LINUX Unplugged - Episode 150: War of the Packages | LUP 150

Episode Date: June 22, 2016

We have a spirited discussion from both sides of the universal packaging issue, take a quick look at maru OS that turns a Nexus phone into your desktop, get the inside scoop on the recent Mycroft update & the new Solus release. Plus much more!

The internet should be simple and safe. We strongly believe that having nothing to hide shouldn't mean having to show it all. This is Silent Keys. A beautiful and secure keyboard that defends your privacy. Open source and plug and play, Silent Keys protects you from viruses, ransomware, mass surveillance and identity theft stop governments employers search engines and credit agencies from seeing what you do online browse without fear and protect who you are what you do and what you like block annoying tracking ads and encrypt all your traffic so you can serve the web privately using? Every year, millions of people experience banking fraud

and lose their savings. Be safe from key loggers when accessing online banking, Bitcoin, and social media. We just stopped you from typing your password

at all. Stream your favorite TV shows from all over the world with no restrictions. What's going on here? Because a simple key logger

can record all your keystrokes and defeat the best security, we created a tamper-proof keyboard. What? An aluminum shell that can't be opened without breaking. It's really easy to use.

Are we serious? Plug it in, press a button, and boot. Silent keys include Satya, a modern and secure desktop. Ah. The best way to protect an operating system against viruses is to make it read-only and have it load from scratch at every use. Is that a cute little dot net I saw there?

This makes it immune to insecure systems and leaves no trace. Just like a CD-ROM! Silent Keys features a secure update switch with free and unlimited updates. To protect your privacy and prevent data leaks, Silent Keys automatically

encrypts your web connection using Tor or VPN. Anonymous browsing on Silent Keys safely counters fake Wi-Fi, ISP snooping, mass surveillance, and hackers. All ads are blocked too.

Secure messaging and end-to-end encryption for your email comes out of the box. That is the slowest email encryption I've ever seen. Okay, if this is legit, if they can make this something that's really simple for the average person, it's not so bad. But look at this, Wes.

616 backers. They were only going for $55,000. They've raised $105,000. So they're scratching some itch. But, yeah, it's essentially a read-only live GNOME desktop with a decent GNOME theme. Yeah. I mean, it doesn't look bad. You know what, Wes? Why is it

a keyboard as well? Well, because then it's the tamper-proof device where it freaks out when it gets broken. You know what, guys? This is what it tells us. It doesn't matter about the technical whiz-bang, because, you know, I mean, this is or it is what it is. It may or may not be hacker-proof, right? Right.

What it does tell you is that we are in the wrong business. This is Linux Unplugged, episode 150 for June 21st, 2016. Oh, welcome to Linux Unplugged, your weekly Linux talk show that is awash in distribution releases. My name is Chris. My name is Wes. How about it, Wes? Like today, this week at least, it giveth. We have some great releases to talk about today that are super fresh.

We've got a Mycroft update that I'm really excited to talk about for multiple reasons. We'll tell you guys all about that. Some cool hardware coming for Ubuntu Phone. A free great game for Linux this week. DockerCon's going on right now, and I've snussed out. That's sussed and sniffed. I snussed out some of the best stuff. And then coming up last but not least, we're going to look at Maroo? Maroo OS.

Yeah. The desktop on your Android phone. Yeah. They claim they've got convergence, and they've based it on Debian, and it's here today. He uses Waylander, some shenanigans. There's some crazy far-out future technology in that thing. I don't know, Wes. You have to dig into that.

Wes has got a load on his phone. He'll give us his take. How it runs on his Nexus 5. And then towards the end, flat packs. They're officially here. Yeah, you know, universal software installation. That's another universal answer. Another standard flat pack.

Now we got snaps, flat packs, app images. They're all competing to essentially do the same thing. But we've already talked about that aspect of it. But since we started that conversation, people have gone to war. It is a fast-moving story. Holy. I mean, I don't want to drop an S-bomb at the top of the show. We were all kind of bitter about Linux packaging or something.

I don't know. I mean, it has gone to full-fledged war over this. I mean, it has gone to full-fledged war over this. And, boy, somehow because I talked on two shows, I have wandered in the middle of this. And I am just getting slammed on all sides. And we're going to just talk a little bit about it today in the show. And we'll see. Depending on how far this beer gets me, I may get a little ranty, but I doubt it.

Because we just got one great beer here from Fremont Brewing just a bit down the road, just a piece from where you probably work. I mean, not too far, all things considered, coming right out of a local Seattle brewery because beer matters, it says. This is the Summer Ale, their American Pale Ale. This looks like a nice treat, Wes. And I got to say, too, it's got a good color to it. I don't know. Do I have the camera on? Yeah.

You guys probably can't make it out yet. It's a little dark, but it has a really good color to it. Oh, I didn't open mine. Gee. Come on, Chris. Let's see. Let's document the hazard and mess right now. There we go. So we will be –

We've come prepared. We'll be enjoying that during the show today. So we've got a lot to get into. Let's bring in our virtual lug. Time-appropriate greetings, Mumbaroom. Hey-o. Hello. Hello.

I really caught – look at that. I caught them off guard. Did you see that? Yes, you did. All right. So, guys, I already asked you on the pre-show, so I already know the answer. We haven't really got a chance to kick the tires yet. I'm going to hold my powder dry for the segment, but Fedora 24 is officially out today.

And there's some really nice things in this one. Mainly the one that I'm really looking forward to is a GNOME 3.20 is in this. Also, talked about just a bit briefly there, Flatpak support is also included in this version of Fedora. So, they have sort of a official unofficial upgrade guide, I'm going to call

it, that we will link in the show notes. Interesting. Yeah, where you can do an update. And I have a, just put it actually recently back into production a bit. I'll tell you about that later in the show. A Fedora cloud installation that I have running a couple of different containers and whatnots that I'm going to do an upgrade from 23 to 24 as part of my. I'll be curious to see what you think about that.

Yeah, I'll give you my thoughts. Part of my review in Linux Action Show on Sunday, I'll talk about Fedora 24. But congrats to Fedora for getting a release out, shipping it with GNOME 3.20 in there, and I think it's got kernel 4.5 or some shenanigans like that. And, of course, looking pretty solid with the Flatpak support. So that's Fedora. We don't have a lot to say about it because we have so much to cover. Why don't we jump into a little Mycroft news because Ryan is joining us today. Wonderful.

Welcome back to Unplug joining us today. Wonderful. Welcome back to Unplugged, Mr. Ryan. Hey, it's good to be back. Thank you, sir. Thank you. Thank you. Thank you. Thank you. Thank you very much for being here.

We really do appreciate it. So tell me about version.7, responsive listening, third-party skills, a new voice, the big update, and better emotions on the front panel too. Like there's a lot to cover here. But what the hell is responsive listening? Can we start there? Responsive listening is a name I came up with. It's good.

Because I didn't have a good name for how to describe it. But for the longest time, and you'll see this in the blog post that I made, we played around with this idea of passive listening. And it worked all right. We played around with this idea of passive listening, and it worked all right. And that was the deal where you can say an entire sentence, put Mycroft's name anywhere, and it will trigger and then take action. But what we found was that lots of people were already interacting with the device in the way that Google now has trained them and Siri and the Echo. And so they were saying, you know, hey Mycroft, and then saying whatever they were going to say. How were they supposed to be saying it beforehand?

Well, it's not that they couldn't say it that way, it's that they could insert the name anywhere that they wanted to. So you could say like, the actual Mycroft part of the sentence the word Mycroft might be four or five words into the sentence,

and the system would have to be able to wake up and scan back to find what was said before the name, right? Correct, and it was pretty cool technology, but people didn't understand the latency that came from that, and it was only about a second, but there are now expectations, especially because people are comparing us against the Echo. And so what we've done is we've gone ahead and put responsive listening as the default. So you say, hey, Mycroft, and then your query, and it comes back pretty much instantaneously. And then what we've done is we've gone ahead and begun working on making the existing passive listening is what we called it,

an option that people can just go in and check and they can take. Oh, that's neat. Responsive listening. That's a good name. And I guess I assumed because of that's how, like you just said, all of the other assistants that are sort of on the market right now operate the other way. I just assumed that would be the way Mycroft would operate.

I didn't even – there's so many little aspects to this. And Ryan, it got me thinking about sort of an advantage that Mycroft might have being open source where the code can be publicly audited. People have a higher level of trust in it than say maybe the Alexa or the Google stuff, because you don't really fully understand their complete motivation behind the product. You have a pretty good idea, though. And with Mycroft, everything being kind of out in the public, it doesn't quite matter what the motivation is, because you can still see what the end result is literally for yourself. And so that gives Mycroft the opportunity to maybe be a little creepier than, say, Google

or Amazon can be. Like it could eventually give Mycroft room to say, hey, would you optionally like that we listen to everything all the time? I mean, there's like a lot of different things that people might be willing to let a fully open source assistant do that they wouldn't want a commercial version to do. Am I making sense here? Yeah, but you're operating in here under the assumption that the echo isn't already listening all the time well yeah that's true

but the uh but yeah i and what we're hoping is that eventually i think that's the way that a lot of these um little smart assistants will be regardless of whether or not they're in a dedicated device in your house or on your phone or wherever the question is about trust and about whether or not because i think eventually when we talked about this at one of our meetings uh one of our you know all hands meetings somebody brought the fact that eventually they thought that that the benefits would outweigh the what a lot of people saw as security concerns. I'm not saying that that's our how we feel. But, you know, it's the idea that it could listen and be smart enough to, you know, actually offer you up suggestions or

take notes for you that you might forget to take yourself and things like that. Those benefits might eventually outweigh people's privacy concerns. But the worry is, you know, who, what company are you going to want to trust your data with? Are you going to want to trust your data with a company? Being an open source company and open source project, we get to be a little more transparent with users and let them know, you know, here's kind of how things are. You can audit the code and make sure that it's only listening when you want it to and not have to do some sort of crazy gymnastics to figure out what exactly is happening with your data and whether or not you just bugged your house. What's their, yeah, and what exactly is it that they're storing and why are they storing it? You know, that's, you make a good point because I think, unfortunately,

and I want to make sure that sinks in because I always get pushback on this. Unfortunately, I think you are right in the sense that the convenience that a lot of these different platforms offer, be it iOS or Android or the entire Google platform, it is, for people who just need a little extra help in life, it is actually very tempting. And it seems to me that if they're willing to make that concession to use Google, like I have recently, kind of started going all in with Google because Google Photos is just – they just happened to nail something that is exceedingly important to me as a father of three young, adorable children.

I would – Google Photos is one of two backup – off-site backups, plus I have a local storage of all my photos. And the ability to search that and say, you know, grandma and hugs and Dylan and get all that mapped out is super useful. And then I started using Google Fit because I'm starting to exercise more and getting all of that information. Being able to use different devices for different tasks and have all of them report back to Google Fit as a central input to get all of this. So now I have like one spot to go to get a lot of different information from weight, water,

sleep, quality walks. All of this information is in one dashboard for me now. Yeah. With charts and all this kind of crap that is very nice to have. And of course I know I, I am fully intellectually aware of what that means they are doing in terms of tracking and the information they're building about me.

And if I had a better solution for some of this stuff, I immediately would probably go somewhere else. So I think even if Mycroft gets to the market in a year or two where they're competitive at some of these levels, be it the artificial assistant where you can prompt it and ask it questions or whatever it is, I think if people are already like me who are very aware of the problem are still willing to make the concession, as soon as something like this comes along, it's going to be like I'll be an early adopter. We've got to go all in on this.

I think it's from people like me to just the average person are going to be really responsive to a product like that. That's a big one. person are going to be really responsive to a product like that. That's a big one. I think that what we have to do as a community is consider what we want the future of this stuff to look like. It's so crazy how locked down this specific technology is with these providers that aren't Mycroft.

is uh with these providers that aren't aren't my craft you know they're they're much more locked down than than let's say like android or which is slowly moving that direction but there's there's with when it comes to these intelligent personal assistants that's really really a black box and what's crazy is they're quickly becoming one of the most intimate things you interact with yeah and uh and so what i think is that uh we should have an answer and you know that's what we're kind of trying to do with my crop one thing i wanted to to get back to is is uh before we talk about all the crazy stuff going on in this space i wanted to talk a little bit more about the release because you talked about the third couple things that were in there yeah one thing i want to know more about is third party uh skills and yeah

some things like that too so yeah absolutely yeah that was i'm actually i'm actually that was the end of end of discussion on that i was kind of clear as to go more back to the 0.7 release yeah so uh for those with developer kits we we got a lot of new face animations um it's a lot more responsive we're looking at trying to present data on the face on that little led array so like whenever you ask for the weather it gives a really nice clean you know interface for seeing weather, as well as other things like that. Then third-party skills, which you talked about. This is the beginning of something we've been planning for a long time, was a way to install third-party skills.

Right now, you kind of have to literally drop the skills into a directory for them to be loaded when Mycroft starts. into a directory for them to be loaded at when mycroft starts but we're building a uh kind of skill management system that talks to a repo uh tool that's awesome people have submitted and so the the idea is that you can just add you know a roku skill and like various other oh man ryan oh man the the thing is there's already third-party skills out there and so like various other. Oh man, Ryan. Oh man. The, the thing is there's already third party skills out there. And so we need to get a list of those so that people can start playing with them now.

And that's why we kind of accelerated this piece because we keep getting people saying like, Hey, I wrote a skill where you accepted in the core and we have to walk this line of, yeah, that's a really cool skill, but I don't think most people use that.

But I know that there are people who will use it. So let's see about getting it, making it installable by people. For instance, there's this one called Diagnostic Skill. And you can ask it all sorts of questions about your system that Mycroft's running on. So, you know, what's the CPU usage? What's the RAM usage? Things like that. Then we also saw, oh boy, there's lots of skills.

So there's, Microp's a little bit more personal now. You can say, how are you doing? Hello. And he'll talk to you. I'm not sure if it's still in there, but for a while, when you asked him how he's doing, sometimes he would tell you a little bit about what's going on. He would say, like, hey, I'm doing good. I'm busy.

And then he would list off a few processes that were running on the market. That's awesome. Yeah, that's really funny. I secretly use Arch Linux. Yeah, and we got a new voice, too. That's cool. In fact, we have an example. Popey posted this.

I think he's pretty excited by this. This is the new... 74, Lawrence, Kansas has few clouds and is currently 76 degrees. Yeah, I'll start from the very beginning. With a high of 78 and a low of 74, Lawrence, Kansas has few clouds and is currently 76 degrees certainly a little robotic ryan but that's that's really cool yeah i mean it definitely sounds like poppy it sure does uh that's amazing yeah it's pretty awesome we're continuing to make improvements to that voice but uh but frankly uh now i'm very accustomed to it. Yeah. And Popey keeps talking to me every day.

Fake Popey. I kind of makes me want to be able to address the unit as fake Popey. You know? Fopey. Yeah. Fopey. Okay. And also, I saw there's a Debian repo.

Oh, I'm excited for that. Yeah. That's great. So what does this mean? Is it just now a couple of app sources away and now I can install all the Minecraft packages? So, yeah, we have an app repo. Aaron, our systems administrator, he's been on here before, ran into a problem that he talked to the packaging community about. So I think that repo will probably be shared out um i would guess

tomorrow yeah i don't know if you got the memo ryan uh we're all about snaps now so not repos no flat packs oh flat packs yeah uh no i i know i'm very very familiar with snaps very very familiar now with flat pack um we. We're playing around with those. We have a Snaps repo up where people can watch as we package that up. And at this point, we're just asking for help on all of our packaging. We want to package in all of the different standards

for every repo, as I know that we've been talking a lot about lately with all these different universal packaging. All these universal solve-all-problems standards, which we're going to talk more about when we get to Flatpaks in a bit. But we're just trying to package MyCroft and get it to as many different people as possible. And that's kind of our focus. So if people in the Juber Broadcasting community

want to help us do that, that would be great. And then finally, we've got... If you look at the bottom of the blog post, which I hope they'll post in the show notes. We will. There's a few other projects that we're working on we're working on a community website um that has a lot of information for those who are interested in contributing to the open source side of

of my crop and then uh and then we're working on uh just all this stuff that you can see in github as far as like we have an issue tracker and we've just been throwing all of our ideas in there. And so we're really getting tons and tons of feedback and lots of people are submitting front ends for Mycroft. We now have a Plasmoid extension, an Electron front end, an Android app. Oh, cool. Oh, man. an Electron front end, an Android app. Oh, cool. Oh, man.

So it's coming in really, really fast, and so fast, in fact, that we're finding it really, really difficult to answer all the questions as people charge ahead, but that's a great problem to have. You know what would be a killer feature in comparison to the Echo and probably even the Google Assistant and Siri is if somehow all of these –

if I had it on an Android device on my desktop and in the hardware device – I know we've kind of mentioned this before, but if somehow all of these were connected in some way where if one of them learned something or could – like if I could say send a telegram to Wes to one of them, all of them could have that functionality without me having to install an app on each one or install a third-party feature on each one. If at the end of it, maybe it's a couple of years down the road, these different Mycroft endpoints all felt like a shared brain, truly got us closer to Jarvis. Because right now, all the other AI assistants, they're each their own individual thing.

If you're using Siri and you have an iPad and an iPhone and now you're going to have it on the Mac OS, you literally – you could say, hey, telephone. And it will – if they're all in the same proximity, they'll all light up and they'll all try to answer the question like a bunch of dummies. And each one of them doesn't have any idea that there's another instance of it running. Now, I don't know what the solution is there, but if there was some way to make it feel like one cohesive intelligent system, Ryan, is that possible? Is that within Mycroft's future? Yeah, it is. It was one of our day one features we're working on, and we're still working to make it so that the units are aware of each other. They live in what we call a domain and so they have they have uh oh they're supposed to eventually have awareness of each other and uh if for

instance two of them pick up the same thing uh go through some very very quick election process to say you're you're in the you're getting the better stream you're in a better position to answer this question, you go ahead and do it. And the other thing too is that we already have shared configuration across instances. I have four Mycroft instances. They have the same settings. And that's all done through our backend, which is currently named Cerberus.

However, a lot of our back-end stuff is going to be getting a really, really big update soon. And so you'll see maybe that name go away and a true back-end that really feels comfortable to use and is really a great user experience comes to the forefront. We're super, super busy, but everything's going great. And we're trying to build as many relationships right now with folks who are packaging software and folks who are interested in working on front ends for Mycroft to just let it permeate

to as many different places as possible and to make that process of getting started with Mycroft to just let it permeate to as many different places as possible and to make that process of getting started with Mycroft that much easier. Yeah. It's great you guys are working on snaps. Do you know if packages started to land in the AUR? That'd be great too. I don't know. The thing is, every time I check out the AUR for a project that I work on, someone has

already beat me to it by the time I go to package stuff. the AUR for a project that I work on, someone has already beat me to it by the time I go to package stuff. So it wouldn't surprise me if there was already my craft in the AUR and I'm running, I've got an arch machine in front of me right now and I haven't taken the time to check, but it wouldn't, it wouldn't surprise me at all. I've been really, I've been working really, really hard on trying to help people get the information they need to finish their front ends. I'm super interested in the gnOME extension in the Android app. Yeah, that is looking really good. So, Ryan, my last question.

Are you ever going to make it out to Washington? Because we've got to Mycroft up the JV studio like nuts. Yeah, I actually was just contemplating this the other day. Now that we've really got so much done i'm interested to come out and not only hook up the jb studio but also lady jupiter yeah um with with mycroft we've got it controlling some things like the nest and um some early smart thing stuff and philips hue stuff so maybe we'll also have to see if we can figure out some smart things to, to pimp your place out with.

Yeah, that'd be really, that, that is, that's going to be, that's going to be so ridiculous, you know, because I actually,

a couple of months down the road, I'm going to go on a road trip to go to a, a developer summit. And I think I'm going to do like a Lady Jupiter technology rover log video for that I probably should have just said that out loud because now I have to do it, damn it Yeah, I'm wondering if we can put like a

what's your internet connection like when you're in Lady Jupiter? It depends on where I'm at very much so anywhere from a megabit you know 700 kilobits a second to up to 30, 40 megabits depending on location. Well, the thing that we could do that I think would be really cool is to take like a NUC and hide it somewhere in Lady Jupiter. Yeah.

And then connect it to like this complex microphone system throughout the thing. Oh, man. So it's just one brain all the time. You know what? I actually have a spot. There is a spot underneath one of my seats that there's an AC outlet underneath the seat, and it's an open space, and I already have a Linux-powered access point and router down there.

So I was just planning to kind of expand that out, and I got a spot in there for a NUC already. If we got a dedicated – somebody who's been working on an Electron app, it would be cool if you full-screened that Electron app on, like, a display. Oh, totally. Totally, totally. And then, like, that was it. Lady Joops, you know what would be really cool is Lady Joops has an outside outdoor television.

And so when we go to summits and stuff. Lady Joops, play the live stream. Put that up. Yeah, put that up on the TV, on the outdoor TV. That would be really crazy cool. And actually, it's all the videos routed to a switcher. So it would be no problem. We just send the feed.

It's easy peasy. So people want to know in the chat room, is there anywhere that they can find the Android app yet? Is that out yet? I would tell everybody to go to the docs.mycroft.ai. And there's a little contributing option there where you can see all the ways ways to get involved and if you join either our irc or our slack channel we've been talking about it a lot lately um the community

member who's working on it i'm not entirely certain that he's popped it into the repo yet but i'm i'll ask him because i know he's collaborating with somebody else in the community so they have to be collaborating somehow. So maybe it exists and I just haven't gone and visited it yet. So it sounds like it's an open source Android app? Yeah, and it runs the full stack on the device. Wow! Mycroft is running on the device. Holy crap!

I want to try that. It's not hitting an API or anything. We got a pull request a while back that took... So this is a core on a Raspberry Pi, so bear this in mind. It took it down from using 100% of one core on a Raspberry Pi to less than 1%, just based on improvements that we made to the listener. And so since we did that, we've been loading it on many, many small devices and it's been flying.

And a lot of phones, a lot of Android phones are actually more powerful than our Raspberry Pi. So from what I saw in the little demo video, and I'll try and share it with you because the guy recorded his son talking to his phone and the Android app answering.

Wow. It seemed to be running just fine. That's amazing. So what you're telling me is you pulled a Richard and you Pied Pipered this thing and you went from like 99, you just cut it. That's an incredible reduction from like maxing out a CPU core to 1% CPU usage. I can just have my craft everywhere with me. That's an incredible reduction from like maxing out a CPU core to 1% CPU usage.

I can just have Minecraft everywhere with me. That's what you get for – It wasn't – it was a community member. He now is an intern, but he wasn't at the time. Isn't that awesome? Isn't that awesome? And so it wasn't – we can't really take – I mean, I guess we can take credit for it now that he works for us. But at the time, we couldn't.

That's a huge benefit of being open source. This is the power of open source. Man. Oh, that feels good. Well, Ryan, thank you for the update. And if you are able to stick around, we will be getting into Flatpak and Snaps in a little bit, and I'd love to hear some of your thoughts on that. While you're just mentioning – yeah, go ahead. I can say something that gets me yelled at.

Good. That's what this show is for. I got to tell you. Don't worry. Nobody's listening. Nobody's going to comment afterwards. Certainly not. Don't have to worry about that at all, Ryan.

There's not even a subreddit for it. No, there's not even a subreddit. We were just talking about the Android phones. OnePlus 3, the new OnePlus phone, which is just getting complicated. Very complicated. OnePlus 3 looks like it's going to become an unofficial Ubuntu phone, though. There is some interest out there by the ubports.com

folks, a group of independent developers. They're going to port Ubuntu Touch to the OnePlus 3. This would make the Ubuntu Touch available on a very nice, modern, capable, fast, and very well-reviewed smartphone. I have been considering a new smartphone.

Maybe the OnePlus 3 should be my future. Maybe, because it's getting really good reviews, too. I mean, I want convergence, right? Like, who doesn't? Well, we'll find out how Maru works on your phone in just a little bit. Speaking of phones, let's also just take a moment right here. Guess who we're going to talk about, Wes. Do you have a guess?

Ting. Oh, it's Ting. Aha! That's right. Go over to Ting and check them out. Ting is a little different than anybody else. They don't work like your regular cell phone provider. Ting keeps rates simple.

We don't make you pick a plan. Instead, you just use your phone as you normally would. How much you use determines how much you pay each month. All right. You can have as many devices as you want on one account. Okay. That's good, because when you use more, you pay less per minute, message, or megabyte of data.

Your usage, plus $6 per active device on your account, plus taxes, is your monthly bill. Simple. That's what we mean when we say mobile. That makes sense. Yeah, it's your minutes, your message, your megabytes. That's your ting math. And then it's $6 for the line.

And then whatever Uncle Sam's cut is, right there, you see, you just got, look, Wes? Boom. That's your Ting math. And then it's $6 for the line. And then whatever Uncle Sam's cut is right there. You see, you just got – Wes? Boom. That is what you get right there. And I got to tell you, Wes, I was sort of all in on Ting just because of the savings when I switched over. I was like, okay, well, that seems obvious. I want to pay less every month. Yeah.

And I have Wi-Fi at home. I have Wi-Fi here at the Studs. I have Wi-Fi at family members' house. I have Wi-Fi at family members' house. I have Wi-Fi at, you know, there's Wi-Fi on the bus these days. Frigging McDonald's has frigging Wi-Fi. So I was like, this is going to be great because I'll be able to game the system. So that was sort of what brought me over to Ting. But then what kept me at Ting, the first, like, bit is I kept having these revelations about how they were actually a really cool company and not just like a bullshit like we're hip.

And so we do blog posts. No, it's like actually there's really cool people and they really get it and they really are not playing a gimmick. They're not playing a game. They don't have to wear a leather jacket and a pink shirt to get you to be convinced to switch over. They just make a really simple mobile service. Every device they sell is unlocked and you own it. The customer support, you get to talk to a real human being. The control panel allows you to control all aspects of your phone.

Oh, that control panel is awesome. It really is super good. Their store, you can buy SIM cards, you can buy feature phones, you can get the latest phones, you can bring your own. And they have GSM and a CDMA network for you to pick from. If you're a small business, this is a no-brainer. Really seriously, independent surveys have been conducted. And not only, by the way, speaking of that, not only does Ting totally kick ass in consumer reports for their category of provider, I got to tell you what, nobody touches them when it comes to

small business plans. If you have like 10 devices and you bring them over, nobody's going to touch Ting. You go 10 devices and more and it's nuts because it's $6 for each line. A lot of places, the minimum is $15 a line, something like that. And in a business account, it could be even more. And that's even when they cut in some big scale deals. $6 a line. One line, 50 lines, $6. If you want to give somebody an emergency phone, it's $6 a line. One line, 50 lines. $6. If you want to give somebody an emergency phone,

it's $6. CDMA and GSM, which means you can repurpose a lot of your existing phones that still work perfectly fine, or you can find a great phone off of eBay or buy one right from the Ting store. Go get it from Google Play. Great controls, great customer service. It's a really good service. I think everything about Ting is top. And tops. And you know, they really do actually have a good blog. They constantly are posting interesting stuff there too. You can check that out. If you start by going to Linux.Ting.com

you support this show. You can try out their savings calculator and just use that as a litmus test. Does this work for you? It might not. And that's why they have the savings calculator. You can just figure that out right there. You don't need to waste your time or Ting's time. Just try the savings calculator. That won't take long.

Linux.ting.com. And then if it does look like it's going to work, like for me, I end up saving like $2,000 every two years. Linux.ting.com. Take $25 off your first device. Or if you bring a device, you'll get service credit, which will probably pay for more than your first month. Linux.ting.com. Thank you, Ting. All right.

So before we go any further, let's take a little break. Refresh. With a free game. Oh, is that a Linux game, I see? Yes, it is. And it's a good one, too. I already own this game. I got it on sale during the sale.

But it's Limbo, which the thing that really stands out about Limbo is it's sort of a platformer. So you have to like the side-scrolling platformers. But it's not like a lot of colors and a lot of whiz-bang. Oh, it's beautiful. It's dark and it's moody and it's mysterious. And it uses lighting and shadows in a way that really no other game has ever before or has since. And it's exceedingly beautiful and also very reasonable to play even like on an Intel chipset.

And it works great with the controller. It's a great SteamOS game. It's been out since 2011. So it's not like it's brand new, but it's one of these overwhelmingly positively reviewed games. It really is. People love it and it's freaking free right now. Can't be free?

Yeah, if you're listening to this show for the next couple of days, it is free right now and you can find it in the show notes. So I think that's just really super cool. Look at that. Did you see that huge creepy spider? That thing is just... Man, that always creeps my kids out.

They love this game because it's slightly creepy. So they love that. All right. So I believe Mr. Ikey is in the Mumble Room. And that just happens to work out great because a brand new release of Solus is out, 1.2. Don't call it 1.2, though. Call it Shannon. And it is looking good.

I've heard people talk. People have been talking about the boot speed this morning. This is what the chat room is saying this morning as I was setting up. They say, Wes, that new Solo S sure boots fast. I've never seen a Distro boot that fast before. That's what the chat room is saying. So, Ike, you're doing something right. What's new and notable in the new shiny release?

Ike, I don't hear you. Did he leave? No. Ike, that's your cue. Somebody bring Ike back. Somebody bring Ike back. I don't even see it. Hey, Wes.

Where's Ike? I don't know. Jeez Louise. But I do want to give this 1.2 Shannon a try. Yeah, Wes! Where's Ike? I don't know. Jeez Louise. But I do want to give this 1.2 Shannon a try. Yeah, it does look nice, although it's such a good setup to have it, to have a fail like that. I just feel like, no, I don't want to talk about it.

I mean, that was like, I think that was my best setup of the show. What do you think? You think that was my best setup of the show? You've peaked, Chris. Yeah, I think I just, I think I peaked, and then the payoff wasn't there. Chris. Yeah. I think I peaked, and then the payoff wasn't there. Hi. Sorry.

Man of the hour. So how's audio working on Solas, Ikey? Audio is working fine. It's great. It's great. So it turns out, like, I spent ages earlier setting up this mouse because it's got all these buttons. I'm going to set that mouse button up there and I'm going to use that one for Mumble. And I'm not going to forget which one it is.

There's three of them, not two. So I'm here now. No worries. Actually gave me a chance to have a little fun. So what is the new shiny in the big release here? So, I mean mean where do we start um i mean one yeah we're 20 days late and you know i can only apologize for that to people um oh so you guys are uh copying fedora uh yeah i mean they we basically turned out the same time

of them and before half-life 3 so i'd say say, you know, we're doing all right there. Yeah, it's quite the difference. It's good. Yeah, I mean, it was later than I anticipated, but I mean, I've not been too well myself lately, but, you know, we made it. It represents over three months of work, over 150 bugs fixed, a completely new replacement Steam runtime and the Linux Steam integration, with roughly 160 of the libraries that we've enabled for Steam now being optimized as well. Yeah, we got the epic boot time, obviously.

It's been a massive, massive release. So tell me a little bit about this Steam stuff because I got a little wind of this earlier. And I don't really – I'm not hip with the Steam stuff. But here's my rough understanding of it, Ike, is if you got all the – if you got all the missing – less problems than using the prepackaged Ubuntu runtime libraries that are based on an old version of Ubuntu that come with Steam itself. And so I guess some people prefer to use the runtime library. Some people prefer to use native libraries. And you guys have tried to solve this problem.

Am I tracking? Yeah, I mean, that's pretty much it. I mean, I know that you yourself like uh go between fedora and art and that so you've probably seen some of the steam issues yourself on ubuntu you know i mean people tend to get it easy on ubuntu up until about 1604 and um i was gonna say 1610 that's not happened yet but these these libraries are like from ubuntu 1204 so i mean the linux terms they are crusty um these are really really old libraries that are built specifically for ubuntu i mean we're talking about libraries that just don't exist in other distributions

and they've got different they've got different soul names you've got like uh curl g and utls which are specific ubuntu and debian things so what we decided to do is like okay one this is nonsense right yeah we build solace and we optimize it for our users it's like okay so if we're gonna have to do some hacks here to get it work well we're gonna make sure we do it properly so we took what would constitute the full runtime and just started building everything and disabled Steam from its own runtime. So we've got something called Linux Steam integration

which basically replaces user bin Steam. So this shim will run instead of that and we made sure it doesn't leak or anything like that and it replaces itself with the Steam process afterwards. So when we disable their runtime and we use ours, you actually get

nice integration, so you'll actually see things like font config is actually working, you get all the right font rendering coming through. I don't know if you guys have seen it on Archer or on other distros, but you'll see the fonts are ever so slightly out of line, the buttons along the top of Steam, they're all kind of drunk as

well. So it fixes those issues. The tray will be natively themed as well which it's those little touches which really really help but the the main thing really was the performance those were advantages that came afterwards um you can disable our runtime and forcibly use steams because ours is new there's there's going to be things we haven't tested yet. We were responsible and tested 100 titles in the first 24 hours. It was our duty, right? But that also sounds like a good idea just from you could see some game developer be, oh, yeah, well, that distribution's doing their own weird thing with Steam.

So that gives the person a reset button to say, okay, well, I'm testing it with the native runtime and I'm still having the problem. Yeah, I mean, they can do either way. And there's even a third option if you like. So one of the games, well, there's two of them. Counter-Strike Go wasn't working for me. I don't remember if you guys saw the update a couple of weeks ago.

Basically, something went wrong when they suddenly switched it to use 64-bit only on CSGO. It still shipped with the 32-bit version, so I thought, aha. So I added a tweak in there, which, you know when you run things via the Linux 32 command line? It basically forces things into a 32-bit environment. So I added that support inside of LSI. So Steam will then be tricked into running only 32-bit environment. So I added that support inside of LSI. So Steam will then be tricked into running only 32-bit versions of the applications. So what that does is for CSGO, I'm able to use the native runtime in 32-bit mode.

And for GOAT Simulator, a very serious strategy game, you're able to run that in 32-bit with Steam's runtime because that uses some special NVIDIA physics libraries. You are doing the Lord's work there. That's great. It's, you know, somebody's got to do these things, right? Why not me? So that's all coming in the new update. And so along with that, do uh package it all up with a um

like an update mechanism or is it sort of a nuke and pave process uh no i mean people can update um so in terms of the packages they've already got installed in the system those will change right so whatever updates you got through there like uh security updates and things like that those are going to come through. Between Solus 1.1 and 1.2, we decided to change how we put certain packages together. New distributions are coming out and they're starting to use Budgie. And we thought, well, maybe those patches we've been doing

aren't quite the best approach. So we'll try and set an example, show people how you can actually ship a distribution or an operating system just to be pedantic and show them how you can actually have all the presets in place and how you do the configuration.

So we added a couple of packages. One of them is Solus Hardware Config, which takes care of certain mouse configurations like the RAT5 mouse and things like this. And Budgie Desktop Branding, which contains all of the live CD configuration certain mouse configurations like the the rat 5 mouse and things like this and budgie desktop branding which contains all of the live cd configuration and all the tweaks that were needed for gsense uh for basically the budgie experience we now offer so those are the those don't come through as an update they're basically going to have to be opt-in so the the package

selection is very much dependent on the ISO you downloaded. You can upgrade and install those things as well. But if you're already on 1.1, you'll just get the normal updates that come through. Of course, you're still going to get all the optimizations that come through for all the lower libraries as well. So you will still benefit. It just won't look exactly the same or won't have exactly the same software installed. It's easy to do that sounds like a lot of thought is going into every single really does thing about this uh release uh is there is there something that you work on that uh you think a lot about that you don't even you

assume probably people don't even notice like is there something in here that people miss like what is it like i always know when i'm when there's something i work on there's something that i just i just sweat the details on and then after i'm done i'm like did anybody actually even notice that uh yeah um well i mean it's going to be that that way for me for for a lot of things i mean you know yourself once you put something out like you say you sweat on something it's not so much when someone doesn't notice that as they notice something far more insignificant yes and that's the killer so i mean not as insignificant so we've had a bit of a ux update and we've actually been lucky enough um i mean i think you guys know yourselves but uh the the guy behind the arc theme um horst3180 we've worked directly with him for a long time now.

And with him, we actually just redesigned the Budgie desktop, and that's how Budgie Desktop 10 came around, like designed by the theme author. Wow. Yeah, so like it was literally called Arc Desktop for a long time. Yeah, okay. That was the code name of it. He's helped us a lot more with getting things together this time.

And we now have the icon theme as well. So within 40 minutes of it having a Git repo, it was the default in Solus. He'd be on the cards for a while. Sam Hewitt has also helped us a lot with UX design this time around. I mean, you all know Sam, right? Sam, I don't know. I probably know him? He – Sam – I don't know. I probably know him by a handle. I probably don't know him by Sam.

So Sam Hewitt, the guy behind the paper theme. Oh, okay. Yeah, okay. Yeah, that guy, right? That guy. Yeah. So he's helped us with designing the software center, giving like little hints and tips and fixes and mock-ups for that,

as well as the installer itself. So how has adoption been? Adoption? I haven't a clue. When did we release? Early yesterday? Well, I just mean as a whole. I don't mean one particular release. I mean, as a community, do you get a sense that there's growth there, that things are picking up? What is your sense? Yeah, I mean, definitely. There's a lot of people really now starting to get it um

i mean solace goes against the grain in so many ways and it's it's never easy to explain exactly in one sentence to people it's like you know this is what solace does and they're like well why not use ubuntu fedorabian, insert something here. It's like, well, because kill me if you want, but I don't think it works. We have to do some, there's so many things that we do that are so fundamentally different that we couldn't do as a downstream.

So explaining those things, but people are now realizing that. And you do see comments now, people are like, I can actually see why you did this from scratch. Like, it was designed to avoid the problems in the first place. It's not like we're somehow superior to or holier than other people. We've taken that experience, and I've taken my own experience in doing this as well, and folded it into something new. That's pretty cool.

And I think it always, after we chat, it always makes me want to download it. So I'll download the new release and give it an install. I'm looking forward to that. So I always like using something. There's a lot of thought and intention that goes into it. Yeah, you get a lot of passion. Well, Aiki, thanks for the update. Anything else you want to touch on?

Maybe that crazy mouse? I don't know. Okay. maybe that crazy mouse i don't know okay well i mean no because it only fires things up i you are you are welcome to stick around i know i think i saw you mentioned something on google plus you're getting lots of questions about snaps in solace so if you want to stick around for that discussion coming up in a few minutes you are welcome okay. Okay. Okay. I'll bite. Excellent. Just really quick, I just wanted to give a mention because it's happening right now as we record the show. DockerCon is going on, and they're integrating something that I'm hoping Wes can explain

to me what it is because I think it's kind of a big deal. Docker is integrating orchestration right into the core Docker engine with their new release. This is some of the big news that's coming out of DockerCon going on in Seattle. So, Wes, you've seen probably some of the shenanigans as a result of DockerCon. I've certainly seen a lot of people with really cute

little whales walking around, which, hey, that's not bad. So you're probably seeing all of the DockerCon traffic. So, container orchestration. If I'm understanding what that there means, Wes, now, you'll have to forgive me. I haven't been a sysadmin for a couple of years now, but that's another fancy way of saying, like, a framework to manage all of your containers and spin them up on different machines and shut them down and delete them and version them and standardize them and things like that, correct? Yeah.

and things like that, correct? Yeah, so it's like, as they say here, container orchestration is what is needed to transition from deploying containers individually on a single host to deploying complex multi-container apps on many machines. Yeah, okay. So yeah, as you're growing your cloud, maybe you're on other people's infrastructure, you're on your own infrastructure,

you need something that will help you take your single little Docker instances and actually scale them, make sure they fit into the roles, integrate with your configuration management. There's a lot of ways to do this, and now there's something built right into Docker. I can even see a scenario where you have four or five DigitalOcean droplets that you spin up containers on demand for, like I would just be playing with Fedora 23 today.

So to me, it seems like what Docker just announced is they're going to take a huge part of the Docker ecosystem and integrate it into the core functionality of Docker. Last time I was at DockerCon was last year around this time. It was LinuxCon DockerCon and the number one industry at DockerCon is container management

and orchestration of containers. They just announced they're integrating that into the core and I read about it. It's some badass stuff. It's like it allows you to do swarms of Docker containers with just a couple of commands and set up clustered Docker containers with a few more commands. And you don't really have to really worry about any of the, quote, unquote, you don't have to worry about it, quote, unquote, underlying technology. Right.

And they're building that into core Docker docker using just a docker command line i mean this is like a massive thing right so we'll probably see it you know rolling out and everywhere i haven't i haven't played with it yet so it'll be interesting i don't know you know how we'll see like will other people adopt it will be wrapped around by other systems like kubernetes or you know all that kind of right yeah that's a that's a good point so we'll just be watching and see what happens and uh i don't know that's probably all we really need to talk about here on this show, I imagine. But I'm sure there's lots of news.

I looked over the talks. There's a lot of interesting stuff. Microsoft's giving a talk about how the containers on Windows are going, if that's your thing. You know what I really don't like about the – so I've been following the DockerConf thing just to see if there's stuff for us to talk about. And so I saw the Microsoft talk. And you know what the really arrogant thing about the Microsoft talk was? Is Microsoft positioned their entire talk as, well, now that Microsoft's getting involved with Docker containers, now Docker's coming to the enterprise. Now it's production ready.

Now Docker's production ready. Meanwhile, those SOBs are running it in a virtualizer on their Windows platform. It's not even a true container. Like, they're such a bunch of posers, they're so desperate to make their Windows platform relevant that they're bending over backwards to make Linux containers run on their platform as easy and thinly as possible using their virtualizer. And then, they

have the sack to say, well, now Docker's arrived to the enterprise. I will say, the Max support is kind of neat, just that they're using Xhive, which is ported over from our friends at FreeBSD. That's pretty neat. Isn't it interesting to see one of the first things that Beehive does in a large production scale is bring Docker games to Mac? Oh, I'm sorry, guys.

Welcome to BSD Unplugged. You know, it's just one of those things. It just happens. Okay, so. Oh, man. happens. Okay, so oh, man. Yeah, you know, speaking of Beehive, it just

got Windows video support, so you can VNC into that thing. You don't have to do it over a serial console, yeah. That's a big usability win. Speaking of DigitalOcean, let's take a moment and talk about DigitalOcean. You got a container you want to deploy? Do it over at DigitalOcean. Now, we've been talking about Docker,

but Wes here, you do LXD containers in production on DigitalOcean. And what do you like about the LXD containers over the Docker containers? Well, you know, they're just a little more... I mean, Docker works well if I was making maybe like a big fleet of microservices and using orchestration tools

and that sort of thing. Docker may be something I would look to be using. LXD containers are nice because they really act a lot more like VMs, which is something I'm familiar with, and I get all of my regular Linux tools for admin, right? So I know how to use SystemD and Journal Control and all that kind of stuff to manage services, et cetera. It's all the same tools, and then I can just spin them up,

transfer them server to server. So it's simpler in some regards. Make copies, take snapshots. It's very simple. And if you don't need a huge enterprise-grade thing, you just want a few containers. Yeah, it's all... I basically set up a system, turn it into an image, I can then just update the image when I need to, I can transfer it, and then just

make as many images as I want. Are there LXD container images? Like, there are Docker? Like, could I go grab a pre-setup LXD container to be lazy, or does it pretty much build it from scratch each time? Well, so they are, like, actual image files, right? So, like, they're tarred up file systems. So they sit on the file system as an image.

Oh, okay. And then is there, like, a place you can find pre-built ones? Yeah, so they have, like, a repository, and then you can also, the nice part is that they all have a REST API, and then they'll all just, you can like, if you have your own, you can then share images with any two servers that you want to let talk to each other. It's also really easy to make. Like you pretty much just take any install that you want. You can trim it down, you know, uninstall packages or build it up minimally. And then the format's really easy.

It's like you make two tar files. One's metadata. One's the file system. You run like load it into an image and then you can use it. So what's process like of getting lxd running on a on a droplet on digital ocean well uh use like ubuntu 1604 when it's there by default and then oh so it's like type lxc yeah pretty much right type lxc and then it generates a client certificate and then you can get started and nice you can pretty much do like lxc launch Ubuntu, I think. I think I'm going to play with that for a bit because I was messing around with Docker, but I was on Fedora.

But I was messing around with Docker, and it did feel like it was way more than I needed really. So that's interesting that you say that. So go to DigitalOcean, and if you use our promo code DOUnplugged, D-O-unplugged, one word, lowercase, that will give us a little bit of support, a little bit of like, hey, yeah, that unplugged show is sending traffic our way. That keeps us going. But it also gives you a $10 credit. And you can try out like a droplet for a while. They have hourly pricing. So the droplet you may use may not need to be on all the time. And then that $5 – or they have a $5 a month if you want to just run it for a month or whatever.

Like they have a really flexible pricing structure. So the $10 credit could actually get you quite a bit of distance. And all their rigs have SSDs, so it's a nice, fast IO. They have data centers all over the world. They run Linux for the host, KVM for the virtualizer, 40 gigabit E connections into the hypervisors, tier one bandwidth at the data centers, and the best interface. I mean, and it's not even like a good interface for a web application.

It is a good interface for managing. Yeah. You know what was sweet is today when I was, I decided to install a whole bunch of crap on this Fedora droplet that I have. And it was really super, super nice to just go in there and have the HTML5 console pop up. And I was able to watch the console, you know, spit out information while I'm SSHed in. So my SSH session feels fast and native and just, you know, in fact, sometimes I even use MOSH. Oh, yeah, totally.

Yeah, which really makes it feel like I'm sitting right there. I even use Mosh. Oh, yeah, totally. Yeah, which really makes it feel like I'm sitting right there. But then to still have like the – to have in a separate window on my second monitor the actual console of the Droplet is just the best, dude. I love the digital ocean of grace. You know if you break anything, it's just right there. And it took four seconds to reassign a DNS host name to the Fedora. I was like, you know what?

I want to change the name of this Fedora machine. So I used hostctl, hostnamectl, whatever it is. Is that hostnamectl to set the host name to the Fedora. I was like, you know, I want to change the name of this Fedora machine. So I used host name CTL or whatever it is. Is that host name CTL to set the host name? And I set the machine name and the icon name. And then in DigitalOcean, I just copy and paste it just because over the interface, when you hover over the IP, it gives you the option to copy the IP, right? So boom, boom, pasted, updated the A record. And because the DNS was already

pointing to DigitalOcean's name servers, it was SSH'd immediately. And I was up and working. I'm installing, just experimenting. I'm hoping to eventually get a nice, secure, open-source location tracking program that we can have for family members.

And also during LinuxFest, we can have crew members install it if they want. You can be like, oh, they're still at the hotel. Oh, they're on their way. Yeah, exactly. But I don't want to use any hosted service, and so I wanted a rig to try these out on. And to be able to install these different packages and stuff, the faster you're going to access that HTML5 console, it's a great setup. You can take snapshots if you need to.

I did. If you're worried about breaking it, you're just like, peace of mind. Yep. DO unplugged to get the credit and support this show. So tell me about Maroo, which seems to promise real big, Wes. For those of you out there that aren't familiar with it, their tagline is, your phone is your PC. Now available for the Nexus 5. It's a new kind of computing experience.

We've talked about it in the past. It's based on Debian. And supposedly it's context-aware. You plug it into a big monitor, and boom, you get a big old Linux desktop. Is it true? In my experience so far, I mean, it hasn't been perfect, and I haven't tried it extensively, but I do have it running right here on my phone. So is it an app on Android that runs in a virtual environment?

So is that Android I'm seeing right now, or is that the phone you are? That's Android. Okay, this is that Android I'm seeing right now or is that the phone you are? That's Android. Okay, but that's Maru OS. So it looks just like what we're seeing in the video right now is exactly what I'm seeing. But this is... I have multi-ROM, so I just booted right into it. So this is Maru OS running Android?

I think it is Android. I thought it was Debian-based. It may just launch through Debian New Zealand. I'm not actually sure how it works. How do I initiate a different desktop experience? Unfortunately, I don't think we can without the right cable, which I do have. So if we want to set up a demo, I can film it and upload it. Yeah, I see what you're saying.

Yeah, because of course you'd have to plug it in. But you plug it in and it just XFCE desktop launches. You need like a Bluetooth keyboard and mouse, which I tested. There's some problems, and I did run into this, where the Bluetooth mouse is like a Bluetooth keyboard and mouse, which I tested. There's some problems, and I did run into this, where the Bluetooth mouse is like a little leggy. They suggest just turning it off and on or turning Bluetooth off and on. That did not fix it for me. I have not been able to fix it, but I have not tried that much. So it may already be fixed, or it may just be something I need to configure. But the keyboard worked perfectly. It was otherwise, it was like fast, it was responsive.

I've seen some videos of people using Blender, using GIMP. So it appears to be running Android 5.1. And it says Maru on the Nexus 5 is the version, version 0.2.3. Kernel version 3.4.0 GEAA8415. And so it's very much right now a pure Android experience. So it does mean you've now had to downgrade from Marshmallow to Android 5,

but in exchange, without a reboot, you plug in the HDMI cable to this son of a gun, and you're telling me it goes, you get a, it just pops right up. It instantly expands to an XFCE desktop because this is not an XFCE launcher on the phone right now.

Nope. Yeah, it just – it will just – And what happens on the phone screen? It stays. You can use the phone too. You can use both. So it is running as a – so Android remains running on the phone. Yep. So it must be an Android app then.

And so you usually just want it to go like fall asleep. What do they call that? Doze? Yeah. And then so it to go fall asleep. What do they call that? Doze? Yeah. And then so it'll keep everything on. Otherwise, if you explicitly lock it, it can sometimes trigger the screensaver in the other window. But otherwise, it really works very well. So in what scenario could this actually be practical?

Because here we are. You wanted to show it to me, and yet without the hardware, you can't actually do anything. So you have to have the cable. Everywhere. You have to have the SlimPort cable. Everywhere you go, you'd have to have somewhere to plug that into. So you also have to have an external monitor ready to go. And a Bluetooth keyboard and mouse. Those two are fairly easy

and I suppose you might have them anyway if you have a tablet or something. This is starting to seem extremely... I mean, now that you actually have to live with this and practice, all the dependencies to actually be able to use this as a desktop machine is really kind of making it limited.

Because at some point you're going to go, well, I could also just go get a Chromebook. You could just get a Chromebook, especially now that they're running the Android apps. Yeah. You know, and I haven't tried it that much, and you would, you know, and you are kind of surrendering, like, if you're not going to go to the trouble of having a computer and you're just using your phone, and you

have, like, you know, you bring your phone to work and then you plug your phone into your keyboard and you know and monitor then you're left with the computing power of the phone here's what i would rather have for me i would rather have a really easy image that allows me to run google's version of android so i can get my monthly security updates that's why i buy a nexus phone and then i would like something like multi-ROM or – but maybe like a fork of it or something that easily reboots me into another OS that can still access the Android data or something to that degree. Oh, yeah. Because I don't want to forfeit having modern Android. Now that you mention this, it's only like a couple of work case scenarios. At best, Wes, you could have a dock and a keyboard and a mouse at work and a dock and a keyboard and a mouse at home. And you could go between home and work.

It's not like you're going to go to a coffee shop and use it. You're not coming here to the studio and using it unless you put a dock here. At a certain point, the overhead of having tooted into a dedicated os that is designed for a desktop computer i don't know and then all i'm doing is using my phone all day which is like i have a lot more powerful computers i'd probably want huh i don't know but it is an interesting idea and i am better than you thought i mean the i mean xfce doesn't really need that much but it was working fine i haven't installed a lot. I haven't tried taxing it.

I'm going to try to play with it more this weekend. Anybody in the Mumble room have any questions for Wes's test drive with Maru OS? I guess, yeah, North Ranger, you were going to say, what's the point of having to reboot? Yeah, I mean, Android is running a Linux kernel, and is it really that much overhead

having the extra Android user land kind of shoved to the side if you're ultimately booting up an X server and running a desktop shell anyway? Probably not. It probably depends on how they do it. I'll have to do some more research to see what I can find out.

And it only supports the Nexus 5 at this point, which is kind of a bummer because... Yeah, I mean, I like it. It's still working for me, but it's definitely feeling old. Yeah. So does this change your anticipation at all one way or the other for a Ubuntu phone? So you mentioned getting the OnePlus 3 because, you know, you're thinking maybe upgrading in the future.

Yeah, I'm thinking about that, yeah. Does it... Now that you've tried this, has this chilled your your um opinion of convergence at all i mean i think i'm still waiting for like the use like you said i'm waiting for like the real use case in my life if there was like a really powerful device and i did have you know it's just not there's just a lot of things to bring around you can't carry a screen a keyboard and a mouse with you everywhere you and the cable everywhere you go one One use case I could maybe think of is like my wife, for instance. She's an artist, so she has a little Cintiq small screen that she carries around with her.

That might be something if you connect it to something like that where like you don't – you're already carrying around a screen for some reason maybe in that case. But that's pretty niche. N3, you had a question? Yeah. Is it running Mirror? Oh, yeah. yeah um is it running mirror like oh yeah what did we figure out what the what the linux display manager was because i don't i didn't see anything on their website about it there's not that much information on the website there is a google group for um like development yeah yeah i don't think they actually have the info but we will link to the page if you guys want to dig through it

uh and then i saw jm you had a question jm, B-B-B-B-B-Z. Go ahead. B-B-Z. Yeah, can you make phone calls and texts and things when you're in desktop mode when you're docked? I have not tried a phone call, but texting, I've been using other apps. That stuff all seems to work.

Yeah, so the full Android OS is working, huh? And so if people have more questions, too, like I'll be happy to film a little thing or film me using it or stress testing it if there's things people want me to try. Is it Google Apps on there not yet no it's pretty minimal it looks like here i'm reading i'm reading a thread right now on how to get the google apps on there though uh in fact it's interesting hmm it's kind of revealing on how they do this

does it feel kind of snappy when you're using it in desktop mode or does it feel like you're we're running off a phone? It really didn't feel very leggy. I mean, the mouse was like a little bit, but I was using the I pulled up a terminal, I was browsing the web, like that part wasn't leggy at all. Yeah, it is. There is, oh boy. You know, the

Motorola lap dock is really the best scenario. That's really what you need is something like that. And those are, you can still find them on eBay every now and then, but that would make that, where what you need is something like that you can still find them on eBay every now and then but that would make that where if you had something that was in your bag that was lightweight that extra battery LCD screen and a keyboard they're out there you should have the Motorola laptops

out there alright so any other notes on the Maru or any other questions no alright then we shall move forward because I do want to talk about the snaps and flat packs here for a second but before we go any further let's take a moment Any more questions? It's a huge part of Linux Academy and not just the core essentials and all the expert things around it. Yes, all that stuff. Yes, it's true. It's true. But also a lot of projects built around it.

So things like Ruby or things like Nginx or things that are tangentially related like the entire OpenStack setup or Amazon Web Services. This is all different aspects. And even to that end, now Linux Academy is introducing Microsoft Azure Essentials. How about that? Because you know what? Azure is a competitive cloud computing platform that runs a whole ton of Linux.

That's right. Yeah. And so they've launched a courseware on that, which I think is pretty awesome. And they've also recently launched a new social note card system where you can fork note cards and get really good learning stacks. They have scenario-based labs that put you in the middle of everyday common day tasks. They have instructor mentoring, which is slick and also a little bit of secret sauce. And the graded server exercises for those of us who just don't test so well.

It's nice knowing, like, getting some verification, right? You're just like, yes, I actually did learn that thing. That's me shaking because I was taking a test. Yeah, it's really nice. It's a good platform, stacked full of Jupyter Broadcasting members. And if you go to linuxacademy.com slash unplugged, you get a great discount. I love the fact that they spin up a server on demand that you get to

SSH in just as the courseware calls for it. You choose the distro, it modifies the courseware and the machine they spin up, which is nice. And the other thing is – and boy, I could have used this a few years ago – is if you're learning some of the Amazon Web Services stuff, they're managing all of that for you. So you don't have to price bills. You don't have to worry about their complicated setup. You don't have to like go get your boss to give you the credit cards.

That way they can pay for it. And out of that, it's all just part of the – it's all just built into the Linux Academy courseware. Check it out. LinuxAcademy.com slash unplugged. All right, guys. So we have the official announcement that Flatpaks were released this week and I thought since it's newsworthy, it came out literally today, it's worth talking about here on the Unplugged show.

And I don't know where to go with this without just putting it all out there. And hopefully people – so on the Linux Action Show, by the time we got to Linux Action Show on Sunday, people had really, really like drawn a line in the sand and picked a side in the new universal packaging war. And there was a full-on response, a lot of it from the Red Hat folks, about Canonical's PR strategy. And there was some attacks, some of which we covered in the Linux Action Show. And it got really kind of ugly fast. A lot of focusing on the PR initiative. A lot of talk about how they use distribution names. And I think, you know, that is worth thinking about and having a discussion about.

I think it is a fair discussion. But I was really disappointed to see they didn't go after the technical merits. Yeah, that's what I'm most interested in. I wanted to take a moment though and just point something out that really rubbed me the wrong way once I came to the realization. A lot of the posts on the mailing lists, on the blogs, on Twitter, even today actually. I'm still seeing them even today. About how Canonical is manipulating the PR in this. One of the Fedora mailing list developers, they said, Canonical beat us to the marketing punch.

And then they focused on this aspect so much that I realized what bothered me about it is, Wes, can you do it? Can you have a web browser on that? I sure do. Can you go to opensource.com for me? I'll go there with you too. Let's all go there. Opensource.com. Yeah, let's give them some traffic.

How would you describe opensource.com, Wes? I do see a big friendly red hat right in the top right corner yeah this is opensource.com an extremely valuable domain name picture the outsider coming into the Linux ecosystem wanting to learn more about opensource this has got to be one of the biggest

pieces of real estate in Linux in the opensource ecosystem this is an outfit that is funded by red hat This has got to be one of the biggest pieces of real estate in Linux in the open source ecosystem, opensource.com. This is an outfit that is funded by Red Hat. Damn near all of the writers write for are employed by Red Hat. The ones that don't are Red Hat friendly. I mean no disrespect to the people. I know some of them that run this and they're fine people. But this is literally, this is literally one of several propaganda fronts

the Red Hat finances to gently nudge the news coverage in the Linux ecosystem on a fucking hourly basis. I just – Wait, which?

This company is among many tech companies that makes and establishes PR outlets, works with journalists, hires Red Hat-friendly people to write things. This is something that all of them do. And I understand that the Fedora project is not directly hooked up to Red Hat. But let's be honest. It's a Red Hat project. So to come after Canonical for just having a successful PR launch, should they have name-dropped distributions? Maybe, maybe not. I think you could quibble about that.

Should have Mark Shuttleworth taken a pot shot at Flatpak saying that 95 percent of contributions come from one person? No, I don't think so. But you can't fault them for working with different outlets, for doing interviews, for people behind the scenes working with projects, making sure they understand the snap package format, troubleshooting problems they run into, directly outreaching to work with them. That's not manipulation. And see, if you read some of the posts, they all essentially go like this. Canonical manipulated the PR. They didn't want to go with somebody else's technology. And then the conclusion is, if they really wanted to make this a community effort, they

should have come around and talked to us all. They didn't follow the process. They didn't follow the rule. That's the same line of thinking that prevents women from joining different groups and industries or singles out minorities. They didn't follow the process. They didn't go through the proper channels. It is a contribution that happens organically. Everybody treats it like it's some master Shuttleworth plan. Like Mark Shuttleworth

is up in his moon tower, up there plotting the global dominance of Linux, and his master plan is to manipulate the tech press and get them to all orchestrate his brand new package format, so that way he can declare dominance

and earn all of the millions of dollars by coming up with this technology. And so when Canonical releases something, instead of going after it on its technical merits, we go after it on the evil Canonical once again, trying to drop something in our laps, just like Upstart and Mirror. It's all part of Schallerwerth's master plan. It just gets so old. And the constant attacks on the PR are so unfounded.

And the Linux Foundation is just as guilty. They also have donor-friendly writers who write puff pieces and don't even disclaim it's being financed by these companies. Again, manipulation. Go ahead. The Flatpak thing was interesting because they were saying that Ubuntu was not working with the community and they're trying to build this stuff because the whole not invented here thing, except for the fact that Snap was created first and there's – the flat pack is the actual not invented here. I don't even – do we really even care which win was created when? No.

Because it seems like. No, we don't care. It's just what they're saying. But the more important part is that they're saying these things about how he's – like when he was saying that 95 percent of the work, he probably didn't actually tally it up. But the people who were saying – it was not 95. It's 75. That's a significant number still. You're caring about a percentage that's not that big a deal.

It's 100 percent canonical contributors to Snap and SnapD and – I mean really. I mean at the end of the day, it is a red, there's a red hat initiative. And there's a canonical initiative. To be honest with you, I think his point was, it was 100%. It was he was saying not necessarily it was 95 redhead. It was 95. One person. I think what's interesting about the snap initiative is the work to get it on the other

distributions. A lot of that was community driven. People were interested in getting snap packages on their distribution. So they put the work in. Again, it wasn't some sort of canonical master plan. Once there was momentum, canonical pushed some of the final bits into piece, made things work, did the outreach. In their interest, for sure. That absolutely makes sense.

You can honestly see how this was an organic like, well, let's create something that solves our problem. Oh, there's some interest here. Well, if there's interest here, we just have to do this, this, this, and then now okay. I mean, you could almost see how it just fell in their lap. Like they're lucky in some sense, and yet it's from their own doing. But the reaction is it's part of this master strategy. And that mentality that comes out brings us all to war, and then everybody picks their side. And I intentionally in the Linux Action Show was like, there's a lot

of different technologies here. I don't know which one is going to work. I think security sucks on all of them right now. And I just think this solution is inevitable at some point. That was my core point Linux Action Show. And I got emails, you have to stop ranting. That was a horrible rant. Chris is supporting Snap Packages.

Chris is a flat Pack yes man. Like, that's just stupid crap that I can't even believe. Everyone knows you're for app image. I mean, come on. I just don't care. Yeah, right. Exactly. I mean, we just need a successful technology.

Hopefully we can all figure it out without too much of the same thing. So, with that said, Flat Pack, I think, is an interesting solution. Yeah, totally. I am a little uncertain that it's going to be the long-term winner for a couple of reasons. It's very desktop-focused, and that's a point I meant to mention in the Linux Action Show, where snaps are definitely more designed for server, Internet of Things devices, and desktops. And if you think about the huge-ass server market, just having all of the momentum that snaps just—

Okay, how can I say this? You know what I'm trying to say, Wes, is because snaps are going to support server installations, there's just going to be more snaps by raw numbers right there. There will be just way more snaps because it supports server stuff and because it works on Ubuntu than AppImage has probably ever just right there. And then the fact that because – I would think all – there's going to be so many people familiar with creating snap packages for the server

that there's going to be a lot of knowledge sharing that is going to just sort of, I think, be a snowball effect for snaps. Well, and that makes a lot of sense too with how they're – Popey was talking last week about how it works with how they're mounting that at boot time. So everything will be there. You can just spin up an Ubuntu droplet, install your snap, and now you're running your binary with all the dependencies handled for you, and it's super easy, right? J6 points out in the chat room that after the Linux Action Show, the blog I was reading, Adam's blog I was reading. I like Adam. I think he's a nice guy.

The blog I was reading, he updated his post to say, oh, and I guess there's this thing called AppImage. Like they weren't even aware of AppImage. Okay. So maybe there's something to say like was prior – enough prior research done. Maybe not. I don't know. I wanted to see – I wanted to give – oh, go ahead. Go ahead.

Let's start. Let's go in the mobile room so the the app image thing is funny because app image has been around the longest yes right yeah and there's like project shipping with app images right now yeah one of them by the uh creator of the kernel we all speak about all the time yeah it's just unbelievable but the funny part is like this it's all about the like the main core problem is that all this hatred towards ubuntu, whether it's even technically justified or not. They just want to jump on as many things as possible. So for example, when Snappy was first released in 1604 and it was big, big press, the first thing people said was, well, they just made it for Ubuntu. They don't care about any other distro.

And then when they show that they're working with the community to actually make it for all other distros as much as possible and as soon as possible, well, that doesn't count anymore. Let's completely ignore the fact that the main complaint we had is now completely voided. Let's find another reason to hate them. Yeah, I think that's it sort of murky is there is also a technical argument to be made and there is also legitimate concerns and questions about the CLA and about their closed back end. All questions I think could eventually be answered but don't have good answers right now. And so I wanted to – let's start with – is Ryan and Ike – yeah, good. So let's start with Ike real quick. Ike, I saw your G Plus post on this.

I saw your G Plus post on this. And I thought it made it a good opportunity to say, you know, we're talking a lot about this in a way that it makes it sound inevitable. But there's probably some pretty common work cases where this type of packaging isn't really necessary. I think you've been getting a lot of questions about snaps in Solus. So I'd be really curious to hear your thoughts about it. Yeah, I mean, it's been a bit of an outbreak so from my perspective i've just seen people asking about it because they've heard about this thing called snap i it's suddenly a thing now everybody needs to ask about it people who hadn't ever heard about it before this is now the future of everything why aren't you using it you human scum and this has pretty much been the the weight

of the tide so far boom and from zero to 60 yeah i mean farfetched be it from me to be outspoken at all um but i mean people started straight up with solace it's like you really really really now need to be using snaps and it's like well i mean i'm probably going to go into a bit of a rant now myself. So may I? Yeah, go ahead. Right. So Solus has been around in one form or another now for the last few years.

It's got a bit of an old code base and it's got this curmudgeonly old bugger behind it for the last few years as well. Now, Solus was designed, as I said earlier, to avoid these design issues, right? Solus doesn't have 14 different versions of a library. We'll have one. You're only using that.

So immediately we don't have the problem of, what if I need this version of a library? Well, you're not getting it. It's as simple as that. How are we going to get LibreOffice? We already packaged LibreOffice for you. Why do you suddenly need LibreOffice packaged? And then there's the fact that Solus doesn't use a traditional packaging format. So if you look at any traditional format, say a Debian package. Now, Debian packaging is hard, right? It's one of the

reasons I first went with an independent base, because I passionately, passionately hate Debian packaging. That if it was on fire, I wouldn't piss on it, right? I hate it that much. So I went as far away in the opposite direction. Now, when you package something, you have a policy. You say, these are development files. These are files the user is going to need. This thing has relational dependencies between them. So I designed a package build system

that would tie all those pieces together and put the policy in place. So I don a package build system that would tie all those pieces together and put the policy in place. So I don't have to do manual file lists. I don't have to do manual things at all. So to give an example, Michael Hall posted up an example of how easy it was to package the Genie editor for Ubuntu and YAML snaps. You saw that post, right? I think it was about 32 lines. Yeah, he said he was editing the YAML file for the Genie snap from within Genie. You saw that post, right? I think it was about 32 lines. Yeah, he said he was editing the YAML file for the Genie Snap from within Genie. Right. The Solus package is smaller. The build file is smaller.

So from the perspective of Solus, just from a packaging perspective, I'm not talking about confinement or sandboxing. Snap is technologically inferior to what that we already have okay so that's one huge reason for solace and not to need to adopt it we go for integration so i don't like say solace is using gtk 318 now. We have the 3.18 theme. We want everything to look good, yeah? So why would I then want to bring in 3.20 packages that then don't quite interact properly or don't look quite well? Solace has an optimized runtime. Why would I then want to use a runtime that isn't optimized or is built or maintained by somebody else or isn't quite secure?

So it sounds like you're worried about it's taking away my curation is kind of what you're saying. I mean, curation is definitely one part of it, but I don't want to water down my system and trust somebody else to build my operating system because if they were capable of doing that, I wouldn't be the person building Solus. Yeah. You are in a unique position there. That is true. That's funny. position that is funny so uh now i want to uh i want to ask ryan who's sort of on the other end of this who's got an upstream project that's moving extremely fast there's lots of features they want to ship they want to guarantee clearly is involved with packaging and the difficulties

there and he wants to get out on as many freaking distros as freaking possible so ryan what are your thoughts about snaps and flat packs and all of that and some of your experience so far oh boy you know that uh ike and i used to work on solos together i used to help try to help out um so uh the problem is when i'm packaging my craft is i i have to it for every single thing. I, it, it's not a enormously,

well, no, it is. It's kind of an enormous task. Yeah. That's what I hear a lot. And you got to do a lot of talking with, you know,

folks who are involved in the distro. Uh, and it's just, it's kind of a mess. Even Debian packaging is a mess. Well, I guess Ike kind of said that. You're trying to be diplomatic about it. Just go and say it's awful. I think everyone knows that.

Yeah, and so the thing is, it doesn't have to be Snappy. It doesn't have to be Flatpak. It doesn't have to be Snappy. It doesn't have to be Flatpak. It doesn't have to be AppImage. But whatever, or maybe it's all three of them. Well, actually, can I make you – I disagree. I think you're being too polite because Mycroft, I mean legitimately, has some server-side components that AppImage simply just doesn't address. Yeah, and I've thought about that, especially when you're running headless or you're running like, you know, with the device. But the thing is, is all we want to do is be able to give

people the newest version of the application. That's it. And we just and we don't want to spend all of our time packaging. I've had a guy who's Aaron who's been working two weeks on just packaging. And that's kind of ridiculous. Like, we just want to get a version that somebody can pull down and install and not have to worry about it. If that's Snappy, I've got SnapD on my Arch machine. Since we were talking about it, I installed it. I haven't restarted yet. So the SnapD isn't running.

It's not in your path yet, yeah. the snap on your path yeah yeah but i was gonna but like the uh but the idea is as an application developer i just want to be able to tell people like here this is an easy way to install it if you want to pull it from your distro that's great but like this is this is a fast way to do it and we know everything's good in there we know the system libraries you don't have to worry about them the thing that really actually made me look at flat pack and look at snappy is i i updated free type two i think i talked about this last time i was on here and free type two broke all my electron apps when it went to 2.6.3 i think yeah and and i was like i can't believe that happened i can't believe that such as just this this font library broke my electron apps

yeah and you can say whatever you want about electron or about anything along that path but it would have been great from a user perspective when I'm just trying to use Slack to tell somebody something if, if it was just all self-contained and I don't know what the security, uh, implications of that are. Um, but it's, it's frustrating, you know, especially when you're trying to get stuff done. Yeah. So, guess I think, Ryan, the reason why I like your perspective is because I think it represents two valid points of view. So I want to toss it back over to Ike. So Ike, I think really what I see here is Ryan makes a case

for certain specific applications to be distributed this way, and I think you're making the case that the base system and all your daily driver applications should probably be maintained by the distro. Do you, what do you think? Yeah, I mean, you do need a domain separation, right? So I have what I call the operating system. And in the case of Solus, that is basically up to the budgie desktop and any, you know, the, the fat stuff and the stuff that we own you know like the the web browser text editor things like that apps on top of that are then separate and in the case of applications so from solace's perspective we're not owned by a corporation it's just me paddy here you have to

deal with right so i can be a bit more flexible than most if i was a company or something it'd be harder to deal with solace, but it's just me. Because I'm not here to make a few without Solus. It's simpler for me. So I'm happy to bend a few rules to help the application developers get in. But then from the other perspective,

I can make it as easy in the world for any of these guys to package their app up for Solus. I'll even make sure their patches get in for libraries, or I'll even make sure they have the different versions of the libraries they need i'm happy to do that from my perspective and sorry ryan but you kind of made yourself a target there now to use my craft as an example now that's something that i look at as having system level components now as for being up in a flat pack, well, I've already gone and looked at the code. Now, Ryan himself will tell you, I've already requested an architectural

analysis of that. I've gone and looked through Mimic because I'm not happy with the code, right? And I've worked with them and they've had a few issues there and I've spoken with them. And they are responsive. I'm good in that. But things like that, I wouldn't want to go in first because

I don't want Solace users in in the dark right because then you're getting into an area now not to use specifically not to use mimic or mycroft in this example but what you then get is a downward spiral that contributes to incompetency of software engineering practices so by allowing everybody to run in a container, by allowing everyone to be isolated, they can get away with having shite code. But don't they already? Yeah, but I'm not one of those people.

I'm the sort of person who runs their code through Valgrind and static analysis and makes sure it doesn't leak. So by allowing these things to just break, like do whatever they want, because they're in a sandbox. We're just going to have this downward spiral of incumbency. It would be like saying if you put your car and limited it to 10 miles an hour and put cushions all around. I think you're caught in a logic loop here because I think we are already in that paradigm of software quality and that's why projects –

People want those projects. And that's why projects want those projects and that's why projects like yours exist in the first place because we're sick and tired of it and we need people who care about some of those small things to sure but i mean there does need to be it's a case of onus isn't it um we all need to agree for first of all we do need the os app separation right so the the os everything up to and including the desktop that you ship, that belongs to you. If your version of a library breaks the applications, you're a bad vendor, right?

If the applications break the operating system, that's a bad application. So I can see why you need the separation. What about the use case that some of these things solve, not necessarily a flat pack, but definitely app image and snaps. What about this really important use case where you just have no control over this?

Somebody discovers Solus, they download it, they've just recently switched over from a Mac or Windows, and the way they install software is they go to the vendor's website and they download it from their website. They're dumb,

they don't know any better, curse them, but that's how they want to do it. That is a one example, but legitimately, let's be honest, it really is solving a problem to be able to say there are better ways to get software on Linux, but if you want to download the Snap, just go here, download these Snaps, or that use case, here's a thumb drive full of Snaps, install these. There are scenarios where it is beneficial to the entire platform as a whole to be able to say this could be installed on linux end of sentence yeah i mean it can be beneficial but

i mean i mean there's so there's so many problems with this as well so when i install a dot x i mean i'm sorry when i install a uh ae, I mean, I'm sorry, when I install a Snap from the internet, because this is exactly what we're emulating. Between Flatpak and Snap, we're just trying to emulate Android, right? Yeah, I guess so. This is exactly what we're trying to do here. We're downloading arbitrary files from the internet

and install it, but this time we're going to do it without the store itself, which is what Snaps are trying to do. Except there's a store and there's a repo with snaps so no yeah except for the store none of our distributions can actually contribute code to this unless we uh the store itself we can't see right that's completely proprietary we're not allowed to see it but we're going to trust code going through though yeah um along with that saying you don't trust it and saying it don't exist. Along with that, if I want to contribute code to get any of Snap fixed or better integrated, I then have to sign a contributor license agreement. I agree.

I agree. So how open is it really? It's open in terms of code. The code is open. That's fine. But if I want to contribute it and I want to integrate into Solace, then I can't. I think you could throw snaps out there and the entire conversation is still valid if you're

just talking flat pack and app image. I agree that I think none of them are technically close yet. I mean, close might not be fair, but... And this is the thing. We're getting into heated arguments nowadays, like everybody is, but neither of them have had a chance to getting into heat arguments nowadays like everybody is.

But neither of them have had a chance to prove themselves yet. It really is. I think it is. And I think – but what I think I really appreciate about your viewpoint is it represents the sort of the person who's willing to go that extra step to think a little bit harder about the way these things are built, to present the user with better options, to make sure that things are built and designed properly and things are running current. And those are going to be the people that are the most disserviced, I think, by the more generic software installation gets on Linux. And so I think the point that you're representing here in this conversation is a huge point of input that we're getting.

It's not exactly refined like that yet, but we're getting a lot of that feedback into the show, is there is a point for the package maintainer. There is a point for the builder, for the assembler. And we have come to these positions and had people do these jobs for very specific reasons. It wasn't just arbitrary that we just one day all decided we're going to have package maintainers and we're going to have repos and we're going to do software this way and we're going to share libraries. That wasn't just an accident.

There was good reason for it, and when you implement it properly, it is a very good system, and it's not going away necessarily. And then I think Ryan does a great job of representing the upstream project that's got something really neat on their hands. They want to get out to as many people possible in the most predictable manner possible in the most user acceptable manner possible. And that's also a completely legitimate point of view. It's pretty amazing. I have a question for Ryan. Go ahead, Ryan. Ryan, if you have a situation where the packaging, the current structure of packaging, let's

say the application level and not the core base was just those three competing. All of the three, the Flatpak, the Snaps, and the AppImage are all equally dominant at some point in time. Would that be an improvement to the current state? If they were all competing but all still – say that one more time. I'm sorry. If you had to make three packages, and depending on the distro, like let's say some distros use Snap, some use Flatpak, but they don't all have a universal. They don't all agree to that. But instead of having to deal with the maintainers,

instead of having to deal with different formats based on the different distros and the different distro versions, you only had to make three, one for each structure, for flat packs, snaps, et cetera. Would that be an improvement to the current state for Mycroft? Sure. I mean, what we're talking about is kind of totally different, I think. Ike is talking about some of these lower level system libraries that frankly,

MyCrop does use. We walk kind of a line, you know, what we've had issues packaging with Snappy in the past because we touch some things that are not, we're not as simple as your normal desktop application. But the thing that we're trying to do, and i think what a lot of application vendors are trying to do is uh put something out there where they're not so worried about a system library changing like free type 2 when they haven't written support for that yet you know there are a lot of things like that where you're you're trying to do your best as an application developer, but you've got so many competing desires from your users that sometimes you need to have that direct relationship with them in order to provide a good, consistent, stable experience. And I think that if we demonize these projects or say like, and I know that's not what a lot of people are doing, but if we don't, there's obviously a reason that they're gaining traction.

And that's because application developers are having a hard time targeting Linux and getting a consistent experience to their users. And to talk about the trust issue, if somebody is willing to go and grab a PPA for a piece of software, they already trust who's providing that PPA. Should they trust them? I don't know. But they already do. And there's a lot of trust that's thrown around anyway. You're trusting the maintainers. Sorry, there's people confusing PPAs with being canonically maintained. Yeah, I think you make a good point, Ryan. And I kind of want to give you sort of the final input here.

You say you have a proposition. Yeah, I mean i'll start off with one bit of confusion actually had with ryan there which i didn't understand um ryan you just asked if there was there was all competing standards would you then accept them all and it would all be a good thing that doesn't seem like it makes sense to me um you would adopt one of these standards because of the universal standard to solve the problem of not doing the other packaging formats. Otherwise, you essentially just took on three more packaging formats. No, I didn't.

No. I never said we would just package one. I never said we would just package one of those. That's not what I just said either. No, okay. Hold on. Stop, stop. This is not really worth talking about. That's not what I just said either. Hold on, hold on. Stop, stop. This is not really worth talking about.

It's just simple. The question was if you could basically get away with using just AppImage, Flatpak, and Snaps and essentially cover the entire ecosystem, isn't that better than creating four different versions of the RPM, two Debs, one for Debian, one for Ubuntu, and submitting it to the OpenSUSE build service and trying to get it in Copper.

And the AR. That's what I was saying. Yeah, yeah. That's all you're saying. It's not bad. I think seriously that the problem is elsewhere. Seriously, the problem is that developers have been doing development relying on system-wide things like package names.

Developers used version control systems appropriately. By appropriately, I mean include your dependencies in your project tree as a sub-module of GitHub or SVN or whatever. These would allow any maintainer to go and compile the project. Now, how they distribute those files in their distribution is their problem. But everyone, there is a maintainer, will be able to just get your source and build it. That would actually fix things. And we wouldn't care if we're deploying 30 different packages,

because at the end of the day, anyone really wanting, they could download the source and build. Yeah, I think... I don't want to wait for a maintainer to do that. Yeah, I think just the complication of that landscape. So here's what I get on the Coder Radio show is a lot of people who want to start distributing Linux server-side apps, and they want to ask Mike and I the best way to go about it. And the number one thing they have a problem with is I can't effing figure out how to even get my app out there.

Like, how do I package it? Which one should I be packaging it for? Which version of those distros should I package it for? And then once I've packaged it, how the hell do I let anybody know about it? Like there's no app store. There's no central news place that people check. Like how the hell do I let anybody know I've created this thing? The entire process is so opaque to them.

It's so nebulous to them that they honestly say, you know what? I got better things to do. I'll just go make a Docker container and call it good. I'm out. I mean, so I think, okay, so we should probably wrap it up. But I guess the point is, is what I think people are fundamentally missing when they're not getting the entire argument is, and maybe I'll send it back to Ike so he can finish his thought here,

but the point is, is there's a problem out there, a meta problem that goes beyond the technology, that goes beyond the sandboxing, that goes beyond who created it. The meta problem is, is there is a wave of talent that wants to create software for a platform, and they simply can't even figure out how to get the applications here. And if you don't think it's true, you haven't been talking to a lot of developers because it is fundamentally just that stupid basic problem is what's holding up a ton of them. As I looked at it, it looks super complicated.

I ain't got time for that. And when you have so many choices, how do you know which one? Yeah, and they're not familiar with it. You know what? I'll just go make an Android app. Or a really fancy app for the Mac OS X because they're defined. Or you end up with own cloud on Debian. Yeah, yeah, yeah.

But okay, Ike, I know that wasn't your main point about the semantics of the packages. You had another point you wanted to make in there, right? Yeah, no, I mean that one I was just literally confused about. I didn't know. Yeah, so finish it up there. Didn't know that was going to happen. So my proposal is basically this. I know probably he's not actually on here at the moment, so I'll let reply on irc so this one's a bit of a freeway um but well moving on so my proposal

is basically this um i know i've raised concerns over parts of the code base in mimic now to be clear this is not me trying to crap on mimic right This is because of the forked code base that it started from. Now, that stuff is already a nightmare to work with. I can see that. So I'm not crapping on those guys' stuff. So my proposal goes freeway. Ryan, feels like this is your life. I will package all of my craft for Solus right and I will

demonstrate how easy this is going to be as that's a great title and I will demonstrate how easy this is going to be now it's something I'm going to be maintaining right and just as long I want you to take the patches coming back so this is you know to get everyone back in the mood of

giving an open source and contributing I will give you the patches back from Solus for the things that I have concerns about, as opposed to just sitting there bitching about them. Now, because there's a lot of that that goes on these days. And for Poppy, there are some things I think that Y package in Solus is already done, right? Now, as much as I picked on it before for having some technical limitations in comparison to Solus' Y package in Solus is already done. Now, as much as I picked on it before for having some technical limitations in comparison to Solus' Y package,

that doesn't mean that Snappy isn't good for Ubuntu. What doesn't work for us will actually work for Ubuntu. This is their solution. That's great. There are some things we have already done in Y package in Solus that I keep going to say flatback. I'm sorry, but it's that.

There are some things that it could really, really benefit from. So some of the things that we took from Clear Linux, it could also benefit from them, like the optimizations, the advanced pattern and logic. So some of those concepts, some of the constructs that we use in our YAML files, I think they could really do this. Now, personally, I'm not going to be able to sign the CLA.

So what I'll do, I will put those contributions out as public domain and I will leave those for anyone in the canonical to take them up and merge them if they wish. Nice. You know what? If my craft in some way, like a Venom extension or some capacity, ended up integrated into

the desktop on Solus. Oh boy, that wouldn't be a bad thing. We have Raven, don't we? I would have to run that on my laptop. I would think I'd make that my laptop. We have the Raven sidebar. How sexy would that be? Yeah. That's my promise. I would like to point out too that

I love Solus. I used to be a part of that community. And I've used YPackage. And it's a good solution. So, Ike and I, we get along in other settings. It's Wes. He brings out the worst in people. I sure do.

Yeah. But what we're just trying to do is find easy ways to get our software, you know, to users. Yeah, I mean, this has got to be a – wherever we end up, even if it's – this is going to be kind of a good thing for projects like yours, Ryan. So I'm excited for it, and I've got to imagine you are. Can you share at all, like, have people from Canonical contacted the project and said, can we get you started? Can we help? Like, have people from Canonical contacted the project and said, can we get you started? Can we help? Like, what's that process been like? Because that seems to be sort of a key piece that they've done that's helped them get an early start on a lot of snaps is

Canonical is working directly with different projects to help them, you know, say, hey, do you have any questions? Are you interested? Can we help you? Have you had anything, any discussions like that? So, first I'd like to real quick thank IKI, and i can't think of who else popped in but some some early folks popped in from from souls and from uh and from ubuntu and

a guy from fedora to provide some uh some kind of day one and day two feedback on the repo when it first went open source the mycroft core repo so first i want to thank those guys that but uh yeah we did we have had the ubuntu guys um reach out to us and and walk us through packaging uh we we have some requirements as far as access to the sound devices that um we that we still do a back and forth with them on. I'm making sure that it picks the right sound device. Yeah, it needs to pick the right sound device. It needs to pick the sound device that's in use by the user.

It needs to have a good default. There's a lot of nuance in picking and managing sound devices. But all around, they've been really good providing feedback. I talked to Adam Miller on the Fedora project, who also provided some feedback around Flatpak. But yes, so far, everybody who I've spoken to on the Snappy team has been really great about answering questions. The same with the Flatpak guys. And then, of course, there have been Wimpy helped us out with packaging for Debian.

Nice. You know, for Ubuntu. So I don't want to like poop on any of the packaging things and say, and I want to say that, yeah, the, the snappy guys and the flat pack guys have been great, um, providing feedback. Um, I, I think that, uh, that what we, what my two cents is that I really like for, um, to start discussing the technical merits of these projects like you've been advocating for, Chris. I'd love to see more write-ups on what the actual difference is and not about canonical, quote-unquote, propaganda.

Boy, that is so great. I think that's probably our ending note right there is I think just maybe future episodes, maybe we'll just start the conversation. Future episodes, I'd like to talk about, first of all, Flatpaks are very much a repo-based technology, which is a totally different approach than, say, AppImage or Snap necessarily can be a totally different approach and is in some ways reminiscent of PPAs, good and bad, and means software updates will be easier. There's a lot of technology there, right?

Their sandboxing is further along for different distributions. Snaps, I think it's only like on Ubuntu and Arch right now. But how are they doing that? What is AppArmor the right choice? All these questions that we have, questions about like re-implementing the Snap store backend or creating your own Snap store. Do you have to rebuild the SnapD program? Like all these – And where will the real community actually flourish?

Yeah. Is uApp browser, whatever it is, uBrowser, whatever it is, is that going to be the place? Like all these little questions that we have, I think now we should start talking about and the technical aspects of them. I think this part of it, it's just – it really has moved to a very partisan discussion very quickly. But I don't know if it's really – I don't know really if there's much more productive conversation to be had there. I will say I think we may end up with two regardless. I think that something you said caught my attention just that really snaps are the only one with any server focus at all. You know, like Flatpak explicitly says that it doesn't really fit with the server and app image has that great desktop

in there so i think really docker or something like that is here to stay on the server side yeah so there's probably at least going to be two the momentum um the momentum and demand around delivering reliable consistent predictable and maybe the software on the server is going to make snaps, at least successful on the Ubuntu platform. I think probably that alone. But all of that is a conversation for another day. Join us on Sunday on Linux Action Show. We'll be reviewing the new Fedora release.

I'm really looking forward to it. This is a release I have been hoping for. GNOME 3.20 is looking nice. It's got Flatpak support. I've had a Fedora installation on the server running for a while now that I'm going to be upgrading to Fedora 24 and reporting on that. Really been looking forward to it. Join us live Tuesday. That's my next Tuesday, my last episode before I go on vacay.

Wes, you may not be here, too. I may not be here. So we may have a special guest sitting in Wes's chair. Find out the time in your local time by going to jupyterbroadcasting.com. Then join us at jblive.tv, linuxactionshow.reddit.com. And we'll see you right back here next Tuesday. I've never installed GNU slash Linux. Zombies don't run Linux. I've never installed GNU slash Linux.

Get it out of here. It's negative in the freedom dimension. It's negative in the freedom dimension. I got the sense right as we were leaving there that Rotten, you had something you wanted to add, but we were wrapping up. Did you have – did you want to complete a thought? I was just going to answer your question about the UAP Explorer. Oh, yeah. That's an unofficial non-canonical thing. Right, yeah.

Remember how the Ubuntu forms were unofficial at one time? Remember that? And they became? That's a long time ago. That's a while ago. Jbtitles.com. Well, the UF Explorer is not just snappy.

It also has click packages, so it's kind of like a mess right now. Yeah, for now. Hey, a big thank you to both Ryan and Ike, who I'd stabbed in the ring. Thank you also, of course, to everybody who contributed.

But there was a lot of heavy lifting there by you guys. Really appreciate it. I don't think I needed to mention it, but it was very nice having different perspectives from different... Yeah, absolutely. Whether it was distro, maintainers, packagers, all that. I wanted to add that

Mycroft is already dockerized. Right, so there you go. There you go. Of course it is. Everything's dockerized. Right? So there you go. There you go. Of course it is. Everything's dockerized. All right. DriverTuttle.com, you guys. Got to go to drivertuttle.com. My little poppy.

Oh, boy. That poppy sounded like a robot, huh? Sorry about the my little poppy one. I got to get Mycroft installed on this machine just so I can make Popey say stuff during the show when he's not here. Right? You have to. That's obvious. Popey's going to end up showing up on a lot of podcasts.

I want the functionality to say, Mycroft, define coconut. Like, I want to be able to just can I do that Ryan will I be able to get like like a word definition because that's what I would like to have right here in the studio with a mic on it and that way when I wanted to find something I can just ask Mycroft and then Popey reads the answer let me check

let me let me pop them up right now I love it so Ikey you spent a lot of effort trying to effort trying to make sure that games work on it, especially Steam games. But are you getting any support from Valve? And doesn't it kind of suck if you're not? Well, I mean, you must have realized by now I'm half-wired wrong.

I mean, you must have realized by now I'm half-wired wrong. So in terms of support from Valve, the way I kind of look at it, it's like, well, I'm doing it because they're not. I mean, eventually they're going to pick up on what I'm doing, and eventually all this stuff will become the norm and everything will be fixed. But it doesn't really bug me, to be honest with you. Do I think they should have been doing it properly first? Yes.

Do I think they should have used Ubuntu 12.04 and used a runtime library like that? No. But it's the same reason why I think Flatpak is doomed as well, because of the whole thing. We're going to provide your API, but we're not going to define it.

So what it should have been is like, this is the API we need, not this is the API you're going to use in. So what it should have been is like, this is the ABI we need. Not, this is the ABI you're going to use in this exact form. And that's exactly the same problem Flatpak has. And it's the reason why Flatpak would never go into Solus, while Snap would never be a primary form. It could be something we could accept as a secondary form in Solus, you know? Because it doesn't push that, it doesn't force that distribution method on you.

That's perfect, because Snap is not intended, it's intended to be a supplement. It's not intended to replace the core at all. Like, they were in charge. Far Shuttleworth was asked about whether Snap would make it where the Debian was not necessary, and he said that was never going to happen, that Debian is a fundamental importance to Ubuntu.

I'm going to pursue that. Hold on, hold on, guys. This is CNN and that Debian is of fundamental importance to Ubuntu. There's no reason for them to pursue that. I mean, you need your next one. Hold on, hold on, guys. This is CNN Breaking News. We have an update from Robot Popey. Greetings, Professor Falcon. Would you like to see some projected kill ratios? Oh, God!

That's awesome. And that is our, oh, God, that is our Poby robot update. Where did that come from? Poby. The Whopper, I believe. Poby's inside the Whopper now. That's amazing. So yeah, I got to get hooked up to the stream.

Yep. Oh, Poby. That is maybe the next most important JB project. Oh, Poby. You know, JBot needs a voice. Yes, it does. JBot. JBot needs a voice. Yes, it does. JBot. JBot needs a voice.

It needs to come alive. Wouldn't it be great if at the end of the show we could turn on JBot and like, wouldn't it be, I wonder if there's a way we could,

you wouldn't want people to like jam it up. No. But you'd want to have certain people, I don't know, because you wouldn't want to limit it either,

but it would be great if we could give JBot a voice at this point in time during the show. Can we ask Ryan right now that if you ask Mycroft, would you like to play a game? That's the answer. How about a game of chess? That was pretty good.

That was good. Greetings, Professor Falcon. Would you like to see some projected kill ratios? You know what I got to do is I got to put that on the soundboard is what I got to do. Because that's going to be handy. I secretly use Arch Linux. All right. So we got to go pick our title and we got to get out of here.

Yep. It's getting ridiculous. This is a long ass show.