LINUX Unplugged - Episode 149: Snaps are Go! | LUP 149

Episode Date: June 15, 2016

Canonical drops a bombshell by making snap packages available for nearly all Linux distributions, Nextcloud has some serious momentum, Samsung is rumored to drop Android in favor of Tizen across all devices & Wes kicks the tires of elementary OS’ new Beta of Loki.Then we try out Snap packages & discuss needs to happen next to really make them take off as the standard universal Linux installer.

So, you know, as part of my job for these shows is I stay on top of the news. And one of the things I've found really handy about Google Now is it's getting a pretty good sense of the news I like to read. And this article showed up in my Google Now feed with wireless conversions on the Pro 5. Ubuntu phone levels up. And I thought to myself, oh, interesting. I hope they have a video of it because I really want to see that I hope that there's something to watch so I can see it in action so I come down here and I hit play

and I think that sounds familiar sounds like a robot try and connect it to my TV sounds like some sort of AI if I knew more people were going to watch this video, I would have tidied my lounge. Yeah, and maybe I had a better mic. I mean, for goodness sake.

So really, you're just in your room there. You hit a button on your Ubuntu phone, and this giant Microsoft display lights up. Now, what's with the Microsoft display? Tell us the story about that. That's a meerkast dongle. So there's numerous companies that make meerkast dongles. And it just happens to be that the first one that we got working was a Microsoft one because I think that's the first one a developer happened to have nearby.

And it was cheap. I think some LG TVs work as well. So I bought one because he said that one works. So I bought one, plugged it into the TV, went and got my phone. And when I came back, it was a Windows logo on my TV. Yeah. Whoops. It's like, you know,

we swallowed that pill, too, when we decided, well, I want a really good game controller under Linux, so I'll get the Xbox. Exactly. And even us, the most hardcore Linux users, eventually buckled down and did it for the most part. So, yeah, this looks pretty good.

How's the performance? So there's a little bit of lag, and it's noticeable if you – like on that video, it's not very good because my thumb is jittering across the display, but I did another video where it's a lot smoother, and I pick up a window with a mouse and drag it around. Is that on your channel too? Yes, the next video, I think. But does that one have nearly 10,000 views?

That's the big question. That's all Chris is interested in. He's not got some fat guy standing in front of a TV. Okay, it's only got 2,700 views. Yeah, okay. Also, the OMG Ubuntu effect helps. So, yeah, you just press a button and it scans your local network

for meerkast devices and it shows them all. So it will show your Sony TV or your sanyo or whatever tv but it might not necessarily connect to them but the cool thing is uh the log files are captured in var log like any you know standard linux device and you can grab those uh the software is called ethercast aethercast and you just grab the log from there and file a bug on Launchpad, and they'll work through all the devices to get others working. So you asked about performance. It's a bit laggy.

There's a bit of a delay, and they're working on improving the encoding and the performance. There's also some funky stuff that I don't fully understand about how there's multiple channels that you can send data to the device. There's like the video stream, which shows the display. And then there's a separate channel where you can tell it where the cursor is. So rather than you sending an image of the desktop and an image of the cursor, you can tell it the XY coordinates of the cursor. And the dongle paints the cursor faster.

So you don't get that perception of lag. The phone paints the cursor faster. Oh, wow. So you don't get that perception of lag. And this is – now, do these two devices set up a Wi-Fi network between each other directly, or do they just have to be on the same LAN? How does that work? Dude, it's magic. I have no idea how the hell this works.

I just press a button, and this stuff appears on my – you can tell the surprise in my voice when it works. It was the first time I tested it. Really? Yeah, if you watch the video, like halfway through, when I look up, I pan the phone up and look at the screen, I'm like, oh, that worked then. Oh, okay then.

I love it when that kind of stuff happens. And the nice thing is is that unlike Chromecast, which is in a lot of things, or Apple's AirPlay, this Meercast, I mean, there's a lot of devices out there that have it built in as just sort of an industry standard, right?

Right. My Sony TV, that TV actually has it built in, but it doesn't work yet. There's some protocol thing. Apparently, it turns out that Miracast is a standard, but they all implement it differently. Yay. Every single dongle is broken. Surprise.

Yeah. That's always how it goes. You know what? The Linux community has seen that before. That is not where you're used to that. Well, cool. The good part is our part, the EtherCast part, is free software. It's on Launchpad.

That's awesome. Launchpad.net slash EtherCast. So I think it's using some components from some other Miracast tools that are around. So if people want to get it working on their TV then that's the code you need to hack on you know it's funny because a lot of times in discussions with people we talked about convergence a lot of people I think Matt

was one of them who was like call me when it's wireless if I have to sit down and put my phone into a dock I'm done and I always thought whatever I'm looking I want a dock I want a hardwired connection. But there is something to the being able to just project your screen up onto a larger screen on almost anywhere

you go. Like, if this took off, you could just see an organization equipping their conference rooms with those TVs that they know, or those dongles that they know are going to work with something like this. And now you have the ability to present in front of everybody immediately by sending your screen up there. Oh, that sounds great.

Yeah. In fact, the first time I saw this, the first time I was at a sprint in Prague a few weeks ago, and the first time I saw it was one of the Unity 8 developers did a demo. We do demos at our internal sprints, so we get everyone up to speed with what's going on elsewhere in the company. And he stood up and connected his phone. I say connected, you know, in the wireless way, connected it and connected his phone i say connected you know

in the wireless way uh connected it to this dongle which was connected to a projector so we're all watching this on a projector and we're thinking wow you could actually give a presentation directly from your device because the phone becomes your mouse and keyboard if you click on a field which is a a an input field the on-screen keyboard pops up on the phone. So you can just use the phone as a mouse and a keyboard. It's not the easiest thing to do. But if you wanted to give a presentation, you totally could do that. Oh, sure.

I've done that with some apps. And you know, now that I think about it, I have seen offices that have Apple TVs and have Chromecasts hooked up to their devices. But in the case of the Apple TV, that's a huge expense because the devices that are capable of communicating with that have a massive premium, which is not always optimal in the business place or especially in the education market, which is a massive field for this, a massive field for this. And the Chromecast, the downside to that, the price is right and the availability is high, but the dependency on a third-party service makes it unreliable and makes it problematic for large institutions behind multi-level firewalls and things like that.

So something that is over your local Wi-Fi network, an industry standard to some degree, and built into potential mobile devices, there is a lot of room for that still. That's really cool. I'm excited. Check. This is Linux Unplugged, episode 149 for June 14th, 2016. Welcome to Linux Unplugged, your weekly Linux talk show that is reporting on the front lines of the Linux packaging cold war. My name is Chris. My name is Wes.

Do you like that? Oh, I like it a lot. You know what I'm referring to. I certainly do. We're going to get into it. I mean, it's unbelievable. As if we had planned it almost every week, there is some big development in Universal Linux packaging. You know, that one installer for all distros.

Maybe someday we'll even be able to use them. Well, Canonical is here this week with a huge bombshell announcement in this category. So we're going to talk about that. We're going to even try it right here on the show. We've been trying it and give you our early hands-on. Speaking of early hands-on, Elementary OS has their new Loki beta out based on Ubuntu 16.04 and full of all of their new shiny. There's also a few hints of things they may have coming. We'll tell you all about that, our early impressions, because it is just a beta, so bear that in mind. And then we have a bunch of epic open source project updates. Stories that broke

just in the last couple of days. Extremely relevant to our interest. So guess what? We're going to cover them right here on the show. Going to do that right here. Heck yes. The rumors of my vacation were false. That's coming later. I'm hanging in there for a few more weeks. Are you strong?

Thinking back, thinking back, if I could have planned it, it would have been I would have left town like a honeymoon after a wedding. As soon as I wrapped up Lass's 10th anniversary, I would have left town because, boy, could I use it. But I'll tell you what. The fight goes on. We continue to be on the front lines just like our virtual lug. Let's bring them in now. Time Appropriate Greetings, Virtual Lug.

Hey, guys. Hello. Hello. Hello. Hello. So I guess I made a comment last week because Wimpy messaged me. He's like, so are you on vacation? I'm like, no, I need a vacation. But no, I am not on vacation.

So if you were surprised that I'm here this week, I apologize for misstating. Well, that's always a good surprise. Yeah. Well, thanks, Wes. Thanks. But I will be going on vacation the first week of July. So Mr. Colonel, I should probably be sitting down with you on the show. Oh.

Yeah. And, of course, we'll probably have – it means I'll probably miss two weeks of last, too. So we'll probably have a guest host in there. Hint, hint. You may already know him. Hmm. Hmm. Hmm.

All right. So that's not what we're here to talk about today, though. Today, we have some really big news. We have like what you might even say is breaking news. I should break it out just for the special event, some CNN breaking news. This is CNN breaking news. Big stories. Two big stories really to start the show off with.

Number one is that NextCloud actually has some code shipping. NextCloud 9 is available, and enterprise functionality is going open source well ahead of the July promise, I guess, the early July promise, I guess. NextCloud makes available NextCloud version 9. They're also releasing all enterprise functionality as open source, building on top of the open source own cloud core and adding functionality and fixes. This release provides a solid base for users to migrate to. Hmm. What do you think, Wes?

Oh, yeah, they've got an Android app here they can get for free, too. Are you surprised at how fast this came? That is kind of fast. I mean, just from what they were saying before, they didn't really hint at that. You know, I expected it in a month or two maybe, but, hey, maybe I'm going to have to spin this up and just play with it. I'm sure there's not that much that's different, right, than the own cloud I'm used to. But I'd be curious to see how it feels.

I mean this release really has got to be about this new enterprise functionality, the new branding, and I think it's got to be about momentum too. Right. They need to start off their new project and really have a good solid release and have a platform they build on. What's your read on it, Daredevil? Basically, I think it's a publicity stunt and smart because you just had a bad impact new. You have to bring new fast impact news

while still people are paying attention to what's going on with you. So it is sort of a building on the momentum thing and a smart one. It shows a good instinct. It shows that they're, you know, they got some savvy behind them. We want that. This next cloud thing is going to work out.

That's what I think. I mean, look at them. I'm tempted to go try it. I'm tempted. Well, I mean. Oh, yeah. Maybe we should have a, get one up and running. I just sort of got my head wrapped around the last release of OnCloud.

So it seems almost too soon to go to something else. But probably the sooner you switch, the better. It's supposed to be a drop-in replacement, too. Traditionally, though, the sooner you upgrade, probably the better, right? I guess you don't want to go too many releases, I mean. Right, probably true. Get it over with. Exactly.

All right. Anybody else have thoughts on the new next cloud release before we move right along? Got a couple other big stories. So I thought this one was interesting, just as sort of a rebuttal to some discussion that was out there regarding the slowdown of Steam machines. Wait, there's still Steam machines? Slowdown.

I thought those were dead, Chris. No, no, they're still around. And in fact, that company Valve has got a good announcement along with Dell. They've just pumped out a couple of more of these suckers. A couple of more of these. They. A couple of more of these. They've upgraded them to Skylake. These SOBs now got Skylake in them with DDR4 memory and some more NVIDIA GPUs. So it looks like they're sort of going at this with a pretty reasonable expectation.

This is at least, I'm just reading the words from Chris Suffen, I think. He's the senior marketing manager at Alienware. And this is a direct quote. He said, we expect that the SteamOS catalog is going to strengthen at the end of the year. What do you think, Wes? He knows stuff, doesn't he? It sounds like he does. What do you think he knows?

I'm not sure, but I'm really curious. I think I have a hint. Because this is a further comment. He says, Vulkan will speed up transition for SteamOS to AAA content. further comment. He says, Vulcan will speed up transition for SteamOS to AAA content. So maybe he thinks a lot of big games are going to be shipping with Vulcan

support at the end of the year. That would be cool. So, is $899 the price point for you? $749 is the low price for an i5. Do you ever go for the low price? GTX 960, 8 gigs of RAM. You know, actually, Wes, what do you really, I mean, so you're going to get two cores with an i5,

and you're going to get four cores with an i7 in reality. But they're both shipping with a GTX 960 GPU. Both shipping with 8 gigabytes of RAM. Both shipping with 802.11ac Wi-Fi and I hope Ethernet. One difference though is 500 gig hard drive in the cheaper one, one terabyte hard drive in the bigger one but to be honest with you I'm more likely to crack that thing up and put an SSD. I assume it wouldn't be too hard

to just replace that. If I could. So I mean part of me thinks how great are these games under Linux that they're optimized for more than two cores? Is that a thing? I mean I have seen that it is a thing but is it that common of a thing that it's worth going from $750 to $900? Maybe not. Especially if I'm not really interested in a spinning rust terabyte hard drive. And you're going to do aftermarket modifications anyway. But what do you think of that price, $749 for the entry? A little steep,

but if you really don't want to assemble something yourself, you want to just have a dedicated appliance and you're in on your Steam library and you don't have a lot of Windows-only games, or the games you do like Maybe We'll Be Here at the end of the year, maybe it would work. I don't know if the i5 is quad-core. Is it? I thought the i5 was dual-core with hyper-threading. No, it's quad. Well, then I definitely would just get the i5, I think.

Yeah, okay. So what's the i7? Is it six- six core it's quad with hyperthreading oh okay yeah i don't really think that's worth it then so wimpy what do you think uh you've got the uh sky lake or not sky like the skull canyon is that what it is the the really fancy like more gamer edition high-end knuck i'm drooling wimpy what are your thoughts uh it's exceeded expectations i'm suitably impressed so um it's got the uh sky lake i7 6770 hq so that's a quad core eight thread pro cpu and it's got the iris pro 540 which i think is the top model at the moment and my last knuck which i got in november last year was

the knuck that was the best at the time so it's the second best and this significantly outperforms it just using the iris pro so the game that i'm most playing is grid auto sport and by comparison that on the old knuck would pump out around 35 frames per second average on low settings at 720p. This new box can do 1080p medium graphics. About 55 frames per second peaks at 75 frames per second. So significantly, significantly better but the reason i've got it the the project for the future is um it's got thunderbolt and i want to experiment with an external gpu connected over thunderbolt later in the year yes that if that worked i honestly if that worked i would be in

almost instantaneously because see for example i i have not had as much success with Iris Gaming. I have played some games, but then there's like a new Formula One racing game that's come out, AAA racing game, and it's punishing even on the highest-end NVIDIA GPU. It's punishing. So there's those edge cases from time to time where you feel limited by the Intel graphics, but there's such edge cases that time to time where you feel limited by the Intel graphics. But there's such edge cases that if I could plug in an external GPU case and then have the capability of playing that game when I need it, but then have a quiet – I assume it's really quiet, right?

Yeah, it's all quiet. It's silent. Yeah, see, I'm all about that. Yeah, the modular. Okay, boom. I like this, Nook. It's tiny, but I've got 32 gigs of RAM in it. It's DDR4, and I've got 128 gig M.2 SSD for the OS and one 1 terabyte M.2 SSD for the home partition. So it has HDMI display port, mini, USB-C, USB 3 Ethernet, and that's it.

Oh, and then a couple of front USB ports. And Wi-Fi and Bluetooth. Right, okay. And a charging USB and Thunderbolt. That's really everything you need, though. Yeah, definitely. Yeah, and the thing's tiny. So if you're familiar with the old form factor NUC, it's a different size.

So if you think of the thinnest NUCs, it's like two of those side by side but a bit thinner and it's got a nifty case so so it goes for uh 650 us greenbacks on new egg see that i'm sold on you think that's a pretty good price that's not bad right actually yeah i suppose so because it comes with the uh i7 processor but I don't think it comes with RAM. Oh, right. Yeah, that's a fair point. You'd have to get RAM and maybe a disk, but I actually, if the performance is there, I think it's worth it to have, these days, especially in here, I'm, like that machine right there, I would love to have it be silent. So interesting, Wimpy.

I will, oh, look at that. I will say, I'm pretty tempted by one of these. Maybe not at the moment because I just don't really need one, but down the road. Down the road, yeah. Yeah, I hope they keep making these. Skull Canyon. Yeah, it's a nice device.

I'm very impressed. And if the external GPU worked, it would also make a pretty good... Yeah, and the only thing about the external GPU, of course, is that you're looking at like five gigabits per second over the thunderbolt interface so when you compare that to pci express it's a lot slower so it's no good hooking up a supremely powerful yeah uh gpu because it won't be able to saturate the bus right so there's i need to do some sums there to figure out what GPU is sensible. The chat room points out too

that it's a mobile i7. But really, Skylake and really even past generation processors, the mobile processors are just incredible. I do notice on the couple of desktops that we built for Linux Fest, they're desktop machines that we're using for different functions

that have six-core processors. That's the new sweet spot. Six cores is so the sweet spot now. I love it. So I'm ready for mobile processors to get that. Yeah, absolutely. So the Intel Nux Skull Canyon. I was thinking that was a good one.

Have you tried any Thunderbolt device, Wimpy? Not yet, no. I haven't got any Thunderbolt devices to plug into it, and I thought I'd wait for the 4.7 kernel to get shipped stable before I start experimenting with that, because I believe that's the version that's going to have all of the new Thunderbolt source in it.

Ooh, man. So I'm going to turn off a couple of filters here on my mic just for a second um so you guys can hear i don't know if you can hear but it probably sounds like probably sounds like rushing air do you hear that can you hear that in the background it is storming like crazy outside right now and i have some filters on my here i just tried to cut some of it out with my mic but uh it is it is and uh if it's following if it's following uh what i believe to be the case what's that for is it forecastio what is that website uh with the like uh what's that website that uh like dark cloud uh pulled yeah forecast.io uh so if i go look at our current location, I think we are about to have a huge thunderstorm hit. Let's hope we keep power, everyone.

The show could be getting disrupted here in a moment. Because I feel like these days, honestly, I feel like I live in a freaking rainforest. I don't know what's up anymore. I just feel like I live where it's constantly storming and rainforest. So anyways, if you hear a loud banging in the background or if we suddenly go away on the live stream, it's because the storming is picking up. It is the Great Flood. Yeah, it's

raining like crazy. Can you see out the window? Can you see out there right now? It's ridiculous. Alright, so we're going to move on. I want to talk about well, a couple of things that came out of Apple's WWDC keynote that are worth mentioning for the open source community. I always feel weird about this after it happens because WWDC comes around and then it's like this big awkward time for us Linux users.

Exactly. Very well said. I feel the same way. Because sometimes some cool stuff is announced. You're like, oh. But you don't want to feel that excited about it. But you're kind of interested and you're still a technologist. So then sometimes you're like motivated to like bash it because you want to like that excited, but you're kind of interested and you're still a technologist. So then sometimes you're motivated to bash it because you want to distance yourself from it?

Right. Is that why you've been Googling ButterFS so much, Chris? Exactly. So there's just a couple of things I wanted to mention. Well, one is a feature that was announced at WWDC that we can already do on Linux. I'll tell you about that. And the other, I bet you can guess what this one is, is a solid response to a rather infamous rant of mine. But first, I'm going to tell you about our friends over at DigitalOcean. Boy, speaking of storage, DigitalOcean has their block storage beta rolling out.

Noah got in on it. Oh, did he? Yeah, I'm not in on it. I'm not yet either. Oh, no, you've got to tell me when you get in. So DigitalOcean is a really easy, straightforward cloud hosting provider. You want to go spin up a Linux rig that's

a bare-bones Linux rig, or one that's got the entire application stacked for whatever you might need. Say you want to go experiment with NextCloud, or say you want to try something like SyncThing or BitTorrent. That's a great idea. Yeah, it's perfect for that. At the price point, starting at $5 a month, the price is easy to swallow, so you could just experiment and test. But then if you want to go into production, well, these systems are all backed by SSDs.

They have data centers all over the world. They have Tier 1 bandwidth. They have 40 gigabit E connectors into the hypervisor. You know, like the NUCs. Or the NUCs. The NICs. The NICs. The NUCs. Not the same thing.

Network interface cards. 40 gigabit E network interface cards. I love that. For $5 a month, you can get 512 megabytes of RAM, a 20 gigabyte SSD, one CPU, and a terabyte of transfer. And then they have the best interface to manage it on top of that. A really slick, easy-to-use UI that's simple for beginners, but powerful enough for actual experts. That's like one of those unicorns that never actually seems to happen.

Whoa, lightning strike. Digital Ocean. There we go. Yeah, Digital Ocean nailed it. You see, that's the cloud right there responding to how awesome Digital Ocean is. They really are awesome. That's the power of the DigitalOcean cloud. Just promise me something.

Come here close. You gotta do me a solid, okay? You gotta use that promo code, DOUnplugged. That way I can pay for that lightning strike hole in the ceiling. DOUnplugged. One word, lowercase, $10 credit. You got that? Thank you. That's all I ask for. Chris needs you, folks.

DigitalOcean.com. Use the promo code DOUnplug for. Chris needs you, folks. DigitalOcean.com. Use the promo code DLUnplugged. And a big thank you to DigitalOcean for sponsoring the Unplugged program. That was funny, man, because I saw that lightning strike out of the corner of my eye. Because it's just right out of the window there. So anyways, by the way, one other thing about DigitalOcean. For those of you that are Kernel Linux crazy, they do have a how to provision and manage Docker with CentOS. What?

It's possible. It's possible over at DigitalOcean. But why? Go check it out. All right, so I teased it just really quickly. Did you guys, did you guys, anybody in the Mumba Room have a guess on what the news out of WWDC is that I actually am kind of happy about? Anybody have a guess? Mumbaroo.

New file format system? Yeah, you got it. The new APFS. Apple's new file system. Copy on write metadata. Native file encryption. Instant cloning. Snapshots.

Don't call it ZFS. It's not ZFS. Don't call it ButterFS. Although, you know what? It's probably very closely more. It's probably more like ButterFS than CFS. Are you saying that it's not going to run well? Ba-dum-bum.

Oh. A couple of interesting things about this. It is, they say, I don't know exactly what this means, but they say it is a 64-bit in-node numbering support system. It has been designed specifically for SSDs and flash storage, not for spinning disks. That's the part I don't quite know what they mean, other than perhaps this is a file system that presumes it will not be using the SATA interface to talk to it. Oh, wouldn't that be interesting? Yeah.

Perhaps it will just talk directly to the memory chips. That's a possibility. It also will not be available for a while. So it's not like it's rolling out anytime soon. But here's the interesting tidbit for the open source community. I normally wouldn't have actually even worth – I wouldn't have mentioned this in the past. But since they actually delivered on Swift, in their documentation right now, they say, just like they kind of promised in the past for other things, a developer preview

of Apple's file system was included in OS X, or now it's supposed to call it Mac OS 10.12. Yeah, come on, Chris. It's their own documentation, Wes. Apple plans to document and publish APFS volume format when the Apple file system is released in 27. Now, they say an open source implementation is not available at this time. But when they publish it in 2017, they plan to open source it. Oh, neat.

Now, the question is, will it be something that already works with Linux? Does it be immediately useful or something that just works on Microsoft SX and is pretty much useless? Yeah, well, probably the best case scenario will be you'll eventually be able to load APFS support like you can load HFS or NTFS support, right? Because you know they're not going to GPL it. Yeah, the best it will be is a FUSE module. Yeah, exactly. And that, I guess, will at least make compatibility a little easier for us Linux users that have to interact with Mac users that somehow mistakenly formatted like a thumb drive or something with this.

I mean, they can't GPL it. We'll pretty much make them make GPL the whole kernel. Yeah, so you know that's not going to happen. But the fact that they are going to open source it's at least notable. And I believe they'll actually do it. What? You don't think so? What?

What? What? They said when they first announced FaceTime that they were going to open source. Oh, true, true. And they never did because they got slapped with a lawsuit. Right. And so it is entirely possible that someone will slap them with a patent claim and say that we own some part of that file system. You can't open source it. and then that shuts that down.

So they've made all this big pronouncement, but they don't actually technically have to do it if someone else comes along and tells them not to. I bet the case with this, though, is they've been working on this for quite a few years. I'm betting that they have been owning a lot of the IP of this as they've gone along because their intention is for this to run from everything from the watch to the desktop and everything in between. So they're probably crossing their eyes and – or dotting the eyes and crossing the teeth. No, you had it right first time. Probably.

You probably did. I find it interesting. Think different, Chris. They talk about doing atomic file transactions. They talk about snapshots and rollbacks. These are all things I like. Trim support. So anyways.

I wonder how long until they think it's production ready. You can't get much worse than HFS. So they don't have a high bar. That's a fair point. Yeah. Is there any indication how long this has been incubating for? I don't know. I think some people speculate.

The number I've heard kicked around is seven years. Wow. No, it has to be less just because I've been actually monitoring their hiring job boards and they've been last three years hiring extensively kernel developers, which I'm assuming

some of them will be for file system stuff. So it's three years tops that I will claim. Yeah. I guess we might know if they open source it because in the case of Swift, when they open source Swift, and some of them will be for file system stuff. So it's three years tops that I will claim. Yeah. I guess we might know if they open source it because in the case of Swift, when they open source Swift, they include the entire commit log. So we might know.

That's a good point. That'd be interesting, yeah. Oh, here comes more thunder. So the other feature, I live streamed the keynote because I figure a lot of people that listen to Jupyter Broadcasting don't have Macs. They have Linux rigs. And the only official stream from Apple, it requires that you use Safari.

Oh, really? Yeah, it's so lame. But always somebody posts like an alternative RTMP stream somewhere. So what I do is I always try to rebroadcast that so that people can just go to the JB Live page on whatever device they want because we're available on the TV. We're available on their mobile. We're available on their web. So I figure just make it easy for our audience that would just want to watch it. Totally.

And so I do commentary along with the chat room and stuff. And one of the things that everybody in the chat room seemed to universally love was this new picture-in-picture feature of Mac OS X or whatever, Mac OS Sierra. Did you see this? Where any HTML5 video could be popped out and then you could drag it and it could stay persistent on top of all applications across all desktops. And fear not, that feature is already available in Linux via

MPV. Nice. Mr. InnoGoGo made a link in the Linux Action Show subreddit where he just gives you the command that you do in MPV, and it can pop up into a little picture-in-picture window like that, and you can move it around. He even shows how you can do it on Windows if you were so inclined. I was going to say, I mean, one of the things we've got, I mean, window managers, man. We should be able to get that down.

I know, right? Yeah. Yeah, so that was like the one feature. I was like, hmm, that's okay. It's okay. This one's cooler, though, the MPV. You don't need to use the window manager to do it. Yeah, right. Mm-hmm. Yeah. Yeah, so that was like the one feature. I was like, hmm, that's okay. It's okay. This one's cooler, though, the MPV. You don't need to use the window manager to do it. Yeah, right.

Mm-hmm. Yeah. I guess other than that, I mean, was there any other features that you heard about? And anybody in the Mumble room, feel free to chime in, too, that you're like, oh, as a Linux user, I kind of would like to have that. Yeah, go ahead.

I feel that they should totally, you know, since we still are dealing with X, maybe show us what they're using. That will be cool. Actually useful, maybe. I don't know. Yeah, I didn't really, I don't want to go, I just want to make a comment. I think the situation for Mac hardware right now is pretty poor.

It really is. And I think there's an opportunity here for Linux because a lot of the things that drive people to – originally that drove people to macOS was a nice graphical interface with a strong Unix underpinning, which is a lot like elementary OS, for example, or something like that, or Ubuntu desktop that we're about to talk about, that people are now aware of these alternatives. And it allows you to buy from any hardware vendor you want. And the problem is, in a lot of cases, Apple sets a pretty high bar for hardware quality, but it's not an impossible bar to surpass. And more and more manufacturers are doing so or

getting close. And the problem is, is legitimately, you look at some of Apple's high-end hardware right now, just as a thought experiment. Let's say you decided you needed to get yourself a very high-end workstation from Apple. Well, first of all, the first thing you'll notice when you go to buy from Apple is the only thing they're really advertising is their iOS devices. I mean it's a little discouraging. You can't really – Oh, people still buy desktop computers? Yeah. So if we go look at the Mac, for example, and say we want to get the Mac Pro, because we want something that has more than four cores. So I want something that has more than four

cores, maybe because I have a large encoding workflow, or there's all kinds of situations where you need a lot of CPU. So if you want to get the Mac Pro right now, you are legitimately forced to use hardware that came out like three years ago, and the starting price is $3,000. Oh, man. And that's with not a very competitive system. This system is significantly less competitive than the computer I am showing it on right now, legitimately. So if I wanted to go – this is just a – I am displaying this on a quad-core – I'm sorry, a six-core Skylake system, and this is three generations old quad-core,

starting at $3,000. Now, what's even more egregious is, besides the 12 gigs of RAM and the only 256 gigabytes of storage, what's actually the most egregious thing on this page is the AMD Fire Pro D300 GPU with two gigs of RAM. It is an ancient an ancient graphics card. And to upgrade, the best thing you can get is Fire D700s, which are like 4-generational GPUs.

They were legitimately a generation old when this Mac Pro shipped. And the best you can get is a 4-generational graphics card, two of them for $1,000. That's crazy you think there'd be a lot of professionals that want to do especially with like you know like machine learning etc it's like there's so many stuff where you would we really do want like a nice gpu for professional work i will claim that this is actually part of the smart move from apple as a developer what has screwed up Windows developers has been that they're constantly shifting. You don't have a stable platform. People choose Ubuntu because they have the LTS and they can be sure that they target that. The fact that Apple doesn't go

crazy on adapting new things so fast allows for someone to master and craft for that platform and then sell on that platform, which means profitability. At the end of the day, and then sell on that platform, which means profitability. I would argue that Windows XP and Windows 7 are pretty reliable targets as far as developer platforms to target. But also, I think your argument would be probably easier to swallow if they reduce the price over time. They're still charging the original price when they launched that thing years ago. And I think the core point here is your scenario may be okay for some developers,

but I think there's a group of people out there, even people who have enough money, who look at that and say, that is just unfair to the customer. And I think I... Okay, go ahead. Sorry, I'm so sorry, but in general, their platform does pay more.

And you develop for Windows, the money you make, you develop for Android, the money you make, and you develop for iOS, the money you make, and you do your math. I think perhaps people like Leo Laporte are a harbinger of exactly what I'm talking about. People who can afford to spend more but look at the commitment that Apple has to the Mac platform and go, commitment that Apple has to the Mac platform and go, Apple is way more committed to the iOS platform and now to their new services division and potentially to even building a car. And they're not committed to their desktop and hardware platform anymore as they have been in the past. And so I can see very easily how people like Leo Laporte would look at that and go, well, I don't want something from the Microsoft camp because they don't like what they're doing with Windows. And I don't like what Apple is doing with their hardware.

So I'm just going to go with Linux. I'll find a desktop eventually that works for me. And it has to be a certain enthusiast that's willing to do this. But I'll find a desktop environment that works for me. And I'll find a piece of hardware that works for me. And I will now just manage that marriage of hardware and software and not outsource it to Apple like I have in the past. And I think that's exactly what's happening right now.

I just think it's a slow trickle that's going to speed up, especially as they continue to ignore their hardware. They just did their only one big event a year, one that is for developers, and they didn't release a single piece of hardware for developers to use to develop for their platform. They're all developers in F4s to buy hardware that has years-old components in it. That's a pretty bad state. I mean, I know it has a cool Apple glowy logo on it, but at a certain point,

that brand attachment only goes so far, and the practical person in you goes, wow, I can go get a ThinkPad or a Sputnik or a System76 or X. That's where the System76 really starts looking attractive. Yeah. Or similar offerings. The devices that people are targeting are also inferior devices anyway.

Aren't they? That's kind of what the point is. It's like if I'm targeting an inferior device and I can test it or run it in the developer machine I have right now, something's very wrong then. I can target those devices that they're building right now, the phones and the TVs and whatever the car system is. They're all going to be inferior, and that's where the clients are. So if me as a developer, I'm buying a Mac as a developer, I don't need the super fancy thing. I need access to the platform.

I need better software tools, which is the services that they are improving. I guess we'll just have to wait and see. Yeah, pretty much. So did you guys hear about this rumor that Samsung is going to completely dump Android? Did you see this? Yeah. How could this be? But apparently this is from Fossbytes.com.

So I don't know. I don't know where they got their source. But they claim that the plan is – Oh, from the Korea Times. Yeah. Oh, yeah. There you go. Thank you. From the Korea Times.

They say that they did a launch in India. They sold lots of devices, millions of devices. The devices are being very competitive, they've said. And they've had a low return rate. Customers are happy with the Tizen-based phones. are happy with the Tizen-based phones. And according to this, they seem to think it's a great opportunity to go from their watches to the phones

to their refrigerators to their television, Tizen, all the things. They say they've had 64 million phones sold that are running Tizen, and that it's proving to be very competitive. That's a lot of phones. I mean, not to the whole phone market, but just as a unit. That's interesting.

I wonder how well they really do work. You think this is legit, Popey? Yep. I called it. I've been saying this for a very long time, that what Samsung are doing with their S apps is building a community and a familiarity with their applications and their app ecosystem and locking people into their app ecosystem rather than the Android ecosystem.

Normal people know two things. They know Apple phones and Samsung phones. Normals don't know about all the others. And when you're on a Samsung phone, you have all these Samsung things. My theory has been for some time that samsung will just swap out the underneath and they'll real bit rebuild all their s apps for tizen and the user won't care because they'll still be locked into all the same s apps that they had on android but on tizen we can stop complaining about touchwiz and if you look at the tizen phone it looks a lot like touchwiz on android

but hasn't the s apps been universally rejected by users don't people hate the s apps um i don't know i mean the the places where these people who apparently have bought millions of tyson phones don't really have the option of all the g apps so what's the size of the people that hate them versus the size of people who just use them because they're there? I don't even know. Yeah, default is game. Yeah, default is game. However, what I would say is I subscribe to a monthly mail that I get about the top 20 apps in the Tizen store. Oh.

And I get this email every month. I signed up to be a developer on the Tizen store just so I can get this email. And it tells me which are the apps. And I've been looking at this for a year, and it's the same 10 or 20 apps in the store for like the last year every so often a new one will come in a new game or a new media player app or something but it's basically the same 20 apps so i don't i don't think they have a massive developer community and i wouldn't be surprised if they're using the android runtime to leverage android applications that's what i was thinking yeah so the number one app in there is whatsapp and i'm pretty certain that's an android app and not they're using the Android runtime to leverage Android applications. That's what I was thinking.

So the number one app in there is WhatsApp, and I'm pretty certain that's an Android app and not actually a native Tizen app. Yeah, that's the response to the chat room, too, that, of course, if they can run Android apps, do you think they could even do the Play Store API? Is that possible? It's possible. How can you run Play Store API on another OS?

The Play Store is Java. Tizen is JavaScript and HTML. But everything all uses the Linux kernel all in the base anyway. Android is pretty much Linux plus Java. Tizen is Linux plus JavaScript. Okay. So they can just shove the Android, or at least their Java runtime in, and it'll be working. It's not going to count the display server either.

Tizen can use Wayland, just like Sailfish. The Surface Flinger, the display server for Android. Yeah, well, I guess they can. Well, I wouldn't want to use it. I would more want to go the Sailfish way and do Wayland on the mobile platform anyway. So, Heavens, there was a report just like a few days ago.

Just came out, said mobile users are not really installing apps anymore. They get 5, 6, 7 apps that they use on a daily basis or frequently, and then they pretty much, they might install something, but they never use it. Do you think that factors into how Tizen could perhaps blow up larger than Android? Because you say you think it could

go bigger than Android. Mainly because of the developer base or the developers for JavaScript versus the enterprise Java who have just migrated over to Android as a benefit. But I guess they could

get even bigger because one, JavaScript applications are all instant apps versus Android's new Android M's or what ends instant app new feature. Huh. Well, I think it's – Anyway, come on. Google's been pouring millions of dollars into JavaScript execution. Better than Java can ever be. And, you know, lawsuit crap going on.

I think we look at it from too much of a technical possibility standpoint. I think maybe outside the U.S. this could happen, but I just don't ever see this happening to – as long as they are trying to – as long as Samsung in the U.S. is trying to go against the iPhone, I don't see it happening. Chris, can I just point out how many billions of people live outside the U.S.? No, I can see this totally taking – that's exactly my point is maybe in terms of scale, essentially they will have done this percentage-wise just by doing this outside the U.S. Right. But I don't think inside the U.S. or any I should say, where they're commercially going after iOS devices. It just doesn't seem like that's going to be happening.

That you can switch to a new platform and be competitive? Yeah. I don't know. It just doesn't seem like it. To target the TV and Samsung TVs are more than the iOS or whatever other TV. Well, that's a very compelling argument that nobody really has right now. Sure, target the TV, target the phones, target the cars, target the watches,

and see where it takes them. And they have their Chromebooks, too. So if they, you know, step there. That'd be interesting. That would be interesting. Maybe soon we'll be running a Tizen desktop fork thing right here on the show. All right.

All right. I'll get into that. I would love to import. Like, even if I had to import it, I'd love to get that. Right? Yeah. It's funny because I say that I don't think Samsung will ever do this in the U.S. Personally, I'd love it if they do it. It'd be fun to try.

Yeah. I would love it. Like, from an enthusiast standpoint, I'd be like, yes, please. A large-scale manufacturer who can build high-quality products relatively and ship them at scale where I can actually go into a store and buy them. Yeah, that would be great. But I don't know. I just linked in IRC the Intel demonstrating Tizen on an Intel i7 Ultrabook in 2013.

All right, Hawkins, would you like to have the last word on this one? I just want to say that with Samsung, if we think back about the Nokia Symbian operating system, which Samsung also had access to, they didn't do any development when they did have access to the Symbian operating system. It's true. Yeah.

Maybe they're going to compete with Google in the like, the ad space and make Advertizen. Ooh. Ba-dum-bum. And on that note, why don't we – Let's move on. Let's talk about Ting. Let's talk about Ting. You know, whenever you get your device, be it a Tizen device, an Android device, an iOS device, a feature phone, or even a BlackBerry.

Heck, I don't care. Just here's – Ting is honey badger about this. It's real simple. They got CDMA and GSM, and you pay for what you use. So you get a SIM card, you put it in your device, and you pay for what you use. It's $6 for the line. You want to have 10 lines?

It's $6. You want to have one line? It's $6. You just pay for your usage. There's no contract. It's really this simple. I can't even believe I, even after all this time, I can't really believe it's that simple.

Hopefully, hopefully you do, and you're willing to try it, because it really is a game changer. I save over $2,000 a year by going to Ting. Average Ting monthly bill per line, $23. That's for your smartphone. They just take your minutes, your messages, and your megabytes, add them up, and that's what you pay. You've got CDMA and GSM to choose from. You can either buy a device from Ting directly, which is unlocked, and you own it. No contract. and you can either buy a device from Ting directly, which is unlocked and you own it, there's no contract,

or you can bring a device and they'll give you a Ting credit if you go to linux.ting.com. Linux.ting.com will also take money off your first device. They have great customer service. You speak to real human beings. They have a super nice dashboard that gives you features you never thought

a mobile carrier would give you right in the webpage. I love that. Same with the Android app too. I don't use the Android app as much or there's also an iOS app. But that web page, I just love it. I love it. And I love the – I just – Everything you need to do is right there.

Just as an example, we have a couple of extra lines we turn on every now and then just when we have people in town. And we can disable them. The process is very straightforward. I love it. And the fact that we have CDMA and GSM means that when we're up north at LinuxFast Northwest, we use the CDMA network. And when we're down here in Arlington, we use the GSM network. And we get amazing service in both locations.

And you get to choose. I love it. Try going there right now. Check them out, linux.ting.com. $6 per line. That's nuts. You can also call them at 1-855-TING-FTW. Talk to a real human being.

It is mobile that makes sense. linux.ting.com. And up on their blog right now, they have an article about using Ting for gamers, which is really cool. So we're going to talk about Snap Packages. We're going to talk about Snap Packages. But first, before we get into the Snap Package discussion, which this is huge news today, there's also a little bit of news this week, just something I wanted to touch on before we run away with the rest of the show. And that's the new Loki beta that's out.

The elementary OS folks put out a beta of Loki. It's totes early stuff here. It is not meant for your daily driver. You shouldn't use it if you are expecting a stable, reliable desktop. It's really to give them feedback. That's what this is about. But it's something that's kind of interesting because it's sort of a closely watched distribution because it doesn't do traditional betas and development in the sort of

daily release tradition that a lot of distributions do. Like when there's a new version of Ubuntu itself coming out, at any point in the development cycle, you could go get the latest ISO. Or you could go jump into Fedora Rahe at any point or Debian Unstable, etc., etc., etc., etc., blah, blah, blah, blah, blah. But that's not how

elementary OS does it. So when they actually put a flag down on the ground and say, okay, the beta's out. You can go try it. We had to give it a look. And so besides it reiterating that beta means beta, they say that they've shipped a new set of toys for developers. They've upgraded the

notifications, better notifications. No unsafe software sources. But also there's a few things like a new app center or whatever they call it in there stuff like that tell me about it Wes what do you think what are your first impressions you've been running it this morning yes I have I'm doing it right here

doing all my show prep etc on it so far mostly good I mean there's like some small things you can tell that you know they're still working stuff out like you know there's no default associated thing with.deb files anymore so if you want to install those you can't double click you need to go into the terminal etc yeah you can change that if you want to uh but i will say it installed easily very quick um it it really didn't break anything

installed nicely on the like three other operating systems i have here loads really fast it has a very smooth like graphical not a lot of like you don't you don't see the the frame buffer terminal or anything like that you know it really goes like right to the desktop loads up and in typical elementary style the animations are smooth you know they're not might not be a lot there and i admit for my own tastes elementary is just enough different from gnome that it's like the the commands are different and the shortcuts are different right like the layout but it's similar uh but it's really nice and like you know just all the little things you can tell it's polished you can tell like every feature that they've put in they really want to be nice and to be complete so do you when you're using it do you have the

sense that it is indeed a um ubuntu family member or does it feel like its own beast and so what i'm asking you is, do you sort of have the background confidence knowing that you're using a desktop that's part of a larger ecosystem or does it feel like it's sort of off on its own island? On the surface, I mean, it doesn't. There's not a lot of clues.

Here I am looking at the app center and it's very minimally branded. It's just kind of – yeah. But as more of a power user, I do appreciate that I can just open a terminal. AppDigit is right there. Let me see that. I'm familiar with everything. So this here is the – they call it the app center.

We don't have a visual of it because Wes's machine isn't hooked up to the capture. We'll have to work on that. So this is their own custom app that... Or is it based on GNOME software? I'm not sure. We should look into that. I think it's actually based on GNOME software.

It does look similar, yeah. Yeah, but it looks like it has much nicer categories, broader... So that's a new feature for sure. They have the App Hub as well. What is the App Hub, Rotten? It's a new thing that sure. They have the app hub as well. What is the app hub, Rotten? It's a new thing that they're going to do, which is like the app center is going to be like the store that you get. The app hub is where everything is going to be provided from, and like a third-party store type thing,

so that you push to the app hub, the app hub pushes packages to the app center that people can install. I see. Wes, try going into the settings or switchboard or whatever they call it there and check out the new sharing settings in there. I guess there's a couple other things. There's new online account, plug-ins, printer support replacing the old one, and there's also new mouse. Oh, yeah. Here we go. Yeah. So what do you see when you go into sharing? The first thing is media libraries.

You can just choose to share music, videos, pictures. They're all turned on by default. Oh, cool. They're all turned on by default. The whole thing is disabled if you enable it, all of those are turned on by default. Have you checked out, have you got any new notifications? I guess a new notification indicator was added to the panel. I have seen a few, and I'm waiting for some

processes I'd run with sudo to complete. It would just pop up a little like, oh, hey, your process is done. That's nice, isn't it? Yes, it's really it feels very slick. Yeah, I like how Fedora does that too. I do. But just for the terminal. So where is their notification tray thingy at? They say it goes up in the – oh, okay, yeah.

Oh, yeah, it's this little bell right here, and it has a nice Do Not Disturb toggle. Have you seen that? Yeah, that seems like it will be very useful. It's really clean looking. It's nice. Yeah, that's the thing. It's like, you know, as someone who does like to tinker and I run Arch, et cetera, et cetera, I can see where some of these things

might bug me or I might be like, I really wish I had that feature or I could do it. But if you just are using the computer or like especially kind of this laptop, I use it for show stuff, but it's not my daily driver. So I don't do a lot of my main project work,

development, anything on it. So if I just want to browse the web and have a laptop that connects to Wi-Fi and suspends and comes back every time, it seems like it might not be a bad fit. Yeah. Anybody else have thoughts about elementary OS and the Mumble before we move along?

I'll probably... Yeah, I think... Yeah, go ahead. Just one quick note. If you're wanting to use PPAs, you need to install the app repository. Yes, you do. Because by default, they disabled it. So there are some considerations like that for power users that may be more inconvenient than convenient for you.

That seems like a pretty clear message there, that it's not for power users. Yeah. And that's okay, too. That's fine. I mean, it's not really, not every distribution out there needs to be for me. No, right, absolutely.

I may not use it, but it might be something I could recommend for my mom to use no problem yeah or you know my son uh is running the last release of elementary os and he's been running it for over a year just totally solid and there's sometimes where like you need that you know you might not like their aesthetic but they do have an aesthetic and there's some users especially you know people who maybe are more designers or artists or like who have have an expectation from other operating systems that it is clean or it is simple or it is just pretty in some way and elementary might fit there too i think when you look at when you look at

elements of it i look at that and go uh if i if i use my computers differently i would i could see myself really being a good elementary os user candidate because i do like the consistent visual um language that's just everywhere through the whole thing. I like that they take care of certain things like the sharing stuff there, like the notification stuff there, the thought that goes into it, the very consistence from the moment it powers up to the desktop. It's a very consistent feeling. That is all indicative of a lot of intention. Exactly. And that intention I respect a lot of intention. Exactly. And that intention, I respect a lot.

And so when I'm using something and it delights me in a way that I go, oh, wow, somebody really thought about that. And I have just a brief moment of insight into, they must have really agonized over this choice. And I think to myself, I'm so appreciative that they did that. I do enjoy using systems that make me feel that way. And that's one of the reasons I like using GNOME, because I often feel like there's things in GNOME 3 that do that for me.

So, there is a part of me that wishes I could be the perfect user for this. But then, of course, the software availability would kill me. Just kill me. Because I would want PPAs, and I would want to add stuff all the time. And I, all that. And you can't do it.

One install. I feel like I'm obviously going against some of those intentions. Are you going to break some things? Are you going to have to... Those very intentions I claim to think were great would seem to be in violation. However, however,

that all could be changing. Right. There are new things on the horizon. There are things afoot. Like we mentioned in the intro, there is apparently a Linux packaging cold war that we have been on the front line. Where do you stand? Where do you stand in the war? Are we being cray-cray? Are we overstating it? Or is there actually really some interesting developments?

All of those things are true. By the way, links to the Loki beta if you guys want to grab it and let us know what you think. is there actually really some interesting developments? All of those things are true. By the way, links to the Loki beta if you guys want to grab it and let us know what you think. Have we had a couple of dropouts? Oh, okay. Yeah, sorry, guys. The storm is picking up. Sorry about that.

Can you hear us okay, Wimpy? Is it all right? I can, but I remember that you lost a load of stuff once. Yeah, we have a backup recording going. I tell you what, this new OBS rig has worked out unbelievably well. I was just talking on the live stream, it runs 24-7 and never crashes. It just runs, runs, runs.

And Wirecast could not live up to that. It has shown itself to live up to a workload that Wirecast could never sustain. That's awesome. Wirecast is really, to use it fully, is a $1,000 product that's been around for a lot of years. And OBS is, in terms of, you know, lifecycle, they rebuilt it not too long ago.

It's still kind of new. Right. And yet when they seem to do releases, at least the versions we've installed so far have been extremely reliable. Maybe not – they don't have necessarily all the whiz-bang features I need at this moment. But the fact that thing – I mean I don't even – it's probably been 20 days that it's ran solid. It's really something. And we had some issues on Sunday and it wasn't that machine.

It was Comcast. And when we do have problems, it tends to be Comcast. Go figure. Open source for the win. Exactly. It's really amazing. It really is amazing. And it helps that –

What a cool project. It helps, too, that it is available on the Linux. A solid operating system. Yep, because Linux is also very solid. So I think a solid piece of software running on top of a solid operating system is just a really great combo. And thank you. Really, the credit there for that system being as good as it is goes really to Rikai and Noah, who stayed up all hours of the night, several nights in a row, making it work. And failed components.

Yeah, and Googling like crazy. So really, the credit goes to the beard and to Brown Bear. But we shall move on to Canonical's big announcement, which may solve my software availability complaints for elementary OS. But first let's mention Linux Academy. In fact, go to linuxacademy.com slash unplugged. All of you right now, go there and watch it. Test the link because we had a bit of an outage the other night.

Oh, no. Yeah, so help me test it. linuxacademy.com slash unplugged and dig around on their website. They have 2,471 self-paced courses where you can obtain experience and learn as you go with scenario-based labs that put you in the middle of everyday common tasks, instructor mentoring when you need help, and graded server exercises that automatically grades your actions so you can see how well you did, which is really nice. Really, really nice,

especially for those of us that don't test well. They, of course, were on Python and, of course, Linux and all the technology around Linux and Android and that DevOps category. OpenStack, yep, yep, that's a thing. Python, really? Oh, AWS, for sure. You know, they've also recently added something they're calling social cards, which— What is this? Tell me more. I don't know. My first thought was

like, what? Some cards? What? No. They're like no cards for learning. And you can fork them and build upon them and there's a reputation system built in and all of this now in Linux Academy. They have a great community of people who are all super motivated and tons of them are Jupyter Broadcasting

members that are now going to be, well, if you want at least, involved in the process of making these cards better. And they're doing a $200 Amazon gift card giveaway for like the first 10 best stacks or something. So you can get in early. What a cool community. Yeah. So get started by going to linuxacademy.com slash unplugged. It comes with your own server, seven plus distros to choose from, modifies the courseware, and the server spins it up on demand.

Enhanced learning plans, detailed comprehensive notes you can download and keep with you. And like I mentioned, instructor mentoring for topics that most other online learning sites would completely gloss over. Linuxacademy.com slash unplugged. Now, I know we have talked a lot about universal software installation, and I hope the audience is not worn out. The sense I get is you all are sort of – the majority of you think, I've heard this one before. I don't have any sense that this is going to go anywhere. Why are we talking about this? That's kind of the general sense I've gotten back from the audience.

However, I have felt for a while now that there is really some serious momentum here, more so than we've seen in the past. XDG app, now known as Flatpaks, and of course AppImage. And we've been talking a lot about Snap packages on Ubuntu. In the abstract. Now today, today, some pretty big news came out. I kind of smelt it coming, and you might have as well. Snap packages are launching on multiple Linux distributions. Essentially, the Snap is going to be available for pretty much all major distros.

Launching with some great partners right out of the gate, including Docker and Telegram packages. And I think Corita is in there and Red Hat Enterprise Linux and Arch Linux. And all of those distros can now install Snap packages. It's snappy time, everybody. Yay! Snap packages are here for everybody. And I look at this and I go, brilliant move by Canonical. In fact, Mark Shuttleworth said, one thing is definitively clear about today's announcement.

I'm paraphrasing. This is not about Ubuntu. This is about all the other Linux distros. Now, I thought that was an interesting thing for Mark to sort of stress. This is not, this is like the antithesis of being about Ubuntu is what he said. Now,

yeah, he also pointed out the Flatpak, the new GNOME XEG app initiative is 95% of the contributors came from one Red Hat employee. I love it when he does that. I love that. He said he'd be delighted if they'd work with us. God, that's so great. That's so good.

Yeah. So the Document Foundation has got LibreOffice. Dell Mozilla is partnering up with them, Samsung. Document Foundation's got LibreOffice. Dell Mozilla is partnering up with them, Samsung. And Arch and Fedora can run and install Snap packages. I installed it right before the show. You have installed an elementary OS.

I do. I'm not sure it's 100% working, but it's installed anyway. Did you get the Telegram pad? You know, it's giving me an error when I try to launch it, but the rest is working. I got SnapD going. I can do SnapList. I can do SnapInstall. Interesting. It's probably user error. user error rotten let's start with you and then i'm going to bounce around some questions uh you've gotten you've gotten your hands on it during the show what are your initial impressions yeah so i started using it earlier in the show and uh snap is great and

two things are really shocking. NextCloud came out. They announced a new version today. There is now a Snap for it today. Wow. And I just installed Krita, which by default when you use apt, it installs like 500 megs of dependencies. The Snap was only 98 megs. Wow.

Okay. All right. I don't know how that happened. Okay. But that sounds very promising. I wanted to ask Wimpy, because I know he's had some experience recently with Snap packages. Wimpy, do you think it's now just come down to a race of who can get the most packages by the biggest names?

Because you've got Flatpak, AppImage, Or app image orbit orbital or whatever it is um and now snaps and it seems like whichever one since they all work across all the distributions whichever one ends up being the sort of go-to to have the most and best software is going to be the winner don't you think maybe yeah um definitely there's some first mover advantage there and there's been a lot of people coalescing around uh snaps and snappy um you know like qnap recently announced that their line of nas devices are going to be moving to ubuntu and snaps yeah that's going to be packaging all of their ancillary services and iot bits and bobs as snaps rather than shipping them in the traditional way. So I think snaps might be edging out at the moment, but time will tell. Yeah, it seems also the KDE project is experimenting with packaging neon as a snap.

Wow. Which is really – That's interesting. Yeah. Yeah. And the other thing that Ubuntu have done is they've reached out to people in the other distributions to get Snap working on the other distros. And I'm not sure if that's happened the other way. So the fact that Snap works in all these different places is because Ubuntu developers and community members have made the effort to try and make that happen, to make it the cross-distribution. So they're reaching out to those distros and saying, hey, would you like to work with us on this? So that's a huge component of it, too, getting more and more adoption.

And not just cross-distribution, but also across the flavors within Ubuntu and inviting people. And, you know, there's been a real effort the last two weeks. It's a shame I've been on holiday for most of it. But, yeah, there's been a real effort to involve people and bring them up to speed and share knowledge and tips and tricks and push the platform forwards and learn from one another. Now, what about the concern that the back backend isn't open source to Snap's packages? Popey, is that a thing? Is that a thing people are complaining about yet? Is that a misunderstanding? Which backend?

So SnapD is open source. So Snap, help me understand. The registry, right? Yeah, is there like a central registry I can search for Snap packages and that's locked up? There is a store and we run the store and i don't know what the license on that is it currently isn't free software i don't know what will happen but the who runs the store doesn't impact who can download a snap and just install

it on their system that's right that's the two things are not dependent uh you can certainly sideload snaps you know you can grab a snap from anywhere so for example the initially the creta developers uh put a snap as a downloadable file on their website or like i installed just like a dev i just downloaded the uh i just downloaded the telegram snap to my desktop right so you can certainly do that and yeah currently the store is non-free i don't know what the plan is for that okay and i don't know I don't know what the plan is for that. Okay. And I don't know what the plan is for supporting multiple different stores either. I know that discussions have been happening there because some people will want to have all the devices within their own estate only talking to other devices within their own estate. And may not want them to reach out of their network and go to some third-party store and allow their users to install random, you know, snaps that could contain games or, you know, stuff they just

don't want. And so I know those discussions have happened, but it's early days and I don't know where that is at the moment. Then we have SnapD, which is an open source piece of software that seems to be the thing that has to be specifically packaged for each distribution to make all of this work, to make the cross-distribution component work. So there is an Arch package, there is a Fedora package, there is an OpenSUSE package of SnapD, and that is the sort of bits and glue.

The shim layer. Yeah, you needed to install snaps. Am I tracking that right, Poppy? Yeah. So there's multiple parts to it. The SnapD part is, yeah, the bit that needed to be there's multiple parts to it the the snapd part is yeah the bit that needed to be ported and that's the bit that you know we already have pre-installed on ubuntu so you know we're halfway there and that's a go demon right

yeah it's written in go um and the the snaps themselves get installed and removed by snapd and there's a front end to snapd which is just called snap so you can do commands like snap list to list all the snaps that you have installed and that would be the same as you using apt or d package to see what you have installed and then you would do something like snap find which is where you go out to find out what packages are available in the store or in the archive or whatever you want to call it. And that's the equivalent of apt search or whatever. I feel like people are going to misinterpret that as,

oh, well, snaps are trying to replace my distribution's package manager. Could you already see how people, oh, look at Canonical. They think they're coming out with a package manager for all distros. It's not really – go ahead. I don't know what i what i think would be preferable is if you see it from the point of view that your distribution is giving your users the the the platform your platform which is you know there's the value that you provide in that whichever that is you know in arch that's package freshness. In Fedora, that's clean upstream GNOME.

And in Gen 2, it's build it the way you want it. However, whatever your platform is, what we're providing is the ability for the users to get absolutely the freshest things from upstream in a secure manner. Yeah, I actually I don't know what universal package format is going to win out, but this to me

If we can even just get one, that would be really, really, I would love to have one that has a company behind it that has a reputation in the open source community. That seems like a pretty important. Linux has a history of resolving these things and only having one of everything. That's right, isn't it? We only have one of everything.

Very good at that. But we do have it. But I could see a dominant one. I could see a dominant one emerging. And the... Boy, excuse me. The launch across multiple distros on day one. Huge. Smooth. Real smooth. The fact that multiple distros on day one, huge, smooth, real smooth.

The fact that I could just – it was just in the AUR immediately and I could just go grab it and install the Telegram snap. I was kind of surprised that didn't leak out, that people didn't spot things appearing in various archives and people say, look at this. What's this doing here? I had heard some rumbling. I had heard some rumbling. Oh, okay, so maybe people didn't know. Some people thought it. Me and my little Ubuntu bubble, I didn't spot that.

I thought, I just, I don't actually know if this is going to be it, but even if it's a well-supported one, the momentum you've already seen behind Snap packages is really good. Oh, you like, what do you got there? What do you got there? Look at that. Are you pasting from the web? Yes, I am. Okay, so read that error message on your there? Look at that. Are you pasting from the web? Yes, I am.

Okay, so read that error message on your screen. So I'm pasting into the terminal from the web. Copying commands into the terminal can be dangerous. Be sure you understand what each part of this command does, and then in big red button, paste anyway, which I am about to hit. Look at Loki warning you about the dangers of pasting commands from the web, which is ironic because a couple of, like LibreOffice was the one. HamRadio was telling me he was downloading

LibreOffice using a snap package and the post just tells you to use wget and has the app get install command and all this. Just copy and paste this into your terminal and you'll be good to go. You'll be fine. Yeah, that's not a good idea. The warning about using, copying and pasting things in the terminal

was actually in the last version, too. I wondered about that. I think, yeah. It's still nice to see. I think that's a pretty huge development, getting snap packages for all distros. I don't know where it's going to go for sure. But to me, I have always struggled with this is really a first world podcaster problem. But when I talk about an app or we do a desktop app pick on the Linux Action Show, I experiment with, well, what do I do here?

Do I link to the open source of build service? Do I link to the AUR? Do I link to a DEB and an RPM? And I have experimented with doing all of them. And it just feels ridiculous. And so what I have defaulted to doing is I generally link to maybe the Ubuntu version somewhere and the version I installed from the AUR or something like that. Right.

Because that's the one I did, so you can reproduce what I've done. And the ability now to say, just go here and download this Snap, and whatever distro you've chosen, it's going to work for you now, like Telegram and HeruPad and Mumble, is going to be a huge shift. Drupad and Mumble is going to be a huge shift. And if this all goes well enough where it's adopted by a majority of application publishers, there could become a point in time where we simply just remember a time before Snap packages and then there's life after Snap packages. Exactly. I will say as well, like I'm just – I've been using LexD a lot, which is also another Go daemon written by Ubuntu. And I've been quite lexd a lot which also is also another go daemon written by ubuntu and and I've been quite pleased with it so I'm I think you know if snapd is of similar quality I could be okay

with this as being the universal package. Hopi do you have any do you know anything about this is go sort of like canonicals uh not to be punny but go to language uh? Yeah it's funny I remember when I very first started at the company, so three and a bit years ago, I remember Mark evangelizing about Go and saying it was the next big thing. And this is obviously pre some of the other new things that have come out from Google and Apple and Microsoft and so on. But he was evangelizing it for a while some time ago, and now a lot of the developers have picked up on that

and said, actually, it is quite a nice language to develop in. And so a lot of the tools, like, but even if someone's just hacking together a little command line tool, they'll do it in Go. Or a daemon, they'll do it in Go. So it seems quite popular with developers as well. Yeah, I get a lot of good feedback about Go and Rust and Dakota Radio. And people seem to feel like Go is really something special

and wonder why Google doesn't use it more themselves. Rod, I guess I'll give you a chance to chime in here as somebody who's packaging up applications for different distributions. This has got to be potentially offering a big relief. Now, how much do you expect users to actually go out and enable Snap support on their distro of choice, though? Well, hopefully it'll become, in a while from now, I expect, but hopefully it will become a normal thing.

Even if Ubuntu, the base of Ubuntu and all the flavors were just Sn and everybody else used flat packs or app images or whatever, I don't care. That's at least tenfold better than what I currently have to do. So for my – every time I make a new release, I have – I make 19 packages for everything. Wow. Dang. That. Dang. That is rough. That's not even the maintenance where I have to contact the maintainers of each distro for packages we don't make. So I have to deal with those too.

And then I have to make sure that the package builds and the e-builds are all up to date and things like that, which they're not always. And then that doesn't even take into account that I have to make new versions. I have to make different packages for each version of a distro. So like Fedora 22, Fedora 23. So every time I make a new release of our app, we also have to make a new package for that distro and every version of that distro

that is still currently supported. So for Ubuntu, we have like four. So we have so many packages we have to make and just and if we have three new universal things that take over i would be i would be so ecstatic yeah sure just to have to have three rather than 20 or whatever wow it's funny on the on the subject of um packaging uh we've done a couple of events last tuesday and this tuesday or a couple of events last Tuesday and this Tuesday, or a couple of Tuesdays ago and this Tuesday. We did a thing called the Snappy Playpen. And the idea was that we would create a GitHub repo in which we just dump the configuration file. So to make a Snappy package, there's a YAML file, a plain text YAML file, that describes the build of the application.

It's pretty straightforward. Anyone can read it, and it's not hard to write up yeah and but we you know snap is quite new and we've been finding all the various edges and pointy sharp places that we mustn't touch and it's a learning exercise for us as well as you know people like rotten who is you know developing his apps and wants to package them up so we've been doing this and packaging a random bunch of applications some of them less random kind of well chosen applications in order to um find those pointy sharp places so that we can educate other people on how to package stuff up so when they say okay i've got my gnome app or i've got my kde app or

i've got my you know c++ open gl app how KDE app or I've got my C++ OpenGL app. How do I package it up as a snap? And we could just point them at this GitHub repo. And in there are examples. So we're not necessarily uploading all of these to the store. They're upstream programs. Like, for example, I packaged up Minetest, the free software Minecraft clone. So if you want to see how to package up an open

gl c++ application and it and it packages packages it from the tip of git so you could run this command snapcraft in order to build yourself a snap package which is you know bleeding edge nightly build so you could run a cron job potentially a cron job which built you a snap package overnight every night and uploaded it to a nightly version to the store so every day your users get the freshest version uh rotten you had a response to uh to that and there's a bit of a but in it i think no not really it's it's more of hopefulness. So the point that Pepe was making about how the upstream is controlling it, that is something I've been begging for for at least 10 years now. Even though – there was never even on the radar, but I've been hoping for something like that for years.

And when I contact a maintainer of an app to get my app updated in the repos, it could take anywhere from a couple days to a year and a half before I get a reply. It's a horrible process that I understand why people have a problem with packaging for Linux, really. But with the fact that Snappy allows for the upstream maintainers and developers to control when stuff gets updated and pushed to the users, that is something I've been waiting for. And I am so excited. I'm, like, beside myself. So what you just touched on is exactly – I've talked about it before, but we've been recently talking about NextCloud and Frank.

And Frank and I had an off-the-record conversation that was sort of on the record in the sense that he knows I've been telling this story for years now. He had a very, you know, sort of unfortunate fight with Debian for a while because they were shipping such old versions of own cloud in the repo that users were installing version 7 when version 8 would be out or something like, you know, or version 5 when version 8 would be out was actually the case. It was really bad. Riddle with security, no longer getting updates.

And he said, you know, really the way our pace of development, the features that we are adding to stay competitive, the methodology in which we publish it, all of it, it's just it's more appropriate if it comes directly from us. It just works better for the end user. It's more secure. We can support it. Plus we don't end up having to answer support questions for five versions of software. So from a developer standpoint, I think people are really encouraged.

Just like that XScreensaver mess that's going on in Debian right now. Yes, exactly. And I think here's what it would really take. Debian right now. Yes, exactly. And, you know, I think here's what it would really take. I think Snap packages, if they could benefit from the Docker effect, I think it would be like wildfire. Yeah. What really

did, what really for Docker, what you were able to do, you could be just the laziest Linux user ever. You could be like one of those, not to insult you guys, but one of those web developers that uses a Mac and then SSHs and do his Ubuntu server. And Docker came along and all of a sudden you didn't have to deal with actually installing

Linux software. You could just pull down a Docker image and boom, you had an Intel. Someone already provisions it for you. Yeah. Oh, you want a Minecraft server. You want a Mumble server. You want an own cloud server. You want a Zimbra server.

Yeah, just pull down the Docker images. Good to go. If Ubuntu, if Canonical can get enough adoption around snap packages where I can install what we as Linux users used to consider extremely complicated installations, things like entire desktop environments. If I can reduce

installing an entire desktop environment down to one or two snap packages, if I want to try out the latest Plasma desktop, I can download a single or maybe four or whatever it is, snap packages, and I have that entire thing, and I haven't mucked up all the depths on my system. I was about to say, the other nice part is when you want to remove it.

Yeah. Done. Gone. Your system's going to be... Another example, Ardour. Ardour is universally known as one of the harder applications to install and get everything working. If I can pull down a single snap package, double-click that, and have a working installation of Ardour,

and you multiply this across all of the little edge cases in Linux where it's hard to get this particular thing set up or it involves trashing your entire system with libraries from something, and you start solving that problem and get enough projects behind it that people want, I think the idea of having to install Snapd first is, well, that's a no-brainer. Because it's – I mean, hell, that's even easier than getting Docker set up. So there is a bit of a Docker effect that could happen. Go ahead. I think I agree with you. What you're suggesting really is there needs to be the killer app. Yeah. So I've been looking at – as part of this Snappy Playpen thing is looking at what are apps that people really want and that

are often updated with a very high velocity a fast cadence like next cloud um that people want the bleeding edge version um but other people might want a stable version so you know you could have two snaps a nightly and a stable and uh one of the ones i i looked at was today was ardor actually and so i was i started building a snap for ardor today and audacity as well and i came across a couple of bugs which are being filed and those will be fixed but yeah those are exactly the kind of apps and the you know the kden lives and the things that move quickly where you want the latest thing i packaged up a snap for ffmpeg because everyone wants you know the latest codecs and all that from everyone thank you so yeah that and mpv as well i did mpv today mpv

was super easy that was like not hard at all so yeah it's it's it's good fun and so now when you install the mpv snap will it automatically how does it get updated so i this is the thing i i when i was creating the the snapcraft configuration the idea is that i would then give this to the upstream developer and say here you go here's the configuration here's how you make a snap because rather than burden them with you know yet another packaging they have to learn i just tell them this is the configuration file you run this command on robuntu system and you'll get a snap out of it or if you want to you can push your code or mirror your code to launchpad and launchpad will build you a snap for you much the same way as it builds deb's in a ppa and then you

can just push that to the store as and when you're ready and if you want you can have two channels you can have a channel for beta tester users and a channel for stable users so some of your people might be wanting the bleeding edge version and you can you know canary out to them and then they can play with it and then you can push it out to your stable users so that's pretty nice so you would get updates when the developer uploads them i i don't intend to upload mpv or mind test to the store i intend to you know put a contribution to upstream and say here's a here's how you here's how you package it you upload it to the store, I intend to put a contribution to Upstream and say, here's how you package it. You upload it to the store because then they control it.

They can see the stats. They can see all that kind of stuff. And they maintain it. That is really cool. So you're doing a few proof of concepts there and saying, hey, if you want to know how I did it, here's all the information. Yeah, we're putting it all on GitHub.

If you just go to GitHub and search for Snappy Playpen, you'll find them all. I don't know. I feel like there's going to be a bunch of naysayers that'll say this competes with obs in this way and this competes with that in this way and this is not invented here syndrome but i part of me i i actually think canonical would have had a good degree of success if this would have been ubuntu only and if this technology never would have left the ubuntu platform. The not invented here one is quite funny though, Chris.

Just as a little nugget of information, if you look at the very first commit done to Snappy, it was done on the 12th of the 12th, 2014. If you look at the very first commit done to XDG app, it was done five days later on the 17th of the 12th.

So they're very close. I don't know who could call that. It's very close. Well, and I think, you know, I think part of it is I think we have to sort of change the mindset a bit. What we have to do is I think it's reasonable to expect Canonical to spend their time and resources developing something that's ideal for their platform. I think it's actually sort of surprising that you guys went to the effort to actually package it up and push it out further because you really never had to do that. That never really had to happen. It could have been –

How dare we be good citizens in the Linux environment? It could have been a good thing, though. That's not our shtick. Never. It could have been a nice competitive edge, I suppose. You guys are a big enough ecosystem that you could have been an Ubuntu specific thing, though. That's not our shtick. Never. It could have been a nice competitive edge, I suppose. You guys are a big enough ecosystem that you could have made it work. I just think it's, even though it's another option, I think it looks like a pretty good one. And I think it has, there's a lot of positives to it.

I'm not saying I necessarily know if it's going to be the winner, but I like the start. The start seems pretty strong, and the partners already on board seem pretty good. And the potential for it just being useful within the Ubuntu ecosystem seems strong. So overall, it seems kind of like a winner just depending on how much adoption it gets from here. I guess we'll just wait and see like so many other things. But I know what I think probably pretty much going forward, I will be installing SnapD on my Arch systems.

Why not? I think that's kind of a big thing. I haven't – I guess I did get of a big thing. I haven't, I guess I did get Flatpak working too. I do have Flatpak support too, but, and I have done an app. I wonder when we'll start seeing like meta wrappers around these things or

like a, like an AUR built in kind of support thing. I feel like someone at home or in the car right now is like pulling out their hair though and wants to argue with us because we haven't heard a lot of opposition in the virtual lug today. So that either means that people sort of universally think it's probably kind of a good thing or that

person who doesn't agree hasn't been properly represented. If that's you, you should join us next week. Join our virtual lug. That's right because we'd love to hear from you or you can leave us a comment at LinuxActionShow.reddit.com We love those too. You'll find episode 149 listed there. Links to the stuff we talked about like snap packages and all the other things will be

in the show notes. Just go find episode 149. We do the unplugged program on Tuesdays over at jblive.tv. You can go to jupiterbroadcasting.com slash calendar to find out when we're live. Probably can look forward to some Noah stopping by in the future too when I go out around the first week of July That's always an adventure Alright everybody, thanks for joining us

Remember you can follow live announcements Also at Jupiter Signal That's always handy as well See you next Tuesday I've never installed GNU slash Linux. You know, I was just thinking about it. Really, the benchmark where I'd be happy for Snap Packages to get to would be if when I get new people to switch over to Linux, I can just say, download this file and install this program.

And it doesn't matter which distro they've chosen. I know there's a high probability that most of the apps are going to want the Snap Package. That's really the only, that's the bar. That's really the bar for me. Do Snap Packages work on GNU slash K Windows? I mean, I... Actually, yes, they will.

Yeah, I mean, why not? Because the subsystem is 14.04, but when they update to 16.04, it theoretically would. Awesome. Are they when they update to 16.04, it theoretically would. Awesome. Are they going to update it to 16.04? There's talk about it, yeah. Does that come down as a Windows, I guess it would be like a Microsoft Store update for your new Ubuntu 16.04?

Yeah, pretty much. Oh, that's gross. My brain hurts. That's gross. All right, jbtitles.com, jbtitles.com. That's gross. All right, jbtitles.com, jbtitles.com. It's funny because people are going to talk about it,

how the Flatpaks are – people are already saying that Snappy wasn't good because it wasn't cross-distro and stuff. So now that that's done, people can pick up other things to complain about. But I think it's fantastic, and I don't care which one is taking over. I just want something to take over. I think it's fantastic, and I don't care which one is taking over. I just want something to take over. And a fundamental problem with the Linux ecosystem is the package distribution. As someone who's maintained and developed on basically every angle, because I work in distros and I also make software, it is one of the most frustrating things of being a Linux user.

I understand when a company says that's too much work, because for the most part, they're right. And Snappy and Flatpak and whatever provides a solution to that fundamental problem we've had for so long that I could not be more excited for the horrible experience of package distribution to be over. Yeah, I think it still remains a well-to-how's-the-right-way-to-say-this. It's the right tool for maintaining the distribution and the core distribution packages. It seems like it's probably not the right tool for a lot of applications we distribute over the web now. Right. Chris, I think you've said before, repos and downloadable distros 15 years ago, Linux and Linux distros were way ahead of the curve compared to commercial OSs.

Totally. We had a little bit of an advancement when PPAs became a thing, you know, but it's been stagnant for a while. And, you know, Apple with the, you know, the iTunes store, Microsoft is trying to replicate things with the Windows store. Right. You know, they caught on, they realized, oh, hey, we need a store, we need a central distribution system. You know, so I guess here's where I'm going to dissent is how do we keep the benefits of a centralized download repository and the security and the assurances that gives users while also easing the difficulties for distribution for package maintainers? I think the way this happens is you have to have two components.

You have to have the front-end component, and I think you already have Canonical's fork, or whatever you want to call it, of GNOME software that is going to work with Snap packages, does work with Snap packages. It already does. Yeah, there's no reason that couldn't be reincorporated back into mainline GNOME software, I would assume. So there's your front-end.

Then I would imagine if this thing really took off and people were really shipping software and businesses who depend on it to maybe earn a revenue are shipping software, I bet the pressure on Canonical to open source any remaining behind-the-scenes bits would be, they would probably just do it. I mean, they would probably do that as part of,

well, this thing has grown beyond our initial vision. It's now sort of become part of the ecosystem, so here's the last remaining bits. And even if not, that'll just get re-implemented if it's big enough. Exactly. So my concerns would definitely be satiated if each of the distros, if they rely on GNOME software, being able to read and understand flat packs and snaps, so that, again, I know, so that again, I guess the biggest thing that I don't want to see come from this is that, you know, Oh,

I heard about this great new app, you know, whether it's, you know, telegram or, you know, new app X. Um, and I just go to a random website and download it as a noob. Um, you know, because even if it is secured, even if it is sandboxed, is it still trustworthy? Yeah, how do you avoid that though? Because I see that every time I switch somebody over to Linux, that's the first thing they do is they go Google it and go try to download it from the website. Right, it's just built in.

They never even think about the app store or a repo. It's just not even... Some even like Mac users and stuff where they're supposed to use that. You know, it's like you type it into Google and you hit download and it's supposed to work. So how do you avoid that? Yeah, so many Windows users that, you know, I see coming over, you know, they're used to, you know, the first thing they do once they get a new Windows box up, they download Night Night Installer or Night Night Installer to go download, you know, 7-Zip and LibreOffice and all of these other random binaries from the internet.

And I show them Ubuntu Mate Welcome, and I'm like, well, this is 10 times better. Yeah. But you see, here's where I feel like. I feel like Snap Packages sort of provide for both, because you have GNOME software. You could easily integrate it into – well, not easily, but there could be future improvements to Mate Welcome to support Snap Packages. into, well, not easily, but there could be future improvements to Matei Welcome to support Snap packages. And also, there is that back-end repository that if you integrate with Launchpad or the back-end index, so if you could convince a user just to use a front-end piece of software

to do all the searching, Snap packages provide for that. But if you can't convince them, Snap packages also provide for at least a semi-streamlined, standardized, and hopefully secure way to do downloads off the web directly when users opt to go that route. So it seems like it provides for both. And I would also agree. So my hope is that Snap Packagers and the upstreams do buy into both pieces, that you know, they see the benefits already with the ability, with the packaging, the ease of packaging, and the improvement that has over working with distro maintainers. But they also buy into the backend integration,

you know, whether it's linking to Launchpad or GNOME software or whatever the case may be. Well, I mean, there could also be something like the uf explorer things i mean right now it's kind of messy but if there was a website that just catalogs everything that's in the snap store um and then all you need to do is you know click a button that would then activate either gnome software or just load it up a thing to start installing from the snap from that that they they don't really care they're just used to going to on the web and finding it. So if you say there's a store that you could use it, like awesome,

great. We'll do that. If you, if they just search by, you know, out of, you know, the normal expectation and they find the,

something like the app explorer and they just click the install button, they don't care if they're downloading or not. they would probably prefer it if it would just start installing so if you provide if you solve both of those issues of their expectation and also you know embrace the new the store concept i think that they would have no problem transitioning whatsoever and probably like it even better