LINUX Unplugged - Episode 223: Fedora’s New Trick | LUP 223

Episode Date: November 15, 2017

A new version of Fedora hits the web and we share our thoughts & chat with a member of the project, Noah joins us to answer your live calls & we’re all excited about Firefox’s new quantum release.Plus Gnome 4’s ambitious goals, a new Linux Kernel that really matters, OpenShot woes & more!

Okay, Wes. I'm almost there. This'll be the last machine. And then they're all gonna be Ubuntu. Are you ready? Oh, yeah. What? Uh-oh. Oh, no. We better ask Noah. This is Linux Unplugged, episode 223, for November 14th, 2017. Welcome to Linux Unplugged, your weekly Linux soul injection.

Now featuring 100% more of Chris, Rikai, West, myself, and of course, the Bumble Room. This hour, we're going to take your phone calls, 1-855-450-6624. And welcome, everyone. Welcome, Chris. Welcome, Wes. I'm welcoming you guys to your own show. I'm kind of the guest, right? No, you're here, Noah, and you're saving the day for us. It's a Fedora 27 extravagant. Blowout. Awesome.

And you know what? It's funny. As a Fedora guy, I'm the only person as one of the hosts that's not actually running Fedora 27. So how pathetic is that? No, well, to make it up for you, all three of us are running 27 right now in the studio. How many people in the mumble room? Hello, mumble room. How many of you are running Fedora 27?

Yep. Yep. Yep. Oh my gosh. A little if long day. Every day, all day, of course. Well, I think this is going to be a super fun show. I think we've only done one other episode

of Linux Unplugged where we've had a bunch of us all in the studio because sometimes it's not even technically possible because there are only so many microphones that we have when we're all together, right? And if I'm filling in, oftentimes you're out of town or you're traveling. So this is the first time we've actually gotten together for this big Linux party, so to speak. Yeah, it's pretty cool. And we have a great turnout in the Mumble Room this week, including, I'm very excited to say, we have a special guest from the Fedora Project

who will be chatting with us a little bit. Dan from Elementary OS just had himself quite the interview, and we'll talk to Dan a little bit about that. But we also just have a great showing. Popey and Wimpy pulled themselves away from Rust to be here with us, guys. I'm very excited about that. That's great.

And Kits and Kitty. Actually playing Rust whilst we're in the mobile room, by the way. Yeah, we didn't pull ourselves away. Yeah, okay. I should have known. And Kits and Kitty's here. Kitty, I can't remember how long it's been since you've been here. It may have even been last week, but I'm excited.

Either way, Noah, it's going to be a good show. I agree. So starting off, the first thing that we're talking about today is firefox quantum so tell me you know as a firefox user you get to enlighten me because i don't know what firefox quantum is oh geez oh oh it's it's quantum day noah it's not okay it's not just firefox 57 today is and shall ever be remembered as Quantum Day. And this is it. This is where Mozilla realized that they could bunch up a bunch of features all at once and actually get everybody excited about their browser for a chance, for a change, and really make a big splash.

No, no. No, no. I'm not being critical. What I'm saying is Mozilla really figured this shit out. They have staged a bunch of fundamental changes to the browser. A new UI. They've switched back to Google for the default search. And the quantum part refers to a brand new way of getting shit done in Firefox.

It's twice as fast as the previous version of Firefox. It has a new CSS engine. It supports multi-process architecture so you can prioritize for different CPUs, different jobs. You can have properly sandboxed and isolated tabs, and it is significantly

faster. For the same... Yes, for essentially the same memory footprint. That's fantastic. It's the version of Firefox we've all been waiting for since Chrome was released. Well, so I don't know if I've been waiting. I don't know if I've been necessarily waiting for that,

but it is nice to see Firefox starting to step up and compete in some of the same ways that Chrome has been competitive. Because as a person, and you have to, but we've both bounced back and forth between Chrome and Firefox. And for me, we're opposite, right? I primarily stick to too, but we've both bounced back and forth between Chrome and Firefox. And for me, we kind of, we're opposite, right? I primarily stick to Firefox, and every once in a while I try that Chrome thing out.

In fact, I'm using Chrome today. But there are legitimately performance things that I notice anytime I'm using Chrome that just don't exist in Firefox. To me, it feels sturdier. It feels faster. It feels like the pages pop faster than Chrome. Right. I don't have all of my extensions yet, which has always been sort of my problem. It feels faster. It feels like the pages pop faster than Chrome. Right. I don't have all of my extensions yet, which has always been sort of my problem.

But I'm trying to just come up with other ways to do that stuff. It's pretty nice, Noah. Huh. Well, that is exciting. You know, I knew that today was National Pickle Day. I did not know that today was National Quantum Day for Firefox. But today I learned. was National Quantum Day for Firefox. But today I learned. You know what?

In the show notes, I put a link to a blog post at hacks.mozilla.org, and they talk about everything. But the things that I like the most is they talk about laying the foundation for what they call coarse-grained parallelism. And it feels like they looked at what chrome's doing and they went shit we really have a problem here we've got to really fundamentally fix this and instead of pulling like a microsoft approach where they just sort of like back port fixes to windows and windows just keeps getting a bigger and bigger beast that can do

more stuff they really did like a house cleaning here. They went through and they broke some stuff and they reinvented some stuff and they replaced some stuff and they redesigned the UI. And then very clever, very clever for the Mozilla project. They saved it all up for this big release, this big quantum release. And they gave it a name, and of course that's from the quantum engine. They gave it this whole spin. We've been talking about it for months now, and it all landed.

And by doing so, it gave them time to test and refine the features so they were ready for launch, and it gave everyone a common target to look at for Firefox. And it gives you a night and day experience. As long as you don't have any extensions that block the quantum engine, it really gives you a night and day experience.

And it uses, at least on my machines, it uses less memory than Chrome. Okay. So here's really the bottom line question. Who out there, and I'm kind of looking at you guys, Mumble Room, who out there is currently a Chrome user and is now, as a result of this news, considering switching back to Firefox? That would be me. Okay, tell me about it. Yeah.

Tell me about it. Well, you know, if they claim to be faster, I really want to look at it and see if they claim to be faster, and they actually are. They've got me switching back after probably several years okay so uh yeah so what what does what does firefox offer you that you're that you're willing to jump ship on on a working platform what does firefox offer you that's above and beyond why is it that if both are doing the same thing basically we're saying is firefox is caught up why then are you going to jump off chrome to come back to firefox well one thing i've noticed is that chrome can get very laggy as it goes, because like if you have a ton of open tabs, it will actually bog down your processor because it is actually literally running each tab in a process.

Chrome seems to be streamlining into like shared processes per tab. You know what i'm saying so if it is if it truly is faster and more efficient on the memory more efficient on the cpu then that gives me more power to do other things like video editing or audio production that kind of thing okay that makes sense anyone else looking at it partly okay who was it that spoke up one of our friends across the pond Okay. Who was it that spoke up? One of our friends across the pond? I think it'd be me. Okay. Sort of a switching over from, I think when they went Firefox 56, which was the recent speed up, I've sort of been about from Chrome going to about 50-50 between Chrome and Firefox.

Some of it is just partly to half de-Google-fy stuff and give support back to Firefox as a competition against Google. Right. And that's my position on it, too, is a large portion of the reason that I put up with some of the things I put up with in Firefox is largely because I like the organization. I like what they do. I like what they stand for. I like their principles, and I want to support that. And I am willing to sacrifice a certain amount of features and a certain amount of usability, I guess, is what it really amounts to. I mean, that's the truth.

I'm willing to sacrifice some of those things. You know what that means? That's bad. That cannot factor in at all in their strategy. Not a single tiny itsy bit. If that's what they're relying on, then they're not going to get adoption. I agree from a business standpoint.

However, I think that they have earned enough good because there was a time before Chrome was ever even a thing where Mozilla, where Firefox was my retreat away from Internet Explorer. And they bought themselves so much buy-in capital with that from me personally. And as I continue – there are so many things that the Mozilla Corporation does that all of their meetings are open meetings. Anyone can attend a Mozilla meeting. I think they have a phone number you can dial up or they have an online web stream and you can just listen to what they're talking about, their line of thinking, how they're thinking. I mean, they're a very transparent organization as opposed to Google, which is very much not a transparent organization. And so today, if we're saying that as Firefox becomes a technically competitive product against Chrome, then I think they really start to hold their own to be able to say,

look, now we have the technical features, plus we have all this really good community offer. Yeah. Well, not only that, Noah, but also it seems like if I'm in the enterprise, this is the horse I want to bet on if I'm building internal applications that are dependent on a browser. Chrome is very – everything is built around Chrome now. Like you look at all these Electron apps. They're really all built around Chrome and Chromium.

But if I'm in the enterprise and I'm looking at building applications around a browser, I would – at this point, I would be much more inclined to go with Mozilla's product simply because what the foundation's purpose is. Don't you think, like— You have a lot more trust and just that they won't suddenly change what they're doing. You have access to more of the secret sauce if you need it. Yeah, exactly. So I'm not hearing much from Rika. I know, Rika, you're a really strong Chrome user, a real strong proponent of Chrome. What do you think about this?

No, I'm not going to change. Okay, that's fair. There's no real reason for me to do so. I like Chrome, and a lot of the extensions that I use on Chrome aren't available on Firefox. Yeah, I think there's a couple people in that boat. I definitely heard some people just around the office chatting about it who I wouldn't say are that interested in browsers. I mean, they are technical people, but I was kind of surprised. I will say for my own usage as well that I already use Firefox and Chrome. I use Google Apps, etc.

But I think Firefox is going to be even more of a part of my life, and there'll be some systems where I probably will just drop Chrome altogether. Same. Yeah. Popey, I don't know if you're willing to talk about this on the air, but you have a very interesting workflow that Firefox is not currently capable of mimicking. Is that something you'd be willing to talk about? Well, yeah. So I use Chrome all day, every day, and I use lots and lots of separate profiles, separate users in Chrome. And I like the way that Chrome lets me click a button, switch to another user, and I'll have a completely separate browser window with multiple tabs,

everything saved for that user, which has a separate Gmail account. Everything is completely separate cookies. And whilst Firefox, I am told, can do this, and every time I try this, there are various extensions and ways and command line options that let you do this. It's never as smooth as it is in chrome and and i what i want is a browser which is my work window to the world and a browser which is my personal window to the world and then another 10 for my other online

personalities that i have and and firefox just can't do that and chrome does and what's as well. Well, and what's funny is, you know, so aside from the puppet account thing, what's interesting about that is I think you are willing to push the envelope a lot further than the average user, even the average technical user is going to do. Because I can tell you, I'm a fairly technical person. There is no way in this earth I'm going to sit there and open up terminals to start launching my web browser just to get you know it's just not practical for me and i think there's a lot of people so i think if if you aren't able to fit that into your your workflow i think anyone else that has that similar you know concept of they want to have online personas and i think a lot of people do i think those people are hosed at least i think i think i'm just as lazy as the next guy and i want i want an easy way to

separate one bunch of work from another set of work i want on a on a friday night when it's five o'clock i want to be able to close a an entire window and all of my work disappears so i can say i don't have to i don't have to worry about email notifications or tabs that have documents in that are related to my work. I can just forget those as soon as I close that tab. Now, I realize I'm a bit of a special snowflake because my work machine is my personal machine. And I know that's not the same use case as most people have a work machine and a personal machine at home. I would say that that landscape is changing. I would say that there's a lot of people that are working from home. And I think a lot of those people are going to wind up with some of the, with some of the same challenges and desires that you have. And I'll tell you what's interesting about working from home because I do it too.

this is home and this I'm spending personal time and I am doing working time. And when you're working from home, the best way to do that, if you can, is to have separate machines. But the problem with having separate machines is oftentimes they are expensive. And so one of the advents that has come along in the past couple of years that has really opened us up to be able to play with things, especially those of us that are doing things from home and are trying to do simultaneous different projects, is obviously the ARM architecture. And up until recently, I have had this issue where I'm deploying Red Hat all day long, and I'm deploying CentOS all day long on all these servers. And then I get home and I'm like, and this just happened just a couple months ago. We were testing an EMR for a clinic, and they said, can you set it up and just show us kind

of how it would work? I said, sure. So I'm kind of setting everything up and kind of got everything set up on a Raspberry Pi because they're cheap and they said, can you set it up and just show us kind of how it would work? So sure. So I'm kind of setting everything up and kind of got everything set up on a Raspberry Pi because they're cheap and they work. The problem was when we went to actually put it into production, we had to end up exporting out all of these, you know, SQL template stuff because I couldn't just move everything over because it was, it was the ARM port of it. It wasn't the, it wasn't, it wasn't binarily compatible with the x86 processor that they were going to use in production. And so what Red Hat has announced today blew me away because it's going to,

in a large way, change a lot of that. And this is from Red Hat themselves. Red Hat has announced Enterprise Linux for ARM. So the article goes on to say, today marks a milestone for Red Hat Enterprise Linux with the addition of a new architecture on our list that fully supports platforms. Red Hat Enterprise Linux for ARM is a part of our multi-architect strategy, the culmination of a multi-year collaboration with upstream community and our silicon and hardware parties. And so the article goes on to talk about how Red Hat basically took this pragmatic approach to ARM servers. And I remember, and Chris and I were talking about this earlier today, Linus said that a couple years ago that one of the things that excited him about ARM was he believed it was a true competitor to Intel in a way that he hadn't seen since back in the old Mac days when they were on a different architecture altogether. And now we have a true competitor to Intel. And at the time, I was kind of skeptical of that.

I was like, are you kidding? Because the ARM, like basically at the time, was like the thing that was running inside of cell phones. It couldn't do anything serious. And then year after year, I've watched these things literally double in power until just last year, I think we were doing – for last, we were doing our review of the latest Ubuntu Mate. And lo and behold, I was installing the same operating system on my Raspberry Pi for $35 Amazon Prime available that I was installing inside of my studio. And it's just, it's just this mind warp that I just couldn't get over. And I know Chris, you, you as a person in the studio, I mean, how many different

computers are you sitting in front of? You can put your hand on probably eight of them right now you know the the thing that i think you touched on there that really is just un it's just it's an impossible force for intel to deal with is because arm processors are in all of these routers and all of these little tiny devices and all of these lady tubes and obviously in all of these routers and all of these little tiny devices and all of these lady tubes and obviously in all of these phones yeah they're lady tubes no i know i just i think of something else until i figure out what we're talking about this is my safe way of not triggering them because people that are listening to the podcast they get really frustrated when you trigger their lady tube so i just refer to it plus it's it can be the google it could be the it could be the there's all kinds of annoying people in tubes

these days. And there's a lot of annoying lady tubes. So, in fact, they're all kind of annoying in their own unique way. Oh, yeah. And the thing is, all of them have ARM processors. Every damn cell phone, though. That's really the thing, right? Every cell phone has the ARM processor in it. And this has

caused a momentum of investment and year-over-year improvement in a platform that Intel and AMD just can't simply sustain on the x86 side. They just can't compete with that. Now, they're so far ahead that it hasn't really mattered. But the thing about these ARM CPUs is, damn if they don't work great together, because you can use modern operating systems to sort of carve out a few of these cores and so you have companies like Cray and Qualicom

that are launching 48 core systems. Oh boy. 48 core ARM systems and they don't even, I mean, Noah, you and I were chatting about this earlier, they don't really cost that much. No, they start out at lowest price I found and this this isn't for the 48-core, I think this is 30-something cores, 20-something cores, 30-something cores, but they start at $900. And then for the 48-core version, you're looking at just under $2K.

48-core is $2K. But if you compare that to what you would pay for just a quad-core Xeon, that's half the price of what I pay for some of the servers that we have that have four cores in it. So the amount of computing power that you're getting per dollar – and then, you know, we haven't even talked about the energy because the power consumption of these ARM devices is just a fraction of what they are on Intel or AMD. And so – and then by proxy then or by translation, the heat that is output by a lot of these ARM devices is a fraction. That's a big thing. And then the fans are quieter. Right.

Yeah. Well, it means less cost for the data center to run cooling. And while these per core – I think Beard was probably about to say this. While these per core don't compete with an Intel or an AMD chip. Yet. The thing is, is you can build core-specific jobs. So you can slice off 20 or 24 or whatever cores for this particular application

that you've custom written for this job. Well, I can also think of, like, any number of servers I've run and or have that it's not like they're at their peak performance all the time. A lot of times they're just sitting there and if you could just have a, especially if maybe you're running some sort of, maybe you're running a chat server or something else that has a high degree of

parallelism but doesn't actually do that much. It's mostly just sending keepalives between the million clients connected to it. These seem like great fits for anything like that. Can I just say all of this aside and much congratulations to the Red Hat folks for getting full Red Hat enterprise support behind this platform, which is, I think, a huge milestone for the industry in general. I also part of me has to say, welcome to the party, by the way.

A lot of us have been here for quite a while now, especially those that are Debian based or Ubuntu based. those that are Debian-based or Ubuntu-based. And then you look at some projects like that Canonical has, like specifically some of the like long-term investments that some projects have made into working on ARM platforms, into really solid upgrades on those platforms and delivering software. I mean, it's good that they're here, but I kind of wonder what took them so long at the same time. Actually, I wouldn't mind commenting on that, actually. Yeah, go ahead,

Christian. Yeah, so as you know, Red Hat, the business model of Red Hat, right, is that you sell a, as called, a standardized operating system that you traditionally have installed on x86 systems. And I think for anyone who worked with ARM in the past, right,

it's historically been, you know, highly custom platforms where you basically like you get some sort of Linux version from usually the harder render. And it's a lot of manual tweaking to get it running and maintain it on a cheaper device, right, with this Linux version. And of course, when you need to do an update, it is usually a bit painful. And, of course, when you need to do an update, it is usually a bit painful. So when Reddit decided to enter this market, one thing that we've spent a lot of time doing was actually working to ensure that there is a standardized platform we deploy towards. Oh. So, I mean, like the fact that this ARM service will be using UEFI, that's partially because we said, you know, we need a standard way to boot these systems. We don't want to deal with 10 different bootloaders on 10 different platforms. So, Christian, would it be an accurate statement

then to say that some of the delay was Red Hat dragging the industry towards some standards? Yeah, I mean, I was working with, you know, Linaro and industry partners to say that, okay, you know, we are happy to put Red Hat's name as an enterprise-hardest software vendor behind this, but we also need you to then step

to the plate and say that, hey, we're going to have a standard platform that you can install a standard operating system on. You don't need to have a bespoke operating system to install it because, I mean, Red Hat is not, I mean, I don't know if I don't even say this, but Red Hat doesn't really have an interest in going into, you know, the bespoke OS market. You know, it's interesting because that seems to align

with what my perception of Red Hat, and a lot of times, if I'm being honest, is a lot of times what is what pushes Red Hat to the top of the list anytime I'm sitting in front of a university board or what have you is that if something is put out by Red Hat, I know that it's mature technology. And I know there's an argument to be had here about system D, but for the most part, when Red Hat puts something forward, they're usually one of the later adopters. They're not one of the, you know, the cutting edge pioneers, at least not on Red Hat Enterprise. So what you're describing

seems to fit with what my perception has been. And again, that's just, I'm just some guy talking about, you know, my experience, but that's what I have seen. Mumble Room, who else in here is interested in who would deploy Red Hat Enterprise, an enterprise operating system on an ARM architecture? Does this change anything for you?

Or is this something that you say, well, like Chris said, you know, Debian's had it for a while. Canonical has had it for a while. So this is really nothing new. This really doesn't give us a tool we didn't have in the toolbox. Anyone? The problem is probably that people are not using ARM architecture on business level right

now. But that might change soon. I think so. I think it is going to change soon. I don't think it's I mean, everybody's been predicting this over and over again, but you now have Cray, you have Qualicom, I mean, you's been predicting this over and over again, but you now have Cray, you have Qualicom. I mean, you have a lot of big players that are saying, we're not replacing x86, but if you have an application-specific task for us, we're ready to go. It also seems like it might lower that entry, right?

Because, like, well, we weren't even considering it before, but suddenly if we can run the same OS, we can use very similar image versions from the people who maintain that. Like, we have a bunch of Raspberry Pis that run TVs at my office. We already had that option. If we were a rel shop now, that would be great. Right. And so that's actually what I was about to say was I respectfully disagree that ARM doesn't have a strong foothold and a wide foothold inside of an enterprise audience. I think where it best fits, though, is in a lot of, you know, appliance type applications. But but but in that same way, and kind of what you were touching on, Wes, is in that same way, a lot of these a lot of these platform systems, if you have software stacks, and you have, you know, management systems for us, that was really big, our remote management

system to log into systems and pull statistics and all of that, when they came up with an ARM port, that was huge for us because we could then enroll all of those ARM systems that we had, even if they're just appliances, into that central management system. And so I would imagine that there are tons of businesses out there, like you said, they're either RHEL shops or they have custom software that runs on RHEL, and they can say, now we had these kind of one-off appliances that were out here that were on ARM. If we switched operating systems, we could then have all of those standardized.

And I think that's hugely beneficial to businesses. Yeah, I just want to know when I can build my ARM PC. You know, as a longtime x86 user, all I really care about is that I can build my own PC system because that's just cool. And when I can start building my own ARM systems, then I'll care. That's really for me.

That's it. Yeah, and I think to us, I mean, if you're happy with a Raspberry Pi, you can do that today. There are more robust ARM architectures out there. In fact, I think some of the best representation we have for that is from the Arch ARM group. And those guys,

and I see them all the time at different Linux events and stuff, and they have a whole table full of, because when we when i think of arm and when you think of our most people we think of these 35 raspberry pies but they have like four and five hundred dollar computers that are arm computers that you can install that are designed uh you know with with tighter tolerances because at the end of the day the raspberry pi was never designed to be a general purpose computer it was designed to be a teaching tool. It turns out it's such a great teaching tool and was built so well and designed so above what they targeted to do that you can use it as a general purpose computer. And for 80% of people, it works just fine. But it does – and I've tried to put it in some critical applications and I've gotten bit.

And so there are other devices out there. And I wish I could tell you what they are. I don't know. I tell you what I do know. Kernel 4.14 was released. OMG Ubuntu came out with an article today, said Kernel 4.14 features a large number of new feature changes and is set to become the next long-term LTS release backed for several years by ongoing maintenance and support. Things that they have a whole list of things that they have added, and those are in the show notes. But things that stood out to me that I think are particularly interesting,

the Asus T100 touchpad support. Now, if you don't know what the Asus T100 is, it is a small laptop, UMPC, not quite UMPC, more like a netbook-sized PC. They sell for about $150 to $200. You can pick them up at, you know, any Best Buy, Target has them, that kind of thing. And it had the computer and the display is all in one.

So it's kind of a tablet style, but the keyboard can be attached or maybe it flips around, I'm not exactly sure. But it is one of the most cost-effective, well-built little netbook computers out there. And Linux was originally very problematic on it. When they first came out, we tried to put Linux on one and we couldn't get Wi-Fi to work, couldn't get Bluetooth to work. The touchpad didn't work, sound didn't work. So

basically you had a display and a keyboard and that was it. And that was still such a good deal that I would run into people. In fact, I ran into somebody at LinuxCon, I think last year, and he had like this just concoction of USB dongles to get all of the connectivity and devices working. And with kernel 4.14, the last of those devices is now supported. That's the touchpad. So I thought that was pretty cool. But yeah, better support for Ryzen processors. If I can make a quick plug, next week on the Ask Noah show, Monday, 6 p.m. Central, we're going to have Wendell Wilson from Level One Techs.

And Wendell is just a really great guy, and he's been doing some really cool things with Ryzen and Threadripper, doing PCI passthrough so that he can run Linux on his bare metal. But yet he can still take advantage of all of the software that Windows has to offer. And he has some really interesting theories about how none of us are, we're all going to be using Linux in just a couple of years. I think he backs it up pretty well. So make sure to tune in for that. But better Ryzen support for Linux means that that is all the more realistic. Yeah.

Yeah. I mean, as somebody who has almost exclusively either Intel or NVIDIA hardware, I got to admit, it's looking really good for AMD GPU users these days. But, you know, no, I got to say, I think the big thing about 4.14 is— Don't say ButterFS. Don't say ButterFS. Nope. No, it's just the fact that this is the new LTS kernel that's going to be around for, like, six years. And if we start getting this in these here Android devices, we might actually have

ourselves like a sustainable, patchable, manageable Linux kernel that's used by millions of devices. Like this could be a really, really good thing for Android, let alone all the other distributions. But just Android alone, this could be a huge thing if OEMs and Google start shipping this kernel at some point. Absolutely. MumbleRoom, what do you guys think? Who's super excited about the new kernel release? Oh! Well, I mean, there's ButterFS.

Produce Pack who comes in with the ButterFS. Yeah, what are you super excited about there, Ron? You liar. New compression support for ButterFS. That's what the feature set is, just so that you can continue to do it. They did also improve their SSD allocation algorithms and other things. Which I will say, we're all pretty down about ButterFS, especially for large storage. But I think maybe OS volumes on SSDs is one of the places it does see some use. I wouldn't put ButterFS on anything.

The one that actually stood out to me is HDMI CEC support for the Raspberry Pi. No kidding? Yeah, let's talk about that. Which I believe is the thing that lets the Raspberry Pi be able to remotely turn on displays. Like here in the studio, the NVIDIA Shield TV supports HDMI C, so when you wake up the Shield, it automatically turns on the TV without you having to grab the TV remote. So it kind of solves the universal remote problem by just not needing it. And they added support for more Raspberry Pis as well. My understanding too, and somebody correct me if I'm wrong on this, but CEC allows

you to not only just turn the device on, but also acknowledge which is the active input. So you can set CEC. So if I have four different devices that are plugged into a TV, when a device starts seeing activity, it will switch the input to that particular device and power the TV on. Am I right about that? Yes. Yeah. It's great. I mean, it works here in the studio with the Android TV devices.

So it makes sense that you'd want this on a Raspberry Pi because you could put this on a Kodi appliance. But I think that's worked with previous kernels because I have AutoArm here on a Raspberry Pi 3. And if I launch it, my TV launches too. So there are two things. There's HDMI CEC which is what Rikai is talking about and that's where the Raspberry Pi is actually talking it's talking via data over the

HDMI cable. Yeah, that works out of the box. But just so we're clear, a lot of TVs just support if there was an active HDMI signal, just turn on the TV. My Western Digital TV, which rest their soul, they're great devices, even though I'm not using them anymore. Those devices didn't support HDMI CEC, but still when I powered the device on, my TV would come on. And that was just because the TV had a feature of if I see an HDMI signal, just power myself up.

So there is a difference. Just because the TV came on doesn't necessarily mean it was doing CEC, but it could have existed in a feature. No, no, I have a Sony Bravia here. So all I had to do, I had to install LibCEC. Now it's introduced in the kernel, but I just had to add LibCEC, install it, and then it worked out of the box. Oh, okay. All right. I mean, if you had to add the it, and then it worked out of the box. Oh, okay. All right. I mean, if you had to add the lib, then it's not out of the box.

Wow. Wow. Okay. I can't. We've got to kill. No, no. Don't feel bad. We've got to kill the term out of the box because when's the last time anybody installed Linux from a box?

Give me a break. We've got to kill that. Well, Chris, it's this giant, nice big monolithic kernel. I would buy one just for the novelty. Have it on the shelf. So funny story, guys. Red Hat 7 shipped with a boxed version

of Red Hat 7 as a promotional item. You could buy a boxed version of Red Hat 7 and they would deliver it in a box to your house. You could open it up and they were like, I don't know what they put inside. I hope it was an optical disk, because I wouldn't know what to do with it in a server these days. We don't have optical drives anymore.

In storage somewhere, I have a nice CD case with Red Hat Linux 5. Yeah. Do you really? Nice. Yeah. I collect all the versions of Red Hat. I bought every single version of Soos and Mandrake that came out. I got the free CDs from Ubuntu as long as I could.

I was all about it as long as I could because there was a celebratory fact or just like I have this physical thing. And one of the things that made me use Seuss early on was this beautiful, gorgeous, detailed, updated manual for every single release. It was a book. It wasn't a pamphlet. SUSE had a book, and I would read through that, and I could feel the texture of the paper, and it would have screenshots, and it just was so awesome that this Linux thing had this really physical book. And I would still do it. I'd buy it for Ubuntu 16.04.

I'd do Solus. you know, I really would, because not only is it a nice way to contribute, but it's, there's just, but what I would want now is I'd want a branded USB drive, like Martin there knows about a, he seems to have found somebody who can make a very nice looking Ubuntu Mate thumb drive, maybe people should talk to Wimpy about that, because damn, that thing's hot looking, it's a nice metal thumb drive. Maybe people should talk to Wimpy about that because damn, that thing's hot looking. It's a nice metal thumb drive. And you could put that in like a packet with like a book or a pamphlet about it. Like I'd spend $65 for something like that. As long as like a good percentage of it went to the project. You know, that's, yeah, that's, that's interesting. That might be a funding source or a funding revenue source for them. You know, I, and I, like I said, I think

Red Hat did it kind of as a promo thing so hopefully we don't you find out if that actually worked out you know it's interesting so when you got that book did you go through and like read it like was there was there useful information in there i really would say i skimmed it i skimmed it it was more about uh converting something that was this hobbyist crazy ass nearly hippie ideal to a physical good that came in a box. Yeah, like six discs and a book. It was more about like, that like reality of this thing that I had. So I never had the Seuss, but I had the Red Hat thing. And the Red Hat, I don't know if anyone out there remembers this, but Red Hat used to come with a little pamphlet. And it gave you like this quick start, like, this is Linux, and it's a different operating system that you might not be familiar with. And here's kind of the basic how to kind of getting

started thing. And one of the things that, you know, the thing I guess that really kind of drew me to it was when I sat down in Linux for the first time, pulling these things out of the box, I had no idea what I was doing. And so every little tiny hint that I could get, including those little pamphlets, you know, that was, that was a big help to me. Of course, today, if I was doing that, I would head over to Linux Academy slash unplugged because Linux Academy is going to do that same thing where you're going to have the ability to just visit a website and get a tutor that, and these people, these people care about Linux. These people understand Linux. They know Linux. And you can tell if you've ever taken a bad online computer

course, a video course, you can hear the instructor kind of like reading through his notes or they'll have bad takes or something and you can tell when you're sitting down and i did the uh the red hat seven one i think most recently uh rather than actually paying for the red hat course and i didn't find myself lost at all and i could tell as the instructor kind of went off on on little real hand stories well when i was working for this client or when i was doing this we found this to be useful, and this is why I'm super excited about this being in Red Hat 7. You can hear that personal commentary, that stuff that you just don't get with somebody who's just a paid voice to read or to explain something, right? So Linux Academy slash Unplugged,

they've worked out a really great discount for you so you can get started cheaper than ever and get started on your next Linux career. We get emails all the time at both at the asinola show and i'm sure you get them here on unplugged chris but people are saying hey you started my career you jumpstart my career in fact is eric still in the chat room with us eric uh eric h maybe not eric i think i think you have to go smoke him in the crack pipe. Yeah, maybe. Eric H, he was telling me just the other day, he was like, you know what? You and Jupyter Broadcasting, and thanks to companies like Linux Academy, you got my new career started.

I started because of what you guys were doing on the network and for what things like your sponsors like Linux Academy have done for me, and it launched his entire career. And now he's on a completely different path, and I think that's just really great, because we spend the majority of our lives, our working years, so you want to be doing something you love. So head over to linuxacademy.com slash unplugged,

and make sure to take advantage of that discount. Sorry, go ahead, Chris. I was just going to say, the thing about Linux Academy, too, that I really love about it as a Linux geek, is they pass the sniff test, because when you start using it, you go, oh, yeah, these guys about Linux Academy, too, that I really love about it as a Linux geek is they pass the sniff test. Because when you start using it, you go, oh, yeah, these guys are Linux geeks. They're geeks.

They love Linux. And that's when it clicks. That's when it clicks. So apparently, I'm talking about Linux Academy like it's something that everyone knows. They have a website. It's linuxacademy.com slash unplugged. Apparently, there's a url that you have to go to to actually get there oh well then you get the free seven-day trial so that's

that's useful yeah oh that's academy.com slash unplugged there you go all right so uh is it is it uh is it chris or rikai who oh no no no no i'm sorry i'm sorry i jumped ahead so one of a new software a new software release yeah i, I know. I know. I know. But I didn't see that until I scroll down a little bit. New software version of OpenShot 2.4 is released. And I tell you what, I played with it for just I didn't have a lot of time to play with it. When I saw it, I saw it just for researching the show. And I downloaded it and I was just playing with it a little bit.

Man, are they making some progress. it and I was just playing with it a little bit, man, are they making some progress. OpenShot, even if, and I'll be the first one to admit, even if OpenShot is not the video editor that I could use today to do the things that I need to get done, even if that's not for me today, OpenShot is definitely the nonlinear video editor on Linux that I want to succeed long term. Yeah. I don't know if that makes sense. No, no, it does. That's the nicest possible way to say it. And I would like Jonathan to get a hold of me, and I would like to have a conversation with him,

because I can honestly tell just by looking at the screenshots that this is never going to work for Jupyter Broadcasting, this is never going to work for me. And if you look at the Linux Gamecast, I think there's a very specific reason that they're on a much older version of this program than this current one. And I completely support somebody working their ass off on an open source project. And I completely want somebody to build something for Linux that does

this. But Jesus, if it's not going to be completely irrelevant by the time he gets anywhere where this thing is usable, Pipewire is now just a couple of years away to be completely feature complete. And this thing's going to have to be completely rewritten to use any of that. And the UI is fundamentally impossible for anything that's more than just a couple of tracks of editing. Which means if it's anybody that's trying to edit anything more than their Thanksgiving dinner, it's never going to work for anybody. That was a really nice dinner. Yeah, right? That's all it's good for.

And I totally respect Jonathan. I totally respect the OpenShot project. I have used past versions of OpenShot, which are really great. This, unfortunately, is like skating to where the puck was about six years ago. And I really am disappointed to say that as somebody who actually backed this project when he did it on Kickstarter, I was a backer of this. Super disappointed. I mean, you have to be realistic. You know, I will say they have done some cool things. Improved playback.

One of the things that Linux has struggled with is both 4K and video that is above 30 frames per second. So this has smooth playback for 50 frames per second, 60 frames per second, and 120 frames per second. So they are making progress. And I'll tell you where OpenShot really fits in is if you just need to clip some video files up. If you just need to drop a video file in and clip off the beginning, clip off the end, or make a clip mashup or something like that, if you want a really easy, simple video editor to do that, I think OpenShot fits the bill there.

But I echo 100% your sentiment. I think when you start getting into any even remotely complicated workflow, you're going to be struggling. I mean, Kdenlive is a great, fantastic open source editor that I think is light years ahead of OpenShot. I agree. Kdenlive is a much better candidate for this particular type of work. And then when you need to go beyond that, Lightworks, it's a bit of a learning curve, but it is absolutely a capable editor, and that's available for Linux as well. And so it's getting...

You have... Go ahead. Yeah, yeah. Talk about that. Rotten, are you using DaVinci Resolve? I have used it. I don't use it typically because I don't like the way they distribute it,

but overall, it's definitely a good editor, and I wish they would make a more reasonable distribution method. Well, what are you using for Tux Digital? Kdenlive, basically. Okay. Almost always Kdenlive. I use Lightworks occasionally, too, but it's almost always Kdenlive. But I just wanted to say, like, I think OpenShot's more in the iMovie realm.

Yeah, I would agree. I would agree. It's one of those things that you can pick up and just throw something together real quick. I'm curious, what do you find egregious about the distribution for DaVinci Resolve? It's a binary that you have to run a rooted SH file to install it. It's not any kind of format.

It uses this thing called make sh make sh and it's just like it's a weird uh binary compiled bash script essentially and you can't see what the script is doing and you have to root give it root access as a script oh and then you and it's and it's so weird because you can't like unpackage it or anything so it's like uh if if anything at least do an app image like do something that's reasonable yeah even openshot does an app image yeah yeah well anyway i i think i you know part of that i think is there's a discussion to be had on you know how software manufacturers are supposed to target distributions because when they go to make an installable file, it's like, well, which distribution are we installing for? So I mean I think there's a discussion to be had there, and I think that snaps and flat pack and stuff are going to go a long way to fix that.

So Chris, tell me about this GNOME Shell 4 proposal. This is really interesting because it's gotten a lot of immediate reaction today on the internet. And some of the reaction has been, are you kidding me? You've just gotten GNOME basically stable and now you want to reinvent the wheel with GNOME 4? And then the other end of the spectrum has been, finally, what the hell is taking you guys so long to address some of these fundamental issues? So we have two links in the show notes. One's to a Pharonix article that summarizes it really well, and then one is to the actual wiki post by the GNOME project that gives a lot of detail.

But essentially, you have GNOME 3 Shell today, which is doing great, right? Everybody's loving it. It's getting deployed in all the places. But it's built around X11. It's built around the era of X11. And with Gnome Shell 4, they want to build it around Wayland, a Wayland-first design, as the project puts it. And the whole idea is to do away with X11-slash-X.org support entirely.

Their words, not mine. The new Gnome Shell would be better fitted for low latency input forwarding, low latency visual input event feedback. Who doesn't love that? And low latency zero copy client forwarding. Those all sound good because they have the word low latency in front of them. I don't like low latency. You hate latency. I kind of like zero copy.

I don't know what that means, but I like it. It's good because it's like you can copy without having to copy on the back end and then forward back to the client. It's good, Rikai. It's good. Basically, it's good because it's like you can copy without having to copy on the back end and then forward back to the client. It's good, Rikai. It's good. Basically, it's synergy. Yeah, it is.

And that's – so that's the – the Wayland first is what's getting the headlines. read the actual wiki post, is the GNOME project is facing a couple of really, really hard choices. Three, I think, to be exact. And they identify a few core problem areas. The number one problem that need new solutions, and man, if I don't agree with every single one of these, number one, low latency input forwarding is a problem that they need new solutions for. Number two, low latency visual input event feedback, like specifically to put that back in user terms, they mean cursor movement feedback. You move the cursor, it should be glued exactly to your mouse movement. Exactly. You move that mouse a millimeter, it should represent that exactly. It should be glued together. Low latency and zero client copy forwarding.

Also input methods in the shell. And this is the number one thing that they put as number five on their list. But for me, this is number one. Stalls on the main thread and stalls on the compositor for frame redraws. Now, that one line item is a really, really big deal because it means breaking GNOME up into a bunch of different processes where you don't just rely on one primary renderer, which is what always ends up being the fundamental criticism of GNOME. Now they have a few different potential solutions for this that they've outlined on their wiki.

solutions for this that they've outlined on their wiki. And some of these are radical. Option A is pretty big, and it's the closest to the, I would say, the Plasma desktop setup. It's a UI process and a compositor process, which is exactly how the Plasma desktop does it today. There's also a few other, like, less dramatic options that they have available to them that they're considering. Another option, potentially being less drastic, but would only solve three out of the five problems I just listed, is to introduce a proxy display server. Yeah, a proxy display server would be somewhat similar to an X server. And then they would essentially have GNOME talk to that, and then that proxy would talk to Wayland. You know, kind of like Mir. But let's not bring that up.

And then they have pros and cons outlined for all of that. And what it left me with is, this is what I was talking about earlier in the year on this show, is the transition to Wayland. And we've all been like, let's get Wayland going. Let's get Wayland by default. Let's get Wayland's get Wayland going. Let's get Wayland by default. Let's get Wayland working on NVIDIA. Let's get Wayland working on Intel. The thing is, is once you get to Wayland, there is so much more work for these desktop projects.

And they, in some cases, are going to have to reinvent some of the core aspects of the desktop. And in the case of Gnome Shell 4, it would break, most likely, and the project's the one saying this, not me, GNOME Shell 3 extensions. So all the extensions that you are currently using to make your GNOME 3 shell usable would likely break with GNOME Shell 4. Now, I'm not saying that's going to be the end of the world. I'm not saying there's not going to be an easy transition or maybe a new standardized stable API out of this. It could end up in something much better.

Stable API. I know, right? Yeah. But it could end up in something that's better than what we have now. But in the immediate short term, your doc, your crazy hack to do XYZ could break. You know, for me, it would be my Bitcoin widget, I'm sure, because I like to break that a lot. And so Gnome Shell 4 is now on the horizon and it's it is just one of the millions of dominoes that are going to fall in this transition to wayland yeah i it so it's going to be interesting to watch. But I kind of resonate with what you were talking about earlier when you said, why not start working with Canonical and Mir to do a lot of these things that they're trying to reinvent the wheel on?

Well, you know. And Mir's got a lot of – we call this thing called drama. I don't know if you guys call it over in Grand Forks, but here on the West Coast, we call it drama. Well, here's the other drama. If we're talking about drama, then what kind of message does that send from the GNOME community up to Canonico, which have basically just banked their business on this desktop? I wouldn't say bank the business. I wouldn't say that.

At least the desktop part of it. Yeah. Yeah. I don't't know that's a good question their desktop they don't seem to actually have like they they care about it but business-wise they don't care about it well i wouldn't say that i don't know about that i wouldn't the desktop part because you can say like the server implementation doesn't really matter if gnome's there or not uh the iot stuff doesn't matter if gnome's there or not yeah but so here's the thing the thing, Rotten, is if I'm Dell or I'm HP or I'm anybody that's going to pre-configure and ship a Canonical-derived desktop, I feel better right now that they're using GNOME because I look at business risk and I evaluate other companies. And compared to me, Canonical is a small shop that is struggling to make a buy. And I'm super rich. I'm worldwide. I want to ship my

products to millions of customers. And I'm a little skeptical that they can pull all that shit off. And so the fact that they're now deferring that work to the community, the wider community, and a huge part of that development is backed by Red Hat. And that's no secret to anybody, anybody in the industry. You know, you look at these facts and you say, well, now we've just deferred our risk because now Canonical is not solely responsible for the development of the Ubuntu desktop. All of these companies are all in and they just have to do a great job of packaging up and make it work right. And that's all I need to ship it on my hardware. You know, and that's the hardware vendor's perspective. But then you have to look at the, and this is not an insignificant amount of dollars, not an insignificant

amount of revenue, and not an insignificant amount of the Ubuntu infrastructure. Look at companies like Amazon and Google. And I'm not talking about Amazon servers. I'm not talking about Google servers. I'm talking about the workstations that sit in front of the human beings that use them. Those machines are running Ubuntu on the desktop. Amazon has, and I've talked to a number of different Amazon employees. One of them is a good friend of mine. He's been on this show and on Ask Noah, Sean. And he's one of the guys that works in the IT department for Amazon. And Ubuntu is by far the dominant operating system, desktop dominant operating system on their computers.

And very much the same thing if you talk to a Google employee. They have a lot of Chromebooks. They have some people using MacBooks. And really, the common denominator for Google, obviously, is going to be Chrome. And they've tooled Chrome to do all the things they need it to do. But the actual workstation that does the calculations, the actual thing that they're remoted into, the actual thing that's doing the thinking. That's a Linux Ubuntu desktop. I'll keep it brief, you know, because we've talked about it a lot recently on the show here. But for us, I was looking at different options for the studio because we were having a different array of technical problems that essentially came down to running multiple Linux distributions.

I mean, not to make – but that's what it kind of came down to. Yeah. And so if we just standardized on one thing thing and i emphasize that by hitting my mic if we just standardize on one thing then we would kind of like have one common thing to work towards and so after after a bunch of testing which i've talked about a ton on the show we landed on ubuntu 1604 for the production os yep now um the – one of the things that did factor into that is there isn't multiple versions for me to figure out. There's one Ubuntu LTS that gets five years. I don't have to go get like a CentOS version of Ubuntu or like some sort of rolling version that's supported by an open source subset of a corporate-backed.

It's Ubuntu LTS. And then later on, after I get it all set up and I go, oh, guess what? This shit's actually working for me and it hasn't cost me anything. I can then go choose to subscribe to Landscape or one of their other subscription services, which then all of a sudden turns them into legitimate systems that I'm centralized managed. And I don't have to play this mental game that feels like it's from the 90s

deciding which version of Linux I need. Do I need the Red Hat corporate version where I subscribe to their service? Do I need CentOS, which is kind of like the same thing, but I support it myself? Do I want Tumbleweed? Do I want Leap? None of this crazy, stupid, abstracted different versions. I get Ubuntu 16.04 full stop. It's Ubuntu 16.04.

It's the same one you can download for free. It's the same one I can download for free. It's the same one I can put on my personal laptop. It's the same one I can put up on a server, on a VPS. It is the same Ubuntu. And the systems that are important to me, I subscribe to Landscape and I get Ubuntu Advantage. And now all of a sudden, they're enterprise systems. And to me, that simplicity that felt like it was more kind of like a basic straight approach was the eventual like, okay, well, you combine the commercial support. You combine the software support. and you combine the selective commercial software-like back support by Canonical where I need it. And it's great because I can have – we can standardize 16 systems in the studio on 16.04, and I can subscribe for support on six of them. Yeah, absolutely. And I, you know, the other thing is you are,

you are very involved in the community. So probably not much comes as a surprise to you, but there's, there are a lot of companies that they just kind of, they kind of just, you know, ad hoc trust that they want a company to, to do the thinking for them and do all the planning. And they deal with all the community minutia and just deliver them a product. And that is a valuable service, and that is also a valuable selling point when I'm going to talk to people about, you know, switching to Linux, basically.

Sure, sure, I bet it is. And when I was working with a company where we were deploying lots of really critical Linux servers, it was no question to buy the $5,000 Red Hat Enterprise support. I can't remember what we called it back then, but they had multiple grades. I don't know if they still do this. And one was like, you got like 10 years and you got like a hotline to call support. And that was what we bought. We bought that really fancy one. You know what I'm talking about? Yeah. And that wasn't even a big deal. Like whatever, That's fine. That's just – it was just included as part of the purchase cost of a server.

And so you spend $15,000 and about $3,500 of that was the server license and support. And it wasn't even a conversation. It wasn't a debate. It wasn't looking at the merits of it. It was you buy Red Hat and it's fine. But now when I'm running my own business, my own small business, it's a completely different perspective I have on it. And so I think that's why Red Hat can make billions of dollars still with this particular setup they have. And Canonical can make a profitable revenue source with their setup because they're essentially appealing to a bit of the same market while at

the same time, super fulfilling different ends of the spectrum absolutely uh anyone from the mumble room have any perspective to add before we move on actually i did uh so in regards to the whole gnome 4 thing uh you know with them the question should be is are they deprecating or are they reinventing? Because this sounds like a process that they've been working on for years, it sounds like, and have been building up to. It just sounds to me like the media is just freaking out in the same way that they freak out with Pop! OS and calling it a fork when they're just changing the theme around.

Okay, one, Pop Pop OS is a fork. System76 said it is. Two, the GNOME thing has been in process for a while, yes, but they also said that GTK4 and GNOME 4 would break all compatibility GTA 3. That's why GTK3 is going to have a very, very long LTS. Yep, and you also have to figure that there's more and more LTS distros these days shipping it. Yeah, so they're going to have...

Yes, they are deprecating it, but they're going to wait a very long time before they deprecate it. Yeah, and I want to double down on... You should double down on your point, too. They have been forecasting this for a very long time. It's not like this is a surprise. This is something that they've been talking about for a long time.

Version 4 is called Version 4 because it's meant to warn developers there's going to be some breakage here. They said that about, I think, three years ago. They started saying that this will have breaks. To chime in a bit here, the wiki post was written by Jonas Odal, who's on my team, and who's been one of the main guys behind a lot of our port work to Wayland. So, I mean, at the moment, no code is written here. This is basically him saying, okay, you know,

I spent a lot of time helping out with the port of Gnome 3 to Wayland, and we still have things we plan to do there. But here's some of the problems I see, and I think here's maybe how we want to look at resolving them in the long run. My recommendation to anyone would be that don't hold your breath for GNOME Shell 4. It's not going to be written tomorrow. It's fair. It's sort of a multi-year thing. So, Chris, tell me about James Nugent.

Yeah, so this is a friend of the show, a podcast they've put out over at advancedtechmedia.org and they just did an interview with James Nugent and James is an interesting fellow. He's across the pond over in, I'm going to try to, I'm going to give an attempt. I feel like

I can get this one. Bath, England. That's a first. Chris got a pronunciation. Everyone give him a round of applause. Thank you. Yeah. He works at Joynet, which we talk a lot about on this network, especially the TexNet program and in BSD now.

They were recently acquired by Samsung Electronics, and they still are super active in the open source community. And he's a core contributor at HashiCorp and EventStore, and he also produces the open source stream database with a built projection system. I don't know what that means, but I would almost listen just to find out. You almost got it, Chris.

It's actually Joyent. Oh, I know. I know. I always get it wrong. I've always gotten it wrong. Joyent. You're right. Someday. Someday. Joynet. Joyent. This looks like a great podcast. Oh, man. Yeah. Fascinating interview.

So, advancedtechmedia.org, and it's put on by Alexandra and Adam, and they both came to our barbecue meetup that we had a little bit ago. So, I thought we should give them a plug. Well, I know what I'll be doing after the show. Plug skis. We'll have a link for that. Or you can just go to advancedmedia.org. It's episode nine with James, and he's all in with Joyant or Joynet. However you want to say it.

I don't judge. I don't judge. Were those the guys that did the sous vide? Yeah. Well, yeah. Yeah. Yeah, they were. And I don't know if we got actually a chance to eat it during the barbecue because it took so long.

That's the problem with the sous vide. But, yeah, they did give the go. And took so long. That's the problem with the soup. They did give the go. And they brought the whole unit with the cooler and all that. It was me and Adam. Yeah, that's right. Bachelor Fruit was here too. Of course, Dustin was here as well. Adam and I just came down together. I love it. So their podcast is at

advancedmedia.org. Check it out. Advancedtechmedia.org. Check it out for episode nine. You know what I would totally recommend to them if they're not doing this already is they should check out TuneIn because TuneIn, it's a really cool service. And what they do is they allow you to basically – you can use their infrastructure to have a mobile app that people can listen to your show or they can listen to a web browser. They can listen to it on the Alexa, all of that kind of stuff. There's a really cool tool that I really like it's called easy stream it's a command line based streaming utility that you can use to send an ice cast uh either a file or an ice cast stream

out to a service you can just take you know a given mp3 and you could put it on a loop people could go to that tune in site and they could and they could listen to it to your show and crazy well here's the thing the problem is when you're first setting this stuff up there's like this And they could listen to your show. You're crazy. Well, here's the thing. The problem is when you're first setting this stuff up, there's like this config file you have to write. And I'll throw the config in the show notes for anyone that wants to play with it. But let me tell you, the first time, here's what it is. It's you install the operating system, you set up the software, and then you try and see if it works.

And then you're not quite sure, well, maybe this thing is wrong or maybe that file is wrong or maybe I put this in the wrong location. I'll just better start over. And then you blow it away and then you start all over again. You know what the easiest way to do that is with something like DigitalOcean. DigitalOcean is Linux on demand. It's like Linux on tap, actually. That's what you could call it. And basically, I can get a Linux server anytime I want it just by going to DigitalOcean.com. I have, I'd say probably, I'd say, I think think at this point all but two of my servers on digital ocean are on the five dollar droplets all of them are on the five dollar drops because the five dollar droplets just get you everything you need of course they can go up to ten dollar droplets

fifteen dollar droplets twenty dollar droplets but it gets even better because if you use the code do unplugged then you get free droplets at least the five dollar ones you can get those for free for two months. So if they, so these guys, they, they want to do a streaming service. So they want to set this up and have a mobile app. They could use easy stream.

They could go over to digital ocean, use that code D O unplugged, and they'd have a mobile infrastructure for free for the first two months. And after that's only five bucks. So how are you getting better than that? Digital ocean.com. Make sure to use the code D O unplugged dashboard forM-P-L-U-G-D. Dashboard for days!

Dashboard. You and your dashboard. You know what? We've never said the word dashboard on the Ask Noah show. I don't think. No. You are doing your viewers a disservice, my friend. It's a good dashboard. It's an amazing dashboard.

You know why I talk about the Ask Noah dashboard? You better be talking about the API then. Oh, that's right! That's right, I did. You say that all the time on the show. Oh, no. Okay, that's right. That's right. I did. You say that all the time on the show. Oh, no. Okay, that ruined it. But finish your thought, though.

Why is it that you always talk about the dashboard? I lived in servitude to really, really, really, really, really, really, really, really, really, hold on, really, really bad virtualization UIs. And sometimes, like, the whole infrastructure, all of it would run on Linux, and then I would have to have a Windows machine to manage it, and it would drive me so crazy. Well, yeah, not the Windows.

It only works in Internet Explorer. Oh, it's ActiveX. Yes. Internet Explorer 5 or 6, maybe. And I suffered through a lot of that in my day, and so now to go to DigitalOcean and have all of it work in whichever browser you want to use on Linux, but more importantly, to have it be better than anything I used. And some of those things, you know, they cost $200,000, $300,000, $300,000, $400,000 maybe, actually.

Big spender. Well, it wasn't me. It was the businesses that were buying it. And you ended up with these total crap UIs. Well, not only that, but it shows that they have total trust in their API because the dashboard is sitting on top of that API. Yeah, it's pretty great. It's pretty great. That's all.

That's why I say dashboard for this. You know what I thought you were going to say? I thought you were going to say, well, I say dashboard because a bunch of people got all upset that I said the word dashboard like twice. And so then they went on this like rampage to like Twitter and Reddit and and all this other stuff so i'll just say that's why i say it now yeah i mean that's why i say it for now that's why i say it yes okay so so interestingly enough in the show in this show that i'm i'm i'm guesting with you guys i'm not the one that puts the the new fedora 27 workstation features in the dock i feel very ashamed about that we're ahead of the curve over here you see noah that's because you're not running fedora because workstation features in the dock. I feel very ashamed about that. We're ahead of the curve over here.

You see, Noah, that's because you're not running Fedora because you hate the distro now. You're such an Ubuntu apologizer. It's embarrassing for us. It took us hardcore Red Hat users here at JB1 to get this straight. And, you know, there's... What? Oh, sorry. Yes, right.

Right, hardcore Red Hat users, yes. You know, we Right. Hardcore Red Hat users. Yes. You know, we're back to the good old days. We're back to the good old days of Linux where Fedora and Ubuntu kind of have the same features, but Ubuntu ships a month early. And so I was like, remember that? That's a great setup for Fedora 27. But do you remember how it used to be like that? Because it's GNOME again. And a lot of the progress comes from GNOME because that's where the Fedora project invests a lot of its time.

And so I joke, but at the same time, a huge part of Fedora 27 is it's everything you love about Fedora with GNOME 3.26. But one of the things for me, as somebody who likes to just follow all of this, is this was really the release that marked the end of the alphas. And there's something else that I just anecdotally noticed that I got to give them credit for. Not only did they kill the alphas with this release cycle, but more than I have ever seen in... Well, I have... Actually, I have followed the Fedora project

since it was the Fedora project. Literally, since I've ever seen this distribution, I have never seen more people anecdotally say across different places that I monitor that this has been the most stable, most reliable beta that they have ever

used of the Fedora project. And so not only did they end the alpha sort of release cycle, but they definitely upped their game on the stability of the beta. Because just anecdotally, and this is just my observation, monitoring a lot of different sources, having followed this project since it was just a glimmer in somebody's eye, this seems to be the most stable release they've ever had. Now, I will say, selfishly, what I really like about Fedora 27 is this is the release that really lays the groundwork for Pipewire, which is the new media. Oh, boy.

Yeah. Oh, boy. Yeah. We're excited about that here at Jupyter Broadcasting because this is something that we think Linux really needs to be competitive with macOS on the video editing side. And it was something just a year or two ago I thought would never happen. And so the fact that

the Fedora project is taking this on, they will for the rest of my life have my respect. And taking it seriously. I mean, they recognize that this is something Linux community needs. And it has been something that I thought would never come to bear because it seemed like it took, it would take like this huge multi-project, multi-distro cooperation.

But I think Pipewire, the way they're doing it is it's humble. It's a good, steady, slow approach. So in Fedora 27, it's nothing major. It's just going to handle screen sharing and screen capture under Gnome Shell. It's like laying the fundamental groundwork. And, of course, it's the groundwork to support these functions under Wayland, which couldn't be done before. So I really like the way they're approaching Pipewire. The KDE version of Fedora hasn't seen a lot. They've switched to 5.91 for Qt 5.

And there's a few other things, but Christian's joining us. He's there. He is very dutifully filling in at the very last moment to join us from the Fedora project to talk about some of the new features. So, Christian, welcome to Linux Unplugged, and I know I've mentioned a

couple of features, but I bet I missed some feature of Fedora 27 that is deserved of mention. Yeah, well, thanks for having me. Yeah, I think you mentioned some important things, and I mean, I think you were talking about that we are a month late, but on the other hand, we spent that last month fixing a lot

of last-minute bugs in Nome, so hopefully it's a bit more stable, too. For instance, there were a lot of multi-month ratios that we were hammering on to the last minute to get sorted. Yeah, I mean, one of the things that really stood out to me when we installed Fedora 27 is it is a straight-up showcase for GNOME, for upstream GNOME. And the background is gorgeous, and the transparent bar up top, it's a beautiful default theme for Fedora. It's a great showcase for Stock Gnome.

Thank you. I mean, I think one of the features I would love to highlight a bit, which is almost sort of on the side, is that this is also the first release really where we have Fleet Commander available, which is our new tool for administrating a large number of desktops.

Oh, okay. so Fleet Commander. So is Fleet Commander... You haven't played with Fleet Commander? No, I've played with Cockpit. Is it related to Cockpit, or is it a separate thing? It's an extension of Cockpit. So when you have Cockpit installed in Fleet Commander, you will actually see a link where you can click into Fleet Commander from Cockpit.

But Cockpit, of course, was written and it's mainly targeted at doing server administration. Fleet Commander is all about, you know, administrating desktops. Oh, nice. And, I mean, the background for Fleet Commander actually was we had a discussion actually a couple of years ago in-house

saying that, you know, if you ever want to see, for instance, Linux make a headway in corporate settings against Windows, we need to come up with something that's a bit more accessible to people than, for instance, Linux making headway in corporate settings against Windows, we need to come up with something that's a bit more accessible to people than telling them to rewrite scripts, figure out how different command and tools work on Linux to configure stuff,

and instead have something that's all graphical and one-click deploy, and it integrates with people's directory services. You know, Christian, I have to be honest. This is the dichotomy I have with the damn Fedora project right here. You guys have Cockpit, you have Fleet Commander, you have Atomic, you have Fedora Cloud, you have all of these things that would make Fedora so frickin' perfect for deploying and production

except for the fact that the Fedora releases are only supported for a few months and then i got an upgrade like this is the dichotomy i just can't get over yeah but i mean what we've done there i mean i mean you're not the first person of course to complain about that but i mean for me fedora is is definitely like you know this story we're targeting towards you know enthusiasts users developers and people who who want who want to get the news right away, straight off the press. And at the same point, of course, I know people were not so happy with CentOS or RHEL because they felt it was like, okay, it's fine, new RHEL out.

But then as the years go by, it becomes a bit long in the tooth for a desktop user. So what we did switch for in RHEL 7 is this model where we're actually updating the desktop in RHEL and CentOS, of course, as a result of that every second release. So that way you get a fresh desktop experience continuously. So that's also what I'm telling people when I said, hey, I want a more long term thing. I said, well, I know that historically, things like CentOS might not have appealed to you or getting a RHEL subscription just because you didn't want to be there at the.6 release until you were using the desktop from 10 years ago. But that's no longer true. With RHEL now, you're getting a brand new desktop every second release, and every other release, we're updating a lot of the major applications there.

Okay, okay. we're updating a lot of the major applications there. Okay. And I think as we go forward, this strategy, we'll call it, coming to focus a bit more, because one of the big goals, I hope that we will achieve with things like Flatpaks,

is that we suddenly can create an application stream that's independent of both Fedora and RHEL, so that you no longer have this sort of tight link between which application version you have, which application version you have and which desktop version you have. Sure. Instead, you just have, you know, hey, I want the latest GIMP, and I can deploy it either on Fedora or on RHEL

or, of course, or on RHELs. Good point. That's a really good point, is that this may be a non-issue once that's in the pipeline. So I have so many things to ask you about, and I don't know which you'd rather answer. So I'm going to throw two at you. Actually, how about three? Christian, I'm going to give you three questions, and whichever one you feel more comfortable answering, you go for.

So number one is what are your thoughts on GNOME's position about the system tray icons? Gnome's position about the system tray icons. To quote the Fedora Project workstation release notes, the quote-unquote antiquated system tray has been removed to reduce visual clutter. So that's question one is what are your thoughts on that? Question two, if you'd rather answer this, is some big changes in Fedora Atomic, including the support for Kubernetes, have been introduced, which seems like a huge, huge... That was one of the things I was most excited about. Yeah, it seems like a huge change there.

And then also some badass support for trim support on encrypted disks. I know that's three totally different topics, Christian, but since you haven't, you know, I just thought I'd toss you those three and you pick which one's the best for you to answer. Well, I think the last one is sort of also on the scope. I mean, just to give you a little background on who I am, I'm actually the senior manager for desktop graphics, i18n and Fedora inside Red Hat.

All right. Great to meet you then. So, of course, even though, of course, the Fedora team that I manage are in charge of all the Fedora additions, of course, due to all the other groups being desktop oriented, that's sort of usually where my focus ends up being. So I don't necessarily have a lot of insight into

these other parts of Fedora. I'm also interested in just your opinions too. So you don't feel like when you come on the show you have to have like, I'm speaking on behalf of the Fedora project. I'm just interested in your opinion as a Linux user, too. Yeah.

Yeah. I mean, talk to the system, Trey. I mean, I think it was something that was single for a long time that they wanted to stop having people use these data cycles because, A, they were using an old extender that was problematic in terms of focus and other things. Yeah. that was problematic in terms of focus and other things. Yeah. And so they tried to do this thing where, you know, we sent up an announcement saying,

hey, you know, eventually this will be deprecated. And, of course, nobody ever moved on those deprecation notices. But what makes you think anybody will move now? Well, the hope is, of course, that once it's not a default feature that you have to, you know, install extension if you want the system tray. It will motivate. And at the same time, we do have an active outreach effort at the moment working with various vendors who are still using the system tray to try to basically say, hey, this is a better way of doing it. Well, that's good. Yeah, that's nice.

I mean, so I'm glad to hear about the outreach. So the thing that crosses my mind is that here's where it gets sticky. And again, I know you're not speaking officially. Just in your opinion, it seems to me like if the position is we need to stop doing this in the meantime using extension, but oh, by the way, that extension is totally going to break in version 4. I mean, that is a particular sticky situation, don't you think? Yeah, no, I agree. And what I guess was not very well communicated was that at least, you know,

speaking for the team I managed at Red Hat, I mean, we took ownership of making sure that that extension works in Fedora. Oh, okay, okay. And thus, I mean, it's a community extension, but of course, if you see that it breaks, we will fix it to make sure it works. So if I install that on Fedora, is that coming from the Fedora repository? No, it does come from the upstream extension

repository, but we would go there and fix it. Oh, the guy officially announced that he was not going to continue maintaining it? Yeah, we saw that. So I mean, as I said, we will... As a project, you're going to keep...

Ah, okay. Do you have any background on that decision? That's awesome to hear that. Thank you for telling me that. But what's the – do you have any background on – did the project sort of – did Fedora debate, you know, this is something we need to do? Or how did that come about? Because I had no idea about this. No, I mean, well, we had a decision that in Udiver, some people who had, you know, an old application where they still acquired a system tray.

I mean, I have to say, though, I have a couple of applications that still use a system tray icon, but I have forgotten about them, right? Because even with old UI, they were sort of hidden in that little weird corner down at the bottom. Yeah, yeah. little weird corner done at the bottom. Yeah, yeah. So the multiple applications that, you know, build their UI in a way that require it are quite few and far between these days. And the few we know about is the ones we're trying to work the hardest with to get moving to a different setup.

But in terms of, you know, the discussion we had inside Fedora and Reddit was like, hey, okay, we know that this needs to be there. And, of course, we need to step up and make sure it keeps working because at the end of the day, a lot of these things end up also in RHEL, and thus we have to make sure our paying customers have things that work.

Having thought about it for a couple of weeks, I agree. They are kind of user hostile in a sense, and the thing is if we're all honest with ourselves, it's a paradigm that was forced upon us from Windows. And it's one that we just adopted on Linux to fit in. And it kind of sucks.

It really actually sucks. And so I'm not opposed to figuring out a better way of doing it. And I think it's pretty decent of the Fedora project to say, we're going to try to work upstream to keep this thing working. And that sounds like it would work for all GNOME users too, not just necessarily Fedora. Yeah, that's pretty cool.

And also going back to the GNOME 3 thing, since that's not going to be deprecated for a while anyway, so the support for this extension will still have a long life even though the guy's not maintaining it. He also said it's going to still be around for a very long time. Yeah, yeah, yeah. So I'll just touch on this point just to follow back because people that heard it might want to know more. So trim support has been around in Fedora for a little while now, but the thing that's really great now is that Fedora 27

has introduced trim support for encrypted disks, which I know Noah's going to care a lot about because it's going to greatly enhance the performance of Fedora 27 has introduced trim support for encrypted disks, which I know Noah's going to care a lot about because it's going to greatly enhance the performance of Fedora Workstation on SSDs that use encryption. And I know that's got to be you, Brown Bear. Absolutely. I don't own a computer that doesn't have an SSD anymore. Everything has either an SSD or is NVMe, which, you know, SSDs.

But, yeah, so everything is going that way too, right? I strongly suspect that for the rest of the time that I ever buy laptops, I'm not going to buy a computer with spinning rust in it. I mean, that's just the reality of 2017. Hey guys, I want to give one last plug, 855-450-6624. That's the phone number.

You can join us live on the air for a special edition of Linux Unplugged. Jay is calling. Hey Jay, you're on Linux Unplugged. What's on your mind? Hey, I'm calling about, well, I'm doing tech support for, well, I have this one client that I do freelance, and I work at the software developer, actually,

but I still have one client that I do freelance that I'm good friends with, the CEO, and I was wondering how I could resolve issues where I make mistakes and I don't feel like I should bill for them. How do you handle that? If you make a mistake and you spend a lot of extra time doing something and then you feel like you shouldn't bill them, but you still spent the time to get it done. Yeah. So part of that, Jay, though, it's an ethics question, right? And the way I answer ethics questions is I simply do the moccasin test. If put the other person's moccasins on and walk a mile in their shoes and see how you

feel. So for example, if you, if they asked you, they said, we want you to come install a router for us so that we can get our forefront desk machines on the internet. So what you should do is take the router out of the box and plug it into the modem and turn it on and plug all the computers in. And they should just come online. That's what you need to do. And you'd build a client for like an hour of time. And instead, you pull it out of the box. And unbeknownst to you, maybe you inadvertently shut the web interface off and you can't figure out how to turn it back on. And you're trying to get in. You're trying to SSH in.

So you're calling the manufacturer. What's the default password for the SSH? Oh, yeah. Well, I don't know what version I have. And you're doing that dance, right? And so come lo and behold, you spent four hours doing it. If you were the client, would you want to get billed for four hours of work while your IT guy learned how to do his job right? No, of course not. But at the same time,

if you get called in and the client says, yeah, our internet went down, we don't know what to do, and you say, well, I've never been here before. Tell me the layout. Where's your network documentation? Who was the last guy that worked for you? And they say, yeah, we don't know anything about that. Listen, is that closet at the end? No, that's where your breaker room is. Well, whatever. Then it's the other one over there, that closet. Yeah, okay, that's the right one. So where's the documentation? We don't have any documentation. There's the owner's manuals. I think if that's what you're looking for, those are in there. Now, if you're spending time resetting username and passwords and you're spending time tracing wire connections, that's not, I mean, yeah, absolutely. You should bill for that. And that comes down to, again, how would you want to be treated. And so, um, and, and, and sometimes you take a hybrid approach to, sometimes you get somewhere, some, you know, so far into something, like maybe you're tracing a

cable and you get to a point and you say, I should, Oh, I'm an idiot. I should have known that this cable would run to that switch room, or I should have looked for if there was another switch closet, that may be your fault. That may be something you feel guilty about. But at the same time, unless you were told about that, uh, you it's, you're perfectly, you're perfectly perfectly within your right, I would say, to build a client. And a lot of times what I'll do in those circumstances is I'll kind of do my no affair billing. So if I look at something and I said, well, I was technically here for six hours, but I think I got about four hours. I think I delivered four hours of work. I'm only going to build that client for four hours. And if I, especially there's a particular client we work for and they pay me or they would pay me if I billed them for it.

They would pay me hundreds of dollars to stand in front of locked doors because they have this idiotic policy that the security guy is the only guy that can come unlock a door for me. And so I walk up to the door and I stand there and I call in the front desk and I say, yeah, I need somebody to come. It's ridiculous. there and I call in the front desk and I say, yeah, I need somebody to come. It's just, it's ridiculous. And so if I was billing them, you know, and I would, I would say within my right, if I'm standing there for 15 minutes, every time I need to get through a door, I would be within my right to bill them our $75 an hour. And I don't partly because, mostly because I value their business and I like the owner and I understand why those security principles are in place, given the nature of their business. Does that answer your question?

business. Does that answer your question? Yeah, yeah. Because I really like to, you know, try and give, give, give my customers the, the, the kind of service that I would like to have. And, and, and, and I knew that if, if someone made a mistake and they were just, you know, spin the wheels for a couple, for a couple hours, I wouldn't, I wouldn't want to get billed for that. And at the same time, I just, I just, I spent time on it and, and I, and want to get built for that. And at the same time, I just spend time on it. And it did get done.

But that's just – Yeah, I hear you. It's kind of an interesting thing to thread. Mumble Room, do you have any thoughts? Is the answer I gave, is that how most of you feel or does somebody have a different opinion? I agree. It depends on how much of your mistake – like how much of a setback is your mistake. If it's something that's not that big of a deal that maybe costs you an extra 30 minutes of time and work,

you could say that that's something you could just kind of throw in. But I think if you cause the problem that gives you yourself an extra four hours, I think if you cause the problem that gives you yourself an extra four hours, I think you shouldn't really charge for that if the mistake was solely on your fault. But it was not a lack of information or something. Like maybe you configured something and you come back an hour later and you realize, oh, you have to start all over again. This is why you build research into your contract. You build a few hours of research in. Yeah. And you, Pat, and that's the is I, and I explain this to people a

lot. I'll say, if we're going to quote something, we're going to quote high. And some people, you know, they're put, I mean, I'm serious. If we're, if you're going to overbill, I mean, overbid and underbill. 100%. 100%. Every time. And no, we don't underbill.

If I, if I go into a place and I'm, I'm being honest, if I go into a place and I think somebody's going to, if something's going to take 10 hours to do, we're going to build that client for 12 or we're going to estimate them for 12 or 15. And the reason for that is because I've done this long enough to know that things come up and on the off chance that something doesn't come up, well, that just offsets the number of times that something did come up and we went 10 hours over. And like you said, after a while, you start to budget that in and you start to get a good feel of it. But if you, we'll do real time, real time billing, we do that. If you say, just come and do this job, fine, we'll do it. And we'll let you know at the end how long it actually took

and we'll bill you for that. But if you want us to stay, because it goes both ways. If I'm going to go in and I'm going to sign my name to something saying we will only bill you for this many hours, well, I'm going to make sure that we don't lose money on it. That's just the way we do things. But you know, more often than that, more often than that, more often than I'm worried about not losing money, I'm worried about saving my clients money. Yeah, and one of the ways I've done that recently, and this was actually with a law firm, we recommended them with Ting. And Ting, if you're not familiar with it, is a mobile service provider where you own the service and you own the device. I remember the first time I switched to Ting, I instantly, I knew, well, there's two things I knew. First of all, I knew that any service that had a URL of linux.ting.com

and when I went to that linux.ting.com, it saved me money. If somebody's going to give me $25 to go to a Linux URL, that right off the bat, I knew that was a good thing. But then after I started to get involved with Ting and I was previously on a big name carrier and they sent me the phone and it felt like my phone because they're not, they're not locked phones. They're not locked to Ting. You can take it to any service you want and vice versa. You can go to Best Buy and buy any phone you want. I bought my son an Asus. Do you remember what that was? Chris, he showed it to you when we were in Minneapolis. Zen phone. I bought him an Asus Zen phone at Best Buy and we, you know, we, and I, I had anticipated this. So I'd activated the Ting SIM card, which I have on

hand. I have like 10 or 15 of them that I keep around. And we get out to the car and my son goes, as soon as he opens a box, I knew this was coming, he goes, dad, this thing doesn't have internet. How do we get an internet? And I'm like, I got you, buddy. I got you. Here's your SIM card. And I just handed him a SIM card. And with Ting's dashboard, yeah, I said it, dashboard, the dashboard allows me to control how much data he uses, how many minutes he uses, and how many messages he sends. And so I've gone in there and said he doesn't need to be texting anyone. There's nobody he knows that he would know how to text. He doesn't need to do that.

We want everything over Telegram where I can monitor his communications. I don't care if he calls people, but I don't want it to last. He never needs more than 100 minutes. If he accidentally inadvertently dials something and just leaves the phone sitting for three days, I don't want it to last. I don't, he never needs more than a hundred minutes. If he accidentally inadvertently dial something and just leaves the phone sitting for three days, you know, I don't want a big bill because I'm going to be calling you, Sarah or my son. Yeah, exactly. Exactly. And so, and so they're, they're only pay for what you use and the ability to, for that granular type control. Absolutely fantastic. Linux.ting.com. And huge thanks to Ting for sponsoring the Linux Unplugged program. So I have a article. I don't know if you saw this, Chris. I threw it in here kind of last minute. But there is a gentleman by the name of Tommy Jordan. Really like the guy. And basically he has a website that he runs called 8 Minutes of Fame. It's a long story that relates to a YouTube video that kind of cast him into the limelight unexpectedly.

But he wrote a very interesting article about encryption. And basically the article starts out and it says, we've seen it plastered all over national news these last two years. The encryption debate between tech companies and civilians wanting to keep their data private while being weighed against the government's desire to gain access to anything and everything that can be hunt to stop the terrorists. Now, he goes on to say, and he outlines this very, very well. He says the government argues that these kinds of data being hidden behind encryption provides important clues that they need to stop terrorists. The data that they're able to retrieve out of a cell phone, tablet or encrypted message can be viewed in simple terms.

And I think that I can give you an inclusive list. And then he lists them. So what things can we get out of encryption? Well, he says we can get where they have been. So the GPS location, when they were there. So, you know, when they were there, the timestamp, what they ran online. So the browsing history, stuff like that,

who they've talked to, messenger, telegram, email, contact lists, et cetera, how they've communicated, analyzing which apps they've used and all that stuff. And then the contextual conversations of the email. And then they show what files that they've downloaded onto the device. But you know, the one key thing that is missing out of all that stuff that we could get if we just done away with encryption, if we just turned encryption off and it is an on off

thing. And he goes so far to explain and to make a real big point about that, that the government keeps saying they want reasonable encryption. There is no such thing as reasonable encryption. It's a light switch. The light is either on or the light is off. There is no in between. And if you say – if your argument is, but no, what they're talking about is they're saying that they want encryption that's weak enough that the government only can crack.

Well, then it's not encryption. It's not actually encrypted if the government can get into it. Try and hold a light switch in the middle of the thing and it's either on or off um but what the important thing is even if we do away with encryption we turn that light switch off you know the one thing we don't ever get why why did somebody do this why did somebody send this message why does somebody feel this way and the why is the most important aspect of it because the why tells you what they're going to do and you know and all of that so good i'll say this um the so this is uh this is a big problem for open source projects and i i haven't really figured out how to talk about it yet because i don't know how to talk

about it without seeming like i'm i'm uh i like i like i need to get like a tinfoil hat and need to run around about the sky falling. Right. Because to me, I don't understand how the U.S. government is going to institute laws about how encryption is implemented that affect open source projects that are developed all over the world. They can't. Yeah. I just don't understand how can open source comply with anything and anything that the U.S. just decrees encryption must follow.

These projects are built by people all over the world. Yeah. And people that largely couldn't care, people that largely not only don't care what the U.S. government thinks, but largely want to combat what the U.S. government thinks and the reason that that encryption exists in the first place.

That's true, I suppose. Yeah. Or have no reason to give a shit. Right. Yeah, yeah, yeah. But there are other governments that they also might have to comply with that want the opposite. Well, and if the U.S. government decides it wants it, well, then how soon after that does

China and Russia and all the other governments want it? it honestly it doesn't matter if every government in the world wants it because what you're fundamentally trying to do is put a genie back in a bottle i mean it is it is it's just this to me seems like the same argument is when a celebrity has pictures that are leaked and then they say we want to start taking them down yeah right don't look at them like i i think what they're gonna ask for and when i say they't look at them. I think what they're going to ask for, and when I say they, I mean the U.S. government, is they're going to ask for some sort of escrow system where they're not going to say give us the keys, give us the back door, put it in an escrow. And then when we have a lawful – when the FISA court gives us the lawful right to peek, we will go to this escrow and we will activate the back door. So they'll go to Apple. Where's my rubber stamp?

It's around here somewhere. Oh, there it is. OK. They'll go to Apple. They'll go to Google and they'll say, with your messaging platforms, and they'll eventually go to Telegram and WhatsApp and all of them that they can. And only the systems like wire and signal that are better designed will be immune from this. Everything else, Skype's already a victim, and Slack, I'm sure, will be immune from this. Everything else, Skype's already a victim, and Slack, I'm sure, will be subject to this.

Do you know, the thing is, though, the people that care are going to know, I mean, it's not going to be a secret. I mean, it's a secret the first time. And then after the first person is convicted or their messages are decrypted, then the cat's out of the bag, and then everyone that cares about keeping their stuff private is going to move to something else.

So it seems like a one-burn match. I would like to hear from somebody from the Mumble Room who maybe is even outside the U.S. a bit, or inside the U.S. is fine, but this is something that's going to affect the whole world. Yeah, to make actually one point there, though, I mean, I have no direct knowledge about this, but, I mean, what the U.S. government or any government can do, right, they can enforce what local companies do. And so, I mean, it's not just about stopping, you know, Osama bin Laden kind of guys here.

I mean, it's also about like, you know, catching white collar criminals, for instance, in big banks or whatever. And of course, if you have policies in place that enforces this bank to, for instance, provide any encryption key to use to escrow, that means that they can get to, you know, the email exchanges, whatever that's happening inside that company.

Yeah, but I mean, I don't mean to be the cynical person of the room, but to me, it seems like anybody who wants to break the law in any significant way is going to be hip enough to realize they shouldn't use any system that's been proactively tapped by the U.S. government. They'll use a system outside of that, and there will always be systems outside of that. And the only people subject to monitoring will be the standards, regular old Joe Schmoes. Let me pose another question to you guys. One of the problems with giving up encryption is that if you had a guy and the article, I'm not coming up with this. This is in the article.

If a guy goes through an airport and he is caught trying to bomb that airport and they take him down and they search his person and they find a camera. I don't think anyone in this room would argue that that camera is not pertinent to the investigation of why this guy was trying to blow up the airport. It might contain evidence, right? We agree on that. But the camera might only contain 30 or 40 pictures. When we start talking about they want access to your Google photos, we're not talking about the last 30 or 40 pictures you took. You're talking about everything you've ever done in the entire scope of your life.

When you're talking about the phone, right? You're talking about every Wi-Fi access point. And so who thinks that who out there legitimately thinks that any government has the right that once you commit one crime because of that one – and actually, really, you're accused of committing a single crime. Who out there actually believes that the government should get unfettered access to every – your entire digital life, which is more and more of our actual lives these days? Who actually thinks that's fair? I would say it's a bigger problem than that, though, because we can all decide as a democracy and the Republic of the United States, we can all decide that's fine. The land of the brave and the free, that we want to be screened, we want our devices scanned, we want everything downloaded. That is totally fine.

100 percent as a democracy, we've decided it's great. I'm talking like in the future, we've decided we are going to trade all of our freedoms for security. The thing is, that doesn't necessarily mean that the citizens of Canada want to live that way. And the problem that open source has to address here is, how do we serve a market as huge as the United States that has all of these fuck crazy laws that are essentially 1984 draconian laws that the rest of the world, except for Iran and Russia and Syria, those countries have these laws in place. But outside of that, nobody, no other modern democracy has these kinds of laws in place. So how do we serve the United States

and everywhere else? And I don't understand how you're going to have people that can contribute to one common code base that have significantly different laws that they have to adhere to can work together. I don't understand how you can ship one distribution for the US and one distribution for Canada when we have significantly different encryption laws that fundamentally mean backdoors into encryption versus no backdoors. I mean, Chris, it's really easy. You just have a setting in the settings. You flip a switch. Right.

Totally crappy encryption. We can have a dashboard. American weak encryption and world encryption. The solution is obviously turning all the U.S. laws into internet laws and everybody has to associate to those. And we'll all go by UTC. Well, that's the one logical thing I've heard in this conversation. But no, it's going to be something interesting because this to me seems like it's largely one of those things that is a driving force in one direction i i i very seldom see anyone the government reversing its opinion and suddenly deciding well actually you know we're not going to push to to

try and break encryption and and the one thing that we have working in our favor is this is the one time when market reality and social reality doesn't really mean anything where really it's about the technical advantages that's really what matters so i and i so it's about the technical advantages. That's really what matters. It's actually a funny thing about one of the recent releases from the current government is that they are saying that they're kind of pulling back on not wanting the encryption broken. But, you know, just make copies of everything so we can look at them later anyway. Yeah. Well, like I said, it's one of those things that I think as long as we have the technical superiority in our corner, I think we're going to be sitting good. Hey, guys, thanks so much for joining us for this episode. I really appreciated letting me guest host with you.

That'll put this episode of Linux Unplugged in the books. You can follow us on Twitter at Wes Payne, at Chris LAS. Follow the network at JBLive. I'm at Colonel Linux, and and of course at RakaiLP. Oh, there's a message coming up. There we go. There's Telegram. Okay. Thanks so much. Make sure, hey, check out jupiterbroadcasting.com. You'll find the schedule for the show as well as

all of the show times. Check me out at asknoahshow.com. And we'll see you next week. Negative in the freedom dimension. No! Oh, my God. something tells me you like having the ability to scream like a madman without actually winding up in a recording that matters. No, no, I like it. I don't think I've heard that long into the outro before. I've never heard it end.

I felt like I was just jiving with it. It was good. What did you say it was? Did I play the wrong one? No, we just never, we cut it off before then. We usually keep, we talk, or sometimes I just hard cut it. Oh, really? Oh, you don't let that nice grad. That's such a good fade.

I know. It's good, though. It's good, right? I mean, I did originally build that fade to just let it play out. But then, you know, in the last 30 weeks or so, we've started talking over it. Yeah. It's just something that happens.

We just can't shut up because everything's too much fun. That's what the thing is. Yeah. Not only did I let it play out, then I gave it a good like one or two seconds before i started talking so her kai has a nice tight cut point all right so we got to pick a title we got to pick oh yeah oh i haven't participated in this process in a long time i'm gonna do this okay yes well the process is kind of uh convoluted now no you just oh is it

really you well you go to the discord you search for your bang suggest and then you come to jp titles for the chat room so you can't do you it's not jp titles anymore i? Well, you go to the Discord, you search for your Bang Suggest, and then you can also go to JB Titles for the chat room. So you can't do, it's not JB Titles anymore? I'm sorry, you can go to both. Oh, you can do both. Oh, okay. So you got to search the Discord for Bang Suggest, because you can just, it turns out Discord has a great search. So you just search for Bang Suggest, and you just see all of them in one result. And then JB Titles for, you know, like the bot action.

I'm not liking any of these titles. What do you think we should go with? Fedora Quantum Leap. That's not bad, but that was a very small part of the total show. Thank you, Christian, for joining us. That's a combo thing. I don't know if there was a big part of the show.

Hey, before we get out of the post show, I, I'm, because I wanted to mention in the main show that, uh, Dan here from Elementary OS was just recently over on Late Night Linux, had a great interview. They gave him a few hard questions, and he was, he had some great answers, so if you want to hear

Dan have to answer some tough questions, go check out Late Night Linux. Uh, do you remember what episode number it was, Dan? That's a great question. You know what I'm going to do, Dan? What I'm going to do is while you and I look it up is I'm just going to fill a little bit of air time. And by doing so, I'm just going to sit here and talk

for a few moments. Because what you want to do, right, is you want to Google it real quick. Until you pull up the page. Episode 24, everyone. Hey, good job. Yeah, episode 24 of Late Night Linux has this here, Dan the Rabbit in there, and he answers the hard questions about elementary OS that have been on your mind.

By the way, we forgot something for the Mozilla review. There's a new chapter in the book of Mozilla right now with the quantum release. It's now chapter 11 colon 14. You know the story about that. Well, chapter 11 here in the United States I think means bankruptcy.

That's not a good thing. You open the Mozilla browser and you type in about colon Mozilla and you will get a chapter which normally tells you about the release and the time of the release. So if you have version 16,

it starts with the twins of Mammon quarreled and now with the new version, the book of Mozilla is 11 colon 14, which started, the beast adopted new raiment and studied the ways of time and space and light, and it goes on like that. Whoa, I'm changed. It's a 15-line. It's some kind of Easter egg.

Mine doesn't say that. Hey, guys. Hey, guys. Jordanism from the chat room suggested member when JB cared about IRC. That really seems like that'd be appropriate for the title, don't you think? That seems like that'd be a really fucking good title for this episode. You're really fucking perfect. I think it really exemplifies what happened in the episode.

I think we talked about it a lot throughout the episode. There was a lot of good discussion. the episode. I think we talked about it a lot throughout the episode. There was a lot of good discussion. It's really respectful for all of the work that 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 people put into this week's episode. And all the

topics we covered. Yeah, that's really, that's great. Thank you, Jordan. Thank you. Boy. I wanted to kick Chris's nostalgia a little bit. You'd mentioned OpenSUSE in the box. It actually didn't come with one book.

It came with two. So it came with the user manual. But then I don't remember when they started doing this, but they put a second book in there with all the printed man pages. That's what it was. I couldn't remember. I thought so. Good memory. Good recollection.

Do you have like a box around it? Have you looked at a box recently? How did you remember that? No, I don't have my box anymore. And that's depressing. I threw it away. Oh, that's sad. Yeah, that's what she said.

Man, printed man pages. That sounds really nice, though. Just keep it right on your desk thumb through it that's about amazing totally searchable i love it well okay so we got to title this thing we gotta we gotta get a name for this thing because this is uh this is going off the rails i tell you what chris i got a real i got a real crash course and installing applications fedora oh. We should talk about that. The beard had to install a few apps on Fedora for the show today.

We didn't even get to that. There's a lot less Fedora than I expected. I installed apps via five different methods, or attempted to anyway. First, you have the normal DNF. Oh, okay. You didn't go GUI first? Who goes GUI for installing software? I don't know.

People that like to install software with ease? It's for people who like to mess with the computer. DNS is pretty easy. And plus the other thing is three of the methods don't have ways to do it via the software. Okay. I was pleased that Slack had an RPM I could easily install. That was nice. Okay. The second thing I tried, and this was actually pretty easy, to get Chrome installed. I was pleased that Slack had an RPM I could easily install. That was nice.

The second thing I tried, and this was actually pretty easy, to get Chrome installed, you just add the third-party Chrome repo from Google. Are you telling me they have a repo for Fedora? Yeah. They did the same thing for Debian as well. I've got to be honest with you. I'm sorry, guys, but Firefox. The third method. Firefox is the sorry, guys, but Firefox. The third method. Firefox is the best, obviously, for sure.

The third method, which I managed to get nothing installed via, is Copper. So, okay, that was no good. No. The fourth method was enabling RPM Fusion. That was to get Telegram installed. Oh, nice. RPM Fusion. That was to get Telegram installed. And the fifth method, because

I could not get Discord via RPM Fusion or Copper was that I also had to use Flatbacks. So I enabled FlatHub and I installed that via the software stuff. Okay, so let's back

up for a second. So what did you say before Discord? What did you install before Discord? So how did the Telegram thing go down? How does that work? So did you... What did you do to get Telegram working? RPE and Fusion. So you added a repo?

Yes. So you did like a DNF command. You said there's a DNF commands? Yeah. And then did you install it via the command line? Or did you install it via the software center? Command line. Okay.

And then the last thing you installed was... Discord. And you installed that via the command line or did you install it via the software center command line okay and then the last thing you installed was discord and you installed that via flatpak but you added a repo for flatpak yeah i enabled a flat hub and did you install it then after that via the command lane or via gnome software gnome software interesting so now it just shows i actually couldn't figure out how to install via via the command line. So you added the Flatpak repo, and then you opened GNOME software, and you searched for Discord, and then Discord showed up. It told me that the source was Flab. Oh, interesting. And when you want to install Flatpak from the command line, you also have to include the repo in the Flatpak command. Yeah, yeah.

Yeah, I'm not a big fan of that. It's a little convoluted. Yeah, that seems counterintuitive. Like, I already added the repo. Why isn't I just searching the enabled repos? Yeah, welcome to PPAs. Yeah. Mm-hmm.

Yeah, I know. This is... Well, PPAs don't require you to tell the repo after you have it installed, though. No, but you do have to do an update, right? So you add the PPA, you add the key, and then you have to do an apt update, and then you do the install. It feels a lot like PPA.

Yeah, but with Flatpaks, every time you install the app, you have to specify the repo URL. That's true. Yeah, that's true. Everyone should use a UI anyway. That's how we designed it. Say again? Everyone should use a UI anyway. That's how we designed it. Say again? Everyone should use GNOME software.

I mean, that's the whole point here, to move away from needing command line tools. It definitely does feel like that. The whole thing is pretty straightforward. There's not a good way to install the repo, though, right? Well, that's true. Yeah, adding the repo, you still drop to the command line. Yeah, the reason I tried to do it via the command line was because I was already

on the command line to add the repo. I don't think there is a way in GNOME Software to add it through and add a flat hub anyway. That does seem like something that needs to get... I guess that does feel like a feature that needs to be added. It'd be nice. Yeah. I actually...

One thing we're looking to do there, actually, is we are going to start pre-populating this database with various repos. So the setup will be that when you start Fedora, since it's meant to be all free software, you'll get a question as part of the initial setup

saying, hey, do you want to enable third-party repositories? And if you say yes, a lot of these things like Chrome and Steam and VitaD Driver and so on will just be pre-populated for you to download. That's wonderful. Hmm. Yeah, it seems like a decent solution. The more I see distributions implementing Flatpak, the more I see how it really does make sense for a lot

of different uses for open-source software. Yeah, I think if I do use Fedora in the future, I'm going to start with FlatHub. Yes, exactly. That's what I would do too now. Because it seems like it has most of what I wanted. I actually just searched Telegram and GNOME software, and now there's two different

versions I have, one from FlatHub and one from RPM Fusion. And the fact is you can still get snaps working on there, too. So that kind of covers some of the other space. And then it's like, well, what do you have left that isn't covered by those two things? And the list is... Well, but thankfully they package it for the distro you want. Yeah.

I mean, they're doing that work now. Right, yeah. I could see if one of these things takes off either way. I could see Google changing that. If Fedora supported both Snap and Flatpak, then that would solve most of the issues of packaging entirely. Mm-hmm. And I'll use AppImage for everything else. Oh, right.

There you go.