trying to be competitive. They want to create a workstation, mobile and desktops, that address creators, developers, engineers, specifically people who are most likely to buy Linux machines, people we've talked about on this very show and said,

let's be real. New users aren't coming to Linux automatically. You have to introduce them to Linux. Otherwise, the people that are seeking out Linux are Windows and Mac OS refugees that just want to get their damn job done. And a lot of times they're writing software, they're in the engineering, scientific, or creative field. And on top of that, System76 makes machines that are exceedingly more powerful than most other hardware manufacturers.

Like we have Doug out there in the living room right now who has the Bonobo workstation. And he pulls that thing out of his bag, man. It is the size of a skyscraper. But sure enough, if it's not a powerhouse with ports for days and power for centuries and the oryx pro is the same right it's nothing like apple makes it's nothing like hp makes it's it's similar to stuff dell makes but it's still in its own category in its own league with certain things plus it's ubuntu now if you're making this hardware and you've got these customers that are coming to you looking for something other than a laptop that has four USB-C ports.

You want to be able to provide an experience that's similar to the competition, a complete hardware and software experience. And – Dare I say synergy? There gets to be a point when you start considering things like super high resolution screens, LTE modems, dual batteries, other things we haven't even considered, that you might just you're just going to have to start making more and more freaking modifications

to Ubuntu that it's almost a joke to call it Ubuntu anymore because you've got a PPA in there that makes kernel module stuff, it's put stuff all over the file system, it changes the way things operate and now you're maybe pre-loading software you've got a PPA in there that makes kernel module stuff. It's put stuff all over the file system. It changes the way things operate. And now you're maybe preloading software like a blender. Maybe you're going to preload a text editor that people really like.

You're going to make this thing more readily available and just out of the box, great workstation for developers. So you're going to maybe put something crazy on there like a Visual Studio Code. Or maybe for strategic reasons, you want to just totally integrate Flatpaks throughout the entire operating system. So there's these things that you're considering. And then there's these back in the mind problems you've had with support over the years. Like a lot of times, we've had to fix problems by just shipping people newer drivers that aren't in mainline Ubuntu.

So that's often the fix. So it'd be kind of better if we could just release maybe every three months, if we could just push out a release every three months with the latest drivers that fix microcode issues that make hyper-threading unstable and fix problems with your Thunderbolt controller and enable the webcam or the fingerprint reader. And we're just going to push this stuff every single month, including that we'd like to update the theme and make it look good and keep polishing the brand because we've got a person over here working on it all the time and they're always making modifications. And geez, you really can't call this Ubuntu anymore. And now all of a sudden you're at a position where what do you do? Do you keep calling it Ubuntu because that starts

pretty much getting you in, I would think, in the sights of trademark issues? Or do you just say, you know what? Fuck it. We can maintain this. We can do the Q&A before we ship these machines. If it's not ready, we won't ship it. We're going to base it on Ubuntu anyways.

We'll be able to take advantage of their repos and we won't have to host all that stuff. So we'll save a ton of money and we'll ship it on our machines and we'll put it out there. And when people complain, we'll say it's Ubuntu. But it's our own spin. And honestly, how could you not do that And honestly, how could you not do that? Seriously, how could you not do that? How could you not do that? I mean it's pretty tempting right now, yeah.

I kind of want to do it myself. I actually started an Ubuntu spin while you were talking. Yeah, I mean if I didn't have the scale to participate in the OEM program where I could just make Canonical do the work for me, this is the route I would go. And that's why you see Tuxedo doing it. Purism actually did it early on. We didn't realize it. But these guys are really copying what Purism did months ago. Purism did this initially and we are all – and I have talked to several sources inside Purism

that have told me that one of the reasons that they shipped their own desktop instead of Ubuntu vanilla – and I've heard this from multiple sources, that they didn't – that System76 was already shipping vanilla Ubuntu. So they had to differentiate. So that's why they didn't ship vanilla Ubuntu with the Librems. Interesting. So it's almost like System76 is going to open themselves up

to more market competition because there is a market for people that really want real Ubuntu because this is what they need for their job and they want their machine, their laptop, their dev environment to mere production. So as I understand it, System76 have said that although they're going to

debut Pop!OS for 17.10, they will continue to offer Ubuntu as an install option at the point of purchase. Oh, very good. I think that's smart. I don't think it's going away at all, but it isn't clear to me what customer Pop! OS is looking to serve. I've not seen anything in the discussions to fully understand that. Well, what if it is just somebody who wants a really great GNOME desktop on Ubuntu? Like if Ubuntu is going to go all in on GNOME and Ubuntu is my jive, then why not run something that's like the mint of Ubuntu's for GNOME? I don't think you want to start labeling Pop! OS as the mint of ubuntu's for gnome i don't think you want to start labeling pop os as the mint of ubuntu's i think that would be an unfortunate but you follow my drift though

right there has been a market demand there yeah so i mean yeah i mean the thing about linux mint and this is another concern i have about theseEMs going their own way as opposed to working with distros, is that there are all manner of considerations that get made. So you were talking earlier about the GNOME JavaScript backend, GJS. That was recently added to the main repository for Ubuntu 17.10 as part of the migration to um the gnome desktop but that came under quite a lot of um security scrutiny by the security team at canonical because that has a short maintenance window so their questions were how do we maintain this thing that is fundamental to the desktop that has such a short lifespan and has is able to pull down code from arbitrary online sources. So that was something that was thought about long and hard at Canonical.

Now, I'm not saying that System76 are about to do this or Tuxedo are about to do this, but somewhere along the line, Linux Mint started making decisions about the governance and the direction of their projects and they ended up with a decision which is you will never get critical security updates by default it's those kinds of decisions that you can work with a wider group of people with security teams with people with experience of putting distros together to help overcome those tough decisions and present some insight that you may not have on your own. Well, it also strikes me as a position of comfort. So these are decisions that are being made both by Tuxedo and System76 that strike me as short-term decisions that don't reflect what we

have coming over the horizon. Linux has got a lot of really big things. And because we're talking about System76 and Tuxedo, I want to keep it focused on the desktop here for a second. But there's so many things I could go into that are server-related. But just specifically on the desktop, you've got Wayland coming down the pipe. Hey, you. But it's not – I mean it won't be that many more episodes until we're talking about GTK4. And wouldn't it – if we were talking about thetk4 and wouldn't it if we were talking

about the transition to whalen and gtk4 right now imagine system 76 and tuxedo now having to take and mint and mint who just got to gtk3 with like a dozen episodes ago imagine okay now imagine witnessing them make the painful transition to gt4. It's a position of luxury in which there has been some refinement and stagnation for the last few years. And it is with that mindset that it seems like, well, this shit's going to be easy. You know what? I just do this. I just do that. And I just put it up here on this. And I got the GitHub community. And what's the problem? And it's not until you have to really start doing some significantly heavy lifting where these things are going to change and then all of a sudden we're right just like

the point was made earlier we're right back in the android system where samsung's delaying their latest os by six months because they got to make sure that the new Samsung UI works with the latest version of Android. Yeah. Now, as far as I know, the transition to GTK4 isn't going to introduce the kind of theming complexities that the not-so-distant past with GTK3 did. But, like you say, just adding a theme is easy, but then maintaining that theme through new releases of the toolkit can be a massive undertaking. And we've seen this in the Linux community. Just look at how many independent theme creators there were six years ago.

And now look at how many are actively making new themes today because they have drifted away because they got screwed over every six months every time they had to rework their themes and it just became too much effort for a lot of people yeah that is actually really true that is something i've watched personally some of my favorite themes have died off um wow but at the same time i i don't deviate from the from the position like i totally understand why they've done it and I almost don't think they had any other option. Especially when you feel maybe that they're – that a bunch of you is less focused on the desktop or on what they're doing. Well, if I could have like gotten a Doctor Who time machine or a Bill – It's in the garage. Oh, yeah?

Uh-huh. Oh, what about the Bill – I think Dylan is playing with it. What about the Bill Shatner bird of prey that I can fly around the sun? Can I – because that's my preferred method of transportation. It's cloaked. I'm sorry. If I could go back just six months and I could – if I could have a little closed room conversation with Carl, I'd go into his office where he's got his system set up there and I would sit down and I'd say, Carl, I would say stay tuned three to five months after this Unity 7 announcement that's coming. And he'd say, what Unity 7 announcement?

I'd say, don't worry about it. But there's going to be something important. It's going to also impact Unity 8 and that phone you've been working on for three years. And you're going to be super upset about it. But if I could just give you one piece of advice, and I'm like Biff right now. I'm giving Biff like the sports book here. I'm saying just give it six months and see where they go. book here i'm saying just give it six months and see where they go because i'll tell you uh i have subscribed to the mailing list for the ubuntu desktop because will cook is submitting stuff

every single week about what they are working on and sure enough if not every single week it has it hasn't made uh something useful for linux action news about what they're doing to make the transition to unity from unity i should say to gnome easier and the things that they're doing to make the transition to Unity, from Unity I should say, to GNOME easier and the things that they're tweaking to make it a, they're adding features that I wish the Unity 7 desktop had. Then they're adding it to the GNOME desktop. So it just, you know, if you just hang tight for a few months and see where things go, hang tight till the LTS, you know, hang tight, ship it, ship it and just wait till the LTS and see what happens. And then till the LTS. Hang tight. Ship it. Ship it and just wait till the LTS and see what happens. And then by the LTS, if you don't

like what you're seeing, rebase off of that. 1804 would be the thing to base off. If you're going to create a distribution at that point, or I'm sorry, don't call it a distribution. If you're going to create an operating system, base it off the LTS and give yourself a buffer to figure this shit out. To figure out

transitions to Wayland. To figure out transitions to GTK4, to wonder if maybe you should have gone with the Plasma desktop to really differentiate yourself. Give yourself that buffer with the LTS. But to go with 1710, to go with 1710, they know better than that. I mean, I know they want to ship the latest hardwares and they want to ship the latest kernels and they want to ship the latest enablements. But my God, go with the go with the 1804. You have watched Ubuntu for 11 years.

You know that they roll this shit out in the in-between releases and they bang things around to see how it works and then they nail it down for the LTS. You know that. You know that. You know that. You know that. You know that. You know that. So why would you base your OS off of a 1710, which is mid-transition? Mid-effing transition. Mid-transition. It is the most awkward stage of what Ubuntu is going to be for the last 11 years that you have been shipping Ubuntu, quite literally. Not even the transition to Unity was this awkward because you nailed that. You got it. But this is it. This

is the most awkward moment in the last 11 years of the Ubuntu desktop. It's 1710 and it looks good. I mean, I'm really impressed, but you based it off 1710. I know right now, I know right now it's 1704. I understand that. I haven't installed. I'm running it, but you're going off 1710? I know right now it's 1704. I understand that. I haven't installed. I'm running it. But you're going for 1710, and that's crazy. That's crazy. Why wouldn't you wait for the LTS?

Why wouldn't you give them a few months to figure out what they're doing? Because you know what? It looks pretty good. They're working with Upstream. They're getting SnapSupport built in to GNOME software so it's not some weird canonical patch set to the GNOME desktop. They're making sensible choices with extensions. They're asking the community for their feedback. It is a clear transition already in its early beta state.

It is a clear transition from the Unity 7 desktop to the GNOME desktop. It's exactly what I would think the System76 customers would want, would be stable, simple transition to GNOME. And what's Pop OS shipping? GNOME! It's shipping GNOME. It's shipping GNOME 3. Pop OS is shipping GNOME 3. What is Ubuntu 17.10 shipping?

GNOME! GNOME 3. It's the same thing. So all of the heavy lifting of a transition, all of the awkward stuff, all of the stuff that's weird and different, it's happening regardless. And they're making it an easy transition. Pop OS is not bad. I mean, the theme is not bad. I kind of like it.

You did the whole show notes and everything on it today. Right. Not bad. Not bad. However, it is nothing like Unity. And the thing is, the GNOME 3 implementation that Ubuntu 17.10 is going to ship with is a lot like Unity. It looks like Unity. It's familiar.

It is an obvious transition path. What Pop! OS is is more like XP to Windows 8. I see what you're saying. I mean, it does kind of feel like unfortunate timing. You know, when they previously talked about, like, the pop theme, et cetera, that felt a little more natural, like an immediate reaction, something like, okay, well, we don't know what's coming now. I think you're entirely right that there is going to be a lot of changes,

and while System76 does need to keep abreast of those, what they have right now, I think, is stable enough. They're not in a huge rush to present this whole new operating system that they suddenly have to support. It's like an entirely change of lifestyle and support cycle and all of that. I guess it really doesn't matter as long as they offer the Ubuntu 17.10 vanilla. If you have that as an option installed and that's what makes your corporate purchaser happy, then have at it.

And then if enthusiasts think it's a good idea, can use Pop! OS and choose that at buy time or build time. That's true. That seems like that's actually a pretty good way to go about it. And then System76 will have the numbers. They'll have that and they can see what's successful. But myself, if I was going to launch Jupyter Broadcasting OS, don't call it a distro. If I was going to launch Jupyter Broadcasting OS, no way in hell would I base it on 1710. I might go 1604 to be honest with you.

But I would probably wait for 1804. Really. So I – so what I – here's where I feel like I'm at is I would probably wait a couple of years to see how this thing plays out. And I would probably buy it with Ubuntu for now. And then I would just see where Pop! OS goes, see if they get a good community, and see if they start shipping some stuff. I'm like, damn, that's a good idea. And you'll know because some of their modifications will start showing up in the AUR. And you'll see Fedora users that are posting guides on how to get that.

And Ubuntu users will have guides on how to convert their existing Ubuntu install into more popular. Like if you start seeing that kind of stuff, that will be a sign that they have some good momentum. They have some success here. And I'm all for that. Like go for it. And then I'll just kind of sit back and wait and look at it and go, OK. All right. I'll wait for it and I'll wait until the 1804 version ships and then that might be when I try it.

I mean I have it running now. But I don't think I would run on anything in production until 1804. Right, I was going to say that. This in the same way feels very much right. It's like they're, just like it's 1710, it's their 1710 except even more so, right? It's like the beta beta where they're kind of playing with this now.

It's just hard to have that kind of announcement and have, you know know the interpretation is always going to be like this is the future no matter what this is what you'll see when really there's a lot to come on both sides to actually find out what we're going to see long term you know what gets me fired up about it and like the reason

why I'm like so passionate still this is like the third show I've recorded about it and I'm still like super fired up about it is it seems like they are just – they're just saying, here, Dell. Here's all the customers that just want a really good Ubuntu integrated machine. You want an Ubuntu desktop that has firmware updates that come through GNOME software and they just ship Ubuntu.

I do want that. Yeah. And then you just want something that has Ubuntu proper with a couple of modifications to make high DPI look right or to make wireless work and Canonical has done that development work. Like it's like the most proper Ubuntu you could possibly get on your workstation. That's now Dell. Yeah. And if you would have asked me a year ago, I would have said it's System76.

But it's not – full stop, end of sentence, that's a huge game changer right there. That's a huge game changer right now. And that is a massive conscious decision that System76 has made. And if the Pop!OS is successful, then it was a wise choice. And if they can really provide something valuable there, it's going to pay off. and if they can really provide something valuable there, it's going to pay off. But if people are buying their hardware because they want a really good machine that runs Ubuntu,

they've really marginalized themselves a bit. And that's why it gets me fired up because I've been a customer for 11 years and I am so passionate about that and I'm so passionate about an idea about a hardware vendor that's pushing this forward. And it's, oh man, it seems like the riskiest time of, because also at the same time they're launching manufacturing.

That's everything, right? That's the whole thing. Woo. Woo. Man. I guess we'll see. Yeah. Yeah.

I think, unless he changes his mind now, Ryan's going to be joining Joe soon on Linux Action News. Oh, is that right? Because I'm going to go to Montana with Noah soon, and I think the beard's going to fill in for me here on this show. How about that? Yeah. Is that good with you? Yeah, heck yeah. Yeah, he offered. I'm like, yeah, man. Although you might have to sit here, and he might have to sit there.

Yeah, because I don't want to mess with his thigh. We'll be staring into each other's eyes. No problem. You'll be staring into his beard. Don't kid yourself. Yeah, that's true. Can't look away. And then we've asked Ryan to fill in for me on Linux Action News for one episode. Neato. And I don't think they're going to talk a lot about it, but it'll probably come up.

So that could be – Linux Action News is like – I think we just – we're close to 10 episodes. Yeah, you are. If you haven't heard it yet, it's like – it is like a direct dose to your brain of Linux News. And what I'm really proud of – and I wouldn't have have it's funny because I wouldn't have thought of this ahead of time but there's a couple of things secret sauce wise that Joe and I are doing for the show that I think are really paying dividends right now and you if you listen to the episodes

half mayo half ketchup that's what it actually no it's not mayo can you guess what else it is Miracle Whip Miracle Whip no there's just some secret sauce in there that we're doing that I think is paying dividends. And I think we do an exceptionally good job of busting through the news and taking hours and hours and hours of research and giving it to you in five minutes. So it's like we do hours and like talking to people behind the projects, emailing people to get follow-ups, running the system, running it on our systems, and then we consolidate it down.

And sometimes we even nail it under 30 minutes. And it's LinuxActionNews.com and a hell of a show. I'm really proud of the way it's coming together. And so Ryan will be joining on that soon. It's like those, you know, like the NPR or other things. It's like the five-minute news update. It's like that except instead of depressing world news, it's like 30 minutes of awesome Linux news.

So it's better in every sense. Eight episodes, Mr. Bitton verifies. And if you haven't heard it yet, I think the show is turning out really good, and you can get all the feeds and everything at LinuxActionNews.com. Also, you can find Mr. Wes, TechSnap, TechSnap it over there. Check out the TechSnap show. It should be out probably right about now as we're recording the show this week's episode.

And then Popey and Mr. Wimpy just wrapped up their live episode from Fostalk Live. And that was a really fun episode. So go check out the Ubuntu podcast. Gentlemen, anything else you want to plug before we run? Not from me. Thanks. No. Well, OK.

Very good. Of course, I'm waiting for poppy's indiegogo project for the mic we will all be jumping on board yeah yeah and if you want to know what i'm talking about go check out what you didn't what you didn't hear from fostalk live is that we had a um a vinyl record made of the no bonus tracks of Popey Mycroft scenes. That's my favorite. That's somebody one.

Well, first of all, I have no idea what you're talking about, and second of all, that's one of my favorite things. So, that's amazing. Very good. Very good. Vinyl records, huh? Do you have a vinyl record player, Wes? I do, actually. I knew it!

I knew it! You and Michael Dominic. You and Michael Dominic. Joe Rissington won that. Yeah. Oh, did he? No kidding. Speaking of Linux action news, go figure.

Yeah, yeah. Good. Does Joe have a vinyl player, though? Everyone but you does. He does. Oh, I've got to give him a hard time about that. We've got to get one here for the studio. Yeah, yeah, yeah.

All right. Well, thank you for turning. It should have been 2.04 for the barbecue, but it just. He does. Oh, I've got to give him a hard time about that. We've got to get one here for the studio. Yeah, yeah, yeah. Alright, well thank you for tuning in. It should have been 2.04 for the barbecue, but it just landed on 2.03. America's birthday. We're going to get out of here. We've got more meat. More meat to eat. More cooking. It's been a lot of fun. Thank you for joining us this week. If you'd like to join us live next week, go over to jblive.tv on a Tuesday. You can go to jupiterbroadcasting.com

slash calendar. Get that converted to your local time. jupiterbroadcasting.com slash calendar, get that converted to your local time. jupyterbroadcasting.com slash contact for the contact page and also linuxunplugged.reddit.com Go there now. Got that stuff, Reddit. Yeah. And we'd love it if you join us live in the mumble room. That's the best.

Yeah, that really is. If you can go over to jblive.tv do bang mumble. Alright everybody, thanks so much for joining us and we'll see you right back here for episode 204 next week. All right, jimmytyles.com. Oh, we got to go bold. Oh, yes, that's right. That's right.

So jbtyles.com. Go check it out.