Untitled Linux Show 175 Transcript

Please be advised this transcript is AI-generated and may not be word for word. Time codes refer to the approximate times in the ad-supported version of the show.
 

00:00 - Jonathan Bennett (Host)
Hey, this week we're talking about ARM dev kits that got cancelled, arm on Linux and the new 4-ARM Raspberry Pi NVMe. Then there is the kernel Russia and the Department of Treasury Boy that's quite the story. Malabal, we talk about Pipewire, obs, alma, linux all kinds of fun stuff. You don't want to miss it, so stay tuned. Podcasts you love from people you trust this. This is the untitled linux show, episode 175, recorded saturday october the 26th, jack the thread ripper.

00:43
Hey folks, it is is Saturday and you know what that means. It's time to get geeky with Linux and open source and software and hardware and gaming and pretty much whatever else strikes our fancy. It's just that kind of show. It's the Untitled Linux Show and of course, it's not just me. We've got Rob, we've got Ken and we've got a set of at least nine stories and three command line tips. We're assuming they're all going to be command line tips today. Well, they're not always command line tips, but that's all right. And we're going to let Rob kick us off and we're going to talk about the Snapdragon. Now the big news this week. I don't know if this is what Rob was talking about, but the Snapdragon dev kit got canceled. Is that related to this, rob?

01:29 - Rob Campbell (Co-host)
No, that was. I believe that was the Windows Snapdragon dev kit that got canceled.

01:35 - Jonathan Bennett (Host)
It's at least partially related to this, like there's going to be some, some overlap between those two things, but dive into it and then we'll. We'll talk about that overlap and maybe what all this means.

01:45 - Rob Campbell (Co-host)
Yeah, so I mean, this isn't from this week. This is actually a story from the week before, because I miss one day and these guys let the biggest news story of last week just slip right on by. And well, for those that follow me on Mastodon, even though it is a week old, I promised I would keep you up to date on the news that these lovely folks seem to have missed while I wasn't here, or maybe they saved it for me because they knew how excited I would be for this.

02:16 - Jonathan Bennett (Host)
Oh yeah.

02:17 - Rob Campbell (Co-host)
This is a good news thing. I don't know about the dev kit being canceled, if that's good or bad or maybe just neutral for us, probably being linuxy, but here on the untitled linux show, we or maybe it's just myself, I have been excitedly watching the progress of qualcomm that they've been making on their new snapdragon x elite, snapdragon X Elite processors, hyped to be the Apple Silicon killers. We were all a little disappointed when these devices came out with Windows only on them first. But we knew Linux would be coming to them soon At least. You know we hoped it would be coming soon. You know we hoped it would be coming soon and not the relatively slow battle slog that slog. Now what maybe I'm on what word I was looking for there that the Asahi folks had to endure with the Apple M1, m2, silicon. I mean, they got to go on eventually. But you know Apple sure wasn't any help to them. But thanks to the folks at Qualcomm being better stewards of their hardware than Apple, most of the elements of the Snapdragon X1 Elite SoC system on a chip support is already upstream into the Linux kernel 6.11, 6.12,. We have seen various device tree files added for different Snapdragon X1 Elite laptops, so now it's just up to the a developer preview of Ubuntu 24.10, with plans to provide an installation image that will quote just work on as many Snapdragon X1 Elite systems as possible.

04:22
There are various feature limitations that remain. Like firmware still cannot be easily redistributed by Ubi2 or other Linux distributions until they have been released or relicensed, I guess. So now users need to jump through hoops of having to fetch the firmware files from their microsoft windows installation for use under linux. But yeah, just be like the old days where nobody ever updated firmware. Anyway, why not? But anyway, uh, I, I joke there kind of.

05:00
But if you're looking, if you are looking to test things out yourself, the best experience experience will likely be with a Lenovo ThinkPad T14 Gen 6. As this has been the devices we've seen the most, testing by Canonical and a quick look on the Lenovo site shows you can get them for as low as right around $1,200. They had like three skews on there. The top one started at 2600 us dollars. So exciting feature. But for that cost I may settle for an m1 m2 apple to put on linux for now. But if if you'd like to see me in one of these nice Lenovo's, all it'll take is a low price of 240 coffee donations and I'll be able to afford one. Maybe Jonathan has some news coming up here too that that I didn't quite catch this week, but he hinted at it a little bit for the show and maybe that'll be a way to get me into one. We'll find out what what's said on that.

06:05 - Jonathan Bennett (Host)
Yeah. So what's real fascinating here is obviously these ARM-based AI chips. They're kind of there for a while it was the new thing for Windows 11. That was what was coming down the pike, and then we got the news this week that the Snapdragon X Elite dev kit is canceled, pulled the plug on it, no longer going to happen, refunding people's money and everything, and so there's even a. There's a story in Pharonix about the kernel devs, because apparently some of them already had their hands on this thing and it's like what do we do now? We'd already started porting it and they decided, eh, essentially, yeah, that's good practice, let's go ahead and port it and make it work, which is hilarious, aren't there?

06:51 - Rob Campbell (Co-host)
already. There's already laptops out there with that. Is this for a specific piece of hardware? This is for, like, a specific Qualcomm piece that was, like their, their Nook equivalent right. Isn't that what it was? It's a dev board patches for the x1e001de snapdragon dev kit for windows so I mean, it's not the end of the world, it's one piece of hardware.

07:19 - Jonathan Bennett (Host)
Yes, but I suppose I'm trying to read the tea leaves here and like why? Why did Qualcomm suddenly cancel their Windows dev kit for their Snapdragon chip? And like that is interesting. Does that signal that Microsoft is moving, taking a step back from ARM? Are we suddenly going to see a whole bunch of these ARM devices that either get canceled or get abandoned? Whole bunch of these ARM devices that either get cancelled or get abandoned?

07:45 - Ken McDonald (Co-host)
It's something, or they do the same thing Apple did to get around the FCC guidelines with their Apple II when they came out with the Apple computer.

07:55 - Jonathan Bennett (Host)
Well, what's that? What's that Ken? What did Apple do?

07:58 - Ken McDonald (Co-host)
Well, they had somebody else build the video connection to go to the TV, the modulator, and they sold that separately. So each could get around the FCC, because it fell on the end user to put up with any interference.

08:23 - Jonathan Bennett (Host)
I got you. I see, yeah, yeah, yeah, I doubt it's quite that sort of a deal, but I don't know. They just it strikes me as being very weird that they're they're pulling the plug on it, rob what do you think?

08:33 - Rob Campbell (Co-host)
so here's my tea leave. I mean, they already have windows laptops out there um running on this. I. I think I think this was like their specific, like their little Mac mini equivalent, their Snapdragon XLA dev kit being an $899 US dollar mini PC. I think this was just their Mac mini equivalent and or maybe their what is that called? Where they make something so other people can kind of copy it, Like a reference implementation Reference.

09:08 - Ken McDonald (Co-host)
They have something to do with ARM, so well, here's what I think the company as far as licensing it to Qualcomm.

09:18 - Jonathan Bennett (Host)
Things are potentially changing with ARM too, it's true.

09:20 - Rob Campbell (Co-host)
I mean, if this is $899 US dollars, you can get a Mac mini for as low as $599 US dollars, $799 for a bigger one. So it's not really competitive from that perspective. If you want a little ARM PC, I mean you got not as powerful, but you got the Raspberry Pi 5s, which yeah, okay.

09:45 - Jonathan Bennett (Host)
So but, rob, you're thinking of this from a consumer standpoint. I'm thinking of this from a business standpoint. If you've got a dev kit, so your dev kit is okay. This is essentially the hardware, this is the chip that we're going to ship. Here's the dev kit you buy so that all of these different companies can do development work for supporting this chip inside of Windows. And when you then say, never mind, we're not going to ship a dev kit, that that is a, that is a huge, that. That is communicating a huge posture change, like a huge change of priorities, and so I just, I just have questions or there's already half a dozen um devices out there running this that they can easily just develop on anyway.

10:31 - Rob Campbell (Co-host)
Why do they need another board when there's already a whole bunch of other ones out there?

10:36 - Jonathan Bennett (Host)
Because dev kits tend to have extra debugging ports on them that laptops don't. That's usually what makes something a dev kit. It's also not a laptop, and you don't necessarily need or even want a laptop form factor to do development work. It just strikes me as being really, really odd, and it makes me think that somebody somewhere put the brakes on and so we may see less ARM on Windows going forwards than we expected to.

11:02 - Rob Campbell (Co-host)
That's your prediction. That is my take on this. My prediction is it doesn't mean much of anything, though I also read the linux. One plans to keep right on going, or whatever the linux port of it. I I didn't read this much, but I skimmed. I saw it out there.

11:20 - Jonathan Bennett (Host)
I didn't go in too deep yeah, the the, the kernel guys basically said, oh, we've got it in hand, we'll go ahead and keep working on it. So that is the Snapdragon Elite. And there is actually something else in the news about Linux on ARM and ARM desktops, and that is that System76 has just released the Thelio Astra. Now they're calling it an official desktop, for this is weird to me. Streamlined autonomous vehicle development, that's. That's just an odd thing to say about this. But it's power, so it's powered by ampere, and so it is a native arm 64 desktop. So, like this has way more power than a Raspberry Pi 5. It's got let's see so you can put up to 512 gigs of DDR4 into it. It's got M.2 slots, it's got PCIe Gen 4. And it's literally an NVIDIA Ampere Ultra in a desktop form factor.

12:20
So I don't see immediately how many cores that is, but it's quite a few. It's like I want to say 32 or 64 cores. It's somewhere in that ballpark, and so this is sort of the opposite extreme. Whereas Qualcomm is dropping things, you've got System76 that's picking up this really exciting new thing. That's picking up this really exciting new thing and they are like I said, they're pitching it as really good for doing development for autonomous driving platforms, and I'm sure that there are companies that will do that. But this thing is going to be great just for doing ARM development, for doing ARM compiling. So if you need to do a bunch of recompiling for ARM, this is going to be one of the places where you can actually do it natively, which there are some big advantages for doing it natively. So I have worries about ARM on the desktop, but at the same time, system76 is saying we're all in, we did it here. It is ARM on the desktop. Go for it, although this is more of a workstation than a desktop, but still here it is.

13:24 - Rob Campbell (Co-host)
Unless I mean from a personal user perspective unless Qualcomm's Snapdragon X Elite is going to outperform Intel in power, kind of like the Mac has outperformed their old intel based chips. Unless they do that from an end user perspective, I'm not too interested in it. But I mean I can the the great use case for, you know, for a desktop the great use case, though, as you said, is being able to develop on ARM, because I mean, obviously it's going to speed it up, you can pack the power on it, but from a desktop, from an end user perspective, I don't care too much about power usage. I just want the thing to be powerful. From a laptop, you know, give me the best of both worlds that you can. But yeah, if you're developing an ARM, that seems like a great option to speed up your development workflow.

14:35 - Jonathan Bennett (Host)
And so the Ampere Ultra is 128 cores of ARM64. Goodness.

14:42 - Rob Campbell (Co-host)
I'd be curious to see how that that actually benchmarks against intel, though oh, definitely, yeah, that would be super interesting.

14:49 - Jonathan Bennett (Host)
I think I think gearling gearling talked about one of these. Uh, he may have one. Um, let's see, I I know he has a video about this because he's he's all about the uh, or he may have a video, but a post anyway. He is all about the whole. Let's put expensive video cards on ridiculous arm hardware, and that's just been his thing for a while. Uh, as soon as there was a pci express lane in the raspberry pi that you could, you could feasibly get to. Um, so, yeah, I I expect to see him trying to get a hold of this and playing with it.

15:29 - Rob Campbell (Co-host)
Is this for sale yet?

15:33 - Jonathan Bennett (Host)
I found it, but I don't see a way to buy it.

15:37 - Ken McDonald (Co-host)
Probably not yet. From the looks of it, not until the 12th of September. That's when shipments start.

15:40 - Jonathan Bennett (Host)
There you go. I'm sure it's going to be a lot of coffees, oh yeah.

15:45 - Rob Campbell (Co-host)
Or a nice christmas gift for somebody hey, you know I have received some actual physical gifts from, uh, a fan out there. I haven't shared them on the show but, uh, you know what fan? Um, this would be awesome not that your other. Not that your other gifts weren't uh humorously awesome, but uh yeah this would be cool.

16:15 - Jonathan Bennett (Host)
Give him an inch and he takes. He wants to. Really, he wants to be a ruler. Give him an inch and he wants to be a ruler. Oh my goodness. Uh, pharonix does have an article where they where, where michael has gone through and benchmarked this, uh, it looks like just against the, the two system, 76. Um, oh, so it gets a 7980x, which that's a. That's what a thread ripper, I think. Yeah, so he did. He benchmarked it against a thread ripper. So I've got the link. I threw the link there in the show notes and you can go and take a look at that. It looks like the Threadripper is, generally speaking, about twice as fast, but in quite a few of these they're pretty close. So, and I imagine once you get to performance per watt, yeah, the Threadripper is like double the speed. But the CPU, the power consumption, is quite a bit better on the ARM-based machine. So interesting stuff there. All right, ken, do you want to talk about Pipewire?

17:22 - Ken McDonald (Co-host)
I want to talk a lot about Pipewire, especially since Bobby Borisoff reported on updates to the Pipewire 1.2.x and the 1.0.x series this week. First is the release of Pipewire 1.2.6 with a bug fix update that maintains API and ABI compatibility with all previous 1.2.x and 1.0.x versions. It addresses several key issues. One involves the filter chain module. We're fixing the bug results in more consistent behavior. We're managing audio effects. Users will notice improved responsiveness when interacting with audio streams now that stream states are based on the underlying node state. Another improvement in Pipewire 1.2.6 concerns exporting nodes, which now have their state changes handled synchronously. This modification allows the server to immediately start the driver, reducing the likelihood of initial X runs or buffer underruns which can cause audio glitches. Jonathan knows well about that. Stream flush handling has been enhanced, ensuring more efficient data processing during playback and recording. Also fixed is an issue in which mixed information will sit to destroyed ports, resulting in errors within jacked clients. This change minimizes unnecessary errors, particularly in complex audio setups connected by multiple clients.

19:06
Now Bobby also wrote about another bug fix update to just the 1.0.x series under Pipewire version 1.0.9. It addresses several key bugs fixes, including the fix for a file descriptor leak and protocol confusion which previously could lead to resource leaks and incorrect memory usage. It also corrected a bug affecting the audio mixer synchronization after port selection has been resolved, sometimes leaving users with muted audio, and issues related to FD leaks and protocol inconsistencies were fixed, contributing to a smoother overall user experience. And last of the ones I'm going to mention is a fix for a use after free issue in the real-time module when stopping a thread. As I've only covered some of the changes in both of these updates, I do recommend reading both of Bobby's articles if you do want more information. I also recommend rating for updates from your distribution, since they will have better integration with your system. Since they will have better integration with your system, unless you like playing with compiling programming and libraries yourself.

20:32 - Jonathan Bennett (Host)
I have compiled Pipewire and installed it and it's actually not too terrible. You know, some of those system libraries are a pain to get working, but Pipewire is not too bad.

20:41 - Ken McDonald (Co-host)
My command line tip later is going to help you find out which version you do have there you go.

20:47 - Jonathan Bennett (Host)
Yeah, I uh, pipeware has come along such a long ways in the past couple of years. It's it's really, it's really turned into quite the quite the good solution for for most most things. Um, there are. There are a few cases where people still need to run jack or, you know, something doesn't work in it, but for, yeah, I, I mean it, I can't I don't know of any case where pulse audio works better than pipewire at this point.

21:13 - Ken McDonald (Co-host)
Uh, honestly, unless it's for uh legacy applications maybe that's why pipewire is so cool.

21:23 - Jonathan Bennett (Host)
When they wrote pipewire, they said, all right, we're just going to take the Pulse Audio API and we're going to re-implement it directly in Pipewire. And we're going to take also the Jack API and we're going to re-implement directly in Pipewire. So any program that was built for Pulse Audio can just talk to Pipewire kind of natively and it just works. And any program built for Jack can just talk natively to Pipewire and it just works. Now Jack is complicated and hairy and you're talking about, you know, very tight timing tolerances. So there's been a lot of work that's had to be put into that, a lot of bug fixes over the years to make things actually work.

21:55 - Ken McDonald (Co-host)
Now you're talking about into Jack In Jack.

21:58 - Jonathan Bennett (Host)
As well as so yeah, so programs that were written for Jack, that's things like Ardor, to make those programs work kind of seamlessly with Pipewire. That has been a big lift, because you were doing things in Jet like in Ardor when you went to render something out, it would put Jack into freewheeling mode where it just sort of ignored the clock on the wall and we just chugged through the audio as fast as possible. And so they had to re-implement that in Pipewire to be able to do this, to put Pipewire in this state, because Jack would actually Ardor, would actually use the Jack backend to process the audio still, and so they had to come up with a way to say, okay, we're going to take part of the Pipewire engine and we're going to also put it into freewheeling mode so that we can chug through all this audio super fast to be able to render a session out.

22:46 - Ken McDonald (Co-host)
Lots and lots of complicated things there and still even with pulse audio or jack, you still had to get awesome house involved yes, yeah, pulse audio jack pipewire as well still runs on top of alsa.

22:59 - Jonathan Bennett (Host)
Also is basically your, your linux kernel drivers for audio um, whereas pipewire sits on top of that and it's the, the thing that routes everything around. Um they they have in the past year or so um they did add the ability to talk to firewire drivers and in pipewire um. Last I checked that still didn't work very well. I need to go back and check that again, actually, um, but I guess in that case it would not talk with alsa, it would talk with uh. Is that f of ado? No, f of ado is the one that's. Yeah, I think it was.

23:34
I think it's what they implemented, the evils covered that when it came out yeah, I, I did, I tried it and, like I said back then, it didn't work very well, I think. I think the community actually sent Wim Tamens some Firewire hardware and it's like look, your code looks beautiful, it doesn't actually work. Have some real hardware to test it on. Oh, my goodness, that's helpful.

23:58 - Ken McDonald (Co-host)
You didn't volunteer some of your hardware, did you?

24:00 - Jonathan Bennett (Host)
No, because I only have one and I really don't want to get rid of it. It's pretty nice audio hardware from back in the day. Oh, okay, Rob, let's talk. Oh, Rob, let's talk about Bitwarden. I don't know much about this story yet. I've seen the headline what is Bitwarden doing?

24:16 - Rob Campbell (Co-host)
So over the past week or so, there have been some concerns in the open source communities and I believe I saw it first on Reddit. Of course, that's where all this stuff comes first that Bitwarden is beginning to move away from open source. So you know, my first rant is they better not? This is a big reason why we love them I am even a paid subscriber that I love them. I, I am even a paid subscriber, uh, that I love them so much. But if you betray us, bit warden, we'll leave you faster than we left last pass when they betrayed us. So that's my warning, uh. But so what happened here is a week or two ago, users began to notice the client introduced a quote Bitwarden slash SDK, internal dependency and that internal SDK having a license clause that stipulated quote. Here it is you may not use this SDK to develop applications for use with software other than Bitwarden, including non-compatible implementations of Bitwarden, or to develop another SDK. So this was raised as an issue in their GitHub, as it essentially makes the Bitwarden client no longer free and open. The comments were flying around on Reddit until Bitwarden founder and CTO, kyle Spearin, made a comment about the situation on GitHub, saying and I'm just going to read his quote here.

26:00
We have made some adjustments to how the SDK code is organized and packaged to allow you to build and run the app. I mean, actually I'm going to back up. First, there was a quick comment about it being a bug. Later on, just a few days ago I think, or very recently, this quote was there. So it says after that issue they've made changes to allow you to build and run the app with only GPL slash OSI licenses included.

26:32
The SDK internal package references in this client now come from a new SDK internal repository which follows the licensing model we have historically used for all of our clients licensing model we have historically used for all of our clients. The SDK internal reference only uses GPL licenses at this time. If the reference were to include Bitwarden license code in the future, we will provide a way to produce multiple build variants of the client, similar to what we do with WebVault client builds. The original SDK repository will be renamed to SDK Secrets and retain its existing bitwarden SDK license structure for the secret manager business projects. The SDK Secrets repository and packages will no longer be referenced from the client app since that code is not used there. So they're essentially saying the inability to compile the free client because of that non-GPL compliant license code that was showing up in there. That was essentially a bug and they plan to fix it and apparently now have fixed it.

27:46
Um, you know, but there is going to be closed source code available. It sounds like in the regular client that most people will just be downloading and using, at least if you just uh download their pre-compiled stuff, whichever operating system use. From what I read, this isn you know, it's not anything malicious, but rather an attempt to add more features to the client that they, I guess, apparently haven't been able to add with their own free code or for some reason wanted to make these extra features that they're building closed. You know, I believe I can accept this answer as long as there is a free option available. Maybe I don't like it too much, but you know, don't push it, bit warden. I don't want to have to take my subscription fee and find another client. I've been enjoying you too much, just the way you are.

28:50 - Jonathan Bennett (Host)
Cue the Billy Joel song, don't co-change it. Honestly, I totally buy their explanation. It is extremely challenging to run a business using GPL code and try to thread the needle of you know what value adds can we give and still respect the GPL. And something like this accidentally leaking out with the wrong license and then as soon as it was pointed out they fixed it. I mean, that sounds like they did the right thing in the end. I would not attribute this to any sort of maliciousness. Now, obviously keep an eye on them. If they try to do something like this in the future, well then maybe we begin to rethink that. But at this point this seems like an accident where they're working on things for internal code and some of their internal code is not ever intended to be external. It's not ever intended to be used under the gpl. So they've got this other license that they use internally and the stream's accidentally crossed that that sort of thing happens I definitely want to cross the stream don't cross the stream.

29:58 - Rob Campbell (Co-host)
It's not in this case as long as they're not going to do something like freeze the free, freeze the features on the free version, which I guess I'm fine with, I'm happy with the features. I don't. I don't know what else I need, but you know they could freeze a feature and say, well, all new features are going into our paid product, our closed source, which I did that once about 10, 15 years ago with the program I worked I I uh, I made but uh and it did so well for you I made about a thousand dollars a year off of it for several years.

30:30
Nice for a 995 one-time fee, so not too bad yeah, so I guess the um.

30:39 - Jonathan Bennett (Host)
The question there with that idea is is there other gpl code from people outside of bitwarden or do they have a uh a license agreement? You know, is there a um contributor license agreement where people have to assign the copyright to bitwarden? I guess if that's the case, then they can change it if they want to I know, like the, the main feature I paid for is the free.

31:07 - Rob Campbell (Co-host)
One doesn't have the otp, the uh, the, the mfa in there, but if you pay for it you get that. And yeah, it's kind of nice having that right there yeah, that's fair.

31:21 - Jonathan Bennett (Host)
So kaira makes a point here that bidwarden would be forked so fast if they went private for essential features. Yeah, that's part of the point of using the GPL.

31:30 - Ken McDonald (Co-host)
And they've already got a competitor in KeePassXC as far as that.

31:34 - Jonathan Bennett (Host)
Yeah, not as good but no, Bidwarden is kind of the premier password manager right now, like standalone password manager at this point, isn't it?

31:43 - Ken McDonald (Co-host)
Especially if you're wanting one that lets you have multiple devices synced without setting up your own cloud service.

31:54 - Rob Campbell (Co-host)
But what's great is you can set up your own server if you want everything to be in your own vault.

32:01 - Jonathan Bennett (Host)
Yeah, okay, we do need to mention. This is fairly important. Bit warden is actually a sponsor of twit. I didn't. Oh yeah, we were not thinking about that. I did not know that for sure until I just went and looked um but I knew that at one time I forgot I was actually gonna say full disclosure. Bit warden is a sponsor of twit but obviously that you know was not a, was not a contributing factor.

32:25 - Ken McDonald (Co-host)
You can tell we did not take that into consideration while discussing this.

32:28 - Rob Campbell (Co-host)
I went from LastPass to Bitwarden while Twit was still, I think, advertising LastPass or maybe they weren't, but it was definitely before Twit started advertising or, I guess, before Bitwarden started sponsoring anything on Twit. So I went to them first. Um, but you know, last pass used to be the premiere and used to also be a sponsor of twit. Um, they betrayed us all in several ways several times.

32:59
Um, if you don't know, feel free to reach out to me and I'll I'll tell you. I we don't have time to go through all this on the show, but that's when I went to Bitward. And then you know, twit was smart and they got them as a sponsor shortly after.

33:18 - Jonathan Bennett (Host)
All right. So I've got two stories here, and they are both these kind of stories that you know it's like, guys, this is not what I want to be talking about. I I would. I would like to talk about the positive things of open source, but no, no, we've got feuds and fights and fussing and all sorts of stuff, and so the first thing we're going to talk about is the shake-up among the kernel maintainers.

33:42
You may or may not be aware of this, but within the past I think it's within the past week a change landed in the Linux kernel and a couple of dozen one or two dozen maintainers were sort of quietly removed from their maintainership, and the thing that they all had in common was they basically all had a ru email address. They were all Russians and they. So the first thing is you may be surprised actually, like this is something we should mention you may be surprised that, wait, there were Russians that were maintainers in the Linux kernel yeah, yes, there were. And some maintainers in the Lennox kernel yeah, yes, there were. And some of these people in fact all of the ones that were removed were employed by or affiliated with a handful of Russian companies, like Bycall being one of them who makes little arm-based CPUs. But Bycall products end up in drones and missiles and things used by the Russian military, and, and so long as they write good code, nobody particularly cared that they were in the kernel.

34:58
There have been some minor things in the past where I think one a maintainer from Ukraine opted to not take a pull request from a maintainer that was employed by Bycall for potentially obvious reasons. So there have been a couple of things that have happened, but they just quietly yanked all these people from being maintainerships and there was a little bit of an uproar and finally we got some clarity on this and so one of the core kernel maintainers came out and said look, the Linux Foundation lawyers gave us these guidelines. They said okay, we finally got it cleared that we can tell you exactly what they said, and it was if your company is on the US OFAC SDN list, subject to OFAC sanctions program. So essentially, the US government puts out these are the companies that are sanctioned for one reason or another, or owned or controlled by a company on that list, our ability to collaborate. You will be subject to restrictions and you cannot be on the maintainer's file.

36:04
So the lawyers at the Linux Foundation came to the conclusion very recently that anyone that we suspect is being, you know, either employed by or collaborating with one of these sanctioned companies, they can't be on the maintainer's file. It's a problem. We can't do it. Legally we cannot do it, and that's fairly understandable, because the Linux Foundation is based in the United States. Linus Torvalds and Greg KH they're both US. Greg is a US citizen, for sure. I don't know off the top of my head if Torvalds is actually a citizen or just a long-term resident, but he resides in the United States regardless.

36:45
And so like, on one hand, well, that's fairly understandable, but then, on the other hand, there were some questions that people had, and one of them was like what about Huawei? What about the Chinese sanctioned companies that are still part of the Linux kernel development team? And I thought that was a really valid question, ted. So one of the other kernel maintainers high-level guy came out, spoke out and said hey, just so you know, there are different rules and regulations within these sanctions, and while there is an exception in place for these Chinese companies, that exception is not in place for the Russian companies. Okay, again, that's understandable. People were annoyed that this was done suddenly, without much communication, but it is what it is, all right. So that's part of this.

37:42
And then there was a statement from James Bottomley, again a core colonel maintainer, who obviously was part of the discussion with the lawyers about this, and here's what he said, and this is the part that's important. Is the part that's important. He said we are hoping that this action alone will be sufficient to satisfy the US Treasury Department in charge of sanctions and we won't also have to remove any existing patches. And when I read that, my head exploded Because this says to me someone from the US Treasury Department contacted the Linux Foundation and said hey, not only do you have members of these sanctioned companies in your maintainer's file, which is a fair. That in and of itself is a fairly legitimate thing to do, it's just. It's just, these are the sanctions, these are the laws. So there's that. That's fairly reasonable to me. But this statement implies that there was a threat that you would also have to remove the code that these people had written, and that is a problem.

39:01
And I will tell you that is a problem because in the United States, code is protected speech. This has been established several times in court cases. Code is, first Amendment, protected speech and the idea that the government can come in and say you're not allowed to say this because it was first said by someone that we disapprove of, is a huge First Amendment violation and a huge problem. So we are now past the whole Russian-China sanctions issue and we're now on to what is the US government allowed to do when it comes to things like the Linux kernel and other software projects.

39:36
I see a huge disconnect between these two. It's two separate issues. You can say I really don't want Russians or people associated with these Russian companies in the maintainer slot. That's fine, that's a discussion to have, that's fine. But to say the US government gets to come along and say you're not allowed to put this into your source code, which is protected speech, yeah, I think that's just entirely beyond the pale and I have literally already called on the Linux Foundation to launch a lawsuit against the US government because I consider this to be entirely unreasonable and illegal and this sort of a precedent cannot stand. We will all be in trouble if the government can come along and say, nope, we don't like that code, you're not allowed to put it into your open source project. So that is my take on it. I am beside myself over this. It's a good thing you got that take.

40:31 - Ken McDonald (Co-host)
I know where you can find some real good lawyers that will help you.

40:34 - Jonathan Bennett (Host)
The Linux Foundation already has really good lawyers. Just so you know, I've sent an email directly to Greg. I have contact information. I've been in contact with him before and I man, I I am unhappy and worried over this. It strikes me as being a terrible thing. So we'll see what happens.

40:55 - Rob Campbell (Co-host)
They don't know that they have to take anything out.

40:59 - Jonathan Bennett (Host)
They are. So the exact statement is that they are hoping that this action alone removing them as maintainers will be enough to sufficiently satisfy the US Treasury Department that is in charge of sanctions. So they're hoping that the action they have taken is enough to satisfy them. They're therefore hoping that they won't have to remove any existing patches. So I mean, just stop and think about that, though. If the US government said, okay, you've got to remove any patches in the Linux kernel that any of these people associated with these companies have written back to the beginning of when the sanctions happened, I guess, like what would that look? Like who?

41:40 - Rob Campbell (Co-host)
I mean, if they said, what would?

41:41 - Jonathan Bennett (Host)
that break, it would break everything. That's the thing. You cannot just remove the patches, that's not so. A friend of mine he's a fellow developer, we were talking about this and he says you would basically just have to roll the entire source tree back to when the sanctions started and then start over from there which the government, being a, a user of Linux, seems like it could be rough for them too.

42:07 - Rob Campbell (Co-host)
But I I wonder, maybe that's part of their concern going forward. But if it's all, good now. But I you know linux. I mean linux isn't like a business per se. I don't know how the organizational unit um holds for like linux itself. So I i'm'm curious what kind of enforcement could actually be done? Who would be responsible, or what entity?

42:44 - Ken McDonald (Co-host)
Or is somebody concerned about what recently happened earlier this year possibly happening with the Linux kernel?

42:54 - Jonathan Bennett (Host)
The XZ hack. Yeah, so that may be what someone is concerned about. That is a ridiculous concern to have in this context, because if someone was going to do that, they would not do it from a dot ru email address. They would do it from a dot gmail address. They would say hi, my name is john smith and I live in the middle of the States, and here's my patch to make things better.

43:16 - Rob Campbell (Co-host)
Yeah, what if they just said no.

43:24 - Jonathan Bennett (Host)
I mean, what's Okay? So the Linux Foundation is a trade organization, so it is a. It is not a non-profit, it is a, I think trade organization is actually what they call it. But they don't control Linux either. No, but they pay the. They, they have a lot of money and they pay the salary of Torvalds and Greg KH and some of the other core maintainers. Like, there's, there's a lot, there's a pot of money there that you could at least go after for fines.

43:52 - Rob Campbell (Co-host)
I mean I'm just curious how that could could all play out. I mean if they just said no, like if there's a way that could play out, you know he goes back to his home country and just runs it from there, gets some other sponsor to pay him, and it's like, what are you going to do?

44:12 - Ken McDonald (Co-host)
Well, I think part of the problem is some of these sanctions are I guess the only way I can think of to put it is copied by some of the other countries.

44:27 - Jonathan Bennett (Host)
There are sanctions in countries around the world. And one other point to this is Torvalds in typical Torvalds fashion, you know, sped it off on the Colonel mailing list and he basically said look guys, I'm from Finland, there's no love lost between me and the Russians, which is like, okay, that's fine. There was one thing he said that I was honestly a little disappointed by and I may be wrong here, but I will just tell you my thoughts on it he, he, there has been pushback, and torvalds mentioned the pushback and he said it's, you know, it's obvious that the russian troll farms are out in force. Because there's so many people that have replied to this. And I have seen this, I've seen this pattern a few times now.

45:17
We saw it with Godot and now we're seeing it with Torvalds that when a company or a person makes a questionable decision that is unpopular in certain places, and the obvious thing happens that a lot of people reach out to them to tell you, to tell them that it's an unpopular decision and they don't approve of it, they immediately jump to oh, it's bots. Unpopular decision and they don't approve of it. They immediately jump to oh, it's bots, it's russian bots. It's like maybe it's not russian bots, and maybe there's just enough of us out here that are annoyed by the decision you made. We think you handled this poorly. I, I'm, I'm a little over the whole. Let's jump straight to. It's russian bots every time, because I don't think it's all russian bots and I don't think it's always russian bots.

45:57 - Ken McDonald (Co-host)
Sometimes it's Russian bots every time, because I don't think it's all Russian bots and I don't think it's always Russian bots. Sometimes it's American bots.

46:00 - Rob Campbell (Co-host)
Sometimes it's real American people that are annoyed Malaysian and whatever else.

46:08 - Jonathan Bennett (Host)
Anyway, that's my piece. I have spoken my piece on that matter and we're going to move on to something else. Ken is going to tell us about OBS and a new versioning system. Yes, I am, and I're going to move on to something else.

46:18 - Ken McDonald (Co-host)
Ken is going to tell us about OBS and a new versioning system. Yes, I am, and I'm going to thank Bobby Borosoff for reporting on the latest news from OBS. This week the team has decided to change the version numbering scheme used. They started following proper semantic versioning to bring more clarity. Starting now, obs will follow a majorminorpatch format, like, say, 30.0.2. Now here's an explanation for those that may not understand the semantic scheming. The major number introduces big breaking changes that might require users or developers to adapt to these changes. The minor number indicates adding new features or improvements without actually breaking anything. Hopefully, the patch number will include bug fixes or other small tweaks that need immediate attention. Security patch anyone. And then, according to the official OBS Studio announcement by and I don't know his real name, just his username that he used War Champ 7.

47:33
We didn't have a strict structure or consistency in deciding when to release a new version or how to label it. Releases would come when we felt there were enough new features or fixes to warrant one and we'd assign a version number based on how significant we thought the changes were. Sometimes a major version would include breaking changes and other times it was simply a feature-packed release. We wanted to highlight this inconsistency could be confusing, and moving forward, we've committed to using major version numbers strictly for updates that fundamentally change how you use OBS or affect developers by modifying the API or dependencies. A great example of why this change was needed is OBS 28.0. That release included an update to the user interface framework we use, which broke compatibility with several plugins. Jonathan, do you remember that?

48:37 - Jonathan Bennett (Host)
I don't actually.

48:39 - Ken McDonald (Co-host)
Now, in hindsight, our inconsistent use of major version bumps muddied the significance of this update, causing frustration among both users and developers. By sticking more closely to semantic versioning, we hope to avoid these kinds of issues in the future. That's the end of the quote from War Champ 7. I'm sure Jonathan could probably provide several more examples if he took time to think about that.

49:10 - Jonathan Bennett (Host)
Yeah, it's neat to see actual semantic versioning here. How was it before? Whatever we feel like versioning, oh, I can't tell you how many times I've been in projects. I think it was in the chat room. I mentioned Meshtastic.

49:29
So we use Platformio, which we compile in Visual Studio Code most of the time, and it does this thing automatically where you say you know, I want to use this library and this version number, but you can put a special character in there I think it's a caret right before the version number. So that's saying this version number or any later non-breaking change which. So in this case that's the dot patch. So you could say, okay, so if I'm running 30.0.1 and 30.0.2 comes out just automatically, next time I compile it, go out and grab that and I cannot tell you how many times compilation has broken because of that. Like, all the time compilation breaks because of that it's ridiculous. Compilation breaks because of that. It's ridiculous. Um, so all that to say yay for properly actually doing semantic versioning where you can do a patch, update and be pretty, uh, pretty confident that things are not going to just break all over the place yeah, they didn't change their sequencing, they just did the major minor.

50:32 - Rob Campbell (Co-host)
Now it looks like I was looking through their versioning, so I was like no, it's just going forward.

50:38 - Ken McDonald (Co-host)
In fact, I'm running OBS 30.0.2, even though they're getting ready to start working on 31.

50:49 - Rob Campbell (Co-host)
Yep, yep, I'm a big fan of the versioning that has the years in it, not enough? Yes, you can do it that way, um, they'll do year, month and do a major minor if you wanted you could do so like what.

51:05 - Jonathan Bennett (Host)
What some companies will do is they'll do year and then month, and that's essentially like. That is essentially their, their major version, and then you do point, releases you do point releases after that?

51:14 - Rob Campbell (Co-host)
yeah, and that works too I like that because you can quickly look at it and know like, okay, this is four years old. It was like I don't know how long ago. If I'm running version 29 of OBS, I don't know if that came out?

51:28 - Ken McDonald (Co-host)
When did OBS 28 come out?

51:33 - Rob Campbell (Co-host)
I don't know. I was just looking at all the versions. I should have looked at the dates. I closed that already.

51:39 - Jonathan Bennett (Host)
The only thing I don't necessarily like about the year and month based, particularly when a business they do that and then they also say, ok, so we have a yearly or a six month release cadence that goes along with it, and that is that not every major feature fits neatly into that release cadence and there is an advantage to saying, okay, this is the next major feature and when we get it ready to go out, that's going to be the next version bump, um. And so then you have like an idea of all right, major feature version bump.

52:11 - Rob Campbell (Co-host)
major feature version bump um as it goes to however much of the version bump have, have the month be the version bump and then have point releases after that for minors well, but when?

52:22 - Jonathan Bennett (Host)
when a lot of these companies go to the year and the month, they are then on a six month cadence.

52:27 - Rob Campbell (Co-host)
But they don't have to, you don't have to, no, but then your I guess the struggle is if you have two major versions in the same month, but that doesn't happen every once in a while it doesn't happen. That doesn't have to happen. They can wait. They can wait a few days for the next month all right, rob, it is time for the rob.

52:52 - Jonathan Bennett (Host)
Talks us all into spending money. Corner of the show.

52:56 - Rob Campbell (Co-host)
Now this time I promised I wouldn't try to sell you, jonathan, to buy anything after your wife donated me a couple of coffees. So now I'm going to. I'm going to work on Ken for a little bit. So what do we know about Ken? Well, we know he likes to stick to a budget aka he's cheap. And he likes to stick to a budget aka he's cheap. And he likes Raspberry Pis and has even ran them as his main desktop computer before. So I have to find something that fits nicely into those categories.

53:40
If he's going to run a Raspberry Pi as his desktop, he's going to want an m.2 ssd rather than the slow sd card uh method. That's been default for years and fortunately, you know, using the m.2 on a raspberry pi isn't all that new, I mean it's. It's new with the raspberry pi 5, which has been out for some time now. But finding one that is 100 compatible and fits nicely on the smaller you know frame, the smaller box, you're looking at that. You want to fit that on. You know you don't want a long stick sticking out. There could sometimes be a challenge, or so I am reading. I haven't purchased a pi 5 yet myself to try it out, but from what I'm reading, it can be a challenge, I guess. Well, raspberry Pi Group is going to make it easy for you by selling a Raspberry Pi branded M.2 SSD. It comes in a nice small form factor, just a little short stick, that fits nicely with your Pi. That fits nicely with your Pi and it comes with the options. You can either get a 256 gigabit for only $30, or a 512 gigabit for only $45. And with that you get a guarantee that it's going to work with your Pi 5. And for the price that it competes with others that I found on sale online.

55:04
I found some that were, a few that were a little cheaper, but I'm not even sure I would trust those ones that I found that were. I found some that were I don't know. You can't get too significant when you're only at $30. I think I found one like $21, but I'm not even sure I trust that. I trust the Pi branded one. So if you don't already have a Pi hat for the M.2 drive, you can now also buy the new Raspberry Pi SSD kit and that comes with one of those drives of the choice and an M.2 hat plus with the SSD, all in one for only $10 more. So you know the other prices it's $40 or $55. You know it's, it's been a few years since I've purchased a new Pi myself. Uh, you know it's, it's been a few years since I've purchased a new pie myself and I think my newest might, might actually be a three, but I really feel like it's it may be time to update my pie gaming machine. Um. So, ken, how is the availability these days for raspberry pies?

56:13 - Ken McDonald (Co-host)
I haven't checked recently, but what I'm really interested in is the MTOT 2 hat and NVMe SSD kits where they both come together.

56:29 - Jonathan Bennett (Host)
You know, there's another kit that I don't remember who dropped it, but it's the SSD and the PoE together.

56:38 - Rob Campbell (Co-host)
Is that Pyoria or something like that?

56:43 - Ken McDonald (Co-host)
Raspberry Pi's got an official kit out that they're bundling them.

56:47 - Jonathan Bennett (Host)
No, waveshare, it looks like. Waveshare is the ones that did it.

56:52 - Ken McDonald (Co-host)
They were doing it before. Yeah, they would offer to provide the M.2 hat along with the SSD.

57:01 - Jonathan Bennett (Host)
Or maybe it's GeekPi. This link is off to GeekPi. Oh, there are several. There are several that are doing it actually.

57:11 - Rob Campbell (Co-host)
Yeah, something like Piori or I don't remember what it is.

57:16 - Jonathan Bennett (Host)
There are multiple companies that have put these out. Waveshare has one, geekpi has one. Geerling has a video about this too. I thought it was just one that was doing it, but apparently there's five or six companies that have put these out, so that's interesting.

57:30 - Rob Campbell (Co-host)
But now you can get it right from the company. If you trust that Pi, you trust their hats and SSDs too.

57:36 - Jonathan Bennett (Host)
I don't think. Does Raspberry Pi have one that's both PoE and the NVMe?

57:41 - Rob Campbell (Co-host)
Oh, the PoE, yeah that's what I don't think. So, no, no, this is just the SSD with SSD.

57:50 - Jonathan Bennett (Host)
Yeah, now it's cool that Raspberry Pi officially has this, but you know, it's pretty interesting that now you can do both, uh, and you can. You can do it without having to do anything weird like stacking hats on top of each other. That's one of the other ways to go about this, and maybe it wasn't ever the greatest, um. So yeah, but now there are, there are options for doing this otherwise, um, and fairly reasonably, like this waist sure is only 32.99, so it's always nice to have options.

58:19 - Rob Campbell (Co-host)
Oh, that would be. I should have brought it years ago. I still have a Pi 1, maybe it's a 2. But I gutted an old, broken Nintendo, I put it in there, I wired it up to the controllers and then I made little like usb adapters. So it looks like you're plugging it in like a usb to controller adapter into the, into the nintendo and and I tried to keep all the um outlets and plugs as much best I could. Uh, obviously I couldn't do anything with. Well, I could have did something with the hdmi. I suppose I put it could have converted. If I didn't, I just drilled a hole in the back for the HDMI. Maybe next time I'll convert it to it to RCA. I think that's what what's on there. But man, I don't know where I was. Oh, you know what? I guess that that the POE one would be cool, because I could like run this Nintendo. I mean, it looks like a Nintendo POE.

59:21 - Jonathan Bennett (Host)
Yeah, one cable Just Ethernet.

59:24 - Rob Campbell (Co-host)
But that kind of breaks away from the somewhat realism I was trying to go for there. It's like, hey, why there's no plug in here, or why isn't it plugged in? And it's still powered up. Magic, magic, yeah, batteries plug in here, or why isn't?

59:37 - Jonathan Bennett (Host)
it plugged in and it's still powered up.

59:38
Uh, magic, magic, yeah, batteries all right, I think it's my turn again. And uh, I've got, I've got another these story. Now, this one is kind of more funny than anything else, but I've got another one of these stories that it's like what, what is, what is going on here? Um, and I saw this directly on Twitter actually on X is where I saw this and then people have written about it since then. But there is a company called Malabal and they make laptops, apparently, and they wanted to. It stands for Modular, adaptable, long-lasting Innovations for Business and Leisure. Malabal what a mouthful.

01:00:22
Anyway, they recently put out a blog post called Don't Support the Coreboot Project, because they wanted to put Coreboot on their own laptops, found it difficult, tried to work with the Coreboot team. Their own laptops found it difficult, tried to work with the Coreboot team and that didn't work out. And so now they have come out and very publicly have attempted to shame the Coreboot team. And so I saw this link and you know I thought, well, okay, that's kind of an inflammatory headline, but all right, whatever, I'll go read through it. And then you know I'm reading their blog post and you know they're talking about how they've had trouble working with the Coreboot team and there were problems, and so my immediate thought is okay, sometimes open source projects are difficult to work with for various reasons. That's fine, that's a fact of life. That happens sometimes. Sometimes developers are a little prickly and are difficult to work with. This is a real thing, this happens sometimes. So that was my viewpoint.

01:01:35
Going into this, like, oh well, let's see what happened and maybe we can learn some things from it. And then and then I get part with you the article and I see this as part of their blog post update system 76 as principal engineer decided to chime in and make a fool of himself, so we banned the entire state of Colorado for life. It's like, oh okay, I see what's happening here. Suffice it to say I think part of the problem here was the people running Malabelle maybe are not quite as mature as they think they are, but anyway, the whole thing is a mess.

01:02:19
There were obvious problems when they were trying to make this happen. They were paying money for it and did not get as much progress as they wanted to do as quick as they thought they should. And then finally they sort of walked away from it, which is one thing, but then they went and they complained, they aired their dirty laundry online, which you know is usually not a great idea. Um you, you could have put out a much more professional now, as they say, as they say, yes, they could have put out a more professional um statement about this, but it wouldn't have gotten nearly as much publicity. So so I'll read from the closing here If we would have said something to the effect of we felt the experience was unsatisfactory and thought they could have done a better job, it would not have conveyed the magnitude of the actual emotional distress that we were subjected to, which can be proportionally measured in our response to ban their states and countries. Where was this all? Posted on Malabellecom? Slash features slash. Don't support the Corbett project on their site.

01:03:26
Yeah, so I've got a, I've got one link. Let me actually I'll put the link to this particular, this particular Blog post right in the show notes. But it's hilarious.

01:03:41 - Rob Campbell (Co-host)
Not a good place for that. I mean those kinds of comments you need to post anonymously on Facebook. That's where you do that kind of stuff. That's what it's there for.

01:03:51 - Jonathan Bennett (Host)
Anonymous the burner account. Yeah, put it on Facebook with the burner account. Yeah, it's hilarious, it's kind of sad. I think it was written somewhat tongue in cheek, but at the same time, it's just it's, it's a, it's an outpouring of things being um broken. Just put it that way things, things are broken and this is what we get the internet is a hard place to hold your tongue yeah, that's one way to put it. Not everybody is as grown up as they would like to think. They are too, I think that's.

01:04:31 - Ken McDonald (Co-host)
I know that's what mutes for yeah, you know.

01:04:36 - Jonathan Bennett (Host)
Uh, all right, let's move on. Let's talk about, uh oh, alma linux. What is, what is in the future of alma linux, ken?

01:04:46 - Ken McDonald (Co-host)
well, I'm gonna ask christine hall, seth aq, seth okumar-Pellini I'm just going to call him SK from now on and Bobby Borosov. They all wrote about AlmaLinux OS developing a preview version of AlmaLinux OS 10, aimed at providing the community with transparency and an early engagement in the development process. Early engagement in the development process. According to Christine, the AlmaLinux devs released AlmaLinux OS Kitten. It's intended for downstream developers building off of AlmaLinux. Now AlmaLinux chair and main spokesperson, vinny Vasquez. Didn't you interview her on FlossWookley?

01:05:37 - Jonathan Bennett (Host)
We did. We talked with benny.

01:05:38 - Ken McDonald (Co-host)
She's great but she explained to christine in a zoom call and I'm going to quote earlier this year we noticed that centos stream 10's code was available on the repos and was getting ready to be shipped. So we started building our pipeline or, excuse me, pipeline for amalyn 10, anticipating that next year Red Hat will release RHEL 10 based on CineOS Stream 10. Now Christine explains that Alma devs can go ahead and start building their next version now, knowing that when RHEL 10 comes out they'll likely only need to make a few minor adjustments before pushing Alma Linux 10 out the door. The Alma folks also decided to go ahead and make the work they're doing now under the name Alma Linux OS Kitten 10 available to downstream developers, instead of letting those developers twiddle their thumbs and waiting for Alma Linux 10 to be released.

01:06:37
Now Bobby explained in his article why Alma chose the codename Kitten. The name is a nod to AlmaLinux's tradition of using cat-themed codenames. So much like a kitten grows into a cow. Almalinux Kitten 10 will evolve into the final AlmaLinux 10 release. All three touch on some of the key features and enhancements of AlmaLinux OS Kitten 10. They're looking at adding Spice and KVM for IBM Power, continued RPM packaging for Firefox and Thunderbird, more hardware supported, re-enabling frame pointers and support for older hardware. Now you can find more details on these features, as well as other information that I haven't touched on, in the respective articles from Christine SK and Bobby. I do have them linked to in our show notes, and I think, jonathan, you may be seeing a kitten in the near future.

01:07:47 - Jonathan Bennett (Host)
Oh, probably you know it's interesting. I was just looking before the show about Fedora 41, which you know, that's a couple of versions beyond RHEL 10 and Alma Linux 10. Kitten Fedora 41 is going to drop on Tuesday I think, and they just had the go-no-go and got the poll to go on that. So you know, definitely a terminal countdown for Fedora 41. We'll definitely be talking about that next week. But I'm sure I'll be running Alma Linux 10 on something before too much longer. That's kind of where I, that's kind of where I settled on my enterprise linux.

01:08:26
I'm doing a lot of alma linux these days so you may may have a few clients trying it they probably won't know that they're trying it, but at some point they will yeah, for I haven't really used either, but from a philosophical perspective I like alma I I've said it before I'm glad that both Alma Linux and Rocky Linux exist, but Alma Linux definitely seems to be the one that is more popular at this point.

01:08:54 - Ken McDonald (Co-host)
Especially with being a clone. That's not really a clone.

01:08:58 - Jonathan Bennett (Host)
Not anymore. It's less of a clone. I think at this point you could say that Alma Linux is a clone of CentOS Stream, more than CentOS itself.

01:09:09 - Rob Campbell (Co-host)
And Rocky is still trying to clone.

01:09:12 - Jonathan Bennett (Host)
Rocky is still trying to be a bug-for-bug clone of RHEL. Yes, so a little bit of different philosophy even in that between the two of them, and that's kind of the point that when I first said that I'm glad both of them are there, that's what I was getting at. There's a little bit of room for them to experiment and go different directions and, uh, time will tell which one is the way that works out, yeah.

01:09:33 - Ken McDonald (Co-host)
And is Rob sushi-ness what this command line?

01:09:42 - Jonathan Bennett (Host)
Oh, oh goodness, Um, something like that. I thought maybe he was hissing like a snake. What is that? Rob, you want to take it away to talk about the first command line tip what?

01:09:52 - Rob Campbell (Co-host)
is ss and it's not the ss minnow, which is, yeah, we're too, yeah, a gilligan's island reference. For those who, uh, don't know, that was the ship that they um crashed onto the island with now. So ss is, uh, it's a socket tool to see what, what sockets you have open, and I'm not going to show you every use case of it, but, uh, let me just uh, flip my flip this over for those watching it. So, the main use case, I mean you can just run SS and it's going to show you a whole bunch of stuff, which is a lot of stuff. And for those watching, I am doing this on my pie hole, pie hole, which is a, a DNS with DNS filtering, for those who do not know. But I thought there could maybe be some interesting things here. So, but the main way I I like to use it is ss space, dash t u l n, and dash t stands for tc, u is UDP, dash L is listening, so it shows the ports that are listening on this and dash, and it's going to show the port numbers rather than the name. So if I do this and I did this earlier, as you can see those, you know ports.

01:11:23
I guess maybe you don't all know ports. You see port 53 there. Well, my head's in the way, so let me just scoot over here. Port 53, that is the DNS port, obviously to be expected. Port 80 is the web port, which that's a non-secure HTTP port, which that's how you interface with the Raspberry Pi. It's on my own home network, not accessible to the public, so it's okay that it's not HTTPS. And there's a port 25 open, which surprised me because that is SMTP. So I don't know why SMp is open on my pie hole. I'm curious about that. I'm going to look into that. Good thing I don't have this out on the internet or otherwise it could become an, a mail relay for people for spam. Um, you know, just an example, if I remove that and you can see what those are a domain meaning DNS, http, that 4711 there is, it's, it's a what's that called Something?

01:12:39 - Jonathan Bennett (Host)
something runs on port 4711.

01:12:43 - Rob Campbell (Co-host)
Yeah, I don't know exactly what it is. It it's not a default port in in in the scheme of that's the, that's the api, that's the pie hole api that's the pie hole api. Oh oh yeah, so I do have the api running.

01:12:57
I forgot because I I have an extension on my gnome which I can uh monitor and stuff, but uh, I don't know if I've ever showed that show some other time, but uh also, but I don't know if I ever showed that show some other time, but also SSH I don't know if I mentioned that, but port 22. So yeah, so there are a bunch of other things you can do. That's how I like to do it. I'm really curious why.

01:13:24 - Jonathan Bennett (Host)
Port 25.

01:13:25 - Rob Campbell (Co-host)
Why port 25 is open on there. But you can see if you do dash. So there's a a bunch of different things. You know. You can resolve host names, uh, display all sockets, a bunch of other things you can. You do only ipv4, ipv6 okay, rob, pull that.

01:13:46 - Jonathan Bennett (Host)
Pull that back up where you've got that list. That shows port 25. Sure, let's take a look at this. I think there's something real interesting here and move your head. Port 25 is listening on IPv4 and IPv6, but they're both bound to localhost. That's the 127.0.0.1 and the colon, colon 1. They're both bound to localhost, which means that it's not listening for anything from the outside of the machine, it is only listening internally, and I am pretty sure that Linux machines, just by default, all of them do internal mail, and that is probably what that is.

01:14:25 - Rob Campbell (Co-host)
Well, if you have it running. I checked others and I didn't have port 25 running. But you don't necessarily have to have SMTP even installed, so I must have installed, I suppose. I think there is in Pyhole. I think you can maybe configure like alerts or reports or something in there.

01:14:48 - Jonathan Bennett (Host)
It's probably for sending out something like that yeah, probably, and so it's running an internal little smtp daemon to do that that makes a lot of sense which I'm not using, that.

01:14:58 - Rob Campbell (Co-host)
So I guess I could just, uh, I could uninstall that and maybe find out what gets broke. Hopefully it doesn't break. I could uninstall that and since I'm running this on a very bare minimum virtual machine, I could have more resources.

01:15:15 - Jonathan Bennett (Host)
keep it running cleaner and slicker, yeah, but at the least it is running. It is running only talking to local hosts, so good point.

01:15:26 - Rob Campbell (Co-host)
I didn't look at that too close because obviously you know it does show the IP that that it's running on too, which I didn't mention that, or the ip that it's listening on um, I didn't mention that, but good point yeah.

01:15:35 - Jonathan Bennett (Host)
So for those that are asking, pie hole is a way to you. You black hole certain dns entries, usually relating to like advertising yeah, generally speaking, it is a network-wide ad blocker.

01:15:49 - Rob Campbell (Co-host)
So, for example, um, if you're on my network using that, as it's a dns server. But it's the purpose. Main purpose is to block ads. So, even like, if you have your phone on my wi-fi at my house, ads are going to get blocked on games and stuff like that, which is probably one of the not being rendered within a browser first.

01:16:14
Yeah, it's one of the greatest features, because phone games these days are terrible, terrible, ridiculous. I go out somewhere, my kids go out somewhere and they're like this game is unplayable. It's like I didn't even know I had all these ads in it. But you could also also use it for your own dns. So I have custom dns in there to point to various systems and machines around my house so I can uh call them by name. So it's also also just a d server. I suppose you could even take out the blocking and just use it for DNS, but that's not exactly the purpose.

01:16:51 - Jonathan Bennett (Host)
Yeah, and then, since we're doing Q&A, steve actually asks why use a pie hole when you can just use an ad blocking DNS service? And I've got the answer for that one the pie hole is not packaging up and selling your data. The pie hole is not packaging up and selling your data, whereas if you use an ad blocking DNS service, they are absolutely recording the DNS names that you look up and selling, that. You collecting that information and selling it somewhere.

01:17:13 - Rob Campbell (Co-host)
And this I mean this is my you are paying for the service. Even if you are paying for the service. I mean, this is my service on my home network, so I can configure it any way I want. You know I can.

01:17:24
I can configure it any way I want. You know I can. I'm sure some. I know a lot of these services you can, especially if you're paying. You know I can whitelist this. I can or I can block this. I can allow this. I can configure it any way. I have dns on your own network, so that way it's basically like a proxy. You know, instead of reaching out to this place on the cloud every time I want to make a dns call, it's just, it goes to my own dns server. It's like yeah, I looked this up already on some other device not long ago. It's cash cache. I'm not even going to reach out to the Internet.

01:18:08
Here you go, here's your IP, or you know you could redirect things as you want. You know, if I want to have fun with people in my house, I could redirect Google's IP to my own custom website and totally mess with them. I haven't done that yet. I've thought about it. I've thought about having some fun like that. But I haven redirect Google's IP to my own custom website and totally mess with them. I haven't done that yet. I thought about it. I thought about having some fun like that, but I haven't, I think.

01:18:32 - Jonathan Bennett (Host)
HTTPS would probably trip you up trying to pull that particular prank.

01:18:36 - Rob Campbell (Co-host)
Oh yeah, the S would these days. Ah, modern technology, but yeah, plus, I just like to have a DNS server. So if I'm going to have a DNS server in my house anyway to point things to my own services around my house, I might as well have an ad blocking one. Plus, it's like easier to set up than like Bind or any DNS solution like that. It's easy to set up. Yeah, absolutely, web interface too to manage.

01:19:12 - Jonathan Bennett (Host)
Yeah, so I've got a command line tip that I ended up using just the other day. It's kind of funny the reason I needed it. So I BitTorrented some things and of course BitTorrent, it has legitimate uses, it's not just for pirating things. I went to download something with with k torrents and it just it wouldn't download, for whatever reason. I I was trying to feel like why, like, maybe the dot torrent file is corrupted, or come to find out? The problem was that I'd already downloaded it in the past, so it was already sitting in my download queue and rather than tell, tell me that Ktorrent was just silently failing, anyway. So I went and I said, well, I don't want to install another full-featured graphical BitTorrent client. Surely there's a command line BitTorrent client. And there is, it's rtorrent and it worked marvelously. And so, just so you know there is is a bit torrent client that works right on the command line, which you know, if you're grabbing, say, an iso on a remote machine that you don't have a full gui on, you might really this might be really really useful. Um, so anyway, it's, it's super simple to use. It's just our torrent and then either the torrent file or the torrent link, you know the magnet link and it will. It'll go grab the thing and, uh, do the download.

01:20:34
Now the only thing that you have to watch out for and this is something that I ran into is how do you get out of it? It's kind of like them uh, it's hard to exit, it's hard to exit the program and uh, apparently it is. You can stop it with control d, which you know. That's another one of those, uh, another one of those keyboard shortcuts that you should really, you should really know. So, whereas control c is to cancel, to send a kill, eventually, control d just sends a um, uh, an end of line um signal actually.

01:21:10
And so control d to stop and control q to quit. So it is. It is one of those programs that you might find yourself stuck in I. In fact, when I was using it for real, when I got done using it, I actually opened up a new terminal and just killed it with the kill all command. Uh, but there it is, our torrent ways to close it. I mean, yeah, right, right, right. So you can either use control q, you can use kill all, or you can use the power button on the front of your computer or just close the terminal window uh, maybe that doesn't always close things it's not always a window either.

01:21:48 - Rob Campbell (Co-host)
Sometimes you're in a tty yeah, yeah, could be.

01:21:53 - Jonathan Bennett (Host)
Anyway, I thought I thought our torrent was cool. Uh, can you've got a? You've got a pipewire tip for us, don't you?

01:22:02 - Ken McDonald (Co-host)
yes, and the tip tip I have is gonna show how you can find out what version of Pipewire you are currently running, as well as how you can change some properties or the configuration file that you're using. Now I've got up on my screen the command. It's pipe wire. Now you can do either dash H or dash dash help, and it'll show the help information which will give you the options. In the case of my Ubuntu studio, it gives me three options, one of which is of how to get the version and this is the important one for everybody Pipewire space dash dash version will let you find out what version of Pipewire you're running so you can see if you've got the latest and greatest. In the case of my Ubuntu Studio, it's compiled with lib pipe wire 1.0, so dot 5. So Ubuntu studio 2404 is currently using the 1.0 X series and I am going to quickly switch to tumbleweed. And there we go, and here it's running version 1.2.5, so I don't have 1.2.6 on it yet, even though I did do an update today With the 1.2.x series.

01:23:41
It gives you a couple of other commands. One is a "-v for or dash dash verbose. This will let you increase how much information or the verbosity by one level. You may specify multiple times to get even more information. This is the one that'll let you do a dash capital P or dash dash properties where you can add given properties to some of the SPA or JSON objects that you may need when you're doing other things with pipe wire.

01:24:15
And of course, the "-c or "-config will let you load whatever config file. So if you've got multiple config files set up, you can use that to easily switch between them. And this may be the first in a series of commands for managing Pipewire from the command line, because the link I've got attached in the show notes will give you a GitHub link that gets you or takes you to docspipewireorg where you can see some of the other commands. If you want to get a head start, it'll wait for me. And in our show notes I've also got attached a link to the Google document that's got these screenshots that I've been displaying up as well.

01:25:05 - Jonathan Bennett (Host)
All right Very cool.

01:25:10 - Ken McDonald (Co-host)
What version of Pipewire do you have?

01:25:13 - Jonathan Bennett (Host)
Ronnie, I was just looking on my machines and I've got 1.2.0 on the desktop and 1.0.3 on the laptop 1.3. Pop OS right.

01:25:27 - Ken McDonald (Co-host)
Yes, three on the laptop and one dot os right yeah yes, yeah, it probably needs an update.

01:25:32 - Jonathan Bennett (Host)
I haven't. I haven't run the updates in a while. It's my uh, I don't do much web browsing on it, but it is the production machine, so I tend to just leave things when they're not breaking.

01:25:43 - Rob Campbell (Co-host)
I have an off-topic question, Ken. Why was there a video of me playing in the background there? I only know it was me because it was wearing my South Central shirt.

01:25:58 - Ken McDonald (Co-host)
Because I forgot to cut that out by doing this, oh.

01:26:05 - Jonathan Bennett (Host)
Hey, look it's me out by doing this. Oh Didn't hey, look it's me.

01:26:12 - Ken McDonald (Co-host)
I used it to resize the my screen so that it wouldn't go over the life of a feed tonight. In the past it was the QR code.

01:26:28 - Jonathan Bennett (Host)
That's hilarious. That's hilarious, alright. Well, we have had our fun, we have covered our topics, we have hit our command line tips and we're now going to let the guys get the last word. If they want to, we'll let Rob go first. Do you have anything you want to plug?

01:26:46 - Rob Campbell (Co-host)
I've got nothing special, just my normal uh called, uh, come and connect with me and um, why is my url not on here like normal? Oh, here we go I. I did the wrong one. There's's the right one, wrong scene. It would have worked, but here you go.

01:27:10
You can come find my information at robertpcampbellcom and that information is also, I think, on the show notes. When they post it, I think there's a link to it. Usually, I think, and when you come there, and that's if you don't know how to spell Campbell, hopefully you know how to spell Robert and P, and it's just the letter P for those not watching. When you come to my website, you will find links on the top to my LinkedIn, my Twitter, my Mastodon, linkedin, my Twitter, my Mastodon and a place to donate those 2,200, no, however many copies. I need to get an X1 Elite laptop. That's where you do that and you know one of you doesn't have to donate all that for that laptop. You can, you know, split that up. If every one of these listeners does one or two coffees, that's probably enough. So I mean just one coffee from everybody.

01:28:17 - Ken McDonald (Co-host)
Just one coffee from everybody a week until Christmas.

01:28:23 - Jonathan Bennett (Host)
There you go. All right, ken you have anything you want to plug.

01:28:27 - Ken McDonald (Co-host)
Yes, this week former Floss Weekly guest, christine Hall writes about the rebirth of the opensourcecom community website idea as the All Things Open website. Under the name we Love Open Source, they are sponsoring the all things open conference in Riley, north Carolina, where on Monday Benny Vasquez will be talking on the state of Alma Linux. If you are nearby, I'd recommend going to check it out, jonathan, the all things open website may also be a good source for some of the future floss weekly guests.

01:29:07 - Jonathan Bennett (Host)
That could be interesting. I will definitely have to check it out.

01:29:11 - Ken McDonald (Co-host)
Yeah, definitely go check out the link I've got in the show notes about all that.

01:29:15 - Jonathan Bennett (Host)
Very cool. All right, thank you to each of you for being here. I do quickly want to plug. Of course, we've got Hackaday. It's where my security column goes live every Friday. That is also where you'll find the new home of Floss Weekly, and we are super happy to be able to continue doing that. Other than that, I just want to say thank you to everybody that watched us, those that got us live and those that get us on the download. We appreciate it and we will see everybody next week on the Untitled Linux Show.