LINUX Unplugged - Episode 168: Linux Shadow Force | LUP 168

Episode Date: October 26, 2016

We get the inside scoop on what happens when Canonical gets a bunch of employees & community members in the same room, discuss the cool open hardware project Chris just ordered for the studio & update you on the big community highlights of the week!

Everybody suffers from downtime every now and then. And there was recently a major DDoS attack. And with that, we got some great fodder. Air, Azure, what is it? AirsAzureThrows.tumblr.com. AirsAzureThrows.tumblr.com. Of course it's a Tumblr. Is a collection of Azure Airs that are really great.

And, you know, recently I've been struggling where I have services where part of them is broken. And I've realized it's because they depend on all these different various microservices that keep going offline. I'm looking at you, Telegram bots. This is a great Tumblr if you want to smile. ErrorsAzureThrows.Tumblr.com. This is hilarious. This is Linux Unplugged, episode 168, recorded on October 25th, 2016. Oh, welcome to Linux Unplugged, your weekly Linux talk show that has a real soft spot for open hardware.

My name is Chris. My name is Wes. Hello, Wes. We'll be getting into a little open hardware. My name is Chris. My name is Wes. Hello, Wes. We'll be getting into a little open hardware purchase that I just made. Very excited. We'll talk about that a little bit in the show. But as is our traditional recipe these days for an unplugged show, we'll be going through some great community updates.

We're going to start with one that I'm personally super excited about. Am I not just dancing right now? Oh, you are dancing off the wall. I was totally just dancing on the intro there because I'm so excited about this. Then you've probably heard about Dirty Cow by now. But did you know it's going after your Samsungs and your Motorolas and your HD? Yeah, your Android phones. It's a train wreck.

And we'll give you a little more specifics about Dirty Cow. And if you really need to worry if you're an up-to-date Linux user, then we'll get into archival of some of our long-term memories digitally. We'll get an update from Popey and Wimpy about their trip to a recent event. That's right. And then we'll get into that hardware surprise. Oh man,

this is something I have been waiting for for two years. Well, yeah. Waiting for this product for about two years. We have had people that have created this product on two different Jupyter Broadcasting shows. And I'm going to play an interview for you today in today's episode of Unplugged. Then I'll unveil what it is that I got. I am so excited.

I'm going to have it right here in the studio. Oh, I know. I know. And it actually relates to something we were just talking about in the post show last week. Oh, no way. It all ties together. It all ties together. But before we can get into any of that, Wes, there's a protocol that must be followed.

A duty that must be followed through on. And that's bringing in the most important part of the show, our virtual lug. Time for appropriate greetings, Mumble Room. Hello. Hello. Hello. Those wonderful people. That is a virtual lug right there.

Okay. Now tell me you can't a virtual logger right there. Okay. Now tell me you can't be a little excited about this. I am super excited about this. Freeablo. Freeablo, a work in progress, free open source replacement for the Diablo 1 engine. Simply import your Diablo assets and enjoy the same old game with better performance, modern resolutions, and first-class support for mods.

Is that Linux I see on the top right of your screen? Yeah, it is, buddy. Yeah, it is. No way. Yeah, yeah. So they love the original Diablo. In fact, they love it so much, they say they're going to spend their precious time developing a free and open-source solution for those that want to play it on a modern computer.

I think I'm down with that. Yeah, and they say it's legal. They won't distribute any of the copyrighted game assets. You have to have a copy of those assets to play. But everything else around it, they're going to open source. See, that's great. That seems like a pretty clever way to do it. Yeah, and they're working on multiplayer right now.

So, I am so excited. I had not seen this. Did you find this? How did I find out about this? Good job, Wes. How did you find out about this? Good job, Wes. How did you find out about this? Where did you see this? You know, I actually don't know, but I had it sticking in one of my open tabs.

I was like, okay, this deserves to be in the show because I had left it for myself, obviously, because I wanted to play it. Yeah. When I saw that come across the link list, I was like, yeah, yeah, we are freaking talking about that. What better way to exercise your left mouse button, right? That's true. That's true.

All right. Well, we did it last week and the tradition continues this week. We're going to open up the show with a little bit of breaking news here. This is CNN Breaking News. So you've probably heard to death by now about the dirty cow vulnerability in Linux, the most serious Linux vulnerability ever. Ars said that, a few others have said that. Now it seems that Android phones are getting rooted by this vulnerability.

This feels like this is where this matters. Yeah, your router, your fridge, your IP camera. But that's still pretty niche to the main consumer. Yeah, so it looks like there's a new method for rooting Android devices that's believed to work reliably on every version of the mobile operating system and, of course, a wide array of the hardware. Individuals can use it to bypass limitations imposed by manufacturers or carriers. They can also use it to sneak apps in for malicious purposes.

The technique comes courtesy of the Linux privilege escalation bug that we talked about on Linux Action Show called Dirty Cow. And so some people are calling the vulnerability, or it goes back to kernels, the vulnerability goes back to kernels since 2007, which pretty much means every damn Android phone ever made. Right. And several independent security researchers are doing proof of concepts right now. Ours interviewed two of them. David Manchouser told ours that the proof of concept code that exploits Dirty Cow on

Android gets devices close to root. And with a few additional lines of code, he can get persistent root access on all five of Android devices he's tested. A separate security researcher asked not to be identified said that several other people have developed separate root exploits based on the SturdyCal exploit. They say we're using a rather unique route that we believe we can use elsewhere in the future. And because of that, they don't want to tell Google about it. Oh, of course. Right.

Because they don't want Google shutting down the vulnerability. So they're keeping it on their own. This is bad, right? This is an example of why, sure, you can get your Fedora machine updated, you can get your Ubuntu machine updated and your Archbox updated. What are you going to do about your Samsung phone? Joe Smith walking around with an S6 or an S5 or an S3 or a Motorola X or whatever. Yep. Especially on one of our big-name carriers here. Oh!

You have to buy a new phone once there's a phone that is patched, I guess. Yeah. I don't know. Anybody in the mobile room think we're worrying too much about this stuff? Because personally, I feel like I can't overemphasize this. I feel like I'm sitting here

yelling down from the mountaintop. You're walking around with vulnerabilities in your pocket that could expose your personal data. A disaster is coming. A flood is coming.

That's something you've been thinking about a lot personally, I think, in the last year or so. Am I overreacting? Sometimes. With the personal data thing, I feel like in the last year or so. Am I overreacting? Sometimes. With the personal data thing, I feel like Android does it pretty well by

just vaguely saying what an app will do with your data. Yeah. And they do have sandboxing. And there is, I mean, and ours does a good job of mentioning it, you know, like there are, there is that double-edged sword of, you know, some users will benefit.

I have already linked this article to some of my friends who are on phones that they've wanted to root but can't. So there are benefits. But I also do worry about that. I mean, realistically, I don't know what other people are going to do, but does it just further the bad press about the Linux kernel? Does it make other people less likely to build a similar platform using that kernel? It also makes me feel like all of those people,

like the black phone, or everybody who's created these ultra-secure Android phones, there's a dozen different models right now, they're all bullshit. Because they're all walking around with this kernel vulnerability. The only phone that's legitimately safe is probably a Nexus that gets the patch soon,

or a Pixel phone. Yeah, that's probably true. Or one that's not safe is probably a Nexus that gets the patch soon or a Pixel phone. Yeah, that's probably true. Or one that's not Android. And speaking of this, I want to give you a little more details about Dirty Cow. Also quaintly known as CV20165195. Rolls right off the tongue. Dirty Cow, which naturally has its own logo and webpage now, which is kind of actually a funny one. It's a sign of our times.

It is kind of a good one, though. They do a very good job of it. Yeah. It's a race condition in the kernel's memory management subsystem. By timing things right, a local attacker can exploit the copy on write mechanism to turn a read-only mapping of a file into a writable mapping. And then with that, a file that should not be writable, you could write to, you could write to, which means change certain parameters, change like the permissions, change the ownership.

I'm in the suit over. Yeah. It doesn't take much in a great – or I'm sorry. It doesn't really probably take much imagination to see how the ability to overwrite files could be used to escalate privilege in a number of ways. The known exploit depends on access to slash proc slash self slash mem, which is not universally available. which is not universally available. Even some systems that don't provide it at all or access to it. But there are still ways to exploit the bug potentially.

The exploit happily bypasses almost every hardening technique out there. Strong sandboxing with mechanisms like SecComp might slow it down. But counting on that is probably not wise either. This is an article at LWN. The only real protection is to upgrade a kernel containing the fix. You have to upgrade with a kernel that has the fix. The workarounds aren't even all that reliable. So Dirty Cow is a real son of a bitch.

And it's not the only one that's being exploited on Android right now. So there is also the Rowhammer vulnerability which allows the attacker to bit flip over to root. Researchers have devised an attack that gains unfettered root access to a large number of Android

phones, exploiting a relatively new type of bug that allows adversaries to manipulate data stored in the memory chips. This is actually, I believe, a vulnerability

more at the CPU level than it is on the freaking OS level. Yeah, it's a hardware problem. Yeah, the breakthrough has the potential to make millions of Android phones vulnerable, at least until the security fix is available,

to a new form of attack that seizes control of core parts of the operating system and neuters key security defenses. Neuters them. Equally important, it demonstrates that the new class of exploit, dubbed RoHammer, can have a malicious and far-reaching effect on a much wider number of devices than previously known, including those that are on ARM. So this seems to be an issue so far tested on the Nexus 4, Nexus 5, and the G4 from LG, the Moto G models from 2013 and 2014 made by Motorola, the Galaxy S4, the S5 from Samsung, the One from OnePlus.

However, not in all cases cases the results are consistent. Only 12 of the 15 Nexus Model 5s that were tested were successfully rooted, and only one of the two Galaxy S5s were compromised. So it's something to do with the chip, and the researchers that are testing this have made an app that you can sideload to test to see if your CPU is vulnerable so they can get a wider sample base. Holy crap! And this they did tell Google in July. Well, that's good.

Yeah, they say the vulnerability, Google, says the vulnerability is critical, which is their highest rating, and have awarded the researchers four grand. So Google thinks this is a legit issue. I like that. Until recently, we never even thought about hardware bugs, and software was never written to deal with them. And there is no quick software update to patch the problem and go back to business as usual. That's a shame.

I do wonder, though, it's kind of making me think. I feel like we don't see the same kind of press about jailbreaking on the iOS side as we do about getting... It's because it's a superior system. Oh. Really? I mean, come on. Let's just admit it. I mean, yeah, you can have your fanboy position,

but you're right. You don't. You don't have this problem on iOS. And if you do have this problem, they patch faster. And that's a freaking reality of the market right now. Maybe another vendor will hopefully do it better soon, but this is not what Google is good at. It's kind of the perfect example of why, you know,

they do have problems on iOS, but they patch them and, you know. My hope is the Pixel fixes this particular gap. I really hope the pixel fixes this in a way that a lot of people have. I just – you ordered one, right? Yep. Yeah. See, I really hope, man, maybe you'll be our canary because this is bad. And to be honest, you know, the other thing that really pisses me off is that comment that, well, until recently, we never considered hardware vulnerabilities and didn't design our operating system.

Bullshit. What is data execution prevention? Come on, give me this. Come on. Really? Why did Apple design the secure enclave so that it doesn't pass any information to the underlying operating system or the ARM chipset? There's a reason they designed it like that.

Of course we know of hardware vulnerabilities. x86 is filled with them. And surely ARM is as well. So just say, well, we... We never thought of... It's like bull crap. And even this one has been, you know, the Rowhammer's been around for a while now.

What do you think, Swift? I honestly think that for the most part, people can't afford the Pixel. So it's asking a lot to expect them to get it they need this security for a cheaper phone yeah i definitely yeah i definitely agree there what do you think mini mac about the pixel yeah i would hope that the pixel gets more than three years of support because of just these kind of vulnerabilities i don't pay a thousand bucks for a phone only for three years and then

have these bucks coming on me and I can change anything. This is probably the number one bit of feedback we got on last week's user error where we talked about the pixel and we talked about the update cycle and so many people wrote it and said, man, everybody

in my extended friends and family that is not – oh, yeah, we also cover this on TechSnap. We've got a bunch of emails on TechSnap. So both shows, we got a lot of people writing and saying, I might update my devices more frequently or I can't update my devices more frequently. But you know who never updates their devices is all of my friends and family who aren't tech. Exactly.

People that aren't into tech. To them, an update is just an annoyance. It might break things. They update their phone when it breaks. Or you go into the store and they update it for you. Or whatever. Maybe when they all die off, that'll change, but that seems to be the way people operate right now. And so these

have to be five-year devices. How do we teach the next generation that updates are good? That updates are something you should do? They sure like buying things. Yeah, right. Maybe they already got it. I think the other thing that really drives me crazy about all of this is I don't accept the answer that it's not possible to update these devices for five years. That's – tell that to the banks that are running – Or just Red Hat in general.

I mean like they're backporting. Yeah, tell that to the servers that have been running a Red Hat distro for seven or eight years. Tell that to the banks that have mainframes still getting support by IBM that are in there for 30 years. Yeah, but Chris, to be fair, the people who are getting updates from Red Hat are paying for them. Once you've gone through the retail experience and you've walked out with a phone, you give zero money to HTC or Samsung. It all goes to Google. I would say that, but it's Google's responsibility to do the updates, not HTC. There's also no option for me to pay anyone to do that if I wanted to.

I guess I— Except for those maybe value-add people. I don't see how this isn't just a technology problem. This seems like it actually—this doesn't—I guess here's what I mean. To me, Popey, this doesn't seem like a technology problem at all. They technically could update the software for as long as they want. I think technologically possible. They simply don't have the will from a corporate standpoint.

No, I'm not suggesting it's not a technical issue. It's a purely financial one. Nobody's paying them to do it. And it's probably also easier for them to, you know. Otherwise, you have to do things separately so they can roll their features. Back in the day of Android 2 and devices never getting updates to, you know, this has been the same for years and years and years that devices wouldn't get updates. And it's slowly got better. But it's always been the case that if you don't pay for someone to actually do the hardware enablement and do the backporting then

you're not getting any revenue from those customers. They've walked out the door, right? You're getting revenue from people who are buying the Pixel phones or the Next phone or the Next phone. You're getting nothing from those people who walked out the door three years ago. Google's not making any money off of Android users besides the hardware, so I'm not

I don't agree. HTC and Samsung and all the others. Well, that's their then they made a mistake shipping Well, that's their... Then they made a mistake shipping Android because that's a separate argument. That's a totally separate issue. The fact that they can't make money off of Android. Well, they've done pretty well off it. I'm not

saying they're not making money off it. I'm saying they're making money off... They're making continual forward revenue off the new devices they're being sold. They're not making any money off the ones that have already been shipped out. You know, Toby, I understand that they're not making money off of it. I think this is Google's responsibility. Just like Canonical

can now live patch the running Ubuntu 16.04 kernels of servers all over the world, I don't see why Google couldn't live patch Android handsets. The technology is there. It's a possibility they just simply don't have the will or the corporate contracts or the bureaucratic structure to's it's a possibility they just simply don't have the will or the corporate contracts or the bureaucratic structure to enable it but yet they still have the

nuts to sell these devices to consumers and literally put their personal lives at risk it seems like a total disconnect to me right and i and i agree with you you know i go to a store if you if you went to a verizon store bought a device, my contract is with Verizon. That's who I made the retail purchase with. I understand. And if Google reached into my phone and flipped some bits on my phone, I would be pretty angry as Verizon that Google are reaching into devices and screwing this up. I think that is an old argument. The carriers have to test the devices to make sure they work on the network.

But yet there are hundreds of millions of iphones receiving updates directly from apple and there are tens of millions of android devices running cyanogen mod and the carriers don't have any how many how many hardware how many different hardware devices are there that apple support like right now not that many look at how many different different devices there are that even one manufacturer samsung has look at the range of devices that are that even one manufacturer, Samsung, has. Look at the range of devices that one manufacturer has. It's a completely different ballgame, Android devices than Apple devices. That is true.

Although I wonder if something like that could be fixed with like an SLA type agreement between Google and Verizon. You know, if they do cause Verizon real problems, support calls increase by a certain amount on these supported phones with upgrades. You could have business arrangements that might get around that. I want to give other people a chance to jump in and kind of push back a bit, too, because I know people have been tagging me. Mosinrath, you had a comment about shipping Android, about going with Android and not making any money. Go ahead.

Yeah, just like a lot of people have been saying in there, such as Demerit, and I believe Sweet Lou has also made the point. You're going to want to get to him on it as well um swift 110 also made a statement about that in you got to take a look at with apple as you see you know they have the enclosed garden setup so they're making money off of the software the hardware you even thinking about it they're making money off sure the fact that you even use the word apple yes somehow they were in your pocket already but with say samsung and htc and all those other companies they ship the product you're buying the hardware from them and as was stated i believe it was by poppy a couple

moments ago that's it for them yeah well is that fair, though? Because Samsung's charging $800 for a Note 7. The Pixel, if you want a decent one, is $750. The 6P is an expensive device. They seem to be making a shit ton of money. The device hardware costs as much, if not more, in the case of Samsung, as an iPhone. Yeah, but as a lot of people have stated, they make the money off of you once. That's it. That's the only cash they get from you. So, of course, they're going to try to get as much as possible in order to develop the next product.

I feel like Android's... They don't get any more money from you. Well, Samsung doesn't, but Google. Google's constantly monetizing the Android platform. Yes, but that's Google that's monetizing the platform. I think that's who should be developing the updates, too. But Google can't develop the updates,

because there's no cooperation between them and the hardware manufacturers, because the hardware manufacturers keep the source to themselves. Well, that's not always... It actually works the other way. Google hold back Android updates from handset manufacturers as well. So Google hold back android updates from handset manufacturers as well so google hold back the latest version of android so they can ship it on their own devices before it goes out to samsung htc even tier one handset manufacturers don't get it as fast as google get it themselves

this is why the pixel is a thing now i mean the only one with that google smart assistant guys sorry it just cracks me up yeah The only way to get it. But I think it just tells you, it just shows you that even Google themselves recognize this is a core issue. This disconnect between the carriers, the handset manufacturers, and Google. It is not a great ecosystem. It is not a great ecosystem. And everybody's paying patent fees to Microsoft that gets in it. Like, everybody's getting screwed at all ends.

And Google's the only one making money. Samsung was, but. All right. So does anybody else tag me again in the mumble room if anybody else wants to jump in before we move on? I think what I feel is I understand how the business relationships are structured today. I understand how the source code dumps over the fence work from Google. today. I understand how the source code dumps over the fence work from Google. I understand how monetization works of these devices and how they make their money and how the carriers want to add their layer on top of it. I get all of that. And I ask you, if you were an alien traveling

from Mars and you arrived here and you looked at the situation, would you not think this is a dumb arrangement and say, well, the only thing that makes it this way is the artificial constructs that you hairless monkeys have set up to do this. And if you just simply did it differently, it would be better. And it would literally be better for consumers. It would protect their privacy better. Mr. Swift, go ahead. The fact that they don't protect people's privacy is something that's going to become a larger issue over time.

And, I mean, if you just have an affordable way to protect people and show that you respect them, I think you'll make more over the long term versus trying to grab, grab, grab in the short term and charge all this money, but you're not providing them with the kind of comfort that they deserve. Right, like you might need that to change the market. Yeah, they're in their walled garden and they feel safe and secure there. If you can offer the same thing in the free software or with regards to the phones, then you'll be in a much better spot. Yeah. In fact, WW, that's kind of what you're thinking, right?

I mean, geez, don't we hope that this is where free software could really help us? Yeah, because if you think about it, console game developers, to put it, to situate it kind of like that, when you see an update on that it's paid for the software development is paid for and they pay for the actual update process to get it on that hardware and to be tested on that hardware and whatever licensing sure and that kind of sucks we're on pc you can do whatever so i'm really hoping that with Ubuntu phone and maybe Popey or would be could answer this that or maybe Popey that would it be open so like if you're a

developer you can solve these issues you can contribute to Ubuntu phone OS and make it better without having to wait and all that stuff because as I see it now the only way this will be ever solved is having an open system where things could be more rapidly hashed out.

WW, when we were talking about, well, it doesn't make HTC any money to update these devices, in the back of my head I'm thinking, that's never been an issue for Linux development. No, you're right. It hasn't. We've developed and people have put in their... It's like the community can solve this problem.

And that would be Ubuntu Phone's advantage, right? Yeah, maybe. May I jump in here? Yeah. So what I know is that manufacturers like Qualcomm and others, they only build devices for the Android platform. So the device is not open. They ship their own drivers. So you can't just build a Linux infrastructure and put Ubuntu phone on it.

That's why Ubuntu phone and Sailfish is based on Android because all these phones only run Android. So it's not as easy as WW thinks. Okay, fair point. Yeah, I have noticed that. And thinks. Okay, fair point. Yeah. I have noticed that, and that is probably a fair point. With the Linux development point you just made, the difference of Linux is

the people who are doing these patches are working for companies that want the code because it helps them. With Android, it's just being done by the companies to help support the device, and they're not getting money after that. So I guess we need google with its deep pockets to step up and incentivize these carriers to or when hardware manufacturers to to consent to updates or um as a market the pixel phone needs to be successful enough that it forces the market to react and that might be what

happens we just keep having these horrible problems and enough that the platform is at risk. Yeah, and like Sweet Lou points out, too, like, not everybody can afford these latest and greatest phones, which also puts them at security risk. I want to address that in just a second, but before I forget, I wanted to mention that I got a note from the developer of the tool we talked

about last week called Magic Device Tool. Oh, really? That allows you to install Ubuntu or Cyanogen or Phoenix OS or Android on your device. And I made the mistake because I think I misread this article. He actually does support the Google Apps. He's like, hey, thanks for the coverage. But by the way, we do support the Google Apps. Hey, even better.

Yeah. Thank you. That's great. Thank you for writing us to tell us. That was really nice. And so what a cool tool. And it's called, again, it's called the Magic Device Tool. And it allows you to deploy the latest Ubuntu touch release, CyanogenMod, with or without the Google Apps, and the factory Android app image, like a backup, including Phoenix OS on the BQ line of devices, pretty much all of them that matter, including the tablets, the Nexus 5, the Nexus 7, 2013.

Hey. Yeah. Which is, a lot of us have that one with nothing to do with it. Also, the 2012 Nexus 7, Samsung Nexus 10, which is a great tablet, and the OnePlus 1 and the Fairphone 2 are all supported by the Magic Device Tool. So it's also just a nice way to create backups to restore your device, and it's nice that Ubuntu Touch is a top-supported option. So basically, you know, I

bring this up because if this is something that you've been chewing on, because it's something I've been thinking about, like Wes mentioned, for a while, I kind of feel like this is at least my boat is now on the beachhead, and I have options. And once the 6P is added to that list, if it ever is,

that'll be the day. The day. I'm really hoping I could... That'd be a great device for it. Oh my god. If I could ever, ever run Ubuntu Touch on the 6P, I would be so happy. Because I am willing maybe one day

to just get an MX Pro type device. Oh man, I would love it on the 6PP because I don't mind the 6P hardware. It's a little big, a little heavy, but... But otherwise, nice. Great screen, great battery, and if you hold it real still, it takes a good picture. So that may be something that happens.

And the Nexus devices you can pick up for super cheap, too, if you want to play with Ubuntu Touch. Yeah. You know, there is another component that is sort of something we have to attack from both ends, and that's the carrier side of this. Because there are a bunch of bastards in this entire thing, too. So let's take a moment and mention our sponsor, which is a carrier that does this differently and does it right. Go to linux.ting.com.

They don't get in the way of your updates. They don't have a Ting experience that they have to develop and integrate into the image. They don't put their bloatware on their apps because they don't have any bloatware. They don't like prevent you from updating over the wireless if you want. Like it's just simple wireless. You pay for what you use, minutes, messages, and megabytes. You pay for each for what you use. So if you don't text a lot, you don't pay for any texting.

If you don't call a lot, you don't pay for any calls. If you don't use a lot of data, you don't pay for data. When you use it, you do pay. It all averages out. You go to linux.ting.com to get $25 off your first device or $25 in service credit if you bring a device,

which will pay for more than your first month, likely, if you average it out. This is something that's really nice because there's no contracts. There's no early termination fee. The devices are unlocked. They're not locked to Ting.

You can move them around if you want, which makes it really nice to bring devices to. When you do need help, you just talk to a person. Not a phone tree, but a person. You don't have to press zero like seven times? You don't have to mash it. And they have

CDMA and GSM, which I know outside the US that's not very common to use CDMA, but in the States with at least... We just love it. Well, especially on the West where you have the Rockies, there's a ton of trees, CDMA has some serious advantages, and, you know,

GSM has been absolutely rock solid for me as well, so it's really nice to have both options. So you can get a little bit of a taste of it by going to linux.ting.com. And then speaking, addressing Sweet Lou's point, speaking of, like, flagship type devices that aren't super expensive,

yesterday Ting just put up a blog post about five inexpensive smartphones with flagship performance. This is brilliant. So you can go get an unlocked phone with great performance, with no contract, $6 for the line, and you pay for what you use.

And you can read the blog by going to linux.ting.com and then click on their blog link. I think this is kind of just to sort of bookend our conversation. This is a huge part of that entire formula. Yes, it is. If I could give you, if you could afford it, and I could give you one recommendation about how to be secure with your mobile device,

get a device from the Play Store and put it on the Ting network. Yep. It's expensive, but you're right. Like, that's probably the best option. You can go this route, too, you know, and if these devices are getting updated by their manufacturers, Ting will not interfere with that. And so go read the devices they've listed here because the prices are, like, under $300, except for the OnePlus 3. But, you know, there's really some serious options, like the Wah the wahe the wahe is that how you say it hawaii hawaii the honor 5x 199 the 5x the 5x still a great phone i mean it would be a big upgrade for what was this guy yes it would in fact you know what i love about ting this is from their

blog right here here at ting we don't want to pressure you to buy a phone from the Ting shop. There's lots of special deals online, and we recommend that you take a look around. Just by waiting a few months and doing a bit of online research, you can get all the smartphone you need at a much smarter price. That is so plainly written, and it is so refreshing to hear that from a mobile phone carrier. They're not trying to create a giant vertical monopoly. They just want you to use their service because it's good. They give

you the couple of tips you need to go do a little searching. They link you to the Ting compatibility checker to make it even easier. They say over 80% of the phones made in the last few years are compatible with Ting. Wow. There you go. I think this is such a refreshing take on this. So get started by going to linux.ting.com.

That gives you the discount, but also lets them know you heard about it right here on the unplugged program. And right there in the middle, what would you save? Just experiment with that. That's fun. And consider, like, if you made a couple of changes. Say you marked your music to play offline and you download your podcast over Wi-Fi. How much data would you use when you're commuting if you did that? And if you switched to Ting and you made those two adjustments, how much could you save?

Check it out. Go to linux.ting.com. And a big thanks to Ting for sponsoring the Unplugged program. So, Wes, there was this post about the XZ. Is that how you say it? Yeah, XZ. The XZ form. It's only two letters.

It seems like we could say them. I know, but I don't know if it's Z or XZ. I actually don't know. I think I always say XZ, but I'm probably wrong. Yeah, well, it doesn't really matter because there's a post over at nonganoo.org that says... We can't just let that go by, right?

It's nonganoo.org? NonganU.org. I've never installed GNU slash Linux. They say that the XZ format is inadequate for long-term archiving. I think this entire post is bullshit, but I was kind of curious what you thought about it because you were the one that found this link. And then I wanted to kind of roll this into a slightly larger conversation with our mumble room about what they do for their long-term data archival needs. And I know a couple of them in there are parents. A couple of them in there probably have keepsakes that they want to have forever, videos, things like that.

The idea of like – before, I feel like a lot of people kind of discovered, right? You had like pictures that you found in your garage or your parents' attic or something. You're like, oh, I have these from when I was was a kid but if it's on a hard drive something like that is not the same in the digital age that is such a good analogy so uh i have photo albums from my childhood that my mom gave me like here and these are yours to hang on to now and you know once every 15 years or whatever i've opened them up and i've gone like holy crap remember when we had chickens remember when we cut a hole in the side of the house and put a window in? That's so neat. I'll look through it and I'll be like, this is really great.

And maybe one day my kids, like I've gone through my dad's photo album. And I found that to be fascinating. Wow, look at my dad as a kid. He looks like a kid. Totally. My dad's just a kid. Those are good memories, right? You get to experience that.

There's nothing like that. There's no book my kids are going to go grab of them. Because it's all on a hard drive. Or it's in some cloud account they don't have access to. And I'm going to be honest with you. This is one of the reasons I sacrifice every sacred freedom and privacy aspect I have by uploading all my photos to Google Photo because I'm just desperate to have some sort of long term archival so that way when my kids are 15, I can look at them when they were four

and five and remember what that was like. I don't want to not have that. So I keep trying to think like, well, how can I do this where I control the data, I have a good archival system, and it is – I guess I want the digital equivalent of a photo album or other things like it. And I'm kind of curious what your thoughts were before we get to that. What are your thoughts about this particular article about how the XZ format's a bastard and shouldn't be used because of checksums? Well, okay. One, I think the Savannah, the non-Gnu website, they support free software. I don't know their long history, but I know they support free software, but they're not associated with the free software foundation,

which, as we've discussed on the show, you can love them or hate them, but they definitely have an ideology. Zombies don't run Linux. Exactly. So this is, fair disclosure, this is written by the author of the LZIP tool, which is kind of a competing tool to XZ. They both use the LZMA compression algorithm, which, let's see, that's a fun one. What is it? Oh, LZMA. Lempel-Ziv-Markov chain algorithm. which, let's see, that's a fun one. What is it? Oh, LZA, yeah.

Lempel Ziv Markov Chain Algorithm. LZA is great because it's not the best compression ever, but it's great speed, works great on a file system level. I've used it in combination with ButterFS in the past. Sure, yeah, yeah. And so this XE also implements that. That's what it uses to compress files inside, but it also implements its own container format.

And so this blog post, there are some things where I think that for the majority of use cases, it doesn't really matter. One big thing I think that did kind of ring true is they talk about how it doesn't support garbage data after the end of the file. That sort of thing where I think if you were going to do things with tape style backups, that sort of stuff, there may be some

validity. I think there are also some things that generally ring true and are good design principles here about you want to keep things simple, you can over-engineer things. There are some plausible arguments here about how XE may be over-engineered, especially for the most common use cases.

I feel like the argument that using XE for backup and archival is bad. The reason why it's bogus, he hangs his hat on things like checksums and whatnot not being built in, but there are so many other tools that if you just use the Unix philosophy, you know, if you use like SHA-256-SUM

and PAR2, there's ways you can do this and supplement the deficiencies of XE that are just built-in toolchains on the Linux system. Right, I think that's exactly. But I wonder, does anybody in the Mamba room – so, Inagogo, Mr. Gogo, you just made a great point. You said, in fact, speaking of you, your thing is next.

But, Mr. Gogo, you had a – I forget what it was, but you combined Tarsnap with something else, and this is your solution? Well, I don't use it, but I guess 4 would be an interesting way of doing it. Using a free NAS to do ZFS stuff, and then putting some of the stuff you want to keep onto Tarsnap, and then also having a photo album just in case any of those two fail. But that's not, yeah, that is a great way to protect the data, but I don't feel like it doesn't have that accessibility that I'm looking for, that a generation from now, when I'm not there to tell them how to get access to it, I'm not there to provide them the login.

But I want them to have access to it. That's what the physical photo album is for. Print them out. Moserath, what do you think? Is there a digital way to do this with like dis probably not like we've been discussing it in the chat room here in terms of like how long things will last and all that but as you can see by my last comment and what other people have said it's like i dare you to try to use your free nas when there's a power outage go ahead let's see you do it you can't and therefore that physical media even

though it is going to fade no matter what it is that you do is still the best one like i got a pile of pictures in a couple of photo albums i'm looking at them right now right i don't have those in any kind of digitized format i had them at one point of time scanned them all lost them all because i formatted the wrong disk. Oh, wait a second. You could do that just about anywhere, any place, any time, couldn't you? Yeah. Because there's no such thing as a setup, as far as I know,

that if you can access it to put stuff in, that you cannot access it and wipe things out. Right, and something that you know they're going to have access to generations later. I kind of like TechMav's solution is create a dead man switch. So maybe once you stop punching the button every few years, it automatically delivers the login credentials and a URL to your kids. Yeah, that would be pretty neat.

You know what? Let's punt this to the listeners that are listening after the fact, and let us know. Comment wherever you're catching this or wherever you prefer to comment or email or tweet, and let me know what you would do for this. And I obviously want it to be Linux-based. Right. I mean, open source, right?

I mean, you need things that you can have if you need to do something 20 years, 100 years from now that you can. Really? Seriously? Yeah, that's a key thing. Right, because that way you could send the code along with the files. So, Mr. Gogo, was it in the post-show last week that I was talking about how I'd love to have analog meters to represent my CPU usage?

I think it was in the pre-show. Pre-show, okay. submitted this to the subreddit, a video where he's connected a VU, or I think it's a voltage meter, I think, connected over USB to his laptop. And you can see he's got GNOME system monitor running here. And the CPU load that's in GNOME system monitor is translated to this voltmeter's analog dial. Have I described this correctly so far? Kind of. the um the gnome system monitor was just me demonstrating that it's quite accurate to where it's pointing sure but there there's a arduino sitting between the laptop and the voltmeter which is having data sent from the okay laptop by a python script which then turns that into a pwM signal that changes where the needle is.

WW says it needs more Nixie tubes. This is pretty awesome, though. I'll be honest with you. When I saw this, the first thing I thought to myself was, God, I'd love to have some bigger versions of these mounted on the wall by the main broadcast machines, the OBS machine and the Wirecast machine, and be able to have these analog dials so I knew how hard these systems were working. I would love to have these as props in the studio

and also actual useful readouts. How long did it take you to do this? A day after getting the VU meter. That is so cool. And then you put the code up on GitHub to do it so other people can play too. See, the other thing this makes me think of is like cloud stuff. Like what if you got your drop

that's running the server for your people? You want to see like any random metric that you just kind of want droplet that's running the server for your people? You want to see like any random metric that you just kind of want. Is that a thing? Could you do that? I mean, if you can make a time series data, I don't see why not. So you could have like your fricking droplet CPU load

represented with a V. I mean, there'd be a lot more intermediate steps, of course, but. That is so cool, Wes. Man, as the sysadmin in me wants that for all the things. Or like load average over the fleet. Wouldn't it be great for the different SATCOMs to know how hard the different SATCOMs are working right now for us?

Or what's your throughput on your bandwidth? Oh, you could do. Oh, man. This is Delta Flyer V2. So here's what we need. Now, I need Mr. Go-Go and others. I need you to come to the JB1 studio, and we need to spend two days out here. I'll film the whole thing.

You guys build these things, and we'll make an episode out of it. We need to do this. That sounds amazing. I think that is so awesome. That is, gosh, that's cool. All right, well, we've got to keep moving because this is the longest episode. This is a big show, yeah, right? So I wanted to give a chance to chat with Mr. Popey, and I want to take a chance to chat with Mr. Wimpy about a recent event they went to.

And it's not too uncommon, I think, but maybe just to recap, we could kind of set the stage. There's these events that Canonical holds from time to time where they bring their team or different members together, and they focus on a few things. And they say, let's really focus on development in this area right now. We have a distributed team all over the world and we're going to leverage having all of them in one space for a short period of time. There's going to be a hyper level of productivity and we're all going to walk away with this with sort of a shared picture, a shared idea of what the next set of goals are. And a lot of things often get produced here, but they don't often get discussed. People don't realize that this is happening. I don't know, guys.

How am I doing so far? Have I kind of set up with the general premises of what these events are? Kind of. It depends. Some of them are development sprints where the developers get together and actually hack on stuff. Others are planning sprints where there's loads of meetings and discussion. And this was more of a planning than a hacking sprint. so you can break them out some hacking that's brilliant yeah well we we have sprints all the time like there was one the week before we were there there's another one

this week and there'll be another one the week after but with different teams so this the one that we went to last week both myself and martin when was um was in the netherlands in the hague and um there was 200 and something people there 250 people there um all from canonical or where no there was mostly canonical people so design team desktop team uh snappy team community team um and others who i don't recall um t18 unity 18 meeting yes well done uh and like loads of other people who i've never seen before and also there were some community people there as in non-canonical people uh there as well who were invited um but it was mostly meetings there was a a room we had set aside for hacking because the snappy team were working hard on getting a release out the door and the release candidate was last week while they were there

and the final release is this or next week and so they they were hacking but mostly it was meetings um and uh i i imagine that for a company that is sort of dispersed i mean i understand there's offices but for a company that has a very dispersed workforce, it seems like these would be pretty essential to getting everybody that's working on this particular thing on the same page. So even if it's not developing code, you're getting everybody to – Making it feel like a team. You're buying it on the idea. Is that accurate? Yeah, it's very much that.

I mean everyone like regularly meets up anyway online you know we all use hangouts and mumble and irc we use all the tools rocket chat you know whatever but um when you meet face to face different things happen like you can have a conversation in a corridor yeah and someone will say wouldn't it be great if this happened and then someone will go back to their hotel room or the bar usually and sit back on their lap and then give a demo a lightning talk later that day saying we discussed this thing and look it's now a thing wow cool that kind of stuff is quite good fun especially when you've got like a couple of hundred um people who are all working on the same in the same direction um you know you can reach out to the right people who have the

you know the knowledge about whatever it is the right people who have the knowledge about whatever it is you want to achieve. Yeah, that's not just productive. What's fun about that is when I've been in those situations, that's been some of the most personally satisfying work I've done is it's just a lot of fun professionally to be working in a team environment that's firing on all cylinders or close to it. And so I kind of want to poke Wimpy's brain for a second because I imagine, except for the snap sprint that happened a little bit ago, but this is Wimpy's first canonical event as a canonical employee. And I would imagine starting at a place, you've heard these names.

You've talked to these people online. You've met some of them, but now all of a sudden you've jumped in the deep end and you're working with them, you're shaking hands. This must have gone from real to very real, I would imagine, for you. Is that true, Wimpy? It is. I mean, I got to meet all of my team in person who I've got to know most of them over the years of contributing that never met before including my manager so that was obviously good but then yes there's lots of people in canonical who've helped me along the way some of whom i got to meet in heidelberg earlier this

year and some who i got to meet for the first time uh this time around um but these sprints are pretty intense um efforts um they're absolutely exhausting for all the right reasons um it's it's pretty sort of breakneck speed during the day and then the exhaustion uh falls into beer and food in the evening um and lots of fun uh with everyone so yeah they're really good fun and uh you know i was a i think a stereotypical first time uh sprint attendee in that i came back with the ubu flu which i hear is something of a tradition yeah of course one thing that's different about you being – this being your first one, which most people don't get, was you were thrown in as a track lead. So you had to manage a track of sessions and not many people have to do that.

The deep end sounds about right. That is the deep end. Yeah, the deep end was quite deep. That's where he thrives, I think. I just stepped on Wes's switch. Wow, Wimpy. Well, that sounds like it must. So now you're back at home.

Other than the sickness, you must really feel like you have a job job now. I do have a job job now. Yeah. And we know what's in store for the next six months now, not just within our team, but in all of the teams and how all of the teams are going to be pulling together. To give you a flavor of what went on, we were all given t-shirts at the end of this event and on the front of these t-shirts it simply says snap snaps snappy snapped snapcraft so if you're in any doubt what the theme was i think that underscores it rather clearly. I think that is – I think it's such a fascinating time right now and Canonical has moved very fast on the Snap packages.

Today, Flatpak got a big update. So this is an area that's getting a lot of attention and a lot of focus right now. Wimpy, I guess going from the outside in, I think you – see, you are sitting in a position where you maybe had suspected that there was people in there that were passionate about this stuff. Or you probably hoped there was people in there that followed current events. that the Canonical group, you know a few names, but maybe you could name four or five, six people at Canonical, the people at the top, the people that join us frequently in the communities. But I wonder, Wimpy, going in now, did you walk away with a sense that people there are hardcore Linux advocates,

they're Linux users? What were some of the senses you walked away with and some of the things you suspected but were confirmed, what you can share at least? Without a doubt, these were the 250, 300 most smart people I've been in a place with ever. Everybody is an expert in something. Wow, that must be exciting. You'll not hear the names of many of the people that were there but they're all doing impressive things and creating great technology and that was my

impression was there are some seriously clever people around me at the moment isn't that a great feeling that must feel very rewarding to work with a group like that yeah it is it's it's uh terrific um so yeah that was my sort of general feeling was what an excellent group of people Canonical have got working for them. And I only got to see this one group cross-section of the people in Canonical. And I imagine it's the same as you go across the other divisions as well. same as you go across the other divisions as well. I think, you know, I universally I have had a an incredible experience when I go meet these companies and I talk to the people who don't come out publicly and I walk away going, holy crap, they're doing some amazing stuff that they're not even discussing.

They have people that are advancing technology that they're not breaking about. Exactly. A lot of these companies, Canonical included, when I meet the people, I'm like, geez, how come you guys aren't telling this story? This is really something. I think it's impressive and it's extremely reassuring to – when you get down below the surface, you get below the spokespersons, you get below the blogs, you get below the Twitter feeds, and you look at the people that are actually contributing, that is some of the, at least if you're a tech head, that's some of the most inspirational stuff about this industry. So that's really cool.

And guys, I don't know if there, is there anything that we should be looking forward to that's coming soon with Snap packages or anything like that that you can tease us with? So there were some lightning talks at the end of the days. that we should be looking forward to that's coming soon with Snap packages or anything like that that you can tease us with? Yeah. So there were some lightning talks at the end of the days, and one of the ones that I was quite pleased to see was Delta updates for Snap packages. Oh, no way. So if you have a LibreOffice Snap, it's quite chunky, but they're adding the ability for the backend to do deltas

so that next time you get a Snap update, it only does a binary delta. It's XDiff or something. And the guy who was talking about this gave a presentation, not just about the technology underneath it. I mean, he gave a quick lightning talk about, his name's Tommy Richards, and he gave a talk about um the lightning talk showing uh

downloading i think it was inkscape or something and then he enabled deltas and then showed downloading the same thing again and it was super fast i mean it was like noticeable difference between many many seconds and a couple of seconds um and then he gave a a talk about how that's implemented and the analysis that they did in order to determine what types of packages compress well. And it turns out there are some that don't compress particularly well, and there are good reasons for that. And it was a great learning exercise for them to figure out what are the best algorithms to use and what's the best methodology to develop this kind of software. And it was quite fascinating to watch. to develop this kind of software, and it was quite fascinating to watch.

Holy smokes. You know, as somebody whose home connection is on a MiFi and somebody who occasionally broadcasts from the road, I thank you so much for Delta Updates. They are my favorite thing when I have a distro that supports Delta Updates. That is, whew, my bandwidth thanks you too. I am legitimately excited about both Flatpaks and Snaps. I am probably personally a little more interested in Snaps because I'm already using a bunch of them,

and I don't have any Flatpaks there, actually. It seems like they reach the practicalness, the easy-to-use level faster, and so, therefore, yeah, that's what I'm looking at. Yeah, and the couple of apps that I wanted to try real quick or the ones I don't want to futz with are already available, and I would love to see more, absolutely. And I could totally see...

I could see a huge place for this in like enterprise or proprietary type deployments of things where you already have all your own dependencies and you just want a little bundle that you ship to and you're going to probably run some sort of Ubuntu cloud type operating system already. No kidding, Wes. I would have used the hell out of that back in the day. Yeah. You're just like, oh, everything works.

It's in this wonderful little snap. Bam. On in the server. You're good to go. Jeez, that makes me want to be a sysadmin again right there. That would be so convenient and straightforward. Damn. Yeah, absolutely.

And you know, Wes, I still have a chance. You still do. There's still a chance. Perhaps I could use my fleet of DigitalOcean droplets. One of our sponsors here on the Unplugged program. Go to DigitalOcean.com and use our promo code D-O-Unplugged. That's one word, and it's lowercase, and you plug it in there on your dashboard and you get a $10 credit. And this is why you want a $10 credit.

DigitalOcean is a simple cloud hosting provider that is dedicated to offering the most intuitive and easy way for you to spin up a badass Linux box on their crazy great infrastructure. They got great connections. They got all SSDs. They have a super sweet interface that's better than it could even... If this was a desktop app that you had to install as a package on your desktop, I'd be like, man, this is just such a great app.

It's a web app. It's all HTML5. It's a web app. It works on your phone. It works on your crappy browser. It's a web app that actually is useful and it's easy to use, but it's also extremely powerful. If you're a total noob or you've been setting up Linux servers for like 15 years, you're going to find this to be a nice UI. Do you agree?

Oh, yeah. Absolutely. And I've got to tell you, when you use our promo code D1Plugged and you get a $10 credit, you could actually get a lot of distance out of that because you can use hourly pricing over DigitalOcean. So if there's something you want to play with like OpenVPN or TinkVPN or maybe you want to play with En OpenVPN or TinkVPN or maybe you want to play with Envy for 15 minutes or NextCloud. I'm using it to compile my kernels when I have a

custom Arch kernel. You just spin up a droplet. Super fast. You run the code to compile the kernel, which is a maker or whatever. Which level droplet do you go with? I usually go to the... What's the two-core one? That's what I usually go with. This is what I was going to say. For three cents an hour, I was going to say because a lot of everybody goes with the $5

droplet, which is a great deal. It's a wonderful, a little web presence, your own domain, a blog, perfect. But, if you're just going to build, so if you go with the two core, so if you go with the West rig, it's three cents an hour, it's two gigs of RAM, a two core processor, 40 gigabyte SSD, they're all SSDs,

and three terabytes of transfer. I don't have a reason to go with the high memory droplets yet, but man, these things, they're crazy. Like 64 gigs of RAM. 224 gigabytes of RAM. And they're a ton of money, monthly. But of course, if you need that much RAM, that's probably not a big deal. But something where you're already doing some caching

or maybe you just need a big Redis that runs or memcache. And look, dude, if you switch it over to hourly, okay. You know what? I could afford 18 cents an hour. Those are all numbers that I'm okay with. Those are close.

Even the $2 an hour, like if I'm doing it for two hours. Right. At most a day. At most a day, maybe. And you're like, okay, well then that's fine. Yeah, it's a pretty cool system.

And then on top of all of it, they have really great tutorials. So you can just deploy a stack of apps or you can use one of their, like here's one, pretty relevant, how to protect your server against the Cal Linux vulnerability.

Well, that might be good to know. Oh, that's topical. Look at that. Look at DigitalOcean over there. So here's what I want you to do. Go to DigitalOcean, set up an account, and use our promo code DEOUNPLUGED. It will give you a $10 credit, and then you can try any of this stuff for free. You can run the $5 rig for two months, or you can go run the crazy memory rig,

or you can run the West rig and build your kernels and stuff like that. Do it. The West rig is like $0.03 an hour. It's not that much. I should do that for the – You know what? After the show, if I remember, let's talk about how I could do that for my – because I got that MacBook run on Arch that you set up during the show.

The kernel builds, they're long. You know what the nice thing about Arch is? It's pretty easy to run your own Arch repo or something you can add. Then you build everything on a big droplet. You transfer it over to your $5 a month droplet because all you need is it to be like an HTTP endpoint. And now you've got your own custom little Chris repo. Man, Wes, I'm going to buy you a beer.

DigitalOcean.com. Use the promo code DEOUNPLUGGED and go build your kernels in the cloud like a boost. DigitalOcean.com. Big thanks, DigitalOcean, for sponsoring the Unplugged program. All right, guys. I wanted to talk about something that I'm beginning to change my tune on.

Do you remember that when I did a road trip to Noah's house to, like, witness his home automation system? I was actually really impressed. I was impressed as well. And it's even cooler than we could capture on camera. That guy's got chaps. And on top of that, he's added on to it. And so I have kind of changed my tune a little bit on home automation. And there's this great—

You're tired of Rikai leaving the oven on? Is that what you're saying? Sorry, Rikai. I catch it. You've listened to user error. I was listening to it on the way up. Isn't that funny? Yeah.

Isn't that funny? Man, and you know what? I've done that too. The pizza situation since that episode has been rough. All right. So this is really cool though because it's building a home automation system where you run it all on your own land powered by Raspberry Pis. And I've been experimenting with something recently. This is going to sound so stupid, but just go with me for a second.

I've been experimenting with during the week weekdays, leaving my monitors on. Okay. Tell me more. Tell me more. Well, traditionally, I've got these two... Yeah, what's the monitor situation? Oh, man. I've got these great monitors. They're 27-inch Asus monitors. I've got two identical monitors. One is vertical and... Full HD, something else?

2K. So they're like... 2K. Nice. What is it? Like 2560 by 1440 or whatever? But one is vertical and one's horizontal. I love it. I love it. For doing research, show prep, or even just having like a terminal that's doing updates and a telegram session. The vertical monitor

is wonderful. It's killer. It's totally, totally killer. However, one's hooked up over HDMI, or no, one's hooked up over DVI and one's hooked up over DisplayPort. Let me guess, DisplayPort. Oh, man. Yeah. And it is, it is, I've actually got a good system, but so when I wake

the system up, the vertical monitor never fires up. Yep. That's the DisplayPort monitor. Yeah. Yep. You've been here. Oh, yeah. I've been dealing, I just got some new monitors at work, so we have six monitors now in a row in the room I'm sitting in with verticals on the left and right-hand most side. So it's a really nice setup.

But one side is all DisplayPort and the other side, I have one HDMI and then one DisplayPort and then a DisplayPort chained to the third monitor. Oh, really? Which the chaining is actually working wonderfully. Huh. But the power save, man. Yeah. That's crazy. I have a trick.

Oh. I have a trick. Oh. I have a trick. So what I do is my main monitor, which I set via GNOME's display manager, because otherwise the display port would by default be the main display. Right. And in fact, when GDM starts up, the login screen is sideways on the vertical monitor. And the horizontal monitor that sits in front of me is off. Perfect. And then I log in and GNOME 3 starts

and it rearranges the displays like a champ every time. See, I'm on Unity too, so it's not even, it's worse. Oh, okay. But that's only if I've booted the system. If I'm waking the system from sleep or if I've just turned the monitors off but left the system running and maybe like let the drive spin down,

when I move the mouse and hit the power button on both monitors only the DVI monitor wakes up the display port monitor never wakes up which is where my telegram is and where my terminal session is and all that I need that monitor so what I have done is I've installed

a GNOME extension. Of course. You click it and it puts your displays to sleep. So what I do is I click it. I have the horizontal monitor on. It is powered up. And the vertical monitor, the DisplayPort monitor, is off. I hit the sleep.

Yep. I turn the Display display port monitor on. I hit the control button on the keyboard. It then sends a new wake-up command to all the displays, and the display monitor, the display port monitor... That one comes back. ...pops right back on. Okay.

But this is about a 55-second process. Yeah. I've kind of timed it. And it's super frustrating. And one of the things I've done recently is during the weekdays, I've decided to just leave my monitors on. And by God, I actually feel more – I just sit down and I immediately start working. All you do is the lock screen and you're good, right? Or maybe not even that.

Boom. Boom. No. I mean, it's the studio, so. If anybody hacks me, there's only one person it's going to be, and it's going to be the beard. And I don't think that lock screen is going to be what prevents him. He's had rude since he got here, Chris. Not only that, but he knows what the password is for my account because all my studio machines have a shared password.

And so I don't bother with the lock screen. I immediately sit down and start working. And I have been, Wes, I got to tell you, it's made a difference. I feel great. But now I got this really annoying problem. I got to turn on my lights. And then there's two different places I got to go to flip a switch. And sometimes I got to turn on the fan depending on the temperature.

All this other hassle I got to do before I get to work. It's really highlighted how much the little things make a difference in your everyday work. This is actually what I'm saying. And so I've kind of been turning around to the idea of automating a couple of things. Like if I could automate the monitor wake-up process so they could be turned off. If I could automate the lights turning on at the right time and the fan turned on so that when I sit down, I am ready to go. If I could have like an NFC checkpoint or something at the entrance of the studio. Or when your phone joins the Wi-Fi.

Exactly. Or whatever. And if this, then that triggers all of that stuff and turns on. I've been thinking more and more like, not only would that just be nice to have, but great. I'll film the entire setup and it would make great content too. So it kind of seems like a double win, right? Yeah.

So I was reading about this setup on how they've done it. And man, this is sitting pretty with me. They said it was a huge project for their office, and they wanted light control, conference call room, occupancy signalization, individual access codes, audio systems with wireless music streaming, manageable audio experiences in different rooms like the bathroom. Honestly, I'm impressed that they were able to take the time to focus on this totally out of, like, this is a major accomplishment for not exactly a business request, you know? I agree. And they worked on, like, light controls and all that kind of stuff, and they wanted to be able to control it via a web application.

What matters is they built it all with this super badass Raspberry Pi system. They have one master Raspberry Pi, which controls all the low-level hardware. And then they have this thing called the distribution board. I'm showing a picture of it. Oh, that's beautiful. It's so beautiful that before the show, I just sent a picture of this to Noah. I'm like, Noah, look at this.

Noah, look at this. This is so gorgeous. This is so great. And then they have a second device called the relay driver board, which does what you would expect. It also runs a web server on the main Pi with a web application that allows the controls. They have occupancy signalization via movement sensors, which is how Noah does it in his house. They have TV panel management with Pis connected to different TVs.

And they have the Pis all centrally located in this rack. So if a TV – they have an HDMI feed that goes out to the TV, but all of the pies are in one place. You don't have to go out to the TV to service it. Genius. That's smart. It's like bringing a patch panel to Raspberry Pies. It's a patch panel of Raspberry Pies, and this is the part of home automation that's really

I'm loving, because I could do Ethernet for like, I could do HDMI over Ethernet for long runs, and i'm just i'm loving this uh i don't know how you'd ever afford all of it but i think it's a great system and they talk about how they set up their audio and all of that i after looking at this it's really kind of turning my head towards home automation again but in a way that doesn't feel like it's a tied to a product or a brand i think that's part of the part like noah uses a lot of interesting things

and i think he's a very cool it's almost like an analog approach but it does necessarily involve a certain amount of products or boxes or and it feels hard to replicate or like a very much like well I've invested in this company's way or at least I've invested in these things and I need this equipment to do it.

But Raspberry Pis are $35 and you know Ethernet is whatever by the spool. Plus I have Ethernet run to all the rooms in the studio already. You probably have the infrastructure to do it. You don't have the time, but you have the Linux knowledge and the wherewithal to do it. Yeah, we need somebody to come out in the studio and work with us, Wes, to do all this kind of stuff. We need some serious projects we can do because I've got like three or four Pis in the studio right now.

They're just sitting around, right? Or like Onion Omegas or who knows what. Yeah. Yeah, or Arduinos. We'll be eating that in a moment. Anybody in the Mumba room experimenting at all with any form of home automation, even if it's like the Hughes brand like I mess around with or anything like that? Anybody in the virtual log?

Yeah, go ahead. I've been actually looking into using two Raspberry Pis to solve an issue in my house, two raspberry pies to solve an issue in my house which is um the thermostat that i have is in one location and my um bedroom is at the other end of the house which means my bedroom in the winter is always cold yeah so i was thinking about replacing my simple stupid thermostat which is with a raspberry pi which i already have sitting and doing nothing um just connecting a relay to it and making it switch the thermostat on and off and then having a second raspberry pi in my um uh bedroom that talks to that other raspberry pi and i can tell you know during the night maybe switch the thermostat to the uh temperature reading from the other raspberry pi in my bedroom and during the day switch it to the living room solves that problem with uh i have to buy a relay

and uh that's interesting yeah thermometers thermometers. Right, yeah. You know, I've been thinking sort of the thing. It's funny you say this because I've been thinking about the same thing. When I finally set up whatever storage solution and server solution I'm going to have in Lady Jupiter for being able to do on-the-road production, it's all going to be inside a dinette booth, which is. Oh, right. This is in the rig. Right. And you've seen it.

They're like plush leather seats. Very comfy. And they're made out of wood. It's like you're putting inside a box that's insulated. And so I've thought about putting a Raspberry Pi or something in there with a temperature sensor that could automatically shut down when things get too hot because, sure, in October in Washington, I'm probably not going to have an issue. But what about in August? In Arizona.

And I'm not, yeah. Or I'm going down to California in a few weeks. Oh, yeah. It could be a little warmer. You park in the shade. I have no idea. Or one day maybe that side of the rig is parked in the sunlight and it's just getting beat with sunlight. The temperature is going to be all over the place.

And I've thought about, well, wouldn't it be great if I had something in there that not only was charting it so I had an idea of how things were going trend-wise, but also a fail-safe that said, oh god, this is getting bad. Shut down the cradle point. Shut down the storage array. Shut down the NUC or whatever it is. Yellow alert, just power down. And then send me a message. That would be my ideal situation.

So that's where I've thought about using it too. Even in the RV. You know, I've thought about using it too, even in the RV. You know, I've thought about it. It seems like there's a lot of applications. Once you've got that kind of like, I mean, if this, then that, that sort of style response, like there's a lot of things, you know, you can handle like pet feeding or care. You can handle turning on the AC, turning on a fan.

Yeah, that could be the route I go to, right? Really, you just want like a home API. Yeah, I looked into it. I really just want like a home API. Yeah, I looked into it. And if you don't like doing things by just using components that you buy on the wheel or whatever, you can buy these relay boards for Raspberry Pis. And they can handle anything from switching 5- circuits to 220 volt circuits.

Oh, that sounds really handy. Anything from thermostat to lights and power. You could literally just shut off power to your entire computer with that Raspberry Pi relay. Just before you do that, make sure that the only board in your mind can handle

the output and the voltage before you bring anything down. That could be bad. They make those and then all you have to do is they will probably be like I2C IO so you can just plug it into your Raspberry Pi and start

hacking. So if you're a software guy and not a hardware guy, they make it pretty easy for you. It's just a few extra bucks. I like it. This is exciting, and I could see... I don't know if I'll get it to the... I don't know. God, I hope I get to it this year.

But with winter, I don't feel like it's the top of my priority list. But I could see early spring as being a project that I actually tackle. It's kind of just interesting to think about it now because as I start to deploy this stuff, I'll have this in the back of my mind of how I want to do this properly. So I'll set the power up and the networking for it. So it's all there when I'm ready to pull the trigger. Get the infrastructure ready and then just, yeah.

And then when it comes time in the spring before it gets too hot, I'll have all the groundwork laid down. So if you've got ideas, feel free to let me know. Please do. Yeah, if you've done this, if you've tackled this particular problem. And you've got a VU meter that'll work with it, right? Yes, you'll have hardware dials. You can get them up on the dash. That'd be great. tackle this particular problem. And he got a VU meter that would work with it, right? Yes, you'll have hardware dials.

You can get them up on the dash. That'd be great. Who cares about the MPH? I'm looking at how my servers are doing. That's so funny. All right, well, before we go any further, if you ever wanted to get really in-depth on a particular topic in Linux or on the back-end infrastructure

or any of the things that people are really making money on Linux, like Ruby or Python or AWS or Azure or OpenStack, any of these kinds of things that you would like to learn more about, I would encourage you to visit our sponsor, Linux Academy. Go to linuxacademy.com slash all unplugged. That's where you go to support the show

and sign up for a seven-day free trial. That's new. I think that's a big deal because you can go in there and get a little taste. Get a little idea. Linux Academy has step-by-step video courses, comprehensive study guides that you can download and use offline. And the courses come with your own server, they'll spin up on demand. I think also if maybe you've been tempted by the certification life, right? They have really good training on that. And I mentioned this because I have been down this path, it's been a while, but I've been down this path.

And you do want training that keeps in mind your end goal. And I think they have a really good path. They have like these learning paths. There are a series of courses and content planned by instructors for specific tracks, which is nice. And they have instructor mentoring available. So if you do have questions, you can actually talk to a real human and get your answer. It's super transparent. Yeah.

You can get your question answered. You get a good grasp. So you really have a good grasp. You really have a good grasp of what you're doing. So when you go in there, you don't have that anxiety about it. It's not that you just prepared for a test. You learned this because you're applying it. You understand it. You know what to do.

Especially if you combine it with their labs and their hands on exercises. You go in there and you get real world experience on real servers. I mean, these are servers they spin up just like you would in production. And you have the whole box yourself. You SSH in and you get real world experience on real servers. I mean, these are servers they spin up just like you would in production and you have the whole box yourself. You SSH in and you're doing the job. That is great experience.

And it's only the kind of thing you get when a company focuses on Linux and open source. When it's not just a thing that they do, but it's what they do. Linuxacademy.com slash unplugged. Go there, try them out. And also you can take advantage of the new public profile feature.

I think this is a really nice feature to let employers know how your training's going. Which, I don't know, Wes, where you work, when I was in IT, training, like there was a part of my yearly, not salary, but there was a part of my yearly budget for me that included a little bit of training. Do you get anything like that? Yeah, there is a general organizational level training budget. Yeah. Yeah, I think I'd be curious to know if people out there have tried this with Linux Academy because to me—

It seems like a great way. They have great team support. Exactly. It's a super easy sell. Yeah, exactly. It's a super easy sell plus it's something you can do where you have full control and it's not just like fluff content. It's by people that really know Linux and care about it.

They're constantly renovating the content. And with the public profiles, it's very easy at the end of the year to show what you've done. Exactly. You know, you can show your employer, you can be like, look, this is what I've done. Here's how I've improved it. Yeah. I think it's probably, if I was still in that field, it's probably my perfect solution. And I wish something like this was around back then. It's also the perfect thing for you. Like, well, I like Linux. I don't really know what I'm doing. Or you have a job that doesn't, you don't get to use Linux like you'd like to. Yeah, yeah.

Linux Academy is like just the golden road for you. For about a year, I was using Linux Academy just to scratch itches because there were so many things I'd want to try. After I got out of doing that as a daily thing, I still had these ideas. And Linux Academy with the labs and the training, it was a great resource for that. And the other thing that's kind of nice is like sometimes there's a topic that I don't feel like, man, I've read the story. I've talked to the people that make it and I'm still not wrapping my head around this. And Linux Academy has courseware on it. And the thing that's really nice about that is I look at it and I go, oh, that's going to be three hours.

OK. I will spend three hours on this. And with the downloadable comprehensive study guides, the audio you can listen to, like that, sometimes I'm podcasting this stuff. And if you have a little bit of downtime when you're commuting, maybe you're a passenger or you're on public transit, I really think the iOS and Android apps could be a great way to learn while you're on the go too. Oh, definitely. So check them out, linuxacademy.com slash unplugged. That's how you support the show. And you can sign up for a seven-day free trial.

And a big thank you to Linux Academy for sponsoring the Unplugged program. So last week, we were talking about the perfect keyboard. And I told you I was a little worried about getting obsessed with the keyboard. Yes. And I walked away from the show and I was like, gosh, you know, I remember, I remember, I remember this Keyboardio project. Does this ring a bell at all to you, Keyboardio? Keyboard, you know, that does sound really familiar. Keyboardio is pretty cool.

And it's, it's a group that Noah and I had a chance to talk to a little while ago where they're embedding an Arduino Linux running board inside a keyboard, a solid piece of wood that they have. Look at that. Are you, yeah. Are you seeing this? Wow. Yeah. This is a really, this solid piece of wood that they have. Look at that. Are you seeing this? Wow. Yeah, this is a really slick device. It's a split keyboard, which is ironic because I was just recently saying

this original Microsoft ergonomic QWERTY split keyboard is how I became a touch typist. Oh, really? And that was what took me up to a really fast typist, and I kind of wish I still had that. This is the model 01 they say it's the best keyboard they can make and it's not like other keyboards they mill the model 01 from two blocks of solid maple it feels good to rest your hand on the wood instead of shallow uncomfortable key switches they use glorious tactile mechanical key switches. They use glorious, tactile, mechanical

key switches similar to those found in the original Apple II. They've custom sculpted each of the 64 individual keycaps on the Model 01 to gently guide your fingers to the right keys. It's

heirloom construction, they say. I love that. I'll play a little bit of their video. This is pretty cool. How hard could this be? And then we built the best keyboard we possibly could. It's not what you'd expect from a keyboard, but people love it. Creating a keyboard is kind of a weird thing.

It's been an obsession. Hi, I'm Jesse Vincent. I'm co-founder and CTO of Keyboardio. I'm Kaya Decker. I'm the CEO of Keyboardio. I'm also married to Jesse. I've actually, before we go any further, you know what, I've talked to both these individuals in person, and I've used the original prototypes, and I can tell you they've been obsessing about this for a very long time. This is a discussion we had with them in 2015. We're here with Kaya with a keyboardio. Is that right? Keyboardio. That's us. All right. Keyboardio. So now there is a lot of talk of different keyboards. There's, you know,

they have the steel series and they have DOSboard. And then there's the people that just use the one that comes with their OEM chip. Right. What is it that sets Keyboardio apart? What is Keyboardio? Sure. So Keyboardio is an open hardware project, which is why we're here at Ascon.

And it is a comfortable, ergonomic keyword. So it's built around the shape your hands work. So it's designed to be more comfortable to type on for long periods of time. And it also was designed by a programmer as the primary designer. So there's a lot of things that you type more often if you're a programmer or a sysadmin or someone else who does that type of work that we've moved into especially easy to hit places. So brackets, for example, I'm sure probably take a huge high priority then? Yeah. So brackets, instead of being off on the side to your pinky, they're just right above home

row. You can hold down the function key and hit U-I-O-P, and that gets you your curly and your straight brackets. WASD moves the mouse if you hold down function. HJKL move the arrow keys, which if you use VI or BIM might be familiar. Yeah, and any keys that you hit and hold down, what's called cording, so control or shift or command,

we've taken off of your pinkies and put onto your thumbs, which is a lot better for you ergonomically. And yeah, it just, it feels comfortable. It's a little bit faster. It's better. So you said that this was designed specifically to be more ergonomic and you used the word typewriter. And so can you give me a little bit of history about how the typewriter keyboard was designed?

Yeah. I mean, so there's a lot of, people written entire theses on the typewriter and the QWERTY layout. Maybe just the cliff notes. Yeah. So it was literally designed in the late 19th century. And so it was based around what you could build then. So everything was laid out in rows. Touch typing was actually invented after QWERTY was invented.

And so there's different stories that I don't know if any of them have been proven around, oh, it was designed so that it was maximally inefficient to type on. One of my favorite stories is that why shift is on your pinky as opposed to sort of in a more convenient place. And the reason is, is originally there'd been an uppercase typewriter and a lowercase typewriter. And they figured out, oh, you can literally shift the plate and the thing that the hammer strikes on a typewriter. But they'd already done their tooling for the little hammers. And so they're like, well, we'll just put it on the side. And so one lazy engineer put shift on the side there, which is why now control and command

and all these other things that didn't exist when the typewriter and the QWERTY layout was invented ended up there. And so we've moved them into a place that we think is more sensible for the way that people use keyboards and computing today. Well, regardless of if it can be proven that the keyboard was designed to be maximally inefficient, as you put it, there's no contesting the fact that if you have a hammer that rises up, that hammer takes a certain amount of time to fall back down. So if you get two of them that are across from each other and they hit, that would be a bad day. So, yeah, I would imagine that some design went into it. It certainly wasn't designed for optimum typing, right? Yeah, definitely.

So tell me some of the advantages of your keyboard versus, like, the programmable ones that I can buy in Best Buy or Office Max or Newegg. Sure. So it's programmable with the Arduino IDE. Sure. So it's programmable with the Arduino IDE. So we're going to have, we haven't built it yet, but have a sort of software GUI configurator. And then it's also, if you want to do really crazy things, you can. So if you want to extend it by making it wireless or Bluetooth,

we give you all the pins that you have in an Arduino, so you can hook up anything that works in the Arduino ecosystem, and it will play nicely with it. In addition, it's mechanical keys, like the comfortable layout and so on, I think are a big benefit. But in terms of programmability, we're trying to give people as much rope as they want to play with, like actually give you firmware source as opposed to saying, well, you could remap the key here and there. It's really, truly fully programmable. So right now, these keyboards, they're being programmed with a config file? Yeah. So it's a couple of text files right now is the way it works. We're going to make it so that it's easy to do through a graphical interface. But, yeah, the way it works inside is actually sort of astonishingly simple in some ways. I think that would appeal to a lot of Linux users.

We're really used to just opening up a text file and changing the things we want. Now, where is that text file stored? Is that stored on the keyboard itself or on the computer? It's stored on the keyboard itself. And, like, one of the things, like, I used to be an Emacs user, and, like, you know, my.emacs file was full of all these, like, weird configurations and stuff. And I liked it because I could take it, you know, with me from computer to computer. operating system and text editor, like I noticed there wasn't really that for, you know, arbitrary operating system, arbitrary sort of key bindings and stuff. And so it was important to us to not build something where you have to install a driver just to use a keyboard, right? Don't

install special software just to, you know, use the keyboard to type. So any changes you make, you can put it onto the keyboard and flash it there and it will go with you from, you know, whatever laptop to desktop to work computer to home computer. Now, is there software that I have to have to flash the new config onto the keyboard or can I somehow plug it in and open a text file and edit it? Right now, I think you would need to install the Arduino IDE, which is free and open and easy. I think we're planning on changing that, but that's probably a question for Jesse in terms of the actual, he's the guy who's doing the software engineering for it. Anything else you want to add or if somebody's interested in purchasing one of these fine keyboards, where they would go to pick one up? Sure.

Go to keyboard.io, keyboardio, or keyboardio.com. We've got a ton of info on our website. If they have any questions, we're real people answering our email and are happy to talk with anyone about keyboards. It's fun to nerd out about it with people. So yeah. Well, thank you very much for taking the time to speak with us. We really appreciate it. Yeah. Thank you. My pleasure. So the cool thing now that they've done with the model that's gone up for pre-order now is they've kind of gone back a little bit. They're doing a QWERTY style to begin with. So it's not all that crazy,

but it still has the Arduino Linux computer in it that is fully programmable that you can reset up and replace the keycaps if you want. And you know what? I have been recently reconsidering my position on keyboards. And since this has been something we have covered across multiple shows, it's got Linux running this thing, I think I'm going to pre-order

this. I'm going to get it here in the studio. I'm going to have it as my main keyboard. And so I sent an email and said, hey guys, I want to buy one. And I think I'm going to do it with it. That is awesome. Yeah, I think it's nice. It's going to look way better than your current setup. Oh my gosh. Yeah, well this is just a free keyboard that came with the case. Oh yeah, really?

Yeah, so it's not that high-end. Yeah, it's just some... Hey, look at this. It's got... See how it says it's like bi-thermal take on there? Oh, you're right. I would never have thought because it has like the blue backlit keys. I know, I know.

It looks like a gamer keyboard. No, it's also got like this Borg insignia-style circuitry that's just lit up. Yeah, where this is like a... This is a proper set piece. Craftsman. When I do videos inside the studio, this is something that'll just lit up. Yeah, where this is like a, this is a proper set piece. This is, when I do videos inside the studio, this is something that'll catch the eye. But on top of that, this is a project we've literally been following for years now. And I got, I don't know, I got

a soft spot for keyboards that run Linux made out of wood. Don't we all? I wonder how it compares to something like ErgoD yeah or or dos keyboard or logitech g19 or yeah yeah i'm a little nervous i am a little nervous well actually i mean ergo docs is a more uh similar comparison right because ergo docs is a DIY split keyboard that's designed to be more ergonomic.

Most designs have the shifted columns, just like the Keyboardio. And they're both designed... The way I see it is Keyboardio is maybe a keyboard designed by people who are passionate about keyboards for people who just like good keyboards, whereas Aerodox is a keyboard that's designed by people who are passionate about keyboards for other people who are also really passionate about keyboards. I think Keyboardio is... Not until recently, I think. At $329 before you pay for fancy shipping, Keyboardio is for people who want something special and unique.

It fits that category. Like I'm looking at their keyboard layout, and it's a little different. The LED light controls where I would expect my Tilda to be, and there's things like that that are going to be a little bit of an adjustment. So we'll see how it goes. But it feels like it's – if you were a collector of fine art, this would be sort of equivalent to that, and it's much cheaper than that. And I like that it's open design, it's open hardware, and it's easily reprogrammable. So there could be a potential. This is my thought.

There could be a potential for this to be a very nice control surface for OBS. The OBS upstream project... You've got your crazy hotkeys keyed in there. Oh, man. OBS upstream is working on some pretty sophisticated hotkey support. And if I end up deciding that this is no good, this is bunk as a keyboard, and the fact that we can replace the keycaps and the underlying Arduino is reprogrammable, this, I feel like, will be a pretty good OBS controller too.

So I kind of feel like I'm kind of in a nice position to test it out and see how it goes. And I'll report back. The keyboard thing has been fun. Mosin-Rath, you found a G19 in Canada? Where? Actually, I can't find it again. I actually found it on Kijiji. I'm currently looking as we speak. It was

in Port Hope, Ontario. I'm completely serious. I found it last week and we were sitting there talking about it because my friends say with some truth that I have a hoarding problem, which has translated to a purchasing problem of specifically Logitech keyboards.

And I have a G15, a G11, a G105. I gave another friend of mine a G105, a G110. There might be another one around here somewhere. I don't know. Love it. But I actually did find one. I don't know. Love it.

But I actually did find one. And if I find it again before the end of the show or even shortly after the show, I will let you know. I'm really in the search mode for it right now. Nice. Stay tuned and find out. You know, I think that's sort of an interesting note to end on right there. And, Mosin-Rath, don't beat up on yourself because it's not that you are a hoarder. It's that you are documenting a precious time of development in our society.

In 20 years, you'll have a museum on your hands. Yeah, you are at the forefront of witnessing some of the most rapid progress in humanity. And you can't feel too bad about that. At least you can tell yourself that and try to self-rationalize with that line. You're welcome. Hey, if you'd like to join us live next week, go over to jblive.tv on a Tuesday. There is some calendar changes coming up in the future because I'll be traveling, so check out jubilabroadcasting.com slash calendar

where you can find the live times and get it converted in your local time. I'm at chrislas. LinuxActionShow.reddit.com. All of that shenanigans and a post show right here next week. Thank you. All right, let's go pick a title, jbtitles.com. Can I just say I'm loving this keyboard thing we're doing in this last episode? I know, it's funny.

It's really fun. It's been totally organic, and it is kind of fun to talk about this because it's something we all use. And I can't believe I found another keyboard that runs Linux. I got to get it. If I find it, it's been years since I found another keyboard that runs Linux. You have to have it. I like it. Cow tipping's not a bad.

Although we've already had cow in Linux Action Show. But jbtitles.com, making IT steampunk, making it snappy. Snap, snap, the snappy snaps. I don't know. Are we feeling that? Are we feeling that? That's a lot of snap. Mosen, did you find it before we get out of here?

Did you find it? Oh! Get it out of here. That'd be a negative. Yeah. We have a couple of different Telegram threads that we're using for conversations that are related to what might work on user error or something like that. And so sometimes things come up and we're like, man, this is a show topic.

But they come up a week before we record user error. And by the time we sit down to record user error, even with a Telegram thread there, we can't find it. We can't remember it. Same thing with Slack and similar things, right? Yeah, yeah. So in Telegram, there, we can't find it. We can't remember it. Same thing with Slack and similar things, right? Yeah, yeah. So in Telegram, there is a to-do bot. Oh. So why not use this bot to say to-do?

Use a to-do list manager as topic show manager, basically, because that pretty much translates. And so I added a bot, and I start working with it, and then my first time I work with it, I get a response. And then after that, I never could get the bot to do anything, and I have no idea if the server the bot runs is offline, if there's some sort of account issue with me. Other bots seem to be working okay. I have no way to troubleshoot. There's no error message returned to me.

Not a phone number to call. And it's like computers work with ones and zeros. So send me back a one or a zero. Like, don't just say nothing. Yeah, right. I don't understand it. You've got to give him credit. He made the jump right as the new MacBooks are about to be announced.

I know. I was going to say. He's like, screw that. I'm moving forward with this. Homey don't play anymore. Is there a big TLDR for why he wanted to switch platforms? The future of platforms for independent software development is probably in the hands of Linux, and many of those servers on the cloud run Ubuntu.

Yeah, there you go. I think he's placing a big bet on AI and bots, AI slash bots, to sell bots as business contracts to develop software. Also, I think he also figures, I mean, you'd have to listen to yesterday's Coder Radio to get the full context, but I think he also figures that the app ecosystem is pretty commoditized now.

Sure. And most good apps that are worth their salt anyways have a huge backend infrastructure. Yeah, absolutely. So why not focus a primary on the backend, on Linux, running on Linux, and then offer iOS and Android apps as part of the entire package? Because so far his business has primarily been around developing apps for iOS and Android and some Mac desktop apps. With backends when needed.

Yeah. And so now he's going to pivot and do backends as the primary focus. when needed. Yeah. And so now he's going to pivot and do backends as the primary focus and, hey, as part of the entire package, we'll also write the web frontend, the iOS frontend, the Android frontend.

And that would seem to be like the way a lot of things are going with progressive web apps or React Native and those kinds of things. And so in his mind, he's like, well, why do I need to spend $2,800 on a MacBook Pro when I could spend $800 on a Lemur and develop on the same OS

that I'm going to be deploying on. And I think the fact that Docker is just right there is a useful thing for him because Docker allows him to experiment and sort of rapidly try things. Yeah. So the cost of changing things up is like nothing. And, you know, I've been saying this for a while. I really think there's going to be a lot of server-side developers and web developers

that look at the Mac platform and say, okay, well, this new MacBook, yeah, it's got a fingerprint sensor. It's got this OLED bar on here that changes for the few people who wrote software that support it. But macOS itself is not going in a direction I want. It's not getting to be stable. It's not taking advantage of the Unix

underpinnings the way I wish it would. Yeah, it hasn't become a better Unix. No, it's not at all. It's getting to be worse. And so I think when you're looking, well, geez become a better unit. No, it's not at all. It's gotten to be worse. And so I think when you're looking, well, jeez, I'm just going to be creating apps for the cloud. Cloud. Anyways, why not get hardware that gets iterated faster, there's more

choice, and the desktop OS is agnostic as to which vendor I go with. Yeah. Because even in that scenario, you could eventually choose to go back to the MacBook hardware if you wanted to. Yeah, sure. Definitely.

So I don't, I'm not, I think this is going to be a huge win for Linux, sort of like a slow-boil scale. Mm-hmm. Slow-boil scale. I like that. We will see.

The Linux slow-boil.