LINUX Unplugged - Episode 54: Microsoft's Munich Man | LINUX Unplugged 54

Episode Date: August 20, 2014

Sam from the Moka project stops by to chat about the business of making Linux look better. Then we get into the role open source plays in self driving cars.Plus we bust some of the FUD around Munich��s much reported plan to abandon Linux and switch back to Windows.

Just make sure I've got it. You've got a workstation with drives in it that has ButterFS? Is it arrayed or is it just straight drives? It was a single disk with ButterFS as the root file system. Okay. And my home partition was on that same root disk. Anything fancy like LZO compression or anything like that? Everything that you covered in your recent ButterFS article, I've been doing that for two years. Okay. All right.

ButterFS article. I've been doing that for two years. Okay. All right. So now, as it happened, fate came to calling and I lost the drive. So I was able to actually restore from backups and I was determined not to lose my Arch install that I'd been nurturing for two years. This is where I'm at right now. So I, you know, rsynced the whole thing over and there were a few files that were complaining that they were, you know, lost sectors and what have you. And it was mostly, you know, Firefox cache and stuff like that. So I went with it and I bit the bullet.

So what I've done is I've got two drives now in that machine. An SSD for the root file system, which is ButterFS and LZzo compression and all that good stuff yeah but an uh and uh second spinning disk which i'm using for my home drive and i've put xfs back on that because i trust it see this is what i was wondering is if xfs isn't maybe the way to go but i until i experimented with butter fs a couple of years ago Arch, I've used XFS on everything for donkey's years, and it's never let me down. Yeah, you would be. XFS all the way.

Yeah, I mean, XFS, back when I really cared about it, and also because I needed extended attributes, XFS was the way I went. This time around, you know, it's like, Linux needs an answer to the ZFS problem so bad, and I think it's way worse than, as a general community, we acknowledge. Somebody who comes from a large server storage backend, the benefits of ZFS and just sort of like the brand that it has now accumulated. So you just say the name ZFS and people kind of associate a lot of factors with it.

Just that alone is going to deploy more free BSD than I think we've seen in a really long time. And so for me, my take was, well, here I am, I'm doing these shows, I want to live in Linux's answer to ZFS, I want to experience it, and now I'm a year and change into it, and I'm pretty disappointed in the result. So it sounds like, Wimpy, your solution was copy it off and reform it. Well, because the disk that it was on was dying, but yes, I took that as an opportunity

to actually switch to XFS. And I was thinking of doing it anyway. And your whole thing there about ZFS is quite correct because I run a completely Linux data center, except I've just bought two stonking TrueNAS devices because of ZFS. Linux doesn't have an answer to ZFS and I think Linux's answer to

ZFS will actually be ZFS. Yeah, it's going to have to be. I think so. I think it has to be. I can't see that ButterFS is going to... Because apart from anything, despite the fact it's flaky and it is flaky because every time you look at the kernel revs, you'll see that it's fixing corruption issues and things of that nature,

which is definitely good to see. The built-in tools can't properly read it. It's supposed to be the native file system of Linux, and I know for technical reasons DF isn't aware of reserved blocks and things like that, but it boggles the mind. It seems so disconnected. And I just can't get over it. It just seems like such a disconnect. And it's so disappointing because there are aspects of using ButterFS that I've enjoyed, like the performance, the compression has been quite good. The feature set I feel like you get does not have a huge

performance penalty. Of course, I've been using it exclusively on SSDs, so I probably shouldn't make that as a proclamation. But at least in my experience, I've gotten using it exclusively on SSDs, so I probably shouldn't make that as a proclamation, but at least in my experience, I've gotten an incredible feature set with very little overhead on a system that has been fairly good except for these final problems that I've now reached to do. And the problem is these ones, there seems to be no solution other than reinstalling. The issue with this is, and this has never really been a problem for me before, is this is really like the one install I really don't want to reinstall, because I really you know, I haven't had a Linux distro that maybe lasted

a year or more. I usually nuke and pave even just to test things out for the show, but because of the hardware setup I have now, I've been able to keep this thing dedicated to it, and it's nice. Don't reinstall. Don't reinstall. I didn't have to. Get yourself a second drive. I mean, I know I could.

And I'll send it over to the partitions. I know, that's the part I'm not looking have to. Get yourself a second drive. I mean, I know I could. And I'll send you over to the partitions. I know that's the part I'm not looking forward to. Yeah, I know. So Chris, I have a mate who works on MetroFS, and he suggested that wiki article. And I just, you know, I'm not teaching you to suck eggs or anything, but have you gone through the steps on that which is probably quite clearly for people who are in your your situation i think so uh in fact i think the only way you're ever gonna stop butterfest from taking too much goddamn space is turning off copy on right yeah i know the cow feature is basically the everything like everything that takes up all

the space but unfortunately that's one of the big huge huge benefits of ButterFS. Yeah, exactly. I don't think so at all. Well, maybe not. Maybe I should change my mind about that. So when I first started running this problem about a month ago, I did a Linux action show special on ButterFS, and this was one of the things I went through. And I did initially reclaim 10 gigabytes of space. I was really stoked.

In fact, I even installed another Steam game to celebrate. And then quickly that has already gone back to like only one or two gigs free. And then I do a scrub or I do a defrag. And then it kind of frees like one, two gigs free. And so every time I keep going back through and take another pass at it and I try something to clear up space or I delete something, clear out the package cache or you know, or whatever. So if I disable copy on...

I haven't done the disable copy on write though. Every ButterFS installation I do, I always disable copy on write. You might as well not be using ButterFS at that point. Well, but then you still get compression and trim. You still get compression and deduplication and trim. I don't use dedupe, but you get it.

So that's not bad. We'll use F2FS then, in which case. Oh, it's actually funny. All you guys want to use ZFS, but here I, on purpose, don't use FreeBSD because the ZFS progs package is not good enough and up to snuff. I stay on Linux because of XFS, even though I philosophically like FreeBSD better.

So why are you running ButterFS with copy on write disabled if you could just use XFS? Testing. That's all testing. I generally use XFS for all my actual installations. Yeah, yeah, yeah, okay. That's kind of why I put it on there. And the reason I still run ButterFS in the root drive is because when I'm doing rebuilds,

the way that the Arch build tools are set up is if you're running butter fs you can create a ch root and then clone from that rather than having to create a fresh ch root from everything each time a bit like the way jails work so it speeds up the build process i certainly recommend dropping by hash butter fs on yeah because they are very. That's what I should do before I go too far. Really, before I go too far down the rabbit hole. Yeah, don't wipe it or do anything. Yeah, yeah.

If you actually want to save it, yeah. Well, what happens is I keep thinking, even though I've been using it for well over a year, I keep thinking to myself, well, this is Arch. Eventually it's going to blow up, and when it blows up, that's when I'll nuke and pave, and I'll just switch to a different distro. But the thing is, it just hasn't blown up yet. And so now it's going to blow up, and when it blows up, that's when I'll nuke and pave, and I'll just switch to a different distro. But the thing is, it just hasn't blown up yet.

And so now it's – Well, it's more reliable for me than people make it out to be. My server is sat on top of my fridge, and the repetitive slamming – That's a great service bot. Well, yeah, it's not, I know. But it's the only space I've got. Yeah, no, I know. The repeated slamming of the fridge door means that the eSATA cables have popped out of the array

on more than one occasion. I SSH into my server and the file system's gone read-only. I just go into the room and I just pull the cables out the back of the SATA array, plug them back in again and remount it and it's fine. It recovers. It's not a problem.

I have not had any data loss with ButterFS. And there is the story of Canonical's build infrastructure. Oh yeah. No, I haven't had any data loss. I'm thankful for that. You know, I just loved the idea, but yeah. I actually, so here's my question, and Heavens, I wonder if you play with this. Once you

disable copy on write, does it free up all of the blocks that it saved in the meantime? Not sure if it'll actually remove all the previous snapshots that it's trying to keep yeah but most of the time or at least a lot of times with time since it's butter b tree based as you do your snapshots the b tree gets bigger and the depth gets broader and deeper so when you you trim it, you're going to get faster because your tree, your B-tree, is going to be a lot less deep. And is scrubbing that trim? Is that count as...

I'm not sure. Yeah, okay. So that would be something I might look... You know, that sounds like questions to toss in the ButterFS IRC. I like that idea. Maybe I can get this thing fixed and not have to reload. Welcome to Linux Unplugged, your weekly Linux talk show that's powered by guacamole this week. My name is Chris.

And my name is Matt. Hey, Matt, I had a little chips and guacamole before the show, and I thought, you know what? I'm drawn from that today. I'm drawn from that guacamole. Hey, we've got a fun show coming up on today's episode of Linux Unplugged, episode 54. We're going to have a little chat with Sam from the Mocha Project. Sam's working to make Linux look a little better, not just Linux, as a matter of fact. He's part of probably a new group of folks out there trying to bring a little beauty to Linux desktop,

and I think they're succeeding in a big way. So I want to talk to him about what they consider to be a theme layer and how that works and learn a little more about the Mocha Project. We're also going to discuss a little bit of Linux-powered robo-cars. That should be interesting, self-driving cars. And then, using our mumble room as our counsel, we will have to discuss the situation in Munich. The reports are going crazy this week that Munich is pulling back

from their 10-year initiative to convert to the Linux desktop and that they'll be re-embracing Microsoft any minute. So we'll discuss that and maybe why it's not as dramatic as some, it would seem to be Softpedia making it sound. So that'll be all coming up in today's episode. But first, like we do, we're going to start with our emails. You guys always send in some great emails to linuxactionshowjupiterbroadcasting.com or go to jupiterbroadcasting.com

and click the contact link. You can choose unplugged, send it right to us. So Heath writes in, we've been kind of covering this whole MacBook phenomenon that we noticed at OSCON and now Elementary OS is targeting MacBooks. So Heath writes in and says, hey everyone, I absolutely have

and will again purchase Apple computers with the sole intention of installing Linux on them. Disclaimer, I'm a developer. You summed it up perfectly. A laptop with a decent resolution, even 1080p, won't cut it anymore. I need the latest processor and SSD in a compact form, and you're left with a very short list. I certainly don't agree that the Apple closed-source walled garden vendor lock in philosophy, but when it comes to high-end laptops, I haven't found anyone that competes.

See, I don't know if I totally agree with that, but I think it's easy to slip into that, especially because the marketing is really pushed in your face. He says, in regard to the similarities between the elementary OS and OS X, all I have to add is that while they may visually be very similar, they offer very different functional workflows. I'm forced to use OS X for work, and I'm constantly frustrated with their outdated window and workspace management. Conversely, I'm annoyed with elementary OS, if I am annoyed. Well, it's just Linux. There's always a config file for that.

Obviously, everyone has their own preferences. That's why we use Linux, right? We weren't happy with the status quo. But if you dismissed elementary thinking it was trying to just be an OS X clone, I'd recommend giving it a test drive because it's really very different. I agree with that ending statement, too. I think some people do just sort of cast it off as an OS X clone and call it good. Oh, yeah.

No, it's much deeper than that. There are other OS X clones that are much more clone-like. This is actually trying to differentiate the experience from the ground up, I think. Yeah. A little follow-up on those Chrome boxes that Noah mentioned, kind of bit him in the butt last episode, I think it was. Bryce writes in, he says, tell Noah he can put those Chromeboxes into production. And remember, Noah had this problem where the gal hit spacebar and reset the BIOS and disabled the ability to run into Linux because she just wanted to move the power plug.

Well, he says, this script will flash the BIOS of a Chromebox, and then you can install normal Linux. My main PC is now an Asus Chromebox with 8 gigabytes of RAM and Linux Mint installed, and everything works. I just completely wiped Chrome OS. The script linked above works on the HP Chromebox also, and we'll have a link to that in the show notes. Interesting. I would ask Noah about that, but he's actually en route right now to LinuxCon.

Wow. He'll be doing coverage for us this week at LinuxCon. So if you're going to be at LinuxCon starting on the 20th in Chicago, stop by and say hi to Noah. He'll be the one wearing Google Glass, so probably not too hard to find. No. All right. Our last email for today's show comes in from Michael,

and he has a question for us, Matt. He says, hey, Chris and Matt, a longtime fan. Thanks for continuing to put out the best show anywhere, and I can't live without it. I'm sure that's very true. I'm just wondering, if you ever thought about doing a segment on flash drive distros. I've got four USB 3.0 drives on my new laptop, and it seems like such a waste not to have at least one of them doing something

as cool as actually facilitating my whole operating system. Which one would you recommend, and is it feasible to make one of your distros full-time, a desktop OS based on a thumb drive? Again, I love the show. Keep up the great work. Sincerely, Michael. So, I was thinking, what do you think, Matt? Should we do like a thumb drive distro roundup at some point?

I've been toying with the idea for a while as it sits now because, I mean, previously it was either going to be damn, you know, like a puppy Linux or maybe damn small Linux or something like that. I mean mean something very lightweight that you can run in the background but boy I don't know that that could be interesting because I've tried a few that didn't really leave a great taste in my mouth so it might be worth trying yeah because it's got to be a right strike so right here I've got yeah it's got to get that bounce so right here I've got a uh a patriot Patriot Rage 64 gigabyte supersonic Rage USB 3.0 thumb drive.

And I've actually ran off this for a couple of days when I had some stuff going on with one of the NUCs, and I didn't feel like troubleshooting it. I installed Fedora 21 on the beta on one of these thumb drives and then ran off that for about two days. And because I think this Patriot USB thumb drive is specifically designed for performance, and it's USB 3.0, and there was nothing else on the USB bus, I actually, I really felt like it was performing very well. And then I started to toy with the idea of a NUC here at the studio, a NUC back at the house, and a thumb drive between them.

That could be really cool, right? Because the NUCs, if they're the same hardware, maybe same monitor on both ends, so it's always the same screen resolution. Huh. I'm just saying. Yeah, that could work. So maybe we'll do a thumb drive review episode. That might be really neat because that might be something both Matt and I could toy around with.

So stay tuned for that. If we do that, we'll give you guys a heads up. Hey, let me give you a heads up about DigitalOcean, our first sponsor for this week's episode. Go over to DigitalOcean.com right now and check them out because we've got a special promo code. If you use this code, you'll get a $10 credit. So you're probably saying, Chris, what is that? I'll tell you. Unplugged August.

All one word, lowercase. Unplugged August will get you a $10 credit over at DigitalOcean. Now, if you're not familiar, and you need to get familiar real quick, DigitalOcean is a simple cloud hosting provider dedicated to offering the most intuitive and easy way to spin up a cloud server. Users can create a cloud server in about 55 seconds.

However, sometimes people get it done a little faster. A little faster. Robert Downey Jr. wrote in. He says, I created my DigitalOcean server in 36 seconds. Hi, he says. I spun up an Ubuntu 1204 VPS at DigitalOcean in 36 seconds. He attached the screenshot, but I didn't grab it.

And he says he'll be turning it into an encrypted email server. What a great use for DigitalOcean. One of the great things about DigitalOcean is they have data center locations in New York, San Francisco, Singapore, Amsterdam, and London. They just announced they're turning on IPv6 support in their third New York data center. So if you know about IPv6 and know why that's a big deal, DigitalOcean can get you a droplet in one of their New York data centers with IPv6 support. Their interface is

amazingly simple and intuitive. Their control panel is powerful and PowerUse can replicate that power on a much larger scale with DigitalOcean's straightforward API. And if you're an Ubuntu user, check out their check out the applet that you can get for your menu bar that allows you to manage your DigitalOcean's straightforward API. And if you're an Ubuntu user, check out the applet that you can get for your menu bar that allows you to manage your DigitalOcean droplets right from the menu bar. You just drop it down.

It sits right up there in the little notification area. It's really cool. And when you use the promo code UNPLUGGEDAUGUST, you get a $10 credit. You can try out that $5 rig, which will get you 512 megabytes of RAM, a 20-gigabyte server, one CPU, and a terabyte of transfer for $5. I've been using a $5 rig for months, and I've spun up a couple of new servers recently because it's so obvious once you start using it.

It's so obvious. DigitalOcean.com, and use the promo code UNPLUGGEDAUGUST. And a really big thanks to DigitalOcean, and congratulations, DigitalOcean, on adding IPv6 to another data center. You guys are rocking it. And thanks to Robert Danny Jr. for sending in his DigitalOcean success story. That was kind of cool.

Good stuff. Yeah. Hey, Matt, joining us now on the Unplugged podcast is the creator of the Mocha Project. His name is Sam. And Sam is joining us on the Mumble line to talk a little bit about Mocha. Hey, Sam, welcome to Linux Unplugged. Hi, thanks for having me. Oh, absolutely. Thanks for joining us.

If you followed any of our shows for a while, you know I'm a big fan of making the Linux desktop more purdy. So I really like following what groups like you and the Numix project are doing and making not just a set of themes that work on one system I use, but across all of my systems so I can finally have a little bit of consistency,

but each device is still sort of meant for that particular task. So I wanted to start with you. What was your motivation for starting with Mocha? How did you get started, and has it evolved into something bigger than you were expecting? Well, I started it for myself, basically. I wanted a look that I couldn't find anywhere in the other themes. And I decided to share it with the community. And from there, it kind of took off as people started using it and sending me requests and such.

And I don't know, it just got very popular, I guess. That makes sense. So you call it, too, I've noticed you, it just got very popular, I guess. That makes sense. So you call it, too, I've noticed you call it a theme layer. A theme layer. What is that? What is that? Well, it's made up, really. It's just that you can, because I do, like, theme, a GTK theme, or an icon theme, shell theme. You can replace your entire default set

with the Mocha-styled one and layer it on top, let's say. And I've made it across RPM-based stuff and Arch and Ubuntu. You can basically put it wherever you want to put it. So I call it a layer. Yeah, I like that. So tell me about I noticed you know

there's also other projects that are coming out right now there's not just Mocha there's Numix and a few others is that a friendly cooperation

are you guys inspired by each other is there any kind of you know maybe animosity because it's kind of the same turf same turf

I don't know there's no animosity I know a few people are in the like Numix, and it's a friendly. It's more a friendly competition, and we make fun of each other sometimes. What do you think about some of these distributions that are partnering with some of these theme groups, like Antegros and Numix, who have come up with the Numix Frost theme and bundled it with their distro.

Anybody talk to you about that? And do you find that to be a potential area of growth possibly? Well, there was a talk with Ubuntu GNOME to get it included in that one. Not as default, but just as an option for users. But I'm open to getting it on as default in some other OSs or distros. There's no reason. Yeah, I think that'd be interesting.

Kind of the really last question I kind of had for you is I wanted to get your take on, I don't know if it's a business model, but a way for these projects like yours, like Mocha, to generate a little money. Is it partially, do you see part commercial, part free? Is it going to be donation driven? Because it seems in the past folks have tried this and have had a hard time making it worth their time. For me, I'm not that, I tried, it's more of an experiment in trying to see if it will commercialize well.

And I just pay what you want or download it for free or just pay that $2 or whatever it is I have. And, well, more often than not, people like the free thing. All right. Well, so if folks want to check it out, go over to mocaproject.com. You can get Mocha. And he's got themes up there for Cinnamon, for Plank, for Gnome, obviously. He's got icon themes up there. Android wallpapers, too, it looks like.

Mocha for Android as well. Icon packs. Really a lot of stuff. You must be working your butt off. And, folks, I have to tell you, this kind of stuff takes a lot of time. So you might want to consider tossing a few bucks this way if there's something you like, if you feel like it makes your Linux a little more beautiful and increases your enjoyment. Sam, is there anything else you want to touch on before we run? No, not really. All right, very good. Well,

keep us posted because I'm a big fan of projects like yours. I think making, I believe, and if you look at Xiaomi's new MyOS, M-I-U-I-O-S-6 that they just released yesterday, to me, I don't know if you've noticed, but a lot of those icons look a lot like some of the stuff you're doing. I think you, projects like yours, are inspiring designers at these ginormous companies,

and they are implementing your ideas. I don't want to say ripping them off, but they're definitely being, I think, inspired by your ideas. And if you look at their new operating system, if you look at their new OS, a lot of those icons have a bit of a Mocha look to them, or some of them maybe have a little bit of a Numix look to them. And I think that's interesting,

and I think you guys are setting a trend, even if you don't have millions of users. So I congratulate you, and thank you for joining us, Sam. It was fun to chat with you, and keep us posted on future developments. Well, I'm flattered. I don't think those guys are influenced by Mocha. I get people compare it to Mui or iOS, but thanks anyway. No, I wouldn't sell yourself cheap. You might be surprised. Well, those have been around longer than Mocha has, I think.

Yeah, maybe. You might be right. I noticed it was sort of a – it seems like a trend that's been picking off. But okay. All right. So, Sam, thank you again. It was really great to chat with you. And keep up the great work, guys.

Go find them at MochaProject.com and kick in a few bucks towards Sam if you like the work he does. And try it out. You might build a – Sam, are you – can I also get it from distro repos? Are there packages available for Arch and Ubuntu and all the others? Oh, I think he's gone. Dang. All right. That's okay.

Oh, shoot. He'll be back. That's okay. I have him up there in staging working on that Echo, so he might be back in a bit and answer it. Or Rotten will get it for us. All right, shoot. He'll be back. Oh, shoot. That's okay. I have him up there in staging working on that Echo, so he might be back in a bit and answer it. Or Rotten will get it for us. All right, Matt. Well, we're going to talk a little bit about robotic cars. Producer Eric went to a conference and got a whole bunch of really good deets that he wants to share with us.

So before we get to that, I'll just take a quick break. And I want to thank Ting, our buddies over at Ting. Go to linux.ting.com. Linux.ting.com. Why? Because you're going to save $25 off go to linux.ting.com. linux.ting.com. Why? Because you're going to save $25 off your first Ting device. Yeah. Or you'll get a $25 credit if you have a Ting-compatible device.

Either way, it's a pretty great deal. So linux.ting.com. That also just kind of lets them know you heard about them here on the show. And then once you go there, you can try out that savings calculator. Take a look at that bad mama-jama. Put your numbers into there. The actual usage. That's the trick.

See, ting's all about what you actually use that's get ready for this you pay for what you use ha right now yeah that's kind of concept right you're not paying into some imaginary number of minutes that you may or may not use and so you always have to budget high so that way you don't get screwed no every ting line is a flat six dollars plus any taxes for the man and then it's just your usage on top of that. They take your minutes, your messages, your megabytes, add it all up, whatever bucket you fall into, that's what you pay. And it's so easy and so straightforward once you switch to that model that you kind of look at the old model and realize what a gimmick it is.

So go over to linux.ting.com, linux.ting.com, try out the savings calculator, and put your actual usage in there. If you are in my range, now, I think the craziest we've had was Rotten Corpse. I think Rotten Corpse was like $3,200 or something ridiculous. Mine's like $2,000 in two years. $2,000 I've saved by switching to Ting. And here's what's great. I've got LTE everywhere I go. Built-in hotspot and tethering, it's included with the plan. Ting's like, hey, you want to use this as an ISP? Go at it, bro. I don't care. Hey, you don't ever want to use our data? Go at it. I don't care. What you got? Bring it. Ting doesn't care. They don't care because they just want you to be happy. That's why they've set up no-hold customer service, too. You can call them right now,

1-855-TING-FTW. We should call them during one of these shows because they're open right now, 8 a.m. to 8 p.m. East Coast time. A real human, an actual human being answers the phone and is like ready to answer your question. Oh, you got a question? Yeah. Oh, you need something solved? I can do that too. The other thing that's really cool about Ting as a longtime customer is their transparency. Over on their blog, they're constantly posting stuff like here's a behind the scenes video with product management.

And you can talk to them. You get an interview with their product manager. They've got the customer service reps on here across the whole board. There's the ICE Challenge, all of that stuff. It's really an awesome company that really is on a mission. They realized, hey, guess what? There's a market need to make the wireless industry better, literally. And they are trying to fill that.

The people behind Ting are also the folks behind some other great properties you might be familiar with. They're two cows, and they have a mission to do this. They recognize a market they can go into and fix things up. I've got to respect that, too. Linux.ting.com, go try out a device because you're going to take $25 off when you use that promo code. You're going to own that device. It's yours. You're only going to pay for what you use. There's no contract.

You're not getting locked into Ting. If you don't like it, you're not locked in. There's no early termination fee. You're not getting locked into Ting. If you don't like it, you're not locked in. There's no early termination fee. And to make things a little easier, they also have an ETF relief program for your existing crappy carrier that abuses you. They'll help you get out of that.

Go over to linux.ting.com to get started. And a really, really big thank you to Ting for sponsoring Linux Unplugged. So, Producer Eric, tell us about this conference you went to. Well, this was a conference that a friend of mine put on. It's called the Think Big Conference, and it took place in Coeur d'Alene, Idaho, at North Idaho College. And it was actually really good. It was just a one-day event on Friday, just this past Friday. And it was where we had basically my friend Nick. It's his vision to put together a basically bring tech companies into Coeur d'Alene, Idaho.

Coeur d'Alene has been kind of suffering from the economy. And he sees a great idea of bringing a whole bunch of different tech companies. And one of the things that one niche that is kind of coming out of this is robotics. So there's a whole idea of bringing the robotics industry into cordelain ah now so so that's just part of it but the first you know we're you're talking we're talking about robocars here and that's that's what uh brad templeton brad templeton is one of the guys who spoke he's the first guy who spoke this guy i, I mean, he helped build Usenet. He was there helping build the internet in the early 70s. And now he's working on Robocars.

Yes, correct. That's kind of what he's up to right now. Now, does he see it as specifically Robocars, or does he see it as, well, everything that's going to be mechanized and automated will eventually be a robot of some form? going to be mechanized and automated will eventually be a robot of some form he the way he sees it as is um in this case specifically he didn't really talk about that he just talked about the cars themselves okay i noticed a macbook up there so not a linux user no that that wasn't his that's my friend nick's laptop okay he was just holding the presentations um and but yeah he he wasn't a linux user from what i could tell um basically his whole point was that uh human drivers like 3400 are killed every year in the u.s costing 871 billion 1.2 million people die worldwide due to driving wow and it's just crazy the thing is we actually do have self-driving cars

that are here now mercedes and bmws both manufacturer cars that drive themselves if there are there's heavy traffic um so you can purchase those they're extremely expensive but it's but they're out there um but one thing that we were talking about was that if people had access to self-driving cars, then people with disabilities could benefit. It would be like a poor man's teleporter. You could call one up almost like you do with Uber. I've also thought about it like it's almost like a cab in a sense. I could potentially tell my truck here at the studio, go home, and then Dylan could get in the truck and it could just automatically drive back to the studio and he could visit Dad for an hour.

Correct. Exactly. There are all sorts of benefits. We wouldn't have to... We spend hours on the road. I mean, myself, driving up to the studio every now and then, that's a 90-minute drive. That's three hours out of my day. Right.

But in terms of if cars could drive themselves, then we could use that time. To drink and drive. Yeah, to drink and drive. There you go. I was blackout. Credit to Blackout in the chat room for that one. Yeah, there we go. To be more accurate, it would be drink and ride then.

Right, of course. Yes, very good point. But what's interesting is that it would potentially use far less energy than we use now um like for you know it wouldn't use if they could all be electric cars they could even be more efficient um the one one thing though that i did question i he there was a question and answer session. I asked, how much does open source play a part? Yeah, that's my question. Yeah, like for instance, what's

driving the car? What's behind the car? What's managing it? Basically, what Google is using is they base their project, their self-driving car, off of the source code from a car that Stanford University

had in a contest and won this project. Google hired those people, the people from Stanford, to develop the Google car that we have been making news. I thought the Google system was Ubuntu-based. It is Ubuntu-based.

That's why I said the source code actually for the card that won that contest that Stanford has, that's available on SourceForge. So you have that. But most of the companies that are developing self-driving cars have proprietary code, and they're keeping it locked down. Yeah. And there's confidential. He said he couldn't disclose too much without basically screwing over his NDA on a lot of these things. So, but he wasn't clear as to what my question was.

I said, well, what if people want to audit the code to know what's driving them around? And personally, I don't know about you, but that's why I wanted to bring to the discussion today is what kind of things we expect from this. And this is coming fast, too, because we talked this morning on Tech Talk Today about the vehicle-to-vehicle communication standard that the U.S. is working on that will be mandated to be in all cars sold in the U.S. by 2020. And it allows for them to communicate speed to each other, potential collision information, things like that. And that's, I would assume, designed to facilitate localized automated driving decision making. And I also deeply am concerned by, it's sort of like, it's not the same thing, but I have the same feelings about voting machines, is I really feel like that code needs to be public.

The public needs access to that code. And in order to have a truly nationwide driverless system where I could be comfortable putting my son in the truck by himself and having to drive him over all by himself, it really needs to be something that everybody's on the same page with. And that means whatever software we pick to manage and run it, i think it's going to be the one that gets used for decades right by everyone right yeah and you know i was basically my question was inspired a little bit by karen sandler who we met at oscon and she was the person who wanted to see the source code for

her pacemaker and was basically overall told no that's right yeah yeah so it when i hear stories like that i'm just like that's a safety issue there are people who are going to want to be able to see that source code so they can trust what they're using more well it's also a safety issue in the sense of bettering public safety by allowing more really smart people to look at that and find problems and fix them right it's It's Lance's law. You know, the more eyes on a project, the safer it is. This is, you know, and the other problem is, and this is something we touched on on Tech Talk today, was the security implications here and how, like, you know, when these vehicles are communicating with each other using this vehicle-to-vehicle communications technology,

they talk in, I think, what did it say, 10 packets a second or something like that? Not crazy. Or 10 times a second. That's what they said, 10 times a second. But they're communicating things like, this is my current speed. I am doing 60 miles per hour. If you could intercept that signal as somebody, right, you're not hacking their car. You're hacking your own car.

And then transmit false information. You could potentially alter all of the cars around you, either bringing them to a stop or speeding them up or causing chaos. And I go back to open source means we could audit this and look for these types of vulnerabilities. Obvious statement. We all agree on that. But we are watching the legislation and the industry, not necessarily always together, steaming down a path that I don't think as Linux users and open source advocates we're going to be all too comfortable with. Yeah, and that's one thing that Brad Templeton touched on was that there is definitely going to be some security implications.

And there is definitely going to have to be some security work done on cars because, like for instance, the the cars that are on the road with the computers that are on board them, they're not built with security in mind. Right. That's going to have to be a reality for whatever is running the cars in the future, especially self-driving ones. Yeah. Big time. Big time. that have the CAN bus where the Bluetooth radio is on the same bus as the brake monitoring system because that's the way the regulations specify it must operate. And you compromise the Bluetooth radio, and now you have access to the CAN bus system.

It's a bad situation. Yeah, exactly. It's just a bad situation if you compromise any part of that. And I guarantee, and this is something we've mentioned before is that there's going to be there's going to be times where people are going to have to do updates to their software you know it like alan says patch your s yeah so yeah i mean or there's going to be or there could there be car manufacturers who eventually ban abandon applying updates to these cars and how will you get people to properly apply updates to their vehicle? Because they're sure as hell not doing it to their routers, even though it could be very harmful not to. People don't update their routers.

People aren't updating, like, a lot of their set-top devices. They just don't bother. Right. So the car is going to have to also, likely, using the argument of public safety, be self-updating without user intervention. Exactly. And then do we want closed-source software that's self-updating without user intervention. Exactly. Do we want closed-source software that's self-updating? Yeah, and that's the other thing I'm seeing. Current cars with

their GPSs, you have to actually pay for a GPS update. I think that's ridiculous because that could potentially be a safety issue especially as cars get more and more automated. Unless the update's his celebrity voice. Yeah, that's true.

Gotta get Arnold. Get Arnold. I think Tom Tom made a Snoop Dogg one. Matt, I don't know how you feel about this. This is one of these where it's like, part of me is totally ready for automated driving. I still, of course, want to be able to drive, but the other part of me is pretty freaked out by a lot of the security and

privacy because another another thing to keep in mind if every person is being driven by a central computer system then everybody's being tracked where they're going does that bother you well so here's the biggest piece of it for myself from my own perspective anyhow is that i think you got to look at it from let's get our house in order before we start handing over more things for people to regulate when they don't understand the technology they're regulating anyway um we can start off with the internet and let them actually learn how it works first then when they got a handle on that then let's let them handle regulating how our cars work that scares the hell out of me because not only are you going to be concerned about closed source and stuff like that but the fact of of the matter is, where is this person from that is in charge of the regulation?

What's their background? Do they understand it? Do they have a vested interest in a past company that is going to benefit from this new regulation? There's a lot of holes in that that concern me. So I think we're way too early on that, not from a technology point of view, but from a common sense point of view. Yeah, I get you there. Derek Doven, what do you think about it from a data security standpoint? From my understanding, even though we don't have as much efforts in the automated cars in Europe, there is already legislation in place.

And I really hope that the data is considered from the user. is considered from the user. And if that's the case, the data is considered yours. Under the current digital laws, you actually have the right to tell them to delete everything, and they have to comply legally. Otherwise, they will get fined like hell. I think as consumers, one thing that maybe we can influence,

and I think we have to demand it from the manufacturers, is we do not want the Android update catastrophe that we have now where Samsung releases one update and maybe Motorola releases another update at a different timeline and LG and et cetera, et cetera. Right. Wouldn't that be awful if Toyota puts out an update that maybe Honda takes another year to put out and maybe exactly Lexus never ships it. Right. And so you have all these different cars and what what will force, that will make the system stupid, right? It will make it vulnerable and stupid. And you could have literally people driving around in 2,000, 3,000-pound vulnerability boxes. You know, I mean, honeypots. It has the potential as you make these cars more sophisticated, and you're actually running an entire operating system on there to get dicey fast.

I'd like to hear the audience's ideas for solutions, and maybe some folks that are out there kind of working on kind of some of this stuff have some insight. So go over to jupyterbroadcasting.com, click the contact link, and choose Linux Unplugged from the drop-down. Hey, one final point. I just overheard this while Brad was talking, and the person behind me said, if cars are driving themselves where's the fun yeah there's

that right yeah that's huge I do enjoy driving that's why I always hope that at least in my lifetime on the main roadways like freeways and highways that's automated and then when you get off the freeway the steering wheel reactivates your seat sits up and it's okay you're driving now human and by the time you're at the

end of the off ramp it's transitioned into human mode. Right, right. That would be because, you know, because really I don't need to drive the freeway. And if the computer can do it twice the speed I can, because, I mean, then we could maybe – then maybe we'd never have traffic jams or very few traffic jams until there was a breakdown. And that would be great. But on the off-roads, let me have my fun. I've got some laws to violate.

Well, in EU – violate yeah go ahead in eu they actually it's mandatory that you um to be able to disable the uh automatic driving and be able to drive the car as well oh really so why are you guys so why why is why are they so on the ball with this when it doesn't seem to be the same stink necessarily behind automated driving? Why are they so ahead of this already? It seems to me that even though we don't have the companies, the companies are struggling a little bit because they're putting their feet in the water. And it's also a way of inviting companies. Look, we have a well-defined framework coming here, and you're ready to set up shop and do your business. Yeah, and I guess, too, they want to get left behind, right? Yeah.

Because I was looking at this proposal that the U.S. is working on that they announced today, and it's U.S.-specific. It only applies to U.S. cars. It's only mandated in the U.S. So it could mean that Europe ends up with an entirely different car-to-car communication system than we use in the U.S. But not the end of the world, but look at the steering wheel situation. We somehow made that work.

It's not like we did that with measurement. Oh, wait, what? Get the king in here. All right. There is an elephant in the room that people have been asking me for days if we're going to talk about it, and it's been snowballing. It's this report coming out of Munich that they're about to abandon Linux

and switch back to Windows. But before we get to that, I want to thank Linux Academy. I think Linux Academy is a great sponsor for this show because you guys know what's up. You know how this works. We only pick sponsors that we think are truly great. And I got a chance to meet the guys behind Linux Academy. And it's really cool because I can relate with a lot of it. You know, they were off on a path, an IT career, which was great, but really felt motivated to do something that could help the Linux community.

And so they combined their efforts. You know, they've got web folks there, they've got educators there, they've got Linux admins there, and they're all working together to create Linux Academy. If you go over to linuxacademy.com slash unplugged, you can take advantage of a special 30% offer. It's the summer of learning over at linuxacademy.com slash unplugged. Go there right now if you would. I'd appreciate

that. And sign up. And here's why you want to sign up for Linux Academy. They're always making it better. They're constantly adding new courseware, holding live events where you can ask the educators directly. They have step-by-step video courses that you can download and watch offline or audio versions if you want to listen while you drive, kind of like a podcast. The study guides are great. You also get to keep

those, download those. And it comes with your own server. So if you get somewhere in the course material where you're ready for a server, they'll spin up the VM and make it part of an integrated course experience. And each course tells you how long each section is going to take, so you know how much time you need to dedicate that evening or that day. Whenever you're ready to just make yourself better, you go there, you log in, and you know, okay, I've got 45 minutes to make myself a little better,

to intellectually challenge or stimulate myself. Why not take advantage of it? Plus, with their 7 Plus Linux distributions available, they'll automatically adjust the courseware to match the distribution you've chosen, which is really slick. And you know that they were thinking, some people aren't even going to need this. Pretty much everybody's going to choose Red Hat, but they're Linux users. So they know. They know what's up. And they're also expanding their OpenStack sections and AWS sections. And they have good scenario-based training in all of those areas. So when you walk away from Linux Academy, you'll really feel like you've actually implemented one of these things in production, because part of the courseware will involve you actually

deploying an application that depends on the services you're learning about. I think that's really smart. Plus, since they automatically spin up a server for you, you're not worried about raising a bill on the Amazon web services while you're just learning because that's all part of your Linux Academy subscription and they're managing that for you. It's pretty slick. So go over to linuxacademy.com slash unplugged. Check out some of their content. Once you become a community member, you can hang out in their forums. People will help inspire you if you ever get stuck and need a little bit of motivation, just that little push. You know, like people recommend like if you're going to diet, you should diet with somebody else. Or if you're getting off the booze, you know, you should find a partner or sponsor.

Go find your sponsor over at LinuxAcademy.com slash unplugged. And a big thank you to Linux Academy for sponsoring the Unplugged program. Nice. Yeah, it is a great one. Okay, let's talk about this Munich situation, because it is snowballing out of heck. One of my favorite headlines came from Softpedia. Munich disappointed with Linux plans to switch back to Windows. Bracketed updated.

And we've also gotten some other ones, like Ed Bot's having... I don't know if you saw Ed Bot, Matt. He's having a heyday with this. Old Microsoft commentator Ed Bot is pleased as punch about this munich news but it sounds like the story might be a little crap is anybody in the mumble room kind of up to date on the the munich situation and want to chime in with some background well isn't it the deputy mayor of munich that is just proposing going back to Windows and nobody else. That one guy, yeah. Yeah, just the one guy. Yeah, one guy. Yeah, that's kind of that.

You just nailed this entire story. And what's so funny about this, kind of proud of myself too, actually, now that I think about this, I was re-evaluating stories Friday, Saturday, and Sunday morning for the Linux Action Show. I always am looking at our mix and constantly thinking, well, maybe I'll add this to the news.

Because sometimes, you know, we might cover four news stories in Linux Action Show. I always am looking at our mix and constantly thinking, well, maybe I'll add this to the news. Because sometimes, you know, we might cover four news stories in Linux Action Show, but I've followed 30 or 40 that week personally and chosen which ones to put in the show. And that's why I feel really bad when we get it wrong, but I feel good when we get it right.

I saw this story. It initially popped up online. And the problem really comes, it's coming from a German media outlet and it's all in German. And so there's a translation problem that happens. The headline's awesome. And then once guys like Ed bought and run with it,

it kind of adds some legitimacy to the story and then it's everywhere. And this is a way where the Linux community eats their own freaking tail because it started here. The rumor started in, not here, like on Jupyter Broadcasting, it started in the Linux community. We started this and then we snowballed it, and then the rest of the media, who now pays close enough attention, picked up on it. And now the meme is that Linux has failed Munich, and they're going to switch back to Windows. Even though it's the deputy mayor, it's the one guy, it does sound like there has been some legitimate customer satisfaction issues from the end user standpoint.

It also sounds like there's been some improper training and any kind of transition is difficult. So I don't mean to discount the entire story. But what I am not so pleased with is just how, once again, we're seeing another story go crazy. Am I missing something? You know, I don't think so. I think you pretty much nailed it because if it essentially really stems down to one guy – Well, Daredevil wants probably – go ahead, Daredevil. You probably want to say that one guy maybe has the power to make the change.

Yeah, and that is a legitimate concern if he's saying that. But I do believe it's a false disappointment. Don't you think that's – even still, shouldn't the headline be Deputy Mayor decides to roll back from Linux, not Munich decides to roll back from Linux? Well, he decides over Munich, I guess. So, for all that matters... No, the MSA council. It's not as flashy. For example, for me, it's a

disappointment because they've done an exchange of programs. There is little time, and from that exchange, they hadn't invested even half of the money they were investing in Windows. there is little time and from that exchange they hadn't invested like even half of the money they were investing in Windows to be actually saying that they did a degradation like they tried to save everything at once and of course

that will bring bad results do you think it do you think maybe Microsoft is influencing this deputy mayor since they just set up an office in Munich? Yes. Yes. And they did mainly because of that guy. Yeah. So this is kind of like... He was recently elected, wasn't he? Yeah. Yeah. This is kind of like how the city of Seattle was planning to deploy fiber internet everywhere,

and then when the mayor who was backed by Comcast got elected, they killed it. Now, it's a little black and white. That's not exactly what happened, but it has that feeling from afar. However, since I'm not sure, and, but this is at least from people I know a little bit, the reaction in Germany is because also these NSA spying things actually lead people to derive a little bit away from these U.S. companies. And as a result, I'm not sure if the citizens are not going to pull up some to revert that situation, if that's the case. Yeah, the chairman's also pointed out that he's been a self-admitted, quote unquote,

Microsoft fan. He says Microsoft has been flexible and helpful. I bet they have. I bet they've been extremely flexible and extremely helpful to the city of Munich recently. It really doesn't matter what happens now. The damage is done. The fact that this has spread like wildfire around the Internet, that Linux was a disaster and Munich made a tragic mistake. It really doesn't matter what the decision is now.

That is the headline the decision makers are going to have in their mind. The correction will never be issued. There won't be a correction ever made. No, and Microsoft will be lapping this up. I'm sure it will be up there on their whymicrosoft.com website before long. Along with that you can practically hear the sigh of relief.

You know, these old quotes that they like to roll out. Yep, and it feels like this is a reenactment of some of that. And I think you got to look at this and think who really loses are the citizens because, you know, this is an opportunity for the output of their government to be stored in open formats using code that benefits the people. And when a government, when a large institution is using things like Libre or OpenOffice, then they have more people using that. That helps that project. So that helps everyone who uses those projects in the entire world. And it is a great way for a community-funded organization like a government to do something that helps everyone in the technology field.

And the fact is, is the things that they are creating for those people belong to those people. And if those documents cannot be opened by the people, then it is a disservice to the citizenry. And so it is not just like saving money. It is deeper than that. And that is, as our buddy would say, the most egregious thing about this. So, Chris, my perspective on this is you don't just switch from from windows to linux overnight well it took them 10 years you don't exactly and similarly you're not going to switch back overnight they're not going to nip down to best buy a pc world or whatever and buy a thousand laptops running

windows 8.1 and say there we go we've switched back again they're going to have a mighty project to switch back again yeah and that's the opportunity at which people who are stakeholders in the Linux community can bid for that work or back up our claims that the total cost of ownership on Linux is lower and that you have more freedom on Linux. I think

one guy making a statement that they're going to move back is not... It's more of a PR hurt. I'm not so worried about... I mean, whatever happens to one particular place happens. It's just more of the – just adding to that narrative that sort of hinders Linux adoption. It just drives me crazy because it feels a little dislegitimate. I'm trying to think of the right word. Illegitimate?

And the fact that the Linux – in inverted commas, Linux-friendly press have jumped on this probably more than the mainstream press have gives it more legitimacy than if this was a press release from Microsoft themselves. That's also kind of like, and isn't that the exact group of people who should be getting this story right? Like, who does it help by getting this inaccurate? Like, it doesn't help anyone. Yeah, research is apparently done. But here's the thing. I saw this story on Saturday before last and decided, you know what? This doesn't look like it's worth running. This is more hype than it is anything.

I recognize that Saturday night, like at 9.30 after putting three kids to bed, and I would manage to suss it out somehow. And I'm not some genius. Clickbait. It's so annoying. Slow news day for many people. I guess. You know, maybe it is that.

Gosh, that's cynical, but maybe it is that. Yeah, maybe it is. Anyways, we'll keep watching it because it could develop into something. I think maybe that's why it does have some legs is because it could eventually, PR damage is done. Like Popey said right now, if they did switch,

it'd probably be another massive transition, which we would then have to report on ad nauseum. But we'll see what happens. But don't believe everything you read on the internets. And it's funny because it does hurt a little more since it started internally first, I think. I don't know, maybe that's also why I'm just a little sour about it is because it started in our community.

It wasn't, like Poppy said, it wasn't started by Microsoft or something like that. So it cuts deep. Cuts deep. Chris, you've got something of a platform here. And Mike Saunders, who is one of the Linux Voice team, lives in Germany. And he's spoken about this and understands the issue

quite well from a German perspective. So maybe it'd be worth getting in touch with him and getting on one of your shows because it's a platform to help air the truth. Boy, it'd be great maybe to have him come on next week if he's around. I'll try to get a hold of him. Let's chat more in the post show about that

because, yeah, that'd be good to get him on. You could join us in Mumble and kind of break it down for us. Before we run, just a couple little pluggy plugs. If everything goes as planned, we might, boy this is a big if, but just check out the calendar, check my Twitter feed. We might have some live feeds

from LinuxCon. Noah's kind of wrangling all together right now, so if it doesn't work out, it doesn't work out. We will have interviews soon from LinuxCon going on starting tomorrow. We may have a live feed from the floor. And if we do, it'll be over at jblive.tv. I don't know when it would be because we're not going to stream the sessions. That's their domain.

But we might stream some stuff from the floor. So just check twitter.com slash chrislas because if it happens fast enough, I might not even get a chance to add it to the calendar. It might just be one of those click-click things. So just keep checking my feed, and we'll also record and have all kinds of good coverage coming up from LinuxCon 2014 happening right now in beautiful Chicago, downtown Chicago. Don't forget, I'm also seeking your runs Linux, your personal runs Linux.

It could just be a quick blurb with a picture. You could take it with your smartphone of your Linux setup. A video is even better if you've got a few minutes to maybe do a – is it called a selfie video? I don't know. But I'd like to see your setup and maybe your face. And you could upload it to YouTube and then send it in to LinuxActionShow at jupyterbroadcasting.com with a link and a description. I'd love to make those my RunsLinux picks for upcoming LinuxActionShows. I've been having a ton of fun doing that. And last but not

least, if you love what we do, if you enjoy a few of the shows over at jupiterbroadcasting.com, please consider going over to patreon.com slash today. That's where we're crowdfunding expansion of the Jupiter Broadcasting Network. It is a Tech Talk Today Patreon in the sense that the show is

associated with it. But the funds go to all of the shows on the network. They pay for physical goods, contract goods, software we have to buy, or whatever it would be. It just kind of depends on the month or LinuxCon expenses, for example, for this month. And that money over at Patreon.com slash Today funds everything, not just the Tech Talk Today show. The Tech Talk Today show is kind of like my thank you.

That's why I do that show. It's like I feel bad taking your money, so here's a show I'm doing for you every single day to say thank you, that's why I do that show, is like, I feel bad taking your money, so here's a show I'm doing for you every single day to say thank you. And you can support any of our shows by going to patreon.com. Slash today we have pledge levels in there that you can just jump in at. Matt, is there anything else we want to cover on today's Linux Unplugged? Well, I think you hit all the important pieces, and that's definitely going to be interesting to see where the whole Linux situation goes overseas.

That whole thing just stinks. Yeah, we'll watch it. We'll watch it. All right, Matt. Well, I will see you on Sunday, so have a great rest of your week. It's nice out. It's really nice out. I know it's weird, but go out there.

Get some sunshine. That big orange thing in the sky. Yeah, now I'm going to go home to Ange and the kids, and we're going to go play in the water park. Nice. Yeah, so everybody say hi to Ange in the chat room. Say enjoy the water park. And don't forget you can join us live, jblive.tv, 2 p.m. Pacific on Tuesdays. Hang out in our chat room or join our mumble room.

It's an open mumble room. You can get the address in our chat room, jblive.tv, jupyterbroadcasting.com, slash calendar for your time zone. All right, everyone. Well, thanks so much for tuning in to this week's episode of Linux Unplugged. We'll see you right back here next Tuesday. Thank you. But I did have a post-show topic. And I'm going to drop it on your brain faces, and you just let it marinate. Let's see, where was it? Good question.

Good question. Oh, yeah, here it is. Yeah, here we go. Here we go. Okay, I promised somebody I would bring this topic up to you guys. I want to get your take. Corky started a thread. He said, is rolling really the future of Linux?

Oh, boy. He says, with Linux Mint Debut Edition and SolidXK kind of announcing plans to go from semi-rolling to more sort of stable, are we seeing the pop of the rolling bubble? Anyone want to take that one? Well, here comes the Arch fanboys. I'd say rolling is... Oh, you cut off, Rotten. Your mic's still on, but I don't hear you anymore.

Weird. Is that better? Yeah. Okay, I was thinking maybe the rolling is not necessarily rolling rolling or stable as in one is better than the other. It's more of a particular type of experience that someone wants. So the fact that Linux Mint Debian is going more stable, which makes sense because the other version is stable, SolidXK going more stable so they can just not have to deal with it. stable so they can just not have to deal with it. And if they're trying to go to a particular type

of the audience or whatever, then it makes sense that they would do that because they're going for the new users or the business area, whereas rolling is typically not associated with that. Chrome OS

is rolling. Chrome OS is rolling. Anything rolling properly vetted is rolling. Chrome OS is rolling. Anything rolling properly vetted is fine for businesses. I mean that's the thing. It's just heavier vetting, just as much rolling, but heavy – just vet the hell out of it. It's great for businesses. I agree. I use rolling in production as well, but I'm just saying like the people who want rolling can get rolling.

And there's really no – there really no which one is better. It depends on the person choosing it. That's true. I like Sam. Sam VDE in the subreddit had a good response to Corky. He says, Hi, Corky. I think the discussion is getting parked by a lot of distros, and perhaps the big, fat super distro story is coming to an end. The real value of rolling is the availability of services and applications. And I think the sourcing

model is about to change. Shouldn't the future be as simple as we finally standardize on one package format, powered by SystemD. Big projects start shipping source code and containers, containers for each iteration. Big ones take the lead,

like Firefox or LibreOffice. I, as a sysadmin, install an application, and I'm able to set up per container, if I have it, auto-update a rolling, thus getting a custom rolling release where the application levels are rolling, but they're totally isolated from the main system. The distro upgrade becomes far less critical. This is the CoreOS philosophy. You containerize everything, and then the damage when you update the CoreOS is much less widespread. Worst case, it doesn't boot. You fix that problem, everything's back up and running.

The libraries, the software, and the containers were never touched. Distro upgrades would become far less critical and easier to manage. This is also something that Fedora Cloud is working towards. Both Fedora Cloud and Core OS, especially Core OS, not so much Fedora Cloud, truly believe in updating as frequently as possible.

Core OS is building an enterprise-grade server OS that's being deployed by Rackspace today on the concept that a server operating system should be rolling. Just containerize the apps. I guess it boils down to, are we talking about desktop users or servers?

Well, I think the technology applies in both cases. Just like we see it right now on phones, you bring in sandboxing, you bring bring in containers you isolate the application off from the system all of a sudden it doesn't really care it might not even care if you switch from qt4 to qt5 like it could be that level of separation yes it adds waste blah blah blah but if it's managed properly it's not so bad hard drives are big Chris, we did exactly that on the phone. Each of the apps is containerized, and we switched from Qt 4 to Qt 5. And how did it go? Comic things?

It should be fine. There were some apps we did an audit, and there were very few, like less than 10, that needed to be recompiled because they had binary components. But the vast majority were QML, so it wasn't a problem. That's got to be the future. It's got to be, right? Yeah. Well, that will be coming to the desktop. There's no doubt about that.

What we're doing on the phone, we're going to push to the desktop as well. So you'll get image-based updates and click packages in containerized formats. I don't know. I mean, the chat room's giving me a hard time. I've had a couple of things break on me. I'm trying to think of the major breakages, and it's been mostly VirtualBox, but I solved that like seven months ago. I'm thinking from an update standpoint where on a server,

you're not going to want to update it nearly as often as, say, a desktop. Therefore, I wouldn't use Arch on a server, even though I did at one point. It's just kind of experimental. But the thing is, you've got to update that thing every so often, otherwise you go to update it and it breaks. I come at this from two aspects. I come at this from a

security-conscious standpoint. Old software that hasn't been fixed by security seems really unhygienic. It's very gross. It's nasty. It's dangerous. I'm a little risk adverse in that sense. It seems risky, not just for yourself, but also as an internet citizen, it's bad for everybody because your machine can get owned and used for bad things that launch attacks against other people. So

there's almost a social responsibility to have my router updated, my boxes that I connect to the internet updated. I almost do feel a bit of a social responsibility there. So to me, getting new code in there, I feel like we are such a bunch of wusses. Like we have somehow locked ourselves into the philosophy that whenever software updates, every update has to be huge. What about just only a rolling distro

that pretty much just is always rolling security fixes or things like just minor stuff? It could be like, I just, I wish we could, I wish we could redefine what rolling means and have one standard definition that everybody agrees on because rolling to one person is

not the same rolling to another person. And then they'll argue at it from their individual standpoints, like, well, rolling means that you could have major feature changes in an application and that's scary. And then another person's like, yes, but rolling means a config file could change and it could break. And then I'm sitting here going, none of that in practice has been a big problem. And I have fresh software that's totally up to date and secure as possible. So there is obviously some pros to it on desktop and especially on server.

Then it just becomes a matter of making it work. Well, guess what? It's just technology. It's just computers. We have the ability to figure it out. So I think the idea idea rolling itself is fine where i where i am becoming increasingly frustrated though like say with an arch based distribution is the fact that i have no problem no matter how bad it gets borked i can get to a command prompt

i can revert the package who gives a rip but when i just updated 1500 of them no yeah that's when you piss me off it's like look i'm reasonable fella, but this is getting to be stupid. It's too much. I guess the Arch problem is like you go two, three weeks, and now it's like 500 packages or 400 or 200 or whatever. Well, and the rub is that here's the real rub. That's my problem. That's my choice. That's my bad.

I'm the one. It's too big of a bonehead to update on a regular basis, and so I got to eat my own dog food. That's not against Arch. That's the fact that I'm really lazy when bonehead to update on a regular basis, and so I've got to eat my own dog food. That's not against Arch. That's the fact that I'm really lazy when it comes to updates. I don't care. I don't want to deal with it. I want to do it occasionally and then leave me alone. Yeah, no, I get that.

And I think that's a totally legitimate use case because that's how a lot of businesses are. That's how a lot of the machines here in the studio are. I'm busy. But you've also got to own the responsibility of how you handle your updates. I own the fact that I am really neglectful there. So I own that. And Rikai points out one of the things we've done here at the studio is we have two Arch servers.

One that we primarily use. And what we just did is, alright, well, these are the roles we envision for this. And if it goes out beyond this role, we spin up a new VM. So that way, each box individually, like, I have gone two months without doing a package update on that Arch server,

and then I'll do an update, and it's like 25 packages. It's not a big, yeah, no, it's not bad at all. And it's like all like, yeah, it's all safe stuff. Well, I think when you're using rolling, you're using your own level of stagnation, depending on what your needs are.

Yeah. And you can control how many packages you're adding are. Yeah. Yeah, and you can control how many packages you're adding to it. I mean, if you're like on Arch, for example, and you've just gone on a shopping spree on the AUR, you know, you're responsible for that, for sure. Yeah, yeah.

That's like my bonobo, I admit.