 Welcome to another edition of Unfettered Freedom, your bi-weekly GNU slash Linux news video podcast. Packing so much freedom into each episode, it ought to be illegal. There's so much freedom packed into this podcast, 4 out of 5 neckbeards recommend this podcast versus the proprietary alternatives. On this episode of Unfettered Freedom, Linus Torvalds has some thoughts about the new M1 Max. He wants to see Linux running on those things. Also, what are the most popular snap packages in 2020? We're going to take a look at some of the analytics and see what people are actually downloading and installing as snap packs. We have two Linux distributions, Fun2 and Cebion, they have decided to join forces. Also, Fedora, they're thinking about switching away from pulse audio as their default audio server to pipe wire. And SystemD, it wants to change how you manage your home directory. All of this and more on this edition of Unfettered Freedom. I am your host, Derek Taylor, also known as DT or Distrotube over on YouTube and on Library. This podcast as well as all the video content on the Distrotube channel is community sponsored. Because of the community support that I received, there are no corporate sponsors or product shilling in these episodes. So if you'd like to support my work, I'd greatly appreciate it. Please consider subscribing to Distrotube over on Patreon. And our first story is the comments recently by Linus Torvalds on the new M1 powered Macs. So Linus Torvalds has always been a fan of Apple hardware. He has actually ran MacBooks before he had a MacBook Air, an 11 inch MacBook Air that he ran for many, many years. And he always spoke very highly of it. Of course, he runs a Linux as the operating system on those devices. He doesn't actually run Mac OS. But the problem with these new Mac lineups, these M1 powered MacBooks is that they are not going to be able to run Linux. And Linus Torvalds really doesn't like that. So recently over at the Real World Technologies Forum, Linus was asked about his thoughts on the M1 powered MacBooks that are coming out. And Linus said, quote, I'd absolutely love to have one if it just ran Linux. And that's the issue. So a few months back, Apple announced that they were going to move away from their x86 processors and move over to using ARM processors, which is gigantic news. And it's kind of interesting, you know, that's really going to push ARM forward. They started making their own silicon chips, these M1 chips. And you're going to say, well, Linux runs on ARM. So why wouldn't Linux run on these new MacBook devices? Well, the problem isn't necessarily their CPUs. The problem is going to be all the other stuff on that MacBook. It's because the CPU is not the entire computer, right? It takes a whole bunch of stuff to actually make a computer work. Linus says, quote, the main problem with the M1 for me is the GPU and other devices around it, because that's likely what would hold me off using it because it wouldn't have any Linux support unless Apple opens up. So this is closed hardware. They're never going to share with you the technical specs of this hardware that Apple's creating. So unless they open that up and share it with the community, no one can really make Linux work on these devices, right? Until you tell me what your GPU is doing until you share with me some of the technical details. I can't write a driver for it. So unless Apple is going to make Linux work on that device, no one can unless they open it up. Now, if they open it up to the community, share all the hardware specs and technical details that they need to share with the Linux team, for example, I promise you Linus Torvalds and his team will make Linux work on those devices. There's too many MacBooks out there and this M1 MacBook, you know, there's going to be millions and millions of these devices out there. They would make Linux work on those devices, but it's up to Apple to open that hardware up. Now, Apple has had a love-hate relationship with open source in the past. They've never been forthcoming with sharing these kind of hardware details in the past. And it's becoming increasingly this way with Apple. Matter of fact, this M1 hardware that they're putting out, the Google Chrome team has been running into what they call showstopping bugs with the Chrome web browser running on the M1 Macs. Linus was asked, will Apple ever open up their chipsets? And he said, quote, it seems unlikely, but hey, you can always hope. So there's his thoughts on that. I guess Apple has not been forthcoming with this stuff in the past, so you don't expect them to ever do this in the future. It's kind of like until these companies show you that they can change, you just assume they're always going to be like that, kind of like Microsoft. For decades, Microsoft was so anti-free software, anti-open source, anti-Linux. For a while, they really had to prove it to us that they were actually serious about working with the open source community and doing good stuff with open source software. And quite frankly, Microsoft is starting to change my mind a little bit with what they're doing. But Apple hasn't even gotten onto that road yet. Linus goes on to say that he's not sure he would buy one of these devices anyway because even if getting Linux to run on it wasn't an issue, he says, quote, my personal hope would be for more cores. Even in a laptop, I don't care about the 20-hour battery life. He says I'd rather plug it in a bit more often, but he wants eight big cores. And what he's talking about here is, of course, his particular workflow, the more cores, the more processes, the faster, probably, you know, compile times for the kernel or whatever it is he's doing on these devices. So for him, it's all about, hey, how many cores can you fit on that thing? I don't need 20-hour battery life. I know Linus in the past has also talked about he doesn't like loud laptops. He doesn't like spinning fans and things like that. He really wants a quiet machine. They asked him about the 16 gigs of RAM that are going to be on these current models. And I believe that is actually the limit. They're going to ship with 16 gigs. And I think that's all you're going to get. He says, quote, 16 gigs is actually OK because I don't tend to do things that require a lot of RAM. All I do is read email, do get, and kernel compiles. So he doesn't need, and he specifically says, I don't need 64 gigs of RAM on my laptop. He has 64 gigs of RAM on a desktop computer that he has. And he says, that's a specialty machine. He does some huge parallel builds on that machine. That's why that's a very beefy workstation. But on his laptop, he says 16 gigs of RAM, more than what he needs. Linus mentions that when it comes to RAM, typically the people that need a ton of RAM are the ones that are doing multiple VMs or working with huge raw file photography and video. And I would agree with that. That's why I have 64 gigs of RAM on my workstation as well. It's because I do a lot with VMs and of course I do a lot with video. So some interesting comments there from Linus Torvalds. Do I personally think Apple is going to respond in a positive way to these comments? I doubt it. Unfortunately, Apple is going in a direction right now where they're actually trying to make their devices more and more locked down. They're not trying to become more and more open. They want you to be completely locked in to their ecosystem, to their hardware. Right to repair is a big deal with Mac right now because you can't even get your MacBooks repaired unless you go to an official Apple repair shop. So I don't see them ever opening things up unfortunately. And our second story is what are the most popular snap packages? So last year in 2019, Canonical released a report stating what the most popular snap packages were and they broke it down by various distributions because of course snaps are just not in a boon-to thing. You can get snaps working on pretty much every GNU slash Linux distribution as long as it runs SystemD. So I thought it would be great to take a look at this year's report because they just released some data about the top five most popular snap packages in 2020 by distribution. And what's really interesting is the list is very different depending on what distribution you're running because some distributions of course are very geared toward just your standard consumer desktop person. Some of them are more geared toward development. Some of them are more geared toward web servers and things like that. So it's going to be a completely different top five depending on what distribution we take a look at. So if I take a look at this article here by Brian Fagioli over at Beta News he breaks down the report here nicely in some columns here. The top five snaps per Linux distro for 2020. And let's start with the far right column here. Those of you watching the video podcast because the far right column is a boon-to. And I think that's the one to take a look at first because of course snaps are in a boon-to thing. A boon-to is probably the most popular desktop Linux distribution. So you would expect because of the large number of users and the large number of users actually using snaps that that's a pretty good representation of how popular these particular programs are. And in a boon-to the most popular snap is VLC. Now that's interesting. I've actually installed VLC as a snap on my Arco Linux that I run on my main workstation and the snap pack for VLC it just works. Why did I install VLC as a snap? I can't even remember why I did it. Obviously VLC is in the standard arch repositories but for some reason I installed it as a snap and it works and it works just fine. Number two on the list for a boon-to code, VS Code. You want to install VS Code and of course it's packaged as a snap because the standard binary builds of VS Code are not free and open source software. They're licensed under a proprietary software. So that's probably why they package that as a snap and they don't have that in the standard of boon-to repositories. Number three is Spotify, again not free and open source software. It's proprietary. Number four is the canonical live patch. I don't know anything about canonical live patch. I think that has to do with updating your system as far as kernel updates and things like that. I haven't actually played with a boon-to lately so I'm not really sure what that's about. Number five and the top five for a boon-to is GIMP. I have installed GIMP as a snap pack and the snap for GIMP works just fine as well. Now there are five other distributions that we have the top five snaps for. We have Manjaro, Fedora, Debian, CentOS and Arch Linux. Now let's start with Manjaro because Manjaro is really for the desktop Linux user much like a boon-to so I would expect the list to be very similar. But there are some different things in here. Number one on Manjaro's list is Spotify. Number two is VS Code. Number three, Skype. Wow, I didn't know that many people were still interested in Skype. Of course, I guess because of the lockdown and the pandemic a lot of people are doing video conferencing and video chat. So for Manjaro, number three is Skype. Number four is Zoom. Make sense. And number five is Discord. Make sense. A lot of video chat stuff for the Manjaro list. Now where things get really interesting, once you get into some of the distributions that are more geared toward development or especially server. So Debian and CentOS are both here. And in both Debian and CentOS, the number one snap pack is Certbot. What is Certbot is this program put out by the Electronic Frontier Foundation that does HTTPS for your website. So Debian and CentOS are often used on web servers. Matter of fact, they're probably the two most popular distributions if you're going to do web servers. And that would make sense. Why Certbot is the number one snap on those particular distributions. MicroKates is number two on CentOS's list. And then we also have LXD, Helm and Debian's list. We have Postman. So these are all server related applications for the most part. Some Kubernetes stuff in here. The Arch Linux list is interesting because Arch is also mainly for the desktop user. Not too many people use Arch on servers. You can, but being rolling release typically is just for your standard desktop user. It's list Spotify, Skype, Discord, VS Code, and Anbox, which is, I believe, for Android development. So that's interesting that that made that list. Now one of the omissions in the list here, these popular snap packs, because so many people got mad when Chromium was only available as a snap on Ubuntu and then, of course, on Mint and then Mint decided we're just not going to do snaps because we don't want to force people to install Chromium as a snap. Chromium is not even in the top five most popular snaps on Ubuntu. It was in the top five in 2019, though. It was number four. Also, I know for a while, Ubuntu was installing a lot of the standard GNOME applications in its flagship edition, Ubuntu. Like the calculator and things like that were all a snap. So I don't see any of them in the list. Overall, my take on this list is that the lockdown, the pandemic and the lockdown has really affected the way people use their computer. I mean, just that Manjaro top five list, the fact that three of them are video conferencing software, Skype, Zoom, and Discord tells you a lot about how people are using their computers in 2020. Also, the fact that VS Code was spotted on several of these lists. A matter of fact, five of the six list VS Code was in the top five snap packs. So that's becoming very popular to use on Linux. I think some of that has to do with Microsoft being a little more friendly to Linux and to the open source community these days. And our third story is two Linux distributions you probably haven't heard of. They are kind of waning in popularity these days. As a matter of fact, a few weeks ago, I remember commenting on one of my videos that I love Sabion. Sabion is a Gen2 based Linux distribution that does some really nice stuff. And it just doesn't seem like I ever hear about it anymore. They haven't had a release in forever. When you go to their website, it doesn't look like there's much activity. Their forms don't see any activity. And Fun2 is kind of the same way. I can't remember the last time. I even heard about Fun2. Nobody mentions it. So both of these projects were kind of stagnating. And what they've decided to do is to join forces. So both on the Fun2 website and the Sabion website, we got this announcement that Fun2 and Sabion have decided to join forces. Now, that doesn't mean that they're going to completely merge into one project or anything like that. I think they're going to keep doing their own separate thing, but they are going to work on some collaboration for some things. For example, I know Fun2. You can use Eqo, which is a package manager, I believe Sabion created, because the standard package manager in Sabion, I think you can also use that in Fun2. I haven't actually looked at Fun2 in many, many, many years. I've never actually looked at it on the YouTube channel. But that's probably some areas where they can maybe make their lives a little easier, as far as doing a lot of duplication of work. Hey, you know what? Why don't you guys do this and we'll do that. And then we can share our work and we can still maintain our own separate distros, but at the same time, you know, not really wasting a whole lot of time trying to reinvent the wheel when somebody else has already solved this problem. I'm just going to briefly read a little bit of the announcement on Fun2's forum. They said, quote, Sabion is a great open source project and has an amazing reputation for delivering an easy-to-use Linux desktop that is built using Gen2. For quite a while, Fun2 has been supporting the Sabion effort because we believe in what they do. Over the past several months in chatting with the Sabion team, it became really clear that we shared a vision. We were also working on several next-generation technologies that were a good complement to one another. We decided it would be more powerful and fun to combine our efforts into one larger effort. So that is what we are doing. And just briefly, I'm going to go to the Sabion website and just very briefly read this line from their announcement, quote, Fun2 Linux is a community-developed Linux meta-distribution based on Gen2. Fun2 Linux is optimized for the best possible performance on the latest Intel and AMD hardware. Fun2 is led by Daniel Robbins, the creator of Gen2 Linux and actively developed by the Fun2 community. And that's one of Fun2's claim to fame is that the person that started Fun2 was the original creator of Gen2 Linux. So what are my thoughts on this? I think this is great news. I love when I hear open source projects actually combining forces, doing collaborative stuff instead of constantly trying to do their own thing and you get all this duplication of work and just wasted time. And I love the fact that Sabion and Fun2 are doing this and I think both of these projects need to do this because I was really worried, especially for Sabion because I really like that distribution, full disclosure, I've been a patron of Sabions before, trying to help them with their development and I really want that particular distribution to succeed because it is a Gen2-based distribution but they aim to be easy to install which we've got a few Gen2-based distributions that are easy to install, but they want to take it to the next step, a clean polished desktop that looks good, easy to use. They want to be what Ubuntu is to Debian. Sabion wants to be that to Gen2 and for a while they weren't succeeding at that. A few years ago, Sabion was pretty popular. You saw people running it, you saw screenshots of Sabion and everything but really in the last two or three years, its popularity has just completely went away. You just don't hear anything about it and I think a lot of that is just development right now is slow on it. There's no new releases when you go to download their ISOs, you get some pretty old ISOs. So I'm pretty excited what Sabion and what Fun2 do now that they're combining their efforts a little bit and I will definitely be taking a look at their new releases when they get around to releasing some new releases for probably take a look at both Sabion and Fun2. And the fourth story I want to talk about is a story I just came across earlier today and it kind of shocked me. It's not really a story but just somebody on a mailing list for Fedora, they are just putting it out there, hey, is it time? Can we go ahead and switch away from Pulse Audio and Jack and go ahead and try to make PipeWire the default audio server in Fedora. And when I read that, I thought, wow, is the guy that wrote this crazy? That's like a medium. But Fedora is like this. Fedora, they're not afraid to experiment sometimes with their releases because Fedora is kind of an experimental distro anyway. They want to just throw stuff out there, see what sticks. They want to try it out in Fedora before it rolls out into something like Red Hat oftentimes. So over on the mailing list, Ben Cotton wrote this summary here. The change proposal is to route all audio from Pulse Audio and Jack to PipeWire Audio Damon by default. And he goes on to say that all desktop audio on Fedora is handled by the Pulse Audio Damon. Applications make use of the Pulse Audio client library to communicate with the Pulse Audio Damon that mixes and manages the audio streams from the clients, yada, yada, yada. The proposal, of course, is to replace the Pulse Audio Damon with a functionally compatible implementation based on PipeWire. So, hey, and then he also talks about Jack because if you're not going to use Pulse Audio on Linux, typically you're going to use Jack. He talks about all pro audio is handled with the Jack client library, which talks to the Jack server. This proposal will install a Jack client library replacement that talks directly to PipeWire. So everything you're doing with Pulse and Jack, we want to just reimplement everything so it's really talking to PipeWire. And I think that's a bold move. I really do. And I, even though I think it's a little crazy to be switching to PipeWire as your default audio server at this point, I love the fact that somebody is actually putting it out there because, honestly, Pulse Audio and Jack, they have problems, right? We know this. Audio on Linux, the audio server situation on Linux has never been good. Pulse Audio has been around for more than a decade now and it still has issues. Jack has been around forever. And Jack also has issues. It's kind of a chore to set up. And even when you use it, the inputs and outputs and routing everything to where it needs to go, it's a mess. And if PipeWire can at least match what Pulse Audio and Jack currently do because I'll be honest, I don't like Pulse Audio. I can't stand Pulse Audio. There's so many issues I have with audio doing what I do with video and podcasting and things. You'll be surprised how often I have problems with Pulse Audio. I'll open up Pulse Audio Volume Control and whatever my inputs and outputs and everything are completely different from where they were the last time. I did something with them. Who changed this stuff? I didn't change it. It's like Pulse Audio. There's a gremlin in the audio server that goes in and changes all of my settings. And I can't stand it. So what are the advantages of moving from Pulse and Jack over to PipeWire? Well, according to Ben in this post, he talks about PipeWire having more flexibility, better integration, better performance, of course. And he also mentions better security, which I'd never considered the security aspects with the audio server, but he goes on to write, quote, PipeWire enforces per-client security. Object visibility and the actions on them can be configured independently per-client. This makes it possible to enforce a security policy for sandbox applications such as Flatpak, which Fedora is heavily invested in Flatpak. So I guess all of that makes sense. Now, this proposed change is for the very next release of Fedora. Fedora 34, which is coming up in, what, eight, nine months. Fedora often doesn't release on time, though they typically release when ready, but sometime in 2021 we'll probably get Fedora 34 come out and will PipeWire be the default audio server? I hope so. If it is, I will definitely install Fedora 34 on my main production machine, because I have extra drives anyway. I don't even have to wipe out my Arco installation. I'll install it on an extra drive and I will make some videos using Fedora 34 with PipeWire because I want to see PipeWire in action. It's been a project I have been following a little bit as far as the news on development, because we've got to move away from pools. We just have to. Now, just one brief aside, I did notice today that Pulse Audio had a new release. Pulse Audio 14.0 is released. And the release announcement here, the release notes is a massive page. Now, when you release something and you have to basically release a novel for the release notes, that just tells you what a massive effort that Pulse Audio is as far as an audio server. And that's why it's hard to replace these things. You can't just up and replace Pulse Audio in a day. And that's what is kind of scaring me with Fedora 34. We're moving from Pulse to PipeWire. Hey, we'll see. And the fifth story is SystemD wants to change how you manage your home directory. So of course, this is about Lynard Pottering, who is the creator of SystemD. Recently, he has been touting this new plan of his to completely change the way we do our home directories on Linux. So for a couple of years, Lynard has been talking about this. He's been talking about what he calls SystemD-HomeD. I guess it'll be the name of the project. So with the name HomeD, I'm assuming it'll be a daemon running. I don't know. Typically when you name processes with a D at the end, it is basically telling you that's a daemon. But anyway, SystemD-HomeD is going to be a new component for SystemD that reworks how we do our home directories on Linux. That's Lynard's quote there. He says, quote, adds strong encryption that makes sense. It supports automatic enumeration and hotplugged home directories and more. And he especially wants to get away from having to do things outside of your home directory for your home user. He's talks about, quote, I personally think that we should go towards systems where slash Etsy is generally not writable. What he's talking about here is for your home user, for example, what if you want to change your password? Well, that's written to a file somewhere in your slash Etsy folder, right? Slash Etsy slash pass WD or wherever it's at. And you have a lot of these configuration files and slash Etsy that have things to do with user stuff, user configuration. And he's like, why is that stuff there? Why is that in the root directory? Why is that in root, you know, slash Etsy? Why isn't that in the home directory? And I think he has a point there. And really when you're talking about the slash Etsy folder, that is specifically for configuration. That's really what that directory is designed for. And Lenard argues that creating or updating users should not be a configuration change, right? It shouldn't, it doesn't make sense why that stuff is there. He goes on to state that the current UNIX architecture, it mixes states with configurations and they're not the same thing. That's what he's trying to separate. Lenard goes on to say that he really wants to see built-in encryption home directories. And that's an interesting point because I do think in the year 2020 it is a little strange that we are not more into encryption as far as when it's on the desktop. Like that's not really a standard yet. Most people don't encrypt their drives. I know I don't. The reason I don't is I fear losing the passwords, losing the keys, and you can't get into that encrypted drive and, you know, if you lose the key to that drive you're just completely hosed at that point. You're never going to be able to recover that data. That's why I don't encrypt my drives. Pottering goes on to state though. One of the problems he has with the way encryption is done now is this weird situation where, quote, imagine booting up a laptop using full disk encryption, supplying your decrypting password, and then being asked to supply another lesser password to access your account on the network. And I agree that's a little strange because the really super strong password that gives you access to all the data on the computer is the one that actually decrypts everything. You know, is the decrypting password to let you have access to that encrypted disk. Why do you then need to add other passwords, you know, logging in as your user, getting access to the network, right? Because those are so much lesser passwords. I agree. It's probably not necessary once you have encrypted home directories as a default. There's no reason to have to worry about all the other passwords at that point because you couldn't even get into the thing unless you were you. Overall, I think he's got some interesting ideas about the home directory. Now, a lot of people are very, very negative with SystemD having anything to do with your home directory. SystemD is already your net system in many cases. It's your network manager and it's your cron job manager and everything else. There's so much that SystemD does on your system. It's not really in a net system. It handles many, many, many important things going on on your computer and a lot of people are like, well, it touching my home directory, that's just too much. Well, if you have a problem with it touching your home directory, you probably already shouldn't be using SystemD. You should probably be using a Linux distribution that uses an alternative, a net system. So I've never been one of these people that's a SystemD hater. I've never been one of those people that trashes Lenard pottering because he created SystemD. No, no, no. I trash Lenard pottering because he created Pulse Audio. And that is it for this edition of Unfettered Freedom. This was Unfettered Freedom Episode 12. Now, I release a new episode of Unfettered Freedom bi-weekly. That means every other week, every two weeks. That does not mean twice a week, guys. Every two weeks. So look for Episode 13 of Unfettered Freedom in two weeks. Now, before I go, I need to thank a few special people. I need to thank the producers of the show. They're my highest-tiered patrons over on Patreon. Without these guys, Episode 12 of Unfettered Freedom would not have been possible. The show is also brought to you by each and every one of these ladies and gentlemen. All these names you're seeing on the screen right now, those of you watching the video portion of the podcast. All of these names. These are all my supporters over on Patreon because the Distro Tube channel and the Unfettered Freedom podcast is sponsored by you guys. And I'll see you in the next podcast is sponsored by you guys, the community. If you'd like to support my work, look for Distro Tube over on Patreon. Alright, guys. Peace.