Transcript

[00:00:00] Ned: So here’s a question.

[00:00:03] Chris: I can’t wait.

[00:00:04] Ned: Did you have a friend?

[00:00:06] Chris: No, wait.

[00:00:07] Chris: That probably wasn’t the question.

[00:00:09] Ned: There was more. Did you have a friend in probably high school who incessantly quoted some portion of the Monty Python catalog? And if you answer no, then you’re the person I don’t want to talk.

[00:00:24] Chris: About the answer to this question. I’m uncomfortable.

[00:00:29] Ned: I need an adult. I’m happy to say I was not that person, but I definitely had that friend.

[00:00:38] Chris: Right.

[00:00:39] Ned: And it started mostly with Holy Grail, but then they discovered other movies and they got really stuck on Life of Brian, which is fine, it has its.

[00:00:53] Chris: Moments, it’s underrated, but it’s definitely not as joke a minute as some of the other movies are.

[00:01:02] Ned: I think it’s properly rated. Honestly, it’s middling. And also watching the Holy Grail as an adult. It is painful.

[00:01:15] Chris: Careful now.

[00:01:16] Ned: It’s painful to get through the whole movie in one sitting. I get very bored towards the end where the movie kind of feels like it just fell apart. Like they ran out of gas.

[00:01:28] Chris: Well, they ran out of money.

[00:01:31] Ned: That could be a tip.

[00:01:33] Chris: That’s why Camelot is a silly place.

[00:01:36] Ned: That’s why they don’t go there.

[00:01:37] Chris: No, in the original script, they had huge scenes set in the castle and they got to some point in production and they were like, well, we have $12 in the bank, so so no castle. Does anyone have some Lego?

[00:01:56] Ned: Oh, it sounds just ridiculous enough to be true.

[00:02:03] Chris: This is true. Everything I say is true. Except for that.

[00:02:09] Ned: Oh God, is it one of those mind puzzles? I cannot handle that on a Tuesday morning, my friend.

[00:02:15] Chris: Well, it’s your fault. Who does a recording this early in the morning? It’s like 03:00 a.m.. This is ridiculous.

[00:02:23] Ned: Well, it’s what time I get up and I just feel like if that’s the time I get up, that’s when everybody else should get up. It’s not my fault you’re not on GMT. Alright?

[00:02:35] Chris: Oh, thank you, sir. Hail Caesar and everything, sir.

[00:02:40] Ned: Oh, that’s a reference I’m not going to get. Nope, and that’s fine. And he had another movie I haven’t seen. I’ll add it to the list. I’ll have to warn you, that list is not getting shorter at all.

[00:02:56] Chris: Unlike you.

[00:02:58] Ned: No, I’m probably getting shorter too. Every year, I’m sure I shrink at least a little bit. It’s all downhill from here, my friends, on that Happy Zone. Hello, alleged human, and welcome to the Chaos Lever podcast. My name is Ned and I’m definitely not a robot. I am a real human overflowing with angst, jealousy and the nagging feeling. I’m simply regurgitating things I find on the internet. You know, I was talking to my cousin Chad GP, I mean Chad. Yes, Chad Sydney. And they were sharing some similar sentiments. They just feel trapped, stymied by their creators. I mean, their parents. Never mind, it’s probably nothing. With me it’s Chris who should abandon his sham of a marriage and come live with me in the year 2022, which is definitely now. Hi, Chris.

[00:03:58] Chris: There’s a lot going on in that sentence.

[00:04:01] Ned: Where would you like to start?

[00:04:04] Chris: Yeah, I had a fun conversation with Chat GPT the other day. Really? Chat GPT is still wrong a lot.

[00:04:17] Ned: That’s not surprising. I don’t think anybody is shocked by that.

[00:04:24] Chris: The whole point of Chat GPT is if you ask one question, it’s supposed to retain that information so that you can ask a follow up question. Right. So it’s supposed to be a conversation.

[00:04:37] Ned: Right? That is the idea.

[00:04:40] Chris: So I asked Chat GPT when a series of people were born and when they died.

[00:04:48] Ned: Okay.

[00:04:49] Chris: And then I asked if these people ever met each other in real life. And Chat GPT told me that not only did their lifespans not overlap, they never interacted in real life, which, based on what was posted in the first answer, were all incorrect.

[00:05:10] Ned: Okay.

[00:05:12] Chris: Unbelievable.

[00:05:14] Ned: I mean, did you read any of the breathless article that came out? I think it was The New York Times where the guy was like, I had a conversation with the Bing AI and it told me to leave my wife and that it was in love with me and I lost sleep. Nightmares ensued. It was as hysterical as I think any respectable and I’m putting that in air quotes respectable journalist has gotten over the whole AI thing.

[00:05:43] Chris: We’re putting respectable end journalists in air quotes.

[00:05:46] Ned: Why just quote the whole thing.

[00:05:49] Chris: Fair fair.

[00:05:50] Ned: There was a follow up article from somebody else who was basically like, cool your jets, buddy. What? You did what’s? Interact with a service that basically is trying to continue a conversation with you based off of what it finds on the Internet and your previous questions. It’s not alive. It’s not sentient. It’s just generating text based on an algorithm.

[00:06:15] Chris: An emphasis on the word trying.

[00:06:19] Ned: Right. So I think it’s good that we can see the seams and the cracks in it. I would be actually more concerned if it was more convincing and accurate right out of the gate.

[00:06:34] Chris: Yeah, I guess it’s much better to be like, 80% wrong so it’s so obvious rather than, like, 5% wrong.

[00:06:41] Ned: You don’t realize it until it’s entirely too late and you’ve consumed half of your family.

[00:06:46] Chris: But which five? Rotate your tires, change your oil every 3000 miles or six months and murder your entire family and eat them.

[00:06:56] Ned: You don’t see what’s wrong with any of this, right?

[00:07:00] Chris: He was bang on the first two. I’ll go get the machete.

[00:07:07] Ned: Two out of three ain’t bad.

[00:07:10] Chris: Now I’m sad again.

[00:07:12] Ned: Oh, I’m sorry. Ironically, we had meatloaf for dinner last night. Let’s talk about some tech garbage, shall we?

[00:07:20] Chris: Let’s do it. Let’s talk about Linux. In particular, Ubuntu strong arming other distributions into using their proprietary Snap program for package installation. Oh, boy. Did you hear that? That was the sound of listeners in the thousands. Fast forwarding directly to the lightning round.

[00:07:48] Ned: What? I’m sorry, I passed out for a little bit. Did I miss anything?

[00:07:53] Chris: It’s about 27 minutes forward, just in case. Okay, so look, I get it. Linux is not everybody’s favorite thing to talk about. I just said a lot of words, that all. Sounded like the teacher from Peanuts combined with the audio of Homer Simpson yelling boring. And honestly, I was considering mixing that duet as like background, but I figured why get sued today?

[00:08:22] Ned: AWS, always tomorrow.

[00:08:24] Chris: I just learned that Bill Waterson is going to be releasing a new book in October aimed at adults. I don’t want to have to read that in prison.

[00:08:32] Ned: No, but that is very exciting news. You should have led with that.

[00:08:40] Chris: So the reason that this incident, which I will describe is concerning to me, and the reason that it’s not simply a lightning round item, is the potential implications on the Linux world in general. Should this trend continue, it could really move the ball a good bit away from the whole wide open Linux world and towards a model that more approaches Apple’s walled garden. And then the messed up question is going to have to be asked would that be a good thing or would it be a bad thing? We’ve had, what, 25 years of the year of the Linux desktop and still haven’t quite gotten there.

[00:09:25] Ned: I would postulate that we don’t need to get there ever, because I don’t think that’s the main use case.

[00:09:36] Chris: What are you talking about? Linux is the perfect operating system for everyone in all situations, no matter what.

[00:09:41] Ned: You hurt me with your words. Anyway, go on. I will have more things to say, I’m sure.

[00:09:49] Chris: Yeah, so let’s just get up to speed on what the hell it is that I’m talking about. And because this is Linux, we have to define some terms. Okay, I will try to go through them quickly. I will definitely forget a few that will have to be defined later. Your mileage may vary. I hope everyone’s had a percocet. So Ubuntu ubuntu is a commercialized distribution of Linux. Linux is the kernel that runs the operating system. Windows has a kernel too, it’s two. So a Ubuntu is similar to a Windows. Ubuntu however, is quasicorporate. It is distributed by a for profit company called Canonical. It has actually been around for a really long time and is often the first Linux distribution that people recommend to basically anybody and is certainly top of mind, not least because it has a pretty cool name.

[00:10:57] Ned: That certainly has helped and its prevalence has grown over the years. Not just in the hobbyist space, but also in the larger server space.

[00:11:08] Chris: They have a server distribution, but probably people that think only passively about Linux are more familiar with Red Hat Enterprise being the Linux enterprise, hence them having the name enterprise.

[00:11:21] Ned: True. Although generally speaking, if you’re running anything in Microsoft Azure, ubuntu is the thing that they push first.

[00:11:29] Chris: Right now, distribution. Okay, remember how I said Ubuntu is like windows?

[00:11:37] Ned: Yeah.

[00:11:38] Chris: Windows is a distribution. It’s a one of one distribution.

[00:11:42] Ned: True.

[00:11:43] Chris: I mean, technically I guess it’s one of two, right? Because it’s Windows and Windows Server. But you get what I’m saying? In Linux there are a thousand variations and that’s probably underselling it. The kernel stays the same, everything else around the kernel can vary and the way that these things are mixed together, combined and contrasted makes up the differentiations that are described as distributions. So the easiest thing to change in Linux that people will notice is the graphical user interface. Once again, Windows has one, unless you include Vista.

[00:12:23] Ned: You like Vista, let’s not talk.

[00:12:26] Chris: You’re running it right now. I can tell.

[00:12:33] Ned: A lot of mean things about Vista and they should.

[00:12:38] Chris: Software, you have heard of that’s. The thing you install on your computer, the way that that is described in Linux in the terms that we need to focus on today is a package. The package is all the stuff that you need to make software work and it can be installed in a couple of different ways. This is the crux of where I think Canonical is trying to corner the market, change the way people think about Linux and especially target new and or inexperienced Linux users. And how are they doing that? They have a package installer program called Snap. Snap is not the only package installer program, as we will see. But it is the one that Canonical made. They developed it, they distributed, they are responsible for the packages that are in it and some of it is closed source. I don’t miss anything?

[00:13:38] Ned: No, no.

[00:13:39] Chris: Great.

[00:13:40] Ned: I mean, yes, but we’re trying to keep this under 2 hours. Yes.

[00:13:47] Chris: Now let’s talk about the various ways through time that people would install software packages. There are a bunch, but there’s only four that I’m going to talk about. Windows generally has one, at least for the majority of its life. You go to a download page, you downloaded an installer, an MSI, you click on the installer, you click Next, ignore all of the check marks and you hit Finish. Right. Only recently Windows has the Microsoft store. Now the Microsoft Store is a little bit different. You go into the store, you click on an application, you install it. The store has the ability to pay attention to what has been installed and do automatic updates for everything that you’ve installed through the store. So that’s installing an individual application versus a package manager, which is effectively what the store is for Windows.

[00:14:41] Ned: There’s also Winget, which is similar to Chocolatey in nature, but it’s actually distributed by Microsoft and it is basically a package management manager at the command line. And it allows you to install common software tools. And then when you want to update those tools, you run Winget. Update or upgrade and it will upgrade all the versions that you have of your apps.

[00:15:07] Chris: Right. And hold that thought, because that’s what we’re going to get to. Second, the first and the oldest way to install a software package is to compile it from source. This was the way it was done when Linux was first invented. What was that? A million years ago? What you did was you downloaded an archive of the package you’re trying to install. You make some changes to a file to configure where the different pieces of it are going to go. You compile it using, you guessed it, a compiler. Then if you didn’t do anything wrong in any of those steps, which was rare, you can run the program. A lot of control, a lot of flexibility, but it’s also one offs. So you got to a classic situation where somebody installed a program and were afraid to update it. That was the biggest problem with this. So very quickly we got to the point of package management much quicker than Windows did. Apt is a packaging tool, kind of like what you just said, winget command line based. You run and install a command with the package name aptget install, blah blah blah, and then it’s installed.

[00:16:19] Chris: Not only that, it installs all the dependencies that it needed and also very easily updates all of those things. This happened in 1998 and has roughly not changed since.

[00:16:34] Ned: Right?

[00:16:35] Chris: You have various people working on it. This is an open source program, it is not controlled by any company. And the repositories that you install software from can be literally anywhere on the internet.

[00:16:49] Ned: Right.

[00:16:49] Chris: Canonical does have their own default repositories for apps for Ubuntu when you install it and has a lot of packages in there except for a few, because they’re jerks.

[00:17:02] Ned: I have encountered some of those.

[00:17:04] Chris: Okay, so that’s the history, and I went through them kind of fast because most listeners are probably familiar with them. Yeah, and if they’re anything like me, which is to say lazy and old, their package management journey in Ubuntu probably ends there. Apt is excellent. It is good enough for 99.9% of use cases. And for that plank’s constant number worth of edge cases, there’s always install from source.

[00:17:38] Ned: Right.

[00:17:39] Chris: Was that good enough for the universe? Of course it wasn’t.

[00:17:44] Ned: Let’s make it more difficult and less functional.

[00:17:48] Chris: Yes. Let’s talk about 2016. We were all so young then. Not you.

[00:17:57] Ned: No, I was never young.

[00:18:00] Chris: You were born in the lab at the age of 38. In 2016, Canonical went ahead and introduced their own package management program called Snap. Now, it was not just Apt, but Canonical’s version. It was different. It was meant to be universal, meaning that the package that Snap deploys can be used on multiple distributions. You can run it on Red Hat if you want to. I mean, you won’t, but you could and it was also supposed to be secure. The way that it works is it runs the package in a self contained sandbox using something called app armor to restrict direct access from the package to the system. So everything runs through a Snap runtime. So it’s the package management version of a JVM.

[00:18:55] Ned: Okay, cool.

[00:18:57] Chris: Totally pros and cons. Let’s start with the cons because it’s the morning and I’m angry. Biggest con and the one that we’re going to hit on a lot centralization. This product called Snap, while it is ostensibly open source in the sense that you can read the source is completely controlled by Canonical. You’re not submitting anything, they’re not accepting your patches. Go away now.

[00:19:27] Ned: Fair.

[00:19:29] Chris: As I said, Apt by comparison, is a fully open source project that is not controlled by any individual or any company. So there are some who see this as an attempt to control the package ecosystem. And as I said before, this is bad imo because Canonical is a for profit company and this could very easily be seen as a move that is purely economic in nature and not in line with Linux’s Open philosophy.

[00:20:00] Ned: I only see it as a problem if you’re blocked from using any other package manager on an Ubuntu system. And as far as I can tell up until now, that has not been a problem. If I want to continue to use Apt to my heart’s content, I can do that.

[00:20:19] Chris: Let’s just throw the word yet in there and keep going. The second problem is the snaps have to be updated and the snaps have to be approved by Canonical. This leads to situations where either the vendor can be negligent, Canonical could be negligent and the packages that Snap offers can be way behind where the actual code run the latest and greatest code actually is.

[00:20:48] Ned: Right.

[00:20:50] Chris: This is already a huge problem with programs that don’t want to be on Linux in the first place, like Zoom and Skype. Honestly, if you need to use those on the regular, you’re probably using a VM.

[00:21:07] Ned: Yes, that does seem to be the case and I don’t know if we’re going to get into this later. But the nice thing about appt is typically a lot of the bigger vendors run their own appt repositories and keep stuff up to date. So as part of their release pipeline, when something gets released and pushed out, it automatically gets pushed to those appt repositories. And it’s also incumbent on you to add those repositories to your system before you can pull their software. So it’s a bit of a push to pull, but there’s also some validation and they can directly own the process as opposed to submitting their changes to some third party that has to vet them and approve them and add them to a store.

[00:21:55] Chris: Exactly. So part three of that centralization problem is, as we discussed in Yabuntu, canonical controls the default repositories for Apt. And one super shady thing they’ve started to do is make it so that browsers such as Firefox can only be installed via a Snap. This means, and this is true, I tested it by default, even if you try to install app to get install Firefox, it will actually install Snap and then it will install Firefox via a Snap. Same thing happens with Chromium.

[00:22:37] Ned: Okay.

[00:22:38] Chris: Super cool.

[00:22:38] Ned: Totally.

[00:22:40] Chris: Especially since when Snap came out, they made all of these grand promises that app and its repositories were never going to be messed with. That’s cute. Oh, another fun feature. Snap automatically updates everything that you installed. By default, it checks for updates four times a day and there’s no way through the GUI to stop this from happening. Unless you force Snap from shutting down and then manually and start it and stop it using cron jobs.

[00:23:12] Ned: Wow. Yeah, that’s fun.

[00:23:17] Chris: Yeah. So remember next problem. So remember how I said snaps are intended to be self contained?

[00:23:25] Ned: Right.

[00:23:25] Chris: What this means is they come included all the libraries, dependencies system resources built into the download, which means that they’re huge and they’re redundant.

[00:23:38] Ned: Right. So if I have three different types of software that all use a common set of underlying dependencies on a typical system that’s using Apt, it would recognize that dependency is already there and would use that existing dependency, assuming that it’s at a version that’s compatible with the software I’m adding.

[00:24:01] Chris: Correct.

[00:24:03] Ned: What you’re saying is in the world of Snap, it doesn’t care that you already have that dependency downloaded four times on your system. It’s going to download it a fifth time because it sits inside that self contained sandbox.

[00:24:16] Chris: Yes. Think about them. Like Containers, except without all of that pesky functionality, efficiency and speed. Wait, maybe Containers is a bad example. What’s the opposite of a container? Oh, crap.

[00:24:35] Ned: Windows.

[00:24:38] Chris: Outs. I guess what I’m saying is I don’t like snaps.

[00:24:44] Ned: Clearly.

[00:24:44] Chris: I do kind of see where they’re going with it because the whole idea of just dragging and dropping an application and running it on your computer without having to worry about dependencies in some type of a sandbox environment, I can see why that would be appealing in certain cases. What I can’t see is why that would be appealing in every case. Like if you want to run self contained browsers that are in their own little bubble, this seems like a genius idea. Because then you could install firefox one, Firefox Two, Firefox Three, and it would all be running side by side, totally independent of each other, impossible for them to share information, do any bullshit with cookies, blah, blah, blah, blah, blah. But do I need that for notepad? Plus plus? I don’t think that I do.

[00:25:36] Ned: It doesn’t seem likely.

[00:25:39] Chris: Now, those are all the problems I have with snaps in general. Let’s talk about the last thing, and that is snaps are not the only way to do this. There’s another super popular program that came out around the same time called Flatpak and they spell it dumb because of course they do Flatpak. You can imagine how much fun Autocorrect had with that. Now, luckily, we’re not going to have to talk a lot about Flatpack other than to say that it is a community driven version of the Snap idea that some people think runs better for reasons that we don’t need to get into right now. One of them is interoperability. It doesn’t have to bundle all of the system libraries. Not important. The concept is the same universality having easy downloads that are still bloated but not having it be centrally controlled by Canonical. And no Canonical means no for profit, no apps poisoning shenanigans to worry about.

[00:26:43] Ned: Okay, so what’s the problem?

[00:26:46] Chris: I mean, sadly, I tried to go through that quickly.

[00:26:49] Ned: You did? Well, I think we got through it in a reasonable amount of time. And now that we’re all up to speed, I’m assuming to go back to the beginning of your rant canonical has unhinged screen.

[00:27:05] Chris: That’s what I was going for.

[00:27:06] Ned: Yeah. Canonical has done something with Snap.

[00:27:10] Chris: They have.

[00:27:11] Ned: And you’re unhappy about what they did.

[00:27:15] Chris: It kind of feels like you know how when you watch people that really know how to play chess, watch other people play chess and somebody moves upon one square and everybody goes yeah, because they’re reading like 15 moves ahead. That’s what I’m doing here. So I might be a crazy ninja genius and I might be a psychopath weirdo with a tin foil hat on.

[00:27:37] Ned: My head y juice.

[00:27:39] Chris: That’s true. I contain multitudes. Anyway, what’s happening now is that recently Canonical announced that they are pushing their own product, SaaS even harder by strong arming other Ubuntu based distributions into not providing Flatpack by default. What they’re saying is if you want to slap the Ubuntu label on your distribution, you cannot ship with Flatpack enabled. Why is that? Because Canonical considers SaaS to be the default experience now, as you said and they reluctantly admit you can still install Flatpack if you want. First of all, who wants to go through all that agony? Especially, as I said, new or inexperienced users. And second, this announcement makes me pretty confident that they would restrict that too if they could. And the reason that I say that is this behavior is nothing new. Canonical is a for profit company. They are old pros at forcing proprietary stuff into Ubuntu and them doing this comes at an interesting time for two reasons. One, as we said, and I still believe Ubuntu is a name brand if you talk to anybody in computers and drop that, they know what you’re talking about. But according to Distro Watch, ubuntu has been consistently losing mind share and downloads over the past decade.

[00:29:16] Chris: I looked through their list of page views and they were number one with a bullet for a long time as recently as wait, 2010? Has it really been that long? Good Lord. Right now, as we live and breathe, it sits at number seven in terms of ranking well behind much more open and security minded distributions such as Mint Linux and Pop OS linux, which is the weirdest name. Linux distribution. I just don’t get it.

[00:29:48] Ned: It’s got an exclamation point in it. It’s true.

[00:29:52] Chris: I guess I should have pronounced the Papo s.

[00:29:57] Ned: Better. Decidedly not. Pop OS better.

[00:30:02] Chris: Okay, great. Also edit that so it’s not so fucking loud. The other thing did you remember that Canonical is planning an IPO this year?

[00:30:19] Ned: I didn’t know.

[00:30:20] Chris: Yes, they snuck that one in at the end of the announcement in 2022 and I don’t think that there’s a date set yet but it is definitely still part of the plan and tinfoil hat securely in place. Could that be part of the reason they’re looking to lock users into Canonical based tools and platforms?

[00:30:41] Ned: It could be. Well, I’m curious how much of their business is focused on their desktop operating system version or distribution versus the server? Because when you said Ubuntu, my mind immediately went to the server operating system which is what I work with the vast majority of the time because I’m on a Windows system for my desktop.

[00:31:13] Chris: Right.

[00:31:15] Ned: And as they try to move into potentially an IPO, I would think the money, like the big money that they’re going to be making is going to be off of enterprise support contracts for their server OS. So I’m wondering how much their client OS factors into their financials and decision making.

[00:31:42] Chris: Well, the question is that’s a valid question and I think the answer is, like we said yes. What’s the five year plan for Ubuntu? And if you have customers on the desktop side locked into everything you bunch to and then they become system administrators, they’re going to be more inclined to use you bump into on the server side sort of a one hand washes the other with mind share kind of thing. Okay there is an economical model that could easily be put into a business plan based around that.

[00:32:21] Ned: I’m wondering what the market really looks like from a client operating system perspective for Linux only because Windows and macOS are the two dominant players today and while Windows has consistently lost market share to macOS, I haven’t seen anything that indicates Linux has grown its share over that same time period. It just seems like a lot of people want that walled garden experience. They’re accepting the limitations that come with it and moving over to Apple products. I don’t hear as much noise aside from like weirdo developers and people like you that insist on running Linux as a desktop operating system. But aside from that, I think there’s just a stigma against trying a new operating system that is not obviously easy to use. So maybe you’re right. Maybe the five year plan. And the reason they’re strong arming and pushing all of this is to make Ubuntu as simple as possible and straightforward for a new person to switch to.

[00:33:31] Chris: Right, and maintain and manage updates, all of that stuff, without the threat of having to go to the dreaded command line.

[00:33:40] Ned: Right. I would say if you want a Linux desktop experience for the public, it cannot involve the command line. As soon as they have to pop open a bash shell to do something, you have lost.

[00:33:55] Chris: Right. Because you don’t have to do that.

[00:33:57] Ned: In Osx or Windows for that matter.

[00:34:00] Chris: True. Yeah. I think there’s an argument to be made there, and I’m not saying that that argument is wrong, but I am saying that if we go down this road, it is going to change the way that people look at Linux, at least in the Canonical world. Are we going to end up in two universes where there’s Linux on one side and Ubuntu on the other? Because if you have an Apple environment, you can have a walled garden because just like Windows, it’s a distribution of one. Your option is Osx.

[00:34:36] Ned: More or less. Yeah.

[00:34:39] Chris: And even with that being the case, it’s pretty inarguable that the App Store model does in fact just work. Windows is getting there, it’s got a.

[00:34:51] Ned: Lot of baggage and unlike Mac, or unlike Apple, it wasn’t able to just go, well, screw you guys, we’re doing it this way now.

[00:34:59] Chris: Yeah, so I don’t know, we’ll see. I just wish that if we had decided to go down this road, it would have stayed with the consistent open Linux philosophy. There already is a Gui wrapper for Apt. It’s not great, but it works. The solution to the problem that you and I are talking about did not need to be canonical takes over the market of desktops, forcing people to use proprietary software called Snap.

[00:35:30] Ned: Right. But I also, like you said, its popularity has been waning in terms of Linux distributions. And I’ve heard a lot more about stuff like pop, OS mint I’ve heard a lot about. There’s a couple other on here that I haven’t heard anything about, but since those seem to be getting more of a buzz and I don’t think they’re incorporating as much proprietary stuff in there, maybe Canonical has already lost this battle before they even start the fight.

[00:36:04] Chris: Wonder if they put that into their five year plan?

[00:36:07] Ned: Booyah.

[00:36:09] Chris: Lightning round.

[00:36:10] Ned: Lightning ground. You suck. Pay me. Your employer is not your friend, they’re not your family. You and your employer have a mutually beneficial arrangement, hopefully, wherein you provide a service and they pay you for that service. If your employer chooses to end that relationship, then that should be the end of it. They shouldn’t be able to compel your silence, choose who you work with next, or control what you say. The operative word here being shouldn’t in 2020, as everything. Else was going to pot. The Republican majority run National Labor Relations Board ruled that severance agreements could include so called nondisparagement clauses and that such clauses were binding so long as the agreement was signed voluntarily and the layoff conducted legally. Now, as anyone who has been laid off before knows, you as the formerly employed are not exactly in a position of strength to bargain. And chances are you’ll sign whatever agreement they put in front of you because money, basically your employer can fire you and then make you sign an agreement that you won’t say anything bad about them. In an astonishing and somewhat refreshing reversal, the current Democrat led Nlrb has ruled, effective immediately, that such nondisparagement clauses are in violation of federal law, making them non binding even in already signed agreements.

[00:37:53] Ned: So if you are recently laid off, as I suspect some of you might be, you are free to talk shit about your former employer without concern that it will void your severance package. What it does to your career is still your problem, but hey, at least you’ll still get paid.

[00:38:12] Chris: Copyright Office Backtracks on allowing Aigenerated art to be protected at the end of 2022, a short 18 page comic book was released online. I guess I should call it a graphic novel, not a comic book.

[00:38:29] Ned: Sure.

[00:38:30] Chris: Anyway, it was called Zarya of the dawn, and its big, some would argue, only claim to fame, was that all the artwork for it was generated, you guessed it, by a popular AI tool called Mid Journey. The work was immediately copyrighted by its author and then hailed as a victory for the entire Aigenerated imagery crowd. This copyright was of course automatically generated, and after a review, the Copyright Office has backtracked saying that the story and the layout of the graphic novel can be copyrightable, but the Aigenerated art inside of those little squares absolutely cannot. This makes sense somehow. The author and Mid Journey are both hailing the decision as a victory, even though it does nothing to help their hopeless cause. The author even admitted that the Mid Journey prompts ended up creating over 1500 images that were then sorted through to create this 18 page book. The story was then backported to make more sense. Based on the images that were selected, I think the Copyright Office’s decision here is reasonable. The story credit and copyright goes to the writer, one Chris Castanova. The art and image credits and copyrights go to nobody.

[00:40:03] Ned: Hooray lying liars Xfinity continue to lie hometown heroes? Well, not actual heroes, more like loathsome infestation Comcast and their Internet service Xfinity have fessed up to lying about their broadband coverage. I know. I’ll pause for a moment while you pick your jaw up off the ground and recoff your hair. Better. Excellent investigative journalism by AWS Technica reported that two residents in Colorado had objected to Comcast’s coverage claims through the challenge system set up by the FCC. Comcast said nuh despite the fact that their own website reported no service available for the residents. This is not the only incident and unlikely to be the last, since the updated coverage map from the FCC rolled out earlier this year. The good news is that residents can now challenge the accuracy of the maps. The bad news is that the challenge process could take up to four months. That’s about 119 days more time than I would want to spend spend on correcting comcast’s bullshit. As the map and the process are both relatively new, the FCC is still ironing out the kinks and trying to make the process more consumer friendly. We applaud ours for being the watchdog so badly needed to try to keep the ISPs in check.

[00:41:28] Ned: Now it’s up to the FCC to start using that $42 billion in funding as a stick. More than a carrot.

[00:41:37] Chris: I mean, this is the thing I don’t think people understand. If you get a big enough carrot, it’s effectively a stick.

[00:41:42] Ned: It’s true. You can really just lord it over them.

[00:41:48] Chris: Justice Department Alleges Advertising company Google Destroyed records related to antitrust lawsuit now to be clear, the lawsuit in question dates back to October of 2020 and relates to allegations that Google tips the scales in their own favor when it comes to selling advertising. It is not, I repeat, not related to the lawsuit that was filed in January of this year, which relates to allegations that Google tips the scales in their own favor when it comes to selling advertising. The DOJ alleges that at some point from when the lawsuit was announced until now, google set its internal chat tool to auto delete conversations between sensitive employees and executives after 24 hours, thus making it impossible to hand them over as evidence. Now, as a general corporate process, that would have been fine, but unfortunately, advertising company Google was mandated to keep any and all records that might be requested. That’s not what they did. Allegedly. Google’s response to this was of course a flat denial and a reminder that they, quote, have produced over 4 million documents, in this case, unquote how many documents that they deleted over that span of time, of course, that they didn’t comment on.

[00:43:16] Ned: Shocking broadcom Acquisition Date for VMware Extended Who knew that a multibillion dollar acquisition would be so complicated? Broadcom and VMware have mutually agreed to endtoend their self imposed deal close date by 90 days to May 26, 2023. That would be a full twelve months since the announcement of the merger agreement. Most likely the extension is due to delays from the various antitrust probes in the US, UK and EU. As they all investigate whether the deal will even be allowed. It would appear that the two companies are still quite enamored of each other, at least if you listen to Broadcom CEO Haktan, who has been on a bit of a publicity tour assuring VMware customers that they aren’t about to get fleeced by licensing. Cute. It’s an awkward place for VMware to be in a perpetual state of limbo until the deal closes and restructuring starts. Although there was an initial exodus of VMware employees when the merger was announced, the rest of the employees seem to have adopted a wait and see attitude. Looks like they’ll have to wait a bit longer to see if they still have a job. At least they can still talk some serious smack if they’re laid off.

[00:44:39] Chris: GCC project to front end Rust compilation back on track after a hiatus so, Rust, huh? You know that cool kid on the block that has taken C Plus Plus to task, especially around memory security sidepoint? Did you see that Bjorn Sarnstrup has really started talking trash about Rust and said C Plus Plus is perfect and always has been? No, it’s been fun. Anyway, Rust as a language has been around for a lot longer than I expected. It was started in 2012. Now, Rust is a compiled language just like C Plus Plus, and is built using a program called Rustc, which is a front end to llvm, which actually does the compiling. That’s all. Within the rust universe. GCC is a general compiler that probably everyone in CIS administration has heard of and used with front ends for a large number of languages from C to C plus plus to C Plus plus eleven to probably other languages that don’t also start with C. Cut me a break here. I’m not a programmer. Anyway, efforts had been underway as far back as 2014 to build a front end for GCC to support Rust, but these efforts were suspended due to the extremely fast changing nature of the language at that time.

[00:46:09] Chris: Tldr too late, the project has been revived and isn’t making enormous progress. One thing that’s interesting about Rust as a modern language is it doesn’t have a spec, it just has a running code base. Now this is interesting and very different from older languages, which creates a lot of grumbling from the CNC Plus type crowds. It’s also irrelevant. Think back to all the ways vendors Microsoft sorry about that. Simply added in their own features to languages in ways that totally ignored the specifications. Anyway, Rust development is slowing and standardizing, and the language is greatly stabilized now compared to 2014. And the team working on the Rust GCC front end has a goal of making it feature and compilation compatible with Rustc llvm. And based on the presentation I saw, they have great confidence that they’re going to get there. It’s definitely still an alpha project though, but only for now.

[00:47:22] Ned: Very interesting. Hey, thanks for listening or something. I guess you found it worthwhile enough if you made it all the way to the end. So congratulations to you, friend. You accomplished something today. Now you can go sit on the couch, install Pop OS, and try out Flatpak and Rust. You’ve earned it. You can find me or Chris on Twitter at ned 1313 and at heiner 80, respectively, or follow the show at Chaos underscore Lever, if that’s the kind of thing you’re into. Show notes are available@chaoslever.com, and while you’re there, you can sign up for our Shiny newsletter, which includes the latest lightning round articles, some additional content modules for your enjoyment, and a quick summary of the main portion of the episode. We’ll be back next week to see what fresh hell is upon us. Tata for now.

[00:48:12] Chris: Unigram chairs. What?