LINUX Unplugged - Episode 268: Elementary, My Dear Plasma

Episode Date: September 26, 2018

We chat with Nate Graham who’s pushing to make Plasma the best desktop on the planet. We discuss his contributions to this effort, and others. Plus we get the scope on a new Juno feature from the source, and the creator of WSLinux a distro built specifically for Windows 10’s Windows Subsystem for Linux joins us. Also some community news, some old friends stop by, and more! Special Guests: Alex Kretzschmar, Brent Gervais, Cassidy James Blaede, Hayden Barnes, Martin Wimpress, and Nate Graham.

Wes found something that I'm going to be using the heck out of potentially. It's YouTube DL web server. It's a web server for downloading YouTube videos. You fire up a Docker container, you give it a YouTube URL, and it's pretty much off to the races. You then are re-hosting your own downloaded instances of the YouTube videos. And I think right now one of the things it does is it re- or transcodes, I guess would be

the way to put it. Not re-encode, but it transcodes them to MP3. So then you can download them from a web UI as MP3 files, which kind of be a cool way to convert long videos into like a podcast. You know, and there are things to do this, various sites out there or bookmarklets or just the command line. But if you're like me, you have others in your life who frequently have sources that are YouTube videos. I can't be bothered to FFmpeg them MP3s all the darn time. So here you go. Pop it in that web UI and you're off to the races. Now, of course, right now you can't really choose, you know, does it give you the full video? And there's a ton of options. There's dash videos. There's, you know, YouTube has a lot of downloading

options. So hopefully those features show up in a later release. I've got a very so shameful Bash version of this that I have been using, so maybe after the show I can replace it. I love YouTube DL, and this just makes it even better and more accessible to friends and family and Lazy Chris,

which Lazy Chris always appreciates making things a little easier. This is Linux Unplugged, episode 268 for September 25th, 2018. Welcome to Linux Unplugged, your weekly Linux talk show that's got a studio full of brown bears. My name is Chris. My name is Chris. My name is Wes. Hello, Wes. We have a packed show today.

Multiple guests stopping by the show, some community news, and something that we don't get to enough. The audience's feedback and emails and questions. We're going to do a little bit of that, too, do some follow-up on some recent topics, and I'll tell you about a new rig that just showed up in the studio. It's a monster. I haven't told you about it yet. Yo, I'm excited. I'll tell you in the show. You'll get to find out.

And, maybe the most important thing on top of all this is after the show, we're going to do an unplugged barbecue. Noah's in town. He's hanging out at the studio. Wes is here. He's brought his pups. We're going to go get Levi, go get some food. He's hanging out at the studio. Wes is here. He's brought his pups. We're going to go get Levi. Go get some food.

There's some meat in the cooker right now. Yeah. This is prep for the big public barbecue we want to do once I'm back up on my feet feeling better. Got to get it right in before the holidays. So the timing is going to be tricky. So this will be our test run at a big barbecue, a big community barbecue.

We'll do a trial. I mean, you got to rehearse if you want to get it right. You already got the sous vide going. Yeah, that's right. All right. Well, before we get any farther into the show, you know what we got to do. We got to bring in that virtual lug.

Time, appropriate greetings, Mumble Room. Hello. Bonjour. Amazing. We have a great turnout in there. Mr. Dar's back. Brent's there, of course. Cassidy from Elementary OS. Longtime contributor Eric in there. Mr. Dar is back. Brent's there, of course. Cassidy from Elementary OS. Long-time

contributor Eric is there. Hayden is here from an interesting project we'll be talking about in just a little bit. Mr. Minimac, true contributor to the, I mean, just here all the time. And returning to the show, Mr. Ryan Sipes. Ryan, welcome back to the show.

Good to be back. We got Sean in there and TechMav as well. So we got a big crew today. We're really looking forward to today's episode. So let's start with the thing that people are sort of the most on edge about this week. We'll just get this out of the way. People are pretty upset that Google seems to be, quote, secretly logging users into Chrome whenever they log into a Google property. So you go into Docs, you go to Gmail.

I do that all the time. You go to YouTube. Wow. Yeah, right. I mean, what don't they own? Starting with Chrome 69, whenever a Chrome user would access a Google-owned site, the browser would take that user's Google identity and log the user into the Chrome browser account system.

Now, with this revelation, obviously there's a large number of users who are just kind of angry. It feels like a sneaky behind-the-back modification. And you worry that it could allow Google to link your traffic to a specific browser and then to a specific account. And as someone with multiple Google accounts, there's some things I search that,

well, maybe not necessarily bad, I don't need on my work account. Right, well, and Dar in the mumble room today was just talking about on the pre-show about how he's on a machine that's shared by multiple people. So how does that work now when multiple people sit down? Is it bouncing between Google accounts every time you change a Google sign-in? Somebody else sits down, logs into Gmail, and now it's a different account signed into the browser? That's a very good question.

Now, there is something we should be clear about here. different account signed into the browser? That's a very good question. Now, there is something we should be clear about here. Google engineers have clarified that this auto-login operation does not start the process of synchronizing local data to Google's servers. That still requires you to explicitly click something. Right, exactly. That's a big difference.

Yeah. Oh, sorry, I forgot to call out Alex. Mr. Ironic Badger is also, sorry, Alex, I didn't mean to skip you. I just don't read the screen very well from that far away. You hurt my feelings. I know, you know, and we and I were just having a nice chat, Alex. Mr. Ironic Badger is also, sorry, Alex, I didn't mean to skip you. I just don't read the screen very well from that far away. You hurt my feelings. I know, you know, and we were just having a nice chat, too. Well, I'm very glad you're here. It's nice to see you. Alex, of course, was joining

us for quite a while after Texas Linux Fest, but then took some time off to travel to the States and move to the States down in Raleigh area. And I'm now dealing with establishing a U.s credit history which is unbelievably pain in the ass territory welcome our sympathies sir um then there's another

thing that's now brewing today uh christoph tavern tweeted out that when you clear all cookies in the new version of google chrome it clears out all cookies except for the Google cookies. And to be clear, what it actually seems to do is it wipes out all of the cookies on your file system and then immediately recreates the cookies for Google's properties and repopulates them. Interesting. Now, I did see some defense out there. There's clearly some privacy implications here that people are rightfully concerned about. I did see some people trying to justify it from the developer side of view,

you know, making the argument that, well, users are confused, that they're logged into Chrome, but they're not logged into our Google sites and vice versa. And then, well, once you, if you are logged into the browser and you do clear your cookies,

on the very next request, it's pretty much going to go check and be like, hey, am I signed in? It's going to re-grab those cookies. That's true, yeah. So you can see them being like, well, it's simpler if we just warm them up front that maybe those cookies aren't, are always going to be there. Yeah. It's, it's sort of like in iOS when you turn on airplane mode now, it leaves Bluetooth on to connect to your watch or your AirPods. There's not like true airplane mode anymore. And what that feels like to me is design and product overruling engineering.

I think the most accurate and cutting criticism about both of these changes is that when asked, Google engineers could not articulate a very clear reason why this change has happened and why it was kind of snuck in, which causes people to lose trust. I think they cannot say what it is because Google hasn't made a public statement about it and generally contracts forbid them to do so. Yeah. So they were limited in what they could legitimately say. I mean, they are making some suggestions, but they all don't really pass the sniff test.

Yeah. Do you think you're right? I mean, at this point, Chrome feels a bit more of like a Google product that happens to have a browser inside of it and less of like a browser first sort of thing. And it was the good engineering of a lot of Google's products that, you know, made me switch to Chrome for a long time, is it was a great browser initially. And that was because of great engineering. Does anyone know, is Chromium also affected by that or only the Chrome proprietary version?

Most likely it will be. Looks like it is. I'm using Chromium now. Now, but I've also heard a lot of people saying they're using a version of Chromium that avoids all of this. Yeah, I think there's some

patches out there to remove that. Know that Debian's version actually disables the Google services unless you go there manually and enable them. Oh, okay.

Only the Debian version or also the Ubuntu version? I can't talk for other stuff. I don't, I haven't tested. Okay. So I have to check that. Now, Cassidy, you point out that this behavior isn't necessarily new for some group of users. Yeah. As far as I know, this is how it's been from day one on Chrome for Android. Like, and it, maybe it was a little more forthcoming on Android. When you sign in, it says, Hey, we're also going to sign you into the Chrome browser to sync your bookmarks and everything. So I don't know if people are freaking out about it here on the desktop, why that's very different from freaking

out about it on Android. Yeah, I guess it does make it more in line with the Android version. That's an interesting point. It's kind of a weird one. Yeah, we have different conceptions on our different devices, I guess. You know, practically speaking, the very first thing I do is I log into the Chrome browser. So this is not necessarily a move that I find personally offensive. I am dumb enough, I guess. I don't know what people would label me who think Google is spying on them constantly. But I am willing to risk it. And I use the sync service.

I sync my history and my bookmarks because I find that to be a very useful feature of the browser. And it's a good sync system, too. So I just have been all in from day one when they turned that on. So I don't really find this to be offensive, but I do understand people that are concerned about it. And it's just good to know and assess when you do have options out there, which is nice. When you're weighing Firefox versus Chrome, this is another data point, and I think it's an important one. Something else that I found really handy in the past

has been Play on Linux. Play on Linux is a wrapper around a series of scripts to make it really easy to install a lot of different Windows applications and use Wine. I've been wondering, where does Play on Linux go now in a post-proton world? How about that, Wes? As a saying, a post-proton world.

I like it. So where does Play on Linux go? And it appears that they're making some massive changes. Alpha 1 is out this week, along with a Play on Mac build as well. And there has been some big changes. PlayOnLinux version 5.0 alpha has been completely redesigned, a new user interface, and a new platform they're calling, I guess, Phoenix?

Am I saying that right? Phonics? I'm the worst. I'm just the worst. I shouldn't be allowed to speak. The new platform is, quote, decentralized, Git-based, and improves their script engine greatly. The new alpha has support for about 135 different games right now. They're also investigating using virtualization or

container technologies to improve the gaming experience, so hey. Hmm, that could be interesting if you want to, like, you know, isolate a series of libraries or a specific version of Wine. There's a bottle system now, but that's more at the software layer. So this would be more at the systems layer. Yeah, so something to keep an eye on, playonlinux.com for more.

I've used it on and off over the years. I've also used Crossover quite a bit over the years. And now it's like, well, how can these projects, Proton, how can these projects, Play on Linux and Crossover Office, integrate Proton? Yeah, right. Can they further the value add and make it even easier to get some great play on Linux and Crossover Office integrate Proton? Yeah, right. Can they further the value add and make it even easier to get some great games on Linux? I suspect there might be a way at least to take

some of the upstream improvements that have been in wine and we'll probably start seeing those benefits. I'm curious to see where that goes. There's another thing I'm also watching to see where this goes is it sort of seems to me like elementary OS is positioning themselves as a family-safe,

family-first distribution. And I started thinking that after I read a post by Cassidy on their Medium blog about app center and content ratings that are coming in Juno. They say, starting with Juno, Cassidy writes, we'll display a content warning when a user goes to download an app that meets or exceeds a certain level of the OARS data. Think of that as like nudity, violence, and language content. In addition, these apps will not be featured on the App Center homepage as recommended installs. They will remain in the search results and category pages, so if you really want to find them, they're definitely still there. And we should note here that they want to be clear that elementary is not intending to be the content police.

They could never make as informed of a decision about what's acceptable as, you know, yourself or a parent of a child. This is just a small step toward more robust content controls and informed consent, as they put it, for their users. So, Cassidy, this seems like an interesting differentiating feature of the distribution, sort of making it safe for parents to deploy elementary OS for their kids. And it seems like this could go much deeper than just the App Center too. Yeah, exactly. And that's the plan. We've had parental controls in Loki now for a couple years. And it's, you know, it's kind of a first take on the concept where you can set up certain restrictions for apps for a managed account. But I think we're looking at these digital wellness features that we've talked about that are coming to like Android and iOS. And we're kind of reconsidering how we could build

this feature into the OS itself for both child accounts, but also for people to be able to control what they're comfortable installing or seeing in the OS. I was talking to somebody recently, and they said that their friend has PTSD from combat-related trauma. And so that's an example I gave here of, you know, if you want to prevent yourself from seeing very violent content, you should have the control over your OS to do that. And it's always in your hands to be able to do that. And Ryan, you also are making the case it's not just for kids. Yeah, I just wanted to pair what Cass said.

We actually had a conversation about this the other day. My girlfriend is really careful about this stuff because she was over in Lebanon assisting Syrian refugees and was exposed to a lot of, you know, violence. And so it really triggers her to see, you know, really graphic violence. And so I think a lot of people benefit from being able to set what they're comfortable with seeing and appearing in front of them on their own machine when they're using it. Yeah. I hadn't thought about that, but that does make, that makes a ton of sense.

It's not an angle we usually think of, right? Like control usually means like I want unrestricted things, but restrictions for yourself, well, that's still control. It's a new idea that we've played with a lot recently, mostly in the mobile context. Like should I start using the built-in OS features to limit screen time in the evening? In some way, things like Secure Boot are almost analogous.

Oh, yeah. Yeah, you're right. We have sort of been building this. It's like, I don't mind optional controls that I have total authority over that I can choose to put in place or not. Like, that's totally fine in my book. It points out problems in human social structures, right? We get a little nervous when we know those controls are there and easy to use, especially when we're not in a situation where we have total control. I also think it's really important that we build these tools to put the control in the hands of the users, because if we don't build it

in that way, I think there's too much of a temptation for people to control it at the service level. For example, if we thought, oh, we shouldn't have explicit content in the store, we might restrict all explicit content from the whole app center. And I don't think that's the right way to go. Ah, I see what you're saying is it also makes room for the other way. So there can be content that is explicit, but there is an easy way for people to opt out of it. Right. So that's a great point.

So, Cassidy, I guess this is one of many new things that are coming to Juno. And we're kind of going to see this in the early stages, it sounds like, with Juno. Will it be specific to the App Center in Juno, but then future releases expanding from there? Yeah, the idea is kind of this tiered progression. At first, we're doing this kind of minimum viable. It's just a warning that says, hey, here's a heads up, according to the data in the app, that it has some explicit content. But after that, I think, yeah, we're going to look at really building in, both in App Center itself, listing out what actual ORS tags are there. And so whether it's explicit due to violence or nudity or language.

And then after that, yeah, in the future is more system-wide controls. So this could theoretically apply to your video player or your music app as well. That's an interesting idea. And I think, like I said at the beginning, it's a good differentiator. I have found elementary OS to be a fantastic OS for my son. It's been on his laptop now for a couple of years at least. Just super solid.

So I really like this. And I plan to, you know, them on m1 trios because it's just worked out perfectly so great cassie well thanks for coming on and explaining that and man am i looking forward to seeing juno land it's got to be getting close you guys must be in like the big final push right um yeah it's it's coming it's sure coming i will say i've been working on the um the big release announcement post you know because there are certain things we know are obviously in this release. And we can start taking some screenshots and things. But you are in that phase.

You're in that phase. It's like a 20-minute read right now. Wow. So it's intense. There's a lot of new stuff, a lot of great stuff. Wow. Congratulations. Looking forward to it.

Thanks. Speaking of other new stuff, we've been covering the development of Pipewire, which has recently had a couple of large growth spurts when it was adopted to provide remote desktop access to Gnome Shell and now to also the Plasma desktop for Wayland. This was always the big question is, you know, I love X11's network capabilities. Wayland has nothing. How are we going to achieve this? And Pipewire is a back-end plumbing system, I guess, for video right now. But that's about to change. Pipewire is going to also provide audio support. And that work has started to be formed.

There is pretty much everything in place they need for video, even including patches for Firefox and Chromium, which are in the development stages right now. So the video part being pretty well solved, and at least at a near 1.0, the Pipewire creator himself, Wim Tamas, and the Pulse audio developer, Core Pulse audio developer, and a few others are getting together at a hack fest this fall to start the core work on building what they're calling a production audio pipeline for Linux. That's what that work is going to start

at the end of October. And they're traveling to Edinburgh, and they're all going to get together. And I think this is a massive, massive moment because macOS has a very competitive audio subsystem, like Core Audio is what they call it. And Linux has gotten pretty good with Pulse Audio and Jack. In fact, we use it extensively in production and are pretty happy with it these days. But I wouldn't say it's a cohesive API that you can easily write to like Core Audio is. Right. I mean, you have different use cases that are

segregated and a lot of your users will end up having to install different software and learn how to configure it if they have a new use case. So Christian here is blogging on one of the GNOME blogs that I got off the GNOME Planet feed. And he writes that he's feeling very optimistic that we can come out of this event with both a plan for what we want to do and the right people involved to make it happen. The idea is we can have a shared infrastructure for consumer-level audio and pro-level audio

under Linux. It really excites me, he writes, and I do believe that if we do this right, Linux will take a huge step forward as a natural home for pro-audio desktop users. Amazing. That is really exciting. Now, of course, it's just early days. That's where it's at.

We do have good audio solutions, but this is what we need. We have had missing abstractions, and I think that's what this is showing. And something to keep in mind is this solves the issue of a proper way of providing audio and video hardware access for sandboxed applications, which we're going to start seeing more and more with. Flatpaks and snaps and Wayland applications. This is going to be pretty nice. Secondarily, it provides a unified solution for user space and pro audio, which means pro audio users will no longer have to read all these crazy how-tos and guides to get jack going on their system, which is what Wes and I are doing right now. It will just all be programmatically achievable.

And third, it provides a video equivalent of pulse audio, which makes dealing with different cameras, screen capture, and a lot of other things, including doing all that under Wayland, possible. of other things, including doing all that under Wayland, possible. So that's what Pipewire is bringing, is a unified audio and video subsystem that works across sandbox applications, works with Wayland, and gives an API for developers to build towards. It's a pretty big deal. I don't want to crap on the initiative, but it sounds like, hey, we have too many standards. Let's make a standard to fit all these standards into one standard. And lay it on top of all the other existing things. Yeah, there is that definite potential. But the thing that seems encouraging to me is this is involving people from the KDE project.

This is involving the core Pulse audio developer, which I always mess up his name, but I became a patron of him. Leonard Puthring? No, no. The original one? No, it's now run by Arun. I can't even take a stab at his last name, but he's on Patreon. And he's now the core developer of Pulse Audio.

And they're coming in together. If it's going to work, this is the right group of people. Yeah, they have the right heads in the room trying to build a coalition here, lay good foundations. It'll probably be a long time, just like Wayland was before we really see it in a desktop OS near us.

A long time before it's totally feature-rich too, right? It's going to be probably a minimum viable product for a while. And then they'll build on top of it. So you and I will probably be writing out Pulse Audio and Jack for probably another year or two. Well, we'll be early testers. I can assure you. I'll guarantee that.

Guarantee that. I still amers, I can assure you. Oh, guarantee that. Guarantee that. I still am just, I don't know. Jack is, the vernacular is rough. It's just different. It's different. But in a year or two, Microsoft will finally have a Linux distro. Come on. Thank you, Tar. Tar, you just never let it go.

Hey, it's three years have passed. Two are missing. I said five years. Come on. That's true. I know, I know. I mean... We can't fault you there. It still could happen. I give you that. In the meantime, I can't believe this new rig

that has come into the studio, shipped to us by Dell for review. It is the Dell Precision 5530. And it is a 15-inch screen, infinity edge-to-edge screen in a 14-inch chassis. Oh. And you can get it with Xeon processors if you want.

You can get it with 32 gigs of RAM. I'll save the spec of ours for the review next week, but it just arrived in the studio today, so I'm going to be loading it up. It has Ubuntu 16.04 installed from the factory. That looks like an XPS. It does. It looks exactly like an XPS 13, only a little bit bigger.

Well, because it's a 14-inch chassis and a little bit thicker. But for that, you get a competent cooling system, which I'll be testing, and, you know, way more horsepower. The desktop replacement version of the XPS you've always wanted. Yeah, and it's, you know,

you can get crazy fast processors in this thing. It's got the infamous i9 as an option, always wanted. Yeah, and it's, you know, it's, it's, you can get, I mean, you can get crazy fast processors in this thing. It's got the infamous i9 as an option, I see. Yeah, yeah. Can that thing really dissipate the heat? Is it really capable of that? I've got the, I'll give you a little spoiler. I've got the i9 version that turbos up to 4.8 gigahertz and I'm going to be testing that very thing.

Don't take it across the border to that crazy Canadian that drops iMAX all the time though. Yeah, I know. Right. I know. I'm pretty excited. Which Dell are you referring to? Well, hello, Mr. Wimpress, the Precision 5530, their 15 inch mobile workstation just got, just arrived here at the studio. Oh, really? Yeah. Very interesting. Can you hurry up on that? Because, um, I've got Bitcoin burning a hole in my wallet and I'm about to slam down on a new laptop and it's either going to be that or the

7350, the new one, or the ThinkPad X1 Carbon Extreme. Oh man, you really are running... I love ThinkPads. I'm seriously considering a return to ThinkPads, having been a ThinkPad user for many, many years prior to, you know, the whole Dell XPS thing.

So, yeah, I'd be really interested to hear what you have to say, Chris, because literally I've got them in a spreadsheet side by side and there's only a few things that are splitting hairs between them. Wow, you've even assembled a spreadsheet for this, you're telling me? Oh, yeah. I need to work out exactly how many one Bitcoin I'm going to spend on each of these laptops. Oh, that's, you know, I completely understand where you're at, because it's like, if you feel like you've got a limited amount of time right now with the Bitcoin, too, now's the time to pull the trigger. This time last year, it could have bought all of those laptops and had change out of one Bitcoin. And this time this year, it's like, spec and had change out of one Bitcoin. And this time

this year, it's like spec those to the max for one Bitcoin. Yeah. Well, I'll be trying it out starting now, even though I just got a brand new laptop. This would be like, this could have been me in the future if I'd stayed into editing and I was doing video editing or I was doing a lot of audio editing under Linux. Yeah. and it's like demanding Linux workload. Perfect. This could have been in my future. So I'll give it a review. I'll load it up.

I'll put it through its paces. It's got a brilliant 4K display, so I'll get a good taste of that. And I'll give you my full report in the next episode of Unplugged. So hopefully then, Mr. Wimpy, you could make your choice. I will wait for one week and one week only. All right.

Well, I had a great chance to chat with Nate Graham. And if that name rings a bell, it's because he's pretty involved with the Plasma Project. And he is the individual who's been posting weekly usability updates about Plasma that really showcases all of the goodies that are coming to the Plasma desktop. that are coming to the Plasma desktop. And it's the kind of things that have gotten me really hyped about Plasma and make me want to keep using it and keep trying out the next version, the next version, which actually ultimately made me end up choosing Neon again. And so I had a chance to chat with Nate about the history of his mission to make Plasma more usable and also funding of that type of work.

I've been wanting to get Nate Graham on the show for a while, and this week we had an opening, and I emailed him late last night, and he took me up on the chance to come on the show and talk about some of the work that he's been doing with plasma. And I love this, Nate, your self-described mission to make KDE Plasma the best and most widely used computer operating system interface on the planet Earth. That's a great mission. Nate, welcome to the show. Thanks so much. It's great to be here. So how are you accomplishing this mission?

I've noticed what really caught my attention was you seem to have a big focus on polishing the usability rough edges of plasma. Polishing the Usability Rough Edges of Plasma. In fact, it seems like this is something that's been kicking around in your head since 2017 at least. I saw a post where you're proposing improving usability. And then in September of 2017, there was a post of KDE's goals for 2018. And in there was a piece from you advocating for usability improvements. What's behind this mission? That's right. So it looks like you did a pretty good job of summarizing the history. I'll just fill in a couple of details. Please do. So back in late 2017, the KDE leadership basically asked the community, hey, what do you think we should focus on for the next couple of years?

And I was a relatively new member at the time. I had done really hardly anything, but I thought, why not? You know, here's what I think we need, and I'm just going to put it out there. So I proposed my usability and productivity goal. And to my pleasant surprise, it seemed to have resonated and people liked that, and it was chosen as a formal goal. So after that, you know, a month or two passed and not much really happened. And I realized if this was actually going to get done, I had to take the initiative on it, you know, because I was the one who proposed it. Nobody else was really going to, going to do it before me. And I'd never really done anything like this before, but I said, you know what, let's give it a try. I'm going to see what happens. And I decided to just start making lists. I made a list of things that I thought we needed. And I decided to start kind of poking people to see if I could get other people

interested in it. And I started submitting some patches myself. And then as this went on and things actually started to happen, I thought, hey, maybe people out there in the community might wanna know what's going on. So I decided to start a blog about it after several other people urged me to and well, we've pretty much been going from there for about, I would say since like January of this year. And there's no stopping me.

I'm gonna keep going forever. Good. And now you're 37 editions in. I know. Each one whets my appetite for the next release of Plasma. That's really like the net effect it's having here is this. Isn't it amazing stuff? Me too.

When I'm putting this together, I'm always thinking, holy crap, this is amazing stuff. Yeah. I just cannot believe how amazing KDE contributors are. I mean, for example, right before we started doing this interview, I got an email notification about a patch that somebody submitted to fix a bug that was filed yesterday. So, you know, so that's that kind of thing is amazing. And I feel like I'm seeing more and more of this as more people get interested and excited. And it's just an awesome thing to see.

Now, are you still pretty heavily involved with the Kubuntu project as well? Yes, I am. Kubuntu is actually my home distro. It's the one that I'm talking to you on right now. And I personally feel that Kubuntu is a really great choice for KDE made sort of accessible for the average person. Rolling release distros are fantastic. And rolling release distros are, in many ways, the best way to get all of these usability and productivity upgrades that I blog about all the time.

But for ordinary, regular people who just kind of want to get stuff done, I think nothing beats the stability of a discrete release distro. And I think it's really important that Kubuntu provide that experience with a really high level of polish and integration so that it can kind of become the de facto KDE distro, in my opinion. I think that's fantastic. You know, when I think about deploying Plasma at like a workplace and office environment, I think Kubuntu is so... Here in our studio, for example, all of our production machines are on Kubuntu 18.04. And it has just been –

Oh, that's wonderful. That's fantastic news. Rock solid. Rock solid. Best – really, it's the best distro I think we've ever used in production. Now, on my personal machine, I've got to admit, I roll KDE Neon. But that's just because of – really, it's your fault, Nate, because you get me excited about stuff. So that's actually totally on you.

Now – You know, sometimes I actually tempt myself to use it. There are many times that I feel like, oh, man, I want to actually use some of these improvements myself, particularly to the input stack, which has gotten a lot of improvements in Plasma 5.13 and 5.14. And I'm stuck on 5.12, and I'm kind of chomping at the bit to do it. Right, right. You know, I kind of feel like if I don't eat my own dog food, it's a pretty shameful thing, so I'm going to definitely keep doing it. Yeah, yeah, there's always ways, like VMs and Docker containers and whatnot. And in fact, that's what I do.

I actually do all my development inside a KDE Neon VM with everything built from source. And I think KDE Neon is basically a near perfect development tool in that respect. It gives you great isolation from your host system and it lets you make all kinds of changes without worrying about blowing stuff up. You get everything that's KDE related pretty much rolling. So, I mean, it's just great. I love developing in KDE Neon. That's fantastic. I love hearing that. That's so great. And that means there's room for all types of KDE plasma powered distros or meta distros. Don't call it a distro. Now, I noticed you in the bio, I was reading your Patreon page, which I want to talk about in a moment, a former Apple engineer. Does that change, you think, the look you have when you come in and look at the Plasma desktop? Did that

influence the way you view visibility as a former Apple engineer? In fact, that was the core reason. In addition to being a former Apple engineer, I had been an Apple fan for pretty much my entire life, starting at age six. And I think that the work that I did at Apple gave me a really great opportunity to see how really focused development teams work to produce software that's of high technical quality, but also exceptionally high usability. Apple has that reputation in the world. And I feel like for the most part, it's justified. And when I left Apple in late 2016, I decided that I just wanted to start working on Linux full time. And I kind of bounced around for a little bit. I started with GNOME. What was that draw? What was the draw to Linux there? I mean, especially coming from a

long history with Apple, what drew you to Linux there, do you think? Well, that's a really good question. There were a number of things. I started to feel like open source, just as a movement, made a lot of sense for me. And I started to feel like the inherent problem with closed source software is that regular people don't have much of an impact on it unless they're on the inside. And a lot of my friends and family had gotten in the habit of telling me their favorite Apple bugs and, you know, file this bug for me, see if you can get this fixed. And for a while, I was actually doing that. And I felt like that was really good.

But after a while, I started to take a longer view. And it was like, well, what is this like for people who don't have a person on the inside? You know, how are they going to get their pet issues resolved? Because despite the fact that Apple products offer exceptionally high usability, nothing's perfect, right? And people are going to find bugs in any software. And if you don't happen to have somebody on the inside, you don't really have that many options when it comes to closed source software. And I just, the more I learned about the open source movement, the more I just sort of liked the idea of participatory software development and an open attitude and the opportunity for anybody to contribute. And not only to contribute, but to go through a pipeline of starting as a user and then becoming a connected user who's engaging with developers and then eventually morphing into a developer themselves. And that was exactly my path. When I started with all this stuff, I mean, I'll be the first to tell

you I was not a programmer and I'm still not that much of a programmer today. I'm a build and release engineer by trade, so I can do some scripting, but being able to learn how to do software development from the KDE community has been absolutely tremendous. And, you know, I feel like if somebody as dumb as I can do it, anybody can do it, really. Well, I mean, it seems like you have gotten your hands on all kinds of different apps. It looks like, I mean, just reading over some of your profile, there's a lot of areas you've touched now recently. And I think that's where the Patreon thing is sort of a fascinating opportunity for open source developers. I'd like to get your thoughts on this because it seems to me – so right now you've got, as we record, 81 patrons. Proudly, I'm saying I'm one of them.

And I think it's a great idea because if that were to grow, you could proportionately dedicate more of your time to learning what you need to learn, to help more, to triage more bugs, to touch more projects. So what are your thoughts on using Patreon as a funding platform? Even in 2018, where there seems to be some Patreon fatigue, people are not as willing as they might have been back in 2017 or 2016, but it still seems like a great opportunity. I don't have a lot of experience with the cultural side, so it's news to me that people have Patreon fatigue. So, you know, if that's the case, that's a little bit unfortunate.

But I feel like this can be a really great supplement. But ultimately, that's all it really can be is as a supplement. Software development is an incredibly challenging and difficult field. And out there in the private sector, people routinely make six figures doing this. And I mean, I think, you know, with some private sector experience, we've all known people who bring home fat paychecks without actually doing all that much. And you compare and contrast that with so many people in the open source community who are just being absolute rock stars and getting paid nothing or very little for it.

So I think Patreon is fantastic. And I also have several other ones set up. I've got LiberaPay. I've got PayPal as well. But I think the best funding model is what essentially the CREDA people are doing. They have a foundation, the CREDA Foundation, and they aggregate donations, and they actually use this to formally pay developers with actual real salaries. I believe the Linux Mint people have done this too. And in my personal opinion, this is really the only payment and funding model that makes sustainable long-term sense, which is that if you want developers, you're pretty much going to need to pay developers. And if you're not going to pay developers, what you're going to get are interested, passionate volunteers, which are fantastic, but it's a hit and miss proposition.

And even the most interested volunteer will eventually go through stages of their life where they're not going to have time for it anymore if they're not paid. I mean, now that I've been doing this for a little while, not anywhere near as long as many people in the KDE community, I've noticed that people who are really passionate at first often just drop out after a couple of months, not because they're bored of it or they don't want to do it anymore, but because they got married, they started having kids, they got a new job that requires them to commute, and they can't be in front of a computer all day, that sort of thing. And those are the generous scenarios. There can be other reasons, too. That's right. Exactly. Many other reasons, you know, and so relying on volunteers is an inherently fragile proposition and it relies on having existing members being willing to

constantly train and retrain new people as they come in and leave and come in and leave. I think if there was a dedicated core of paid developers, you could see that they develop a real culture of a sort of self-sustaining culture where other people help out as well. And the knowledge becomes distributed and you don't see so much churn. That sort of happens right now in the KDE community with Blue Systems. Blue Systems sponsors a whole bunch of Plasma developers. They're all paid, they work, they do it all the time. And that's one of the reasons why the quality of Plasma is now so high and constantly getting better. And these guys aren't just working on Plasma, they're working on frameworks, they're working on apps, they're doing all kinds of stuff. So I think that's the

sort of payment paid development model that we need to try to make happen everywhere else. It does take some admin, but I could completely see what you're saying, how it's more sustainable. And it's more like a real job to the developers involved. It takes a large funding initiative and a foundation to admin all of that. But if major projects like KDE seem like a good size project for something like that, and others as well, That's a fascinating idea. So Patreon in a way is sort of not a stop gap, but it's an intermediary solution where direct consumers of the product can still try to help fund the development until something, a larger solution comes along, perhaps. That's exactly how I see it. You just

phrased it perfectly with about a thousand times fewer words than I did. it. You just phrased it perfectly with about a thousand times fewer words than I did. Well, Nate, I really appreciate all of the work you've been doing. I will link to your blog, if folks haven't seen that yet, linuxunplugged.com slash 268. We'll have a link in there to Nate's usability posts. Also, you can go to planet.kde.org if you want to see some of the other developers' posts to their blogs. It's a great aggregator of all of that stuff. And we will have a link to Nate's Patreon, which is ngram, right? It's ngram, patreon.com. That's right.

Yeah, if you want to just go there directly, but we'll have a link in the show notes. Nate, thanks so much for coming on the show and stay in touch. I'd love to chat again in the future. Thanks so much, Chris. It was fantastic. I had a great time. He had some really interesting ideas around how to fund long-term, sustainably large open source projects. And I thought that was a particularly new and good insight on a potential solution there.

And, you know, it's sort of interesting timing, too, because it's lining up right with a new venture that's launching that I want to tell you about in just a moment. But I just want to say thank you to Nate. He also is stopped by the live show. So he's chatting away with folks in the chat room, which is awesome. Just so great. It's incredibly exciting to have someone like that, you know, with the commercial experience he's got with Apple coming into the open source community and bashing on something like KDE. You know, that's fantastic.

Yeah. on something like KDE, you know, that's fantastic. Yeah, and I think the posts do a lot for encouraging people to check out and stay tuned into what Plasma Desktop is doing. Just, you know, in a sort of, I don't know, it's not like a tease, but it's a bit of, you get, you're able to appreciate all of the work that's going into it. That's probably what I'm trying to say.

Like we were just talking with Cassidy about how so much stuff is going into Juno that he's struggling to even write up all to outline it all. But this way, this weekly approach that Nate's taken gives you the space to sort of enjoy some of the finer, smaller little things like new icons and when you right-click in the menu and things that are nice to know about but would otherwise get lost in a much larger noise around a new release and so it it helps build appreciation for all of the work that these people are contributing to a project and it creates hype and i think it's in part responsible for as wimpy puts it the ascendancy we're seeing of the plasma desktop right now and it's got people like us going on a podcast and talking about it,

which amplifies the hype a bit. And we're using it in production. We're using it in real life. Yeah, it seems like it's a good position, right? They've got a great community going. They've got active users. And their software these days is just great. I keep getting a steady stream of one or two, one or two, at least a week.

And this is, you know, months after we did our plasma episode of people saying, all right, I'm trying it. I'm taking the plasma challenge and I'm really liking it. I'm still getting that. I'm still using it over here. I did have, I've been using it for about three months every day. I had my first lockup today. But other than that, it's been fantastic.

Really, really fantastic. It could have made it one more day and then that wouldn't have gone on out there. Yeah, you know, that's how it always works. Well, it's better than none. Yeah, well, I suppose so. It does last a little bit longer. So the conversation there at the end was kind of about funding and what's a sustainable approach for somebody who's making a nice contribution, doesn't even need to be their primary revenue source. Of course, that'd be great.

But, you know, just as like a secondary to really justify the time investment. Well, a new startup called Tidelift hopes to help these developers achieve just that and get paid with a business model the company compares to Netflix. So Tidelift is a startup right now. It just got $30 million in funding, at least I think. And the idea is that a company pays a subscription to Tidelift, which then takes a cut and then distributes the remainder to open source projects. And the subscribers will, in exchange, get assurances that the software will be properly maintained, quote unquote. So customers pay Tidelift. And then their pitch here is developers can focus on code instead of sales and marketing, getting their customers pay Tidelift and then their pitch here is developers can focus on code instead of sales and marketing, getting their income from Tidelift. And then when a customer

signs up for Tidelift, they go analyze your code base, go see all the neat open source projects you're using, and then bill you based on funding each of those projects. Right. Do they contribute any of that work back upstream or is it, you know, you pay for it, you will then commit to maintaining the software for X period of time and to also keep Tidelift and our customers informed of major changes. And then Tidelift goes back to the subscriber, the customer, and says, all right, here's the pledge by the open source developer to maintain these super important projects that you depend on. That's kind of the idea, right? Have they got any track history in how much they're paying developers for their commitment to these projects?

Yeah, that has been the very thing that Wired tried to get from them, and they didn't really have much to offer there. They've made a couple of blog posts about developers that they've got brought in on the team, but they haven't really been clear on the amount of money that developers get. They say, so, okay, here's the numbers. They got 15 million in venture capital. Sorry, it's not 30. And they announced last week that they set aside a million of that 15 million for new developers that join the program, and the developers will be paid at least $10,000 over a two-year period, at least. That's all they've said publicly in terms of funding. $5,000 a year.

Yeah. Between hundreds of people. It's not even a cup of tea. Here's the other big problem with this. That's just called a consultancy company. And of course, I can see why investors would put money into it. If they acquired the developer base, of course, they will be a powerful consulting company because there is a brain shortage of skilled developers. So eventually they would be able. And what are the conditions for the developers to be tied into them? Because otherwise I don't see them making money long term anyways. It seems to me this is a middleman in a business negotiation with open source developers

that doesn't need to be there and if companies are serious about engaging with a project that they're dependent on just go and contact it's very easy to find the uh the key people behind a project or a foundation or whatever just Just find those people, approach them directly, and that's probably the best way to invest your money behind a technology that you're dependent upon. And more foundations on the open source side would help, I think. I think the other thing to consider here is the pitch. This is what raises a red flag for me.

Netflix for open source, that just makes my skin crawl. What does that even mean? Well, that's a pitch for investors. That's not for the developer side of things. Come on. I wonder how Netflix feel about their brand being associated with this as well. We shall see. I thought that was interesting, though.

It's another crack at this. And I don't know. I think Nate's idea is a better one. And then Wimpy's right. Then companies could just interface with that foundation, and there would be an established process for that. It does seem is a better one. And then Wimpy's right. Then companies could just interface with that foundation and there would be an established process for that. It does seem like the better solution. So let's talk about something new that's getting some attention,

a Linux distro that's optimized for the Windows subsystem for Linux. It's based on Debian. And the core developer, Hayden, is joining us today. Hayden, welcome to Linux Unplugged. Hi, thanks for having me. Hello, Hayden. It's good to have you here. So you had a chance to chat with Joe Resington, my co-host and Wes's co-host, who's been filming for me on Linux Action News, to answer a few questions so that way Joe could accurately cover the story in LAN. And he recorded

that chat so we could refer to it for notes. And I thought it was so great that I would just play it here. It's a bit rough, but it was a great deep dive into why WLinux has been created and why the Windows subsystem for Linux might need a distro that's built specifically for it. Thanks for joining me, Hayden. Thanks for having me, Joe. So you're here to talk about WLinux. What exactly is that then? me, Joe. So you're here to talk about WLinux. What exactly is that then? WLinux is a Debian derivative that is designed to function on Windows subsystem for Linux. It is a distro

that is customized for that environment. And so this is based on Debian then? Yes, this is a derivative of Debian Stretch with a handful of packages from the Debian testing branch that have been brought down, mostly Python and other dev tools that have a more updated version in the testing branch that are fairly stable. And so what makes this different from the version of Debian that's already available in the subsystem? So a free distro of Debian is already available. subsystem? So a free distro Debian is already available. And if you're first, you know, getting into WSL, which is short for Windows Subsystem for Linux, you might want to give it a shot. But WLinux has a lot of customizations and additional packages that make working in this kind of hybrid environment a little bit easier between Windows and between Linux at the same time.

So one of the big differences is no systemd then. Why is that? Well, part of that is because systemd doesn't work well on Windows Subsystem for Linux. So it was kind of a no-brainer to pull out. I'm a big fan of the Devuin project and appreciate in it freedom from Unix philosophy standpoint. But mostly it was just taking up space on WLinux. So we took it out. SysV is there. If you install some programs that have a hard system D dependency. Lib systemd will come in. But it was mostly just a matter of size. Okay, but why didn't you just go with dev1 then? Because isn't that just Debian with no systemd? I did do testing with dev1. It added some more complexity to the build process with no apparent user, no improvement to the user.

And kind of being a Unix-y guy, believing in Unix principles, I didn't see needlessly interjecting complexity for something that we can actually just do on the Deian from the debian sources ourselves also um you know we do know that people might want to bring in packages uh you know from upstream uh gnome and kde to run those apps on windows and if we had a hard block on systemd um they wouldn't be able to run the most recent version of those. Okay, that makes sense. So talking of GUI stuff, I mean, the subsystem is not supposed to be for GUI applications, but you've made it easy to run Google Chrome and Visual

Studio and other GUI stuff. What's all that about? So this isn't an officially supported feature of WSL, but it is available. And there's a few tweaks we do to make that work with an external X client. There's a few, there's a couple of free ones available. My favorite is X410. It's available in the Microsoft store for $5. And we just made some tweaks so that the, for example, the libGL is rendered on the Windows side where you can get some more acceleration out of those apps. And it's very functional. In fact, there are people, although I don't recommend it, who do boot WSL into a full-fledged XFCE, KDE, or GNOME desktop.

You know, sometimes in a separate window, like you would a virtual machine, uh, and sometimes right next to their windows, uh, apps on the same screen. People are doing interesting things with this. I heard about Flatpaks up and running on WSL this week. Um, presumably they work reasonably well. And what about Snaps as well? So far, Snaps are a little bit of a challenge. Flatpaks are working better. So we'll

see where that goes. Snap has some dependencies that we have to work through, see if we can't make those work. I thought the chat that Joe and Hayden had was fantastic. In fact, it went on for quite a bit, but I wanted to take a moment and bring Hayden in live since he's here and chat about WLinux a bit with him. And I wanted to maybe double down before we've gone too far on some of the issues with systemd under the Windows subsystem for Linux. I have never really spent

much time with the Windows subsystem for Linux, Hayden, but looking at bug reports, it seems that the WSL has its own process that it wants to run at PID1. Is that one of the issues with SystemD on WSL? The issue with SystemD on WSL is that it's not meant at this point to run servers. It's not meant to run ongoing processes that you access from the background. That may change in the future as Microsoft just announced Windows Server 2019 with WSL as a feature. So that may change and we may see more growth on that use of WSL to run Linux processes. We'll see where it goes. I found it really interesting,

some of the work you've put into running the graphical applications like Chrome. I think I just recently saw, I can't remember where, you were talking about fixing an issue with Chromium from running it from WLinux, which I think is great because that's been an area I've been really curious. If I was ever to play around with WSL, I would love to do that. In regards to the Snap issue you ran into, I invite you to chat with Mr. Martin Wimpress here in our virtual lug. He could probably help with any issues there because he's working on that at Canonical.

I guess probably Snap must be dependent on something that's not in WLinux now? Yeah, well, the main thing is the dependency on systemd support, but we do have a shim that can work around that, which is how we delivered Snap support in 14.04, which was built upon Upstart. So, Hayden, I'm Martin Wimpress. I work on the Snapgraph team. Find me on Twitter at M underscore WinPress,

and I'll be happy to help. Absolutely. I will do that because I want to work with you on that. Flatpaks as well. Yeah, we're testing an issue right now with Google Chrome, the latest version, 69. We're working through some LibGL acceleration issues. We have an open issue on the GitHub, and we're troubleshooting that.

As far as we know, it affects most of the WSL distros available. But, you know, we're the ones actually doing the investigation, you know, pulling logs, you know, seeing how we can fix this for WSL users. Yeah, it seems like that could be an interesting impact that your distribution could have is by really focusing in on this like no other distribution is, you're going to push up against these edge cases that could drive future improvements to the subsystem. So I think that's great. And I wonder if it's also going to be revenue sustainable potentially because I noticed you're selling it for $19.99.

It's on sale right now for $9.99. And that's in the Microsoft Store, which is, of course, the way you get it into the subsystem for Linux. Do you think this might be at some point revenue sustainable? So that's our long-term goal. And we've talked a lot about different sustainability models for open source projects in this episode. And I think what we need to realize is that people have been paying for Linux on a commercial and enterprise level for a long time. So what I see is a democratization of that process with Krita, with elementary OS, uh, with other projects, uh, not so much the one you mentioned in the interim, but, uh, where the users, uh, the open source users themselves are being entrepreneurial, uh, you know, taking ownership of these, pushing them along and commercializing them so that you can push these developments out to other people.

Another good example of this is CodeWeavers, who's worked on Crossover and have worked with Valve on Proton. So there's always an angle sometimes to commercialize these, and we give back. an angle sometimes to commercialize these, and we give back. One of the examples is we already identified a package that's just not available anywhere as an app in an app repo, and we're hosting it. So you're welcome to use our app repo in any existing Debian or Ubuntu distro to install that package. Debian or Ubuntu distro to install that package. Ah, I noticed also that you mentioned one of the features is faster patching than other upstream distributions for subsystem specific bugs, and that you're specifically monitoring

subsystem related CVEs. Is that delivered through that repo as well, those fixes? So the fixes aren't delivered through the repo. The repo is currently simply delivering packages and package updates to our custom packages. We are in the process of rolling out a new onboarding experience for users. We had so many requests for packages, it simply was not possible to ship everyone's editor and everyone's tools in one. So we're going to create a front end where you can go through when you launch WLinux the first time and specify, I'm a Python developer.

Install all my Python. I want to use NeoVim or code. And we will take care of that for you. And then we're going to build auto updating to that going forward so that everyone, regardless of whether you're building it off the GitHub or downloading it from the store, will always have the latest availability to the features. So there won't be a gap there. So one example of this was a problem in a relatively, so WSL doesn't implement IP tables yet or very well at all. And there was a package in Ubuntu that broke called EB tables and it's an

obscure package and it did something in the install script that broke apt and basically prevented everyone on Ubuntu 18.04 on WSL from updating for a few weeks. Well, we worked, you know, and I give credit to Ubuntu for, you know, helping us get that fixed, but it took a few weeks, you know, and we could have pushed that out in a day or two. Yeah, and again, I pushed that out in a day or two. Yeah. And again, I think that's going to be one of the great side contributions or direct contributions really of the project existing is pushing on those edges. Well, Hayden, thank you for taking the time to chat with me and for chatting with Joe to get the coverage right

on Linux Action News. I'll put links to a bunch of information about the project as well as the Microsoft Store link because like I mentioned, it is on sale for 50% off right now if you want to play around with it. I know we have Windows users out there in that audience, Wes. Yeah, you're stuck. Maybe you don't want to run a VM. You don't need a VM. Hey, now you've got a custom tailored distribution. Yeah, yeah. So good work, Hayden. I hope

it keeps up. I know some people have been a little like, what is this? But I say, good on you. And I'm curious to see where the project goes in the future. Any hints where we're going after this? What's next? So we're looking right now, after we get the onboarding experience done, we're looking at internationalization. WLinux is unexpectedly popular in China and Japan.

It took me completely by surprise. So we're adding in foreign language support as quickly as possible. And then we're going to be adding ARM64 support to support Windows on ARM devices. I love it. That's interesting. Well, do keep us informed on how it goes. Keep us up to date. I'd love to follow the progress of things. And you're welcome back any time to update us on any major developments. Thank you so much. I do appreciate it.

You bet. Well, let's do something that we should do a lot more of, to be honest with you, and that's read folks' emails and follow-ups and all those kinds of things. How about I take this first one here, Wes? Because John writes in to address you, so I'll read this one here.

He says, I took note of Wes mentioning a possible root on ZFS setup for his new ThinkPad. I would definitely recommend giving this a shot. I don't know what distro you plan on using, but in the past few years, I've been using nothing but root on ZFS on Arch with binary updates. And he points you out to, there's an Arch ZFS GitHub page.

But he says, I also wasn't quite happy with the way Linux supported boot environments, which is a nice feature when you use ZFS. It really is, yeah. And you could really dig this, I think. So he wrote his own. He calls it ZD environment, or Z environment. ZD. And he

links to it in the email, and he says it removes the worry of kernel updates or system changes causing issues since you can create a new boot environment and you make your changes there. If you happen to give root on ZFS a shot and are looking for a boot environment client, please take a look at it and best of luck with your new computer. John, that is awesome.

I'm also really just thinking I'm going to double down on this Root on ZFS, maybe also encrypt it. So we'll have more to talk about in that arena soon. That could be fun. So Omkar writes in and he says he loves the show. In fact, he also seems to be a TechSnap fan because he says, I find Wes knowledgeable,

and the way he explains stuff on TechSnap and Linux Unplugged is fascinating. Oh, thanks. That's really nice. Much appreciated. He says he's got a few questions for us. What kind of hardware setup have you configured for Proxmox VE,

and which version of Proxmox? Oh, man. That's a you question, Chris. It's a three-year-old install, or maybe even four years old now, so that would be... I couldn't tell you what version that is. That's a pretty old install. Maybe I could

bring in our IT guy. Maybe he could tell me, Hey, Brown Bear, could you come in here for a second? I got a question for you. Hello. Hey, Brown Bear, do you remember what version of Proxmox we're running on that i7 rig in the garage? Well, I've got good and bad news for you. Which would you like first? Well, okay.

Let's start with the bad news because I think I know where you're going. The bad news is you're no longer running Proxmox on the i7 in the garage. Ooh. Yeah, I know. It's running now CentOS, isn't it? It is, yeah. Alan and I set up CentOS and then we put, we're actually, we installed LibvirtD

and set it up just like Red Hat would, as a, you know, it's kind of like a KVM hypervisor kind of a deal. And that's actually been working really good, I can tell you the truth. It's been really solid. Proxmox was great, though. But things, times have really changed. The options have changed.

I don't know if you've ever been a Proxmox user, Noah, but I know that you're big on the CentOS virtualization stuff. Yeah, I have. In fact, I've gotten crap on the Ask Noah show for not giving enough good juju to Proxmox because it really is a fantastic project. They really do some really great work. And if you want an enterprise-grade virtualization system

that has a really fantastic UI, man, Proxmox is really the way to go. It's just when you start looking into things like I want the ability to, you know, have a BSD server that's on here. And then when I go back to Grand Forks, I want that BSD server to fall over to my virtualized host in Grand Forks. And to come back here, I want it to go back over here and those kinds of things. It's really nice to have something a little bit meatier. That does sound really cool. things. It's really nice to have something a little bit meteor. That is nice. That does sound really cool. Okay, so

Omkar's next question was, what droplet specs do you guys recommend for a GitLab and a PeerTube instance? Well, those are two tough questions. So GitLab, you'd probably be okay with the $20 a month or less droplets. GitLab's not going to be. It just depends on how busy you are. It depends on what you do with it, I guess. There's a lot of extended

features, but if you're just dumping some commits to it, probably not. Yeah. Now lot of extended features, but if you're just dumping some commits to it, probably not. Now, PeerTube, here's what's tricky about PeerTube, is if you think about it, you're kind of re-implementing what YouTube does, and there is a transcoding process that PeerTube goes through when you upload a video. And the speed of your droplet

is going to directly impact how long it takes for those videos to be available on PeerTube. If you don't care, if you just want some place to throw up videos and you don't care if it takes an hour for the video to be available, then you really could get away with a $5 a month droplet because you're not using much CPU because the decoding is happening on client side for the video and the bandwidth is still fast regardless of which machine you get.

And then he asked a question that I'm going to punt to you, Wes. He asked if either of us have ever tried out Traffic, which I think is how you pronounce it. It's spelled T-R-A-E-F-I-K. And for those of you that don't know, I didn't. It is what they call a cloud-native edge router. It's a reverse proxy and a load balancer that's easy, dynamic, automatic, and open source.

So it seems like it's pretty cool. Have you ever heard of Trafficfic or Traffic, Wes? Yeah, so you can front, basically, if you're running a bunch of stuff on your fancy Kubernetes clusters, you can put this in front of it and it will auto-discover. It'll talk to Kubernetes or Swarm

or some other technologies. Oh, cool. Find what services you're running, connect to them, and integrate with metrics. It can do HTTPS at the front. Also ties in with some of those open tracing systems. You can get logs and sort of analyze how a request flows through all of your services.

Yeah, it's a nice looking web UI to manage this. This is, I see. I see. So it ties in basically with the different orchestration APIs and then automatically configures the reverse proxy, attaches statistics using open source projects, and you're good to go. Yeah, so there's a lot to like there,

but if you're not using some sort of orchestration software, it's probably not necessarily what you use. Maybe something just like Envoy might work or just Nginx even. There's a lot of solutions that won't have all those fancy features, but if you don't need them,

if you just have two apps that are running on two different droplets, well, you're probably fine. Yeah, so there you go. Thanks for the emails. We got a bunch of others. Some of them, which were eight or nine paragraphs long, read them all, but can't cover them all on the show. They've got to be a certain length.

And we would love to get your feedback, your questions, your suggestions for the show. LinuxUnplugged.com slash contact. And links for anything we talked about today, LinuxUnplugged slash 268. So, Mr. Noah Chalaya, what are you doing in this part of the woods? Are you just coming out here to hang out for the

barbecue? Dude, I am always here for just the food. I mean, I say nice things like I came here for your stunning personality and I wanted to hang out and stuff, but really, it's the barbecue. How many minutes was it that you walked into the door? I think it couldn't have been more than five minutes. We were already packing up to go to lunch when you got here.

You know, I really had a... I mean, here's the thing. We can talk about food. I'm not saying that we can't. But I feel like you had an agenda before I ever came here, before I ever set foot in the door. I just – I felt like it was your premise to take me to Indian food, and that decision process had been made before I ever entered the premises. Yeah, you're welcome. Is that your way of saying thank you? Yeah, that's a long-winded way of saying thank you.

So what are you doing over there? Getting ready for the Ask Noah program? I am, yeah. I'm sure you guys have heard. Google has been making some weird decisions, forcing people to log in, and we're going to ask the audience if they think this enhances their experience using Chrome, or if this time Google has gone too far. Have they taken privacy and thrown it out the windows, Google tends to do.

We got a number of other exciting stories that we're going to get to. Project Triton has adopted a code of conduct, so we're going to talk about that, a little bit of follow-up to last week's episode. Chrome OS 71 has launched. Now Linux apps are available on non-Pixel devices, also the ability to share files with Linux. Yeah, it's absolutely fantastic. So we're going to talk about all those things,

and then towards the end of the show, if we have time, because this is like the least important thing in the world, but Microsoft has launched the ability to have multi-user virtualization on Windows. We're going to explain how that can be leveraged in an all-Linux environment. If you're just virtualizing Windows, if you want to be Linux on the hardware, Windows in a virtual environment, how this is actually going to benefit your Linux environment. It's for people who like to mess with computers.

That's right, Leo. Well, I look forward to it. Yeah, you know, I say, I just was making this joke earlier in the show. I look at the stats and, you know, I'd say there's 30%, well, yeah, it might be around 30%-ish, somewhere in that ballpark of the people

that download the show are on Windows. Oh, really? Yeah, it's somewhere in that area. And we love you too, guys. Yeah, so we appreciate it, you know. Maybe they're just checking us out before they make the big switch, you know. They're looking for all the guides and the know-how.

Yeah, I look at myself. Which distros are hip. I look at myself as Jesus Chris, Jesus, you know. Wow. He wouldn't go into the places where everybody had become a follower. Wow, like the Beatles. There's not a lot to be done there. So I need to be in the homes of the sinners so I can bring them to the wonderful world.

This is really just falling apart as you keep going. My Lord and Savior, Lenny Storvold. Jeez. This really explains a few things, actually. This worldview now explains a few things. I understand you a little better now. There are memes going on. Apparently, I said one time on an episode, I said that Linux was a religion.

I was just going to give in. And so I just started calling myself the Pope. And now it's become memes. Well, yeah, Pope might not be too far off. I've been memefied. Well, now that I'm sure we've enraged somebody, we should probably wrap this thing up. Go get more of Martin Wimpress over at the Ubuntu podcast. Go get more of them.

Great show. Not just me, but also Mark. Everyone remember Mark is not Martin or Alan or Popey or Wimpy. He's Mark. Mark is one-third of the Ubuntu podcast. Yeah. He visits on occasion. So that's why he doesn't get probably the mention as often

because you're right here front and center. But it is a great trio at that. And also, go check out Cassidy James' work. The Medium post that we chatted about earlier will be linked in the show notes. And stick around. We've got the post show coming up. And then right after that on the live stream

is the Ask Nowhere program, which will be live taking your calls. We might talk about some of the same topics, but you actually get to chime in via the phone. You don't even have to bother installing Mumble. Wes Payne's available all the time online. You can just hunt him down. He's at Wes Payne.

That's right. And then he's on the TechSnap program. New show just launched. Actually, I kid. It's not a new show. It's been around for a while. But it's coming out new all the program. New show just launched. Actually, I kid. It's not a new show. It's been around for a while. But it's coming out new all the time. It's pretty great.

And we just had John the Nice Guy on the show last week. You should go check it out. He's got some hot tips. Yeah. Yeah, especially if you want to dive a little deeper into interplanetary file system. All of that is over at TechSnap.Systems. All right. That's all.

Yeah. What? You had John Spriggs on? Yeah. Yeah. That's nice, right? Wow. I Spriggs on? Yeah. Yeah. It's nice, right? Wow. I've been away too long. John is not just John the Nice Guy Spriggs.

John is a super nice guy. He is. He really is. Yeah, yeah. He's the organizer of OpCamp. I know. I know. I've been wanting to have him on the show forever. So go check it out. TechSnap.Systems. All our links, LinuxUnplugged.com. I'm at ChrisLES. Thanks for joining us.

See you next Tuesday. I wanted to pick Wimpy's brain. I don't know exactly what the right way to ask this question is, Wimpy, but is there a way to tell what the difference would be in terms of patches or additions to Ubuntu's 18.10 GNOME versus, say, stock GNOME? So is there a way to actually detail patch sets that might be be better for performance i'm trying to think about how to review the new distro releases and performance of the gnome desktop is something i think i kind of want to take a look at but i'm not quite sure how to tell differences in terms of patch sets that have been applied for performance yes so if you want to see what patches are being applied to what is, you know,

the upstream version of GNOME Shell and the patches that are being carried in Ubuntu, what will become 18.10, then yes, you can see all of that in Launchpad because all of that Debian packaging that happens, happens in Launchpad and you can see those patches directly. happens in launchpad and you can see those those patches directly what i would say is that i know that uh daniel van vote who's uh part of the ubuntu desktop team all of the patches that we're carrying as a as vendor patches right now have been submitted upstream are and are in various stages of review and will eventually percolate into you know the upstream gnome project but whilst they're being negotiated we're carrying the patches that he's been working on