Untitled Linux Show 196 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)
This week we're talking about the upcoming Ubuntu 25.04 release and the big XZ release. You may not know why it's a big deal, but it is, I promise. Then we sort of scratch our heads over EUOS. We talk about the new Torvalds rant of note in the Linux kernel mailing list and a whole lot more. You don't want to miss it, so stay tuned. Podcasts you love From people you trust.

00:35
This is the Untitled Linux Show, episode 196, recorded Saturday, march 29th, running with safety scissors. Hey folks, it is Saturday and you know what that means. It is time for some Untitled Linux Show. I almost announced the wrong show there. I've been doing this for a long time and every once in a while it still sneaks up on me. It's like oh, are you doing something else? Anyway, we are doing the Untitled Linux Show. It is the three of us today. I've got Ken and I've got Jeff with me and we're going to have a lot of fun. It's maybe, maybe, a lighter news week than normal. I'm not sure. It kind of seemed nothing, nothing earth shattering this week, at least that I'm aware of.

01:13 - Jeff Massie (Co-host)
No, I'd agree it was a calmer, calmer sailing. But you know, calm is good.

01:19 - Jonathan Bennett (Host)
Every once in a while you need some calm. Calm is good, yeah yeah, there is something sort of big in, uh, in ken's world over in ubuntu land and I'm gonna jump over to jeff, who also is going to talk about ubuntu, but something nice. So we'll just do these two stories out of order. We'll let jeff go first, we'll have ken disconnect and reconnect and then we will jump back in with ken's story yeah.

01:41 - Jeff Massie (Co-host)
So, as you're soon to learn, ubuntu 25.04 beta has been released and its full release is less than a month away. It's time to drive into whether there are any notable regressions or improvements in performance. Michael Larable from Phoronix has done extensive benchmarking to analyze the performance of the updated code, and he tested Ubuntu 2410 alongside 25.04 beta using an AMD Ryzen 9, 7900 X3D, which is a 12-core processor or, sorry, that should have been dyslexia there 9700 X3D, sorry, 9900. Because I specifically remember it was kind of weird that he used the brand new 12-core system that we did the benchmarking on last week, but sorry. 9900 X3D, 12-core processor, 24 threads. His setup included two 16 gigabyte DDR5 sticks at 6,000 megatransfers per second and he used an AMD Radeon RX 7900 XTX GPU. Now, if you check the link provided in the show notes, you'll find a detailed list of version changes between the two for Java, python, security updates, compiler flags. You know all the nuances between the two operating systems.

03:10
Michael noted that these benchmarks focus on desktop usage, not server or high performance computing workloads. Those tests are likely to follow later using enterprise or server grade hardware. While the AMD graphics hardware used here is slightly older, it's far from obsolete. Importantly, though, or even impressively. There's still some gains coming for RDNA 3 graphics cards, especially when it comes to Vulkan, which saw a notable uplift in performance. Rdna 4, of course, is still an involving driver's state and getting better all the time, but Michael picked RDNA3 because it not only shows what more people would have, but RDNA4 wasn't out even wasn't even available when 24.10 was released. So it wouldn't really be a fair comparison to compare an operating system for hardware that didn't exist yet and because the changes in the driver. He was picking something a little more stable, hardware-wise and driver-wise to see what subtle improvements came in.

04:17
For gaming benchmarks, the improvements were evident, but generally minor and not always noticeable in real-world gameplay. Still, you know, faster is faster and overall most benchmarks were closely matched, but the 2504 snagged a few small victories. Out of 129 tests, 2504 Beta claimed 91 first-place finishes compared to 2410's 38. However, when you look at the geometric mean and the rankings, the wins for the 2410 were narrow and while 2504 excelled in areas like compression and Python performance, with significant improvements, most results were neck and neck.

04:56
One of the key takeaways is that Ubuntu 2504 is you know it has a pretty clear improvement. You know, you know it has a pretty clear improvement. You know but I mean realistically we can't say Ubuntu as much as we can say you know, thanks in no small part to the kernel maintainers and you know all the people that are involved in the graphics, drivers and everything else that goes into making the core of Linux. You know, solid, not that Ubuntu doesn't have a part to play as well, but we should give love to everybody in the community because it's a large community project, not just canonical. So basically, with the updated graphics, newer kernel, newer software versions, linux performance is steadily advancing. For the detailed results and benchmarks, check out the article linked in the show notes. But in summary, there are no major regressions and overall the outlook's very promising yeah.

05:49 - Jonathan Bennett (Host)
So something to keep in mind is that this is the up-to-date ubuntu and we got ken back. Um, this is ubuntu 2410 that has been fully updated, and then compared with 2504, so, uh, I bet it would be a different story if it were based on the uh, the stock, like 2410, off of the original iso. That's that's going to be.

06:13 - Jeff Massie (Co-host)
Oh yeah, it's going to be a different, different conversation altogether yeah, because ubuntu now is more leaning into I won't say cutting edge, but they're're a lot closer to the edge than they were. So they're really being more aggressive about updating their packages using newer kernels, not holding back quite so much. Yeah, yeah, absolutely.

06:39 - Jonathan Bennett (Host)
Interesting stuff.

06:39 - Ken McDonald (Co-host)
How do I sound?

06:41 - Jonathan Bennett (Host)
You're good, Ken. Why don't we talk about Ubuntu Studio? What's new there?

06:45 - Ken McDonald (Co-host)
How do I sound? You're good, ken. Why don't we talk about Ubuntu Studio? What's new there? Well, we're going to be talking about both Ubuntu and Ubuntu Studio, and maybe even touching on Kubuntu. Plucky puffin entering public beta testing. Gotta love live. The Ubuntu Studio team also announced the beta release of Ubuntu Studio 25.04. Now Marius reports that the recently released Linux 6.14 kernel will be powering Ubuntu 25.04.

07:29
Featuring the GNOME 48 desktop environment, ubuntu 25.04 promises many goodies like the triple buffering feature from Ubuntu papers as the default document viewer replacing events, papers as the default document viewer replacing events and beaconDB-powered geolocation services. Dual boot users will see improvements with a focus on BitLocker protected Windows systems. Users can now install Ubuntu alongside existing BitLocker partitions if enough unallocated space and dual boot support are available for encrypted installations and other advanced scenarios. The Ubuntu desktop installer even received a new option to replace an existing Ubuntu installation, so that should make it easier for some of y'all if you're already running Ubuntu and don't want to wait to download it, but sharing some of the same desktop features as Ubuntu. Ubuntu Studio will also ship with the latest Qt 6-based Plasma 6.3, kde desktop, firefox 136, snap as the default browser, and LibreOffice 25.2 is provided as the default office suite In the full installation. Of course, ubuntu Studio's most notable change is that the default panel icons are back and that default is dynamic based on which one of the applications you have installed.

09:15
For audio applications, we'll see Pipewire upgraded to 1.2.7, audacity upgraded to 3.7.3. Ardour upgraded to 8.12.0. On the graphic side, we're going to have Digicam upgraded to 8.5.0. Of course, gimp is going to be upgraded to 3.0. Krita will be upgraded to 5.2.9. On the Blender video applications, we'll have Blender upgraded to 4.3.2, not 4.4 yet hopefully by 2510. Kdn Live will be upgraded to 24.12.3, 24.12.3, and free show will be upgraded to 1.3.9. Now in our show notes I have linked to the Ubuntu Studio blog about this release as well as Marius's article. The Ubuntu Studio blog includes a list of multimedia package versions if you want to see everything that will be current in that one. I also want to remind you this is a better release. It should be finalized sometime later in April, but we could see changes before that final release in April. And, jeff, thank you for showing us how the performance will compare with the 2410. I'm currently running A tiny bit better.

10:44 - Jonathan Bennett (Host)
In a lot of cases performance will compare with the 2410 I'm currently running, hopefully A tiny bit better in a lot of cases.

10:50 - Jeff Massie (Co-host)
Yeah, but the biggest thing is no regressions. But there were some compression in Python that really took a jump forward.

10:59 - Ken McDonald (Co-host)
Yeah, Maybe Chrome will be able to support 9,600 sample rate. That's what was causing it earlier.

11:10 - Jonathan Bennett (Host)
Maybe that might be exactly what it was. Yeah, so it looks like the the upcoming Ubuntu release is going to be is going to be pretty solid. I am. I'm rather excited Now 2504, that is not one of their LTS releases.

11:24 - Ken McDonald (Co-host)
No, it's only supported for nine months.

11:26 - Jonathan Bennett (Host)
Yeah, is 2510 going to be an LTS or do we wait until 26?

11:31 - Ken McDonald (Co-host)
No 2604.

11:33 - Jonathan Bennett (Host)
Okay, that's what I was thinking. So this is where they get to do all the crazy stuff, right, they get to try everything, try all the nutty things, although I don't know if there's anything too crazy coming in in 2504.

11:48 - Jeff Massie (Co-host)
Wayland.

11:50 - Jonathan Bennett (Host)
I suppose that's true. We've been talking about Katie.

11:52 - Jeff Massie (Co-host)
We're playing around with Pipewire.

11:55 - Ken McDonald (Co-host)
I'm looking, I'm waiting for them to catch up with the OpenSUSE Tumbleweed. It's running 1.4.1 already.

12:08 - Jonathan Bennett (Host)
I kind of think Ubuntu is. Ubuntu is never going to quite catch up with tumbleweed or fedora, because those are just, they are the bleeding edge distros, right yeah, they're, they're not gonna they're.

12:17 - Jeff Massie (Co-host)
They're kind of back from the edge they're. They're the safety scissors of uh distribution. You know, now you, you have some of the Debian and some of those that they're in red red hat enterprise. Now those are the full bubble wrap. You know, there's the encased in foam and all protected.

12:37 - Jonathan Bennett (Host)
But I wondered where you were going to go from? From safety scissors, scissors.

12:45 - Jeff Massie (Co-host)
Yeah, no, that's pretty accurate. Or you just say they just fully take the scissors away. They don't even let you have the safe ones.

12:53 - Jonathan Bennett (Host)
I don't know about that. I've broken some rel installs pretty badly over the years.

12:58 - Jeff Massie (Co-host)
But in general they run in pretty old stuff because it's tested, it's known, it's you know.

13:08 - Ken McDonald (Co-host)
Yeah, old stuff, because it's, it's tested, it's known, it's you know. Yeah, and even with ubuntu studio I've done some damage, it's true, and y'all you can.

13:13 - Jeff Massie (Co-host)
You can always do damage, but if you're running straight rel or you're, you know, debian stable, you're pretty solid. Yeah.

13:25 - Jonathan Bennett (Host)
Yeah, that's fair all right, so so, speaking of distros and interesting, unique things, uh, there is a bit of news on the gaming front. I came across this article that, uh, apparently hp hewlett packard is looking at making a Linux based gaming handheld, and that that really surprised me. So we know that Lenovo is working on their Legion Go S and it's going to be SteamOS based. There are some other companies that are sort of looking into doing this. We might see something from ASUS and MSI, but one of the executives from over at HP said that they are looking at let's see how exactly do you say it. Thanks to Valve adding the beginnings of support for non-Steam Deck handhelds so SteamOS running on decks that are not the Steam Deck it is now possible to make a gaming handheld an HP gaming handheld running SteamOS.

14:34
And then he said something really interesting here. It says that they struggle with the experience of Windows, and that sort of caught me off guard, although I am reminded of another story in the news this week, and that is that in Microsoft's latest beta of Windows 11, the next Windows 11 release so like the developer preview, I think they call it they've removed the script that you can run at your first install. To say I really don't want to do a Microsoft account, thank you. It's like bypass NRO, I think is what it's called, and they've removed that. And so now Microsoft is trying even harder to make everyone get Microsoft accounts, which, you know, is something that most of us really don't want. We have to do a Windows install is something that most of us really don't want. We have to do a windows install, we don't want to do a microsoft account.

15:28
Um, so all of that is interesting. I I have to wonder if that's the sort of thing that hp is talking about, because you know hp is going to want to be able to give their gamers their own uh, you know first run experience, and on linux, that's easy. On windows, that's much less easy, that's easy. On windows, that's much less easy. Um, so yeah, it'll be. It'll be fascinating to see, to have yet another player in this space and we'll see if HP makes something that's worth playing on?

15:55 - Jeff Massie (Co-host)
Well, I bet they will. And I kind of wonder if the steam deck was less of valve wanting to get into hardware and more of a proof of concept to show it can be done. Sure, and then why make another one if you can have the HPs, the Lenovo's, the other hardware manufacturers who are well versed in hardware? Because it's one thing to write code and have very good code. Manufacturing hardware is a whole other world. I mean, you can have a great idea, but mass producing it is a monumental task and expensive and expensive and you've got to have quality control and all this sort of integration. It's a headache. And if Steam could go, okay, we did it. Now we let all these other manufacturers who are going to fight for it run our, our operating system, which you can.

16:45
Linux can very easily tailor to whatever the hardware you have and you can add and subtract what you want, versus Windows where you're kind of stuck with the whole ball of yarn. You know. I mean it's kind of all or nothing so many times with Windows and I'm sure there's ways you can cut it back, but they don't like to and there's probably a lot of extra licensing to have that custom version. If you want a custom version, and versus linux, it's like I want to do it okay, have at it, yep yep, so when will we see a console version?

17:20 - Jonathan Bennett (Host)
that's an interesting question. Um, I personally I think it makes sense to have a SteamOS console, maybe even in like the Chromecast kind of form factor, the little thing that just hangs behind your TV. I mean, there's nothing that says that a console has to be a huge box.

17:39 - Ken McDonald (Co-host)
Somebody comes out with Bluetooth or Wi-Fi capable controller that would hook up to it, I mean.

17:54 - Jeff Massie (Co-host)
Well, I think you'll see a console when you see the full OS SteamOS PC release. Because really, what's a console? It's a simplified PC. It's just a small form factor. It's one of the powerful things of consoles, it's just they're very dedicated hardware. So, okay, we'll just make up an example. You have an AMD chip with 3D cache and an internal GPU, integrated GPU. Well, now you don't need any of the NVIDIA driver stuff. You don't need Intel driver stuff. You don't need. You know there's so much stuff you don't need. Because here's your hardware, I can put it in a smaller package and only support these limited things.

18:44 - Jonathan Bennett (Host)
Hasn't there been some? Speaking of which, for those of you listening, I held up my steam controller that just happened to be on my desk with a significant look on my face. Uh, is it? There are some some like next gen steam controller rumors that's been going around a little bit.

18:59 - Ken McDonald (Co-host)
They didn't start them.

19:01 - Jeff Massie (Co-host)
Yeah, I, I don't think it would not surprise me. There's not going to be a second steam deck. They they said they were going to have a second steam deck when they had a very major jump in hardware performance. I'm thinking, with all the hardware manufacturers getting on board, why compete with your audience? Because their steam's main business is selling games and if they're partnering with a whole bunch of other hardware manufacturers, they're in it to win it, and I think it's. I bet we don't see a second, second one, it's very possible, like.

19:40 - Jonathan Bennett (Host)
So you kind of? You kind of think through the economics of that and what does? What? Does steam make its money on? Steam? I mean, they've made money, I'm sure, on the Steam Deck, but it's not the money printer, that the store itself is right. The Steam store where they just have to transfer bits and they get 15% or whatever. However much they get on the store, they just get a percentage of the cost of the bits and that's all they have to do, whereas when you make hardware, there's all kinds of liability and investment and, as you said, making hardware is hard, selling software is easy.

20:17 - Jeff Massie (Co-host)
In comparison, Well, I mean, they might not have even made any money on the Steam Deck.

20:21 - Jonathan Bennett (Host)
It could be a break-even thing. It could be.

20:30 - Jeff Massie (Co-host)
That's very possible, because a lot of times they make it up with the software sold through it. Yep, yeah, exactly.

20:33 - Ken McDonald (Co-host)
Playstation and xbox have done that for over a decade now they sell the hardware at a loss, but they make it up in the software it's funny you mention that, because I was noticing that a major company behind a very popular streaming service. It looks like they're partnering with Microsoft to let you stream Xbox games or play Xbox games through their streaming service. Which service is that? Amazon's gaming service?

21:12 - Jonathan Bennett (Host)
Yeah, that makes sense. Amazon has been trying to do that sort of thing. Various companies have been trying to make game streaming like that work for the longest time and I just, personally, I don't see it. The, the value proposition does not make sense to me. Um, because most people I I don't know that I'm ever gonna buy the idea that you can meaningfully play a game you know, a high performance game at internet latencies, because internet latencies are like 20 milliseconds and that's a lot, especially when it's added on top of all the latencies you already have. Um, it just it never did make a whole lot of sense to me that would sure be a lot of help in pushing our uh.

21:57 - Ken McDonald (Co-host)
At least here in the states. Are uh people who are responsible for our infrastructure are improving it?

22:07 - Jonathan Bennett (Host)
I'm not going to touch that one. I have opinions, but I'm not going to touch that one. Instead we are going to go on. Ken is so optimistic. We're going to go on to RescueZilla and that is Ken's story too, because we did things topsy-turvy out of order to start with. But we'll let Ken pick it up and tell us what is new in the newest version of RescueZilla.

22:30 - Ken McDonald (Co-host)
Well, yes, we're talking about RescueZilla, and thank you, marius Nestor, for writing about the latest stable version of RescueZilla, version 2.6. He states this Swiss Army knife of system recovery distribution includes a brand new base derived from Ubuntu 24.10 oracular Aureole to provide users with the best possible hardware support, as well as an updated UEFI Secure Boot shim package to support Windows 11 machines Under the hood. Rescuezilla 2.6 includes the latest part, clone 0.3.33, open source partition, clone and restore tool, and the memtest86 plus version 7.0, open source memory testing tool. It also includes many new and significantly updated translations, alongside other minor improvements and bug fixes.

23:37
Rescuezilla version 2.6 release note states RescueZilla creates backups that are fully compatible with the industry standard CloneZilla tool and works with images created by, in addition to CloneZilla, virtualbox's VDI, vmware's VMDK, qmuse, qcal2, hyper-v's VHDX and the raw image files dd or img, redo, rescue, fox Clones, fog Project, fs Archiver. Now, rescuezilla will only do restore only. It won't let you explore those FSR archive for files as well as apart GTK and redo backup and recovery. Now, as always, I do recommend following the link in our show notes for more information about RescueZilla.

24:41 - Jonathan Bennett (Host)
Yeah, so just kind of thinking about that as you.

24:54 - Ken McDonald (Co-host)
RescueZilla. Is it synced up with CloneZilla, Like, do they do releases at the same time?

25:06 - Jonathan Bennett (Host)
Uh, it seems to be close to the recent CloneZilla update, doesn't it? I was just thinking through what the release cadence of those two projects look like, because CloneZilla is the tool itself, which is also bootable, and then RescueZilla is the Linux distro that has a lot of that built into it. So, yeah, it's an interesting pair of projects the way they work together.

25:24 - Ken McDonald (Co-host)
So you can use CloneZilla for backing up, but if you do have used another thing or you've got virtual devices that you've been backing up, you can use RescueZilla for recovering those if you have to.

25:46 - Jonathan Bennett (Host)
Yeah, yeah, that's interesting, All right, so probably one of the things that you use in RescueZilla to not take or well, in CloneZilla, I suppose, to not take up as much space on the disk is things like XZ and XZUtils and we just had a release of XZUtils and this one is kind of a big deal. I'm not sure if Jeff is going to tell you why it's a big deal, and if he doesn't I will, but we'll let Jeff take it away and tell us about XZU Tills.

26:15 - Jeff Massie (Co-host)
Well, now I'm interested to hear what you think is a big deal, because I'm keeping a little high level, because this can get pretty deep, pretty quick. So XZutils 5.8 has been released, bringing performance improvements to the LZMA and LZMA2 decoder, which sounds like a mouthful right. Let's take a step back and look at the bigger picture Now. Maybe you can argue not quite as critical today as it was years ago. Compression algorithms have been around for a long time, and not just for storing data on block devices, but also for transferring data across phone lines and the internet. Some of our audience might even remember in the days when computers transferred data with the speed and efficiency of two tin cans connected by a string and their storage capability was akin to a small stone tablet. Over the decades, much of the compression we use today has become built in, and there's still a need for high efficiency compression and decompression. So it hasn't gone away. And this is where XZ comes in. Xz is not only a highly efficient compression suite of utilities, but it's also a file format, or a compression format sorry, the XZ file format. To be precise. It improves compression and decompression using the LZMA and LZMA2 algorithms. Now I won't dive too deeply into how these algorithms work, but it's the kind of stuff that lands you a doctorate. But for those unfamiliar with compression, think of it this way Imagine I've got a text file with 100 spaces. To make it smaller and transfer it faster, you could use a special character to represent that there's a compression coming. Then you follow it with a space followed by the number 100. So instead of storing 100 characters, you only need five Three for 100, one for the space and one for your special character. Well, when decompressing the program it'll see that special character and the number and the space and it can expand it back to its original file. While actual compression methods are far more complex, the concept is similar. So the dictionaries and everything. It's basically getting rid of redundancy and patterns, so it can get very deep very quickly, but the concept is solid.

28:33
On the x86 platform, going back to our update, xzutils 5.8 updates the libZMA compression library, boosting decompression speed by up to 5% on systems built with GCC runtime library exception and up to 15% on systems using the MUSL libc standard library for highly compressed files. Additionally, powerpc sees improved compression speeds for both big-Eddy and little-Eddy systems, as well as 64-bit RISC-V processors that support unaligned access. For those wondering, eddiness determines whether the most significant bit in a binary number is stored on the left or on the right. The release also deprecates the LZMA utility script aliases, including LZCMP, lzdif, lzless, lzmore, lzgrep, lzegrep and LZFGREP. It also rewrites the x86-e2k-clmul-crc code for a performance boost. Crc code for a performance boost as a general purpose data compression library and command line tool suite.

29:49
Xe Utils 5.8 includes several other updates, and these include support for CMake 3.20, minor tweaks to ARM64 CRC32 code and improvements to BCJ filters. Be sure to check out the article link to the show notes, along with a link to the official website. There you'll find detailed documentation, links to the official GitHub page and there the official LZMA software development kit which XZ Utilities is built on. So from there you can dive as deep into the subject matter as you like. So happy researching.

30:30 - Jonathan Bennett (Host)
All right. So as to why this is an important release, that is because XZutils uses the even odd rule they have just they've made it standard and so your like 5.7 was their unstable beta releases. So you had 5.6 as a stable release and now 5.8 as a stable release. It is significant because this is the first major release after the backdoor, which I think that is actually fairly significant. Not necessarily anything directly to do with the backdoor in the code, but just I find it a good sign for the health of the project that they've been able to move past it and do another major release after all of that happened, so it's sort of a significant release for XE.

31:20 - Jeff Massie (Co-host)
And in their documentation they actually talk about some of the CVEs and what they've done, and they they do cover some security stuff in their official documentation, thankfully the 5.7 and 5.8 branches have been much less interesting.

31:36 - Jonathan Bennett (Host)
When it comes to all of that, when you're in the security world, interesting is not necessarily a good thing, yeah.

31:44 - Ken McDonald (Co-host)
Unless you're trying to write a book.

31:46 - Jonathan Bennett (Host)
Well, yeah, I mean I like interesting security things, but, yes, I cover this stuff. So I do not want interesting security things in any of my projects, just in my writing about Just in your spy thrillers. Yeah, pretty much yeah.

32:01 - Ken McDonald (Co-host)
Jeff, do you think Michael Larrabelle might do a performance check of it to see if it's any faster than the older one benchmarking I bet I don't think he will, because it's very specific for the most part.

32:17 - Jeff Massie (Co-host)
I mean, it's it's compression, so I don't know as you'd get a whole article out of it, because you compress, you know how you compress and decompress a few different ways and and it's it's kind of a evolutionary release, not revolutionary. So I I don't think he would probably do it, but but I could be wrong.

32:42 - Jonathan Bennett (Host)
Yeah, I would. I would almost expect this that kind of a thing to be. Well, honestly, it'll get covered in something like the uh, the ubuntu 2410 versus 2504 roundup. You might have an xz test in there that is going to use the two different versions of it, and that's sort of where that would, where that would get picked up or I could do a comparison between it and I'm trying to.

33:06 - Jeff Massie (Co-host)
What's the one I'm using went to uh, the uh, I can't think of the name of it well, and you know I say like in the xz utils, but the lzma and lzma2 compression algorithms they actually get used in a lot of different you know. I think seven zip uses it and there's some other other ones as well. So it's. You can't even just say well, it's only xz, it's. It covers a lot of different stuff and for the most part at least on the computer data compression transfer, that L, the LZ algorithms pretty much have kind of been semi-standardized. I mean, there's other like compression programs out there, but a lot of them are based somewhere in the LZA or LZMA tree somewhere.

34:09 - Jonathan Bennett (Host)
Yep, yep, and the differences. So like, when you talk about optimizing things, it's going to be little differences at this point. It's things like oh, we have AVX 512 everywhere. There's this one AVX 512 instruction that we can use to do this, this one part of the decompression, and it's stuff like that that, most of the time, it's not going to make a huge difference. Sometimes it will. Most of the time, though, it's going to be just a little incremental improvements.

34:35 - Jeff Massie (Co-host)
And it's for those servers that it's maybe streaming some data and that's all it does. So that small improvement actually turns out to be pretty big because that server is got you know yeah 256 cores continuously compressing things going out or something, and it but for most of this audience. Unless you're running enterprise, you'll.

35:00 - Jonathan Bennett (Host)
You won't feel it, you just have to know it gets a little better yeah, there's one other place where you might actually notice little little changes like that, and that's things like in the kernel compression itself, because on almost all modern desktops the kernel that is on the drive is actually compressed. A lot of times it's compressed with LZMA, and so if a big win shows up in the kernel itself, then your machine might boot a little bit faster and you may notice that. That's one of the other places that we all sort of interact with these compression libraries a lot. Or in tar yeah, sure, we do a lot of tarring, all right, well, let's talk about something else new that is coming that we all interact with, and that is display slash desktop managers. And this has been quite the story.

35:54
Over the last years the number of years that I've been running Linux I've been through quite a few display managers. I think there was a K display manager once upon a time, and then there's GDM, the GNOME display manager, and then the folks from kde and maybe fedora helped, I don't remember eventually went to sddm, the simple something display manager, and um, it turns out that sddm is maybe a little too simple or maybe just doesn't have the right architecturing. There is now a very work in progress slash, experimental plasma login manager, the PLM, I guess, and the folks at KDE are working on this and, yeah, it's going to be Wayland first. It's going to allow some theming things that the other ones didn't. But one of the big things, one of the things that really intrigues me that they're wanting to do with this is have like BNC and RDP, so remote desktop from startup. I don't know how many people have tried to crack this nut before. I've worked on this problem, but having remote desktop on a remote machine and being able to maintain remote desktop even after, say, a reboot of that desktop or power off and power back on, is really challenging. So being able to have a desktop manager, display manager that has like VNC or RTP support in it from boot is really intriguing. It's also going to have things like automatically log into Wi-Fi, which is important if you're doing something, like you know, user management, ldap or one of those You're going to be able to pair with trusted Bluetooth devices. So that's kind of important if you have a Bluetooth keyboard, screen readers, power management management display and keyboard brightness. Um, even chinese, japanese, korean, vietnamese input uh, apparently sddm does not handle that very well. Virtual keyboard is something again not handled very well, um, and even the high dpi and hdr support right. So, like all of these things are not done very well in SDDM.

38:10
They talk about GDM, the GNOME Display Manager, being the gold standard in this, and one might ask the question well, why doesn't KDE just use GDM? And I think, for the longest time you could, I think it may be that you still can do that, um, but the problem is that it just it doesn't have all of the, all the integrations with the rest of kde that you would want, um. So ideally you want to be able to set your, say, your your screen layout in kde and then have your display manager use the same screen layout, and so that's the sort of thing that hopefully we're going to see with the KDM, the. I don't know, that's not what they're calling it. It's not KDM. Kdm was the old one, but anyway the new KDE-based utility here. So it's not ready for primetime. It apparently actually has enough code together that it works, but this is something to look forward to in a couple more releases, I don't know, maybe around KDE 6.4 or 5. Maybe it'll be usable in Fedora 43, somewhere around there. Really, reading the tea leaves on those.

39:24
I wonder how much it has to do with getting rid of a lot of X than maybe simplify some of the Well, that's actually a really interesting question, because KDE I think we talked about this last week or the week before KDE is intentionally splitting their X and Wayland support into two different projects and they are going to continue supporting X, but it's basically frozen at a standstill because nothing new is happening there, right? And so I don't know if the, the new manager, the, the plasma login manager, I don't know if it's going to have X 11 support or not. I could see I could see that going both ways.

40:06 - Ken McDonald (Co-host)
And isn't it going to end up being tied to uh the k-win the their wayland, uh compositor um, it will, it will reuse some of those, some of those things, I'm sure. Um, that's an interesting question, I think because one of the thing they mentioned is that when they were trying to do it as agnostic as far as the compositor, they couldn't yeah, so it will certainly reuse a bunch of KWIN code, but I don't know if it's going to be directly based on KWIN or not on.

40:46 - Jeff Massie (Co-host)
K-Win or not? I guess where I was kind of going was would it make it easier to just say, okay, we're only going to take care of Weyland and whatever's working on X is working on X and we're just going to leave it at a standstill and that way, you just have one less set of problems you have to work through.

41:03 - Jonathan Bennett (Host)
I mean, that's a reasonable way to look at it. I'm not sure if that's the plan or not. I'm not sure if they know yet if that's the plan or not. There were still early days on this, um, but uh, yeah, it's. Uh, it definitely makes sense to do it that way.

41:22 - Jeff Massie (Co-host)
Yeah, so very interesting something we will see in the future as it makes, uh, makes progress and I don't know know, I'm just happy how far we've come, because I remember some of the old ones too on Unix, like CDE, common Desktop Environment, I think that was a. Sun thing Boy. That was pretty clunky compared to what we have now.

41:43 - Ken McDonald (Co-host)
There's a couple of people that are wanting to try to come up with a Common Desktop. You know that's been a thing.

41:48 - Jonathan Bennett (Host)
A couple of people that are wanting to try to come up with a common desktop. You know that's been a thing that people have wanted for the longest time and every time you try to do it there are problems with it, and one of the big ones is that KDE and GNOME are just different. They do things differently. You know you could say that Wayland itself is the common desktop, like Wayland is the set of protocols that is common to the desktop. So I don't know, maybe you could make a login manager specification and put it into Wayland.

42:22
I don't know if that would help anybody, though you kind of get to that joke about. Well, the problem is there's seven specifications. We should just make a new one so everybody can get on the same page. The problem is that there's eight specifications. We just make a new one.

42:38 - Ken McDonald (Co-host)
Don't. Don't go with making a new one, Just do a proof of concept.

42:42 - Jeff Massie (Co-host)
Yeah, and sometimes that works too, yeah, except the problem is when. Oh, we're going to standardize in this. Well, I don't like it, I've got to fork it. I'm going to make a different one.

42:51 - Ken McDonald (Co-host)
I've got an example of somebody trying to do that.

42:55
With Linux distros right With a Linux distro and using your favorite distros as a starting point. Oh, this is coming from the it's FOSS community. They posted about a new community-led initiative called EUOS. It is led by Robert, and I do apologize for mispronouncing your last name. I want to put the emphasis on the I. He's a physicist and computer scientist by education, currently working in Brussels on data protection policy.

43:34
Now the EUOS is, as I'd mentioned earlier, proof of concept for the deployment of a Fedora-based Linux operating system with a would you believe it, a KDE Plasma desktop environment in a typical, is there anything such as a typical public sector organization? But it is specifically created to address the unique requirements of the European Union's public sector organizations. This is not the first time a Linux-powered operating system has been proposed for the EU. Similar goals have been inspired in earlier projects, like France's Genbuntu for law enforcement use and Linux for government administration. Now, despite the name, EUOS is technically not a new operating system, Since it is going to be based on Fedora. It's just going to be, I would say, a spin. At the moment. That would give you the added value of. Euos is standardized, gives you standardization. It would standardize on a common Linux OS as a base for all EUOS users with options to their own top modifications.

45:10
You could think, maybe a national layer, regional layer, sector specific layers, maybe even organizational specific layers, Similar to the way that Fedora and Ubuntu currently have. Spins like Kubuntu Ubuntu Studio, the two examples we've had tonight. Spins like Kubuntu Ubuntu Studio, the two examples we've had tonight. You'd also, as we were talking earlier, we were talking about the Plasma Lock-in, a common desktop environment to make it easier for someone to move from one hardware system to another without having to relearn desktop behavior. Jonathan, have you moved from KDE to GNOME and then back again? Do you find your muscle memory getting in the way?

46:00 - Jonathan Bennett (Host)
Not for a very long time. It has been a long time since I've actually done anything with GNOME.

46:06 - Ken McDonald (Co-host)
Another area that it would help to standardize would be a common method to manage users and their data with GNOME, Another area that it would help to standardize would be a common method to manage users and their data, software and devices. Now, what do you think drives EUOS?

46:22 - Jonathan Bennett (Host)
A government grant.

46:25 - Ken McDonald (Co-host)
In a way, yeah, it's the concept of public money, public code. I think we've talked about that in the past, haven't we?

46:33 - Jeff Massie (Co-host)
Supposedly, this, though, isn't an official EU-driven project, but to me it doesn't make a lot of sense. Why wouldn't they start with like Sousa, or even Ubuntu? Ubuntu is from the Isle of man.

46:52 - Ken McDonald (Co-host)
Robert's got a website. It's a link to in the show. There's a link to it in the article. I've got in the show notes that some of y'all can go and it's even got where people have come, put in suggestions about that and discussed it.

47:10 - Jeff Massie (Co-host)
But yeah, you're right because like, yeah, cash eos, manjaro there's. I mean, there's a lot of german french distributions there's, and why reinvent it? I mean, if you could just say you know what we're going to take susa or ubuntu or manjaro or whatever you pick one and like, okay, this is going to be our standard and help develop that for the rather than kind of recreate a bunch of stuff. It to me it just doesn't seem like it makes a lot of sense, or else I'm not fully understanding what the goal is basic.

47:53
Uh think of a minimal fedora installation that would meet all the eu compliance requirements and then but if you have a european distribution that already is there, like say SUSE, it's already in the EU, it would need to meet EU compliance. Why not just tweak that a little bit? And then you could even then say, look, we have our own distribution we're using, not based on another country's um distribution. I mean, I understand they'd like to standardize it, to have kind of a common base, but it seems like you could be further ahead by just you know you.

48:39 - Ken McDonald (Co-host)
you create the susa manjaro ubuntu eu edition but be honest, my personal opinion on why robert went with uh fedora. He already knows it there's.

48:54 - Jonathan Bennett (Host)
There's one other thing. I'm doing a little bit of digging into the same question. Um, and they're actually using it sounds like fedora, uh, kind of night, and it's because they immutability it's about the only immutable distro that has a stable release.

49:14 - Jeff Massie (Co-host)
Okay.

49:16 - Ken McDonald (Co-host)
Yep.

49:17 - Jonathan Bennett (Host)
That also checks all their other boxes.

49:21 - Ken McDonald (Co-host)
And I think by going with a stable release that checks all the other boxes, it'll make it easier for them to convince the European Commission to host them at codeeuropaeu.

49:38 - Jonathan Bennett (Host)
Yeah, I'm still not entirely sure what I think about this project. There are some interesting ideas Like take a standardized base and install, or then have like layers that you put on it, you know so one layer for one country, one layer for another country, and then you kind of descend down the organizational tree. You have, you know, different. Here's my Americanism coming out. You have different states inside of that country and you have different layers for them, for their needs, like that's an interesting idea, I suppose interesting idea, I suppose, and those layers then would be the responsibility for managing them would fall at.

50:17 - Ken McDonald (Co-host)
Whoever does that creates that layer, yeah, and then lee you. Uh, the european union itself wouldn't have very minimal coding to worry about, probably be able to just go with a commercial entity or a subcontractor in Europe to provide the maintenance on it.

50:40 - Jeff Massie (Co-host)
Yeah, and there's a lot of Linux that comes out of Europe, that's for sure. A lot of projects Europe's heavily involved in it.

50:49 - Ken McDonald (Co-host)
And I bet Fedora's getting a lot of that European assistance already.

50:54 - Jonathan Bennett (Host)
I'm sure Lots of the developers are from Europe, from all over the place.

51:01 - Ken McDonald (Co-host)
Just want to let everybody know about this. Maybe we should keep an eye on it. But check out the link. Be careful you don't go down the rabbit hole I went down reading about with the EUOS website and then the codeeuropaeu website, though you may find some interesting projects there that you may want to take advantage of.

51:26 - Jonathan Bennett (Host)
You just may, jeff. What about DRM? Do we want to take advantage of? You just may, jeff. What about DRM? Do we want to take advantage of DRM, or is there something disgusting about it where we wouldn't want to touch it? You know, that was such a terrible and perfect tee up at the same time.

51:45 - Ken McDonald (Co-host)
Jeff, just doesn't know what to do with it. Which definition of DRM are we talking about?

51:50 - Jonathan Bennett (Host)
Director rendering manager.

51:52 - Jeff Massie (Co-host)
For HDR, which is Jonathan loves. So what we're kind of fumbling about with is what would an untitled Linux show be without a good Linus Torvalds rant? Now, the rants may be a little more politically correct these days than they used to be, but they're still fun nonetheless. And the rant comes about because of some testing code for an HDR test that Linus is not very happy with. The code is for the Intel XE kernel driver and it's to make sure DRM header files are self-contained and pass kernel doc tests. These are basic maintenance checks to ensure header files are in good shape. This isn't anything really advanced or super complex. It's basically housekeeping code. So I'm just going to read Linus's quote, or what he responded to directly.

52:47
So, grr, I did the poll, resolved the trivial conflicts, but I noticed that this ended up containing the disgusting HDR test crap that A slows down the build because it's done for regular all-mod config build rather than some simple thing that you guys can run as needed, and B also leaves random HDR test turds around the include directories. People already complained separately about this and it should have never made it to me in this broken form. Why in the heck is this testing being done as a regular part of the build and dammit. We don't add random turd files for dependencies and then make the source tree and then make the source tree nasty. The stupid that made me notice it was that it was still. The thing that made me notice that it was still there was that git status complains about the stupid turds not being removed but, more importantly, those turds also break file name completion. So no, adding it to git ignore doesn't actually fix the problem. It would just have made me not have noticed it as quickly. This thing needs to die and that's in. Got some asterisks around that.

54:01
If you want to do that HDR test thing, do it as part of your own checks. Don't make everybody else see that disgusting thing and have those turds in their trees. I'll disable it by marking it broken for now. You guys can figure out what you want to do, but no, forcing us to see those things is not the answer. I would suggest you not make this part of the K config setup and normal build at all, but be somewhere else. You can run it as part of your tests. Ie, do it as a make DRM, hdr test kind of thing, not as part of regular builds. Linus, so I thought that was a fun rant to go over. I don't have a lot to say about it, other than I agree that you shouldn't leave garbage everywhere for a specific build and force everyone, even if they don't have the hardware to deal with it. So any any thoughts from my co-hosts.

54:57 - Jonathan Bennett (Host)
The kernel needs more automated testing, but this sounds like it was not the right way to go about it?

55:05 - Jeff Massie (Co-host)
Yeah, and it literally was. They were forcing every time you build the kernel that it would run this test, whether you needed it or not.

55:13 - Jonathan Bennett (Host)
And running the test was creating extra files that didn't get cleaned up.

55:16 - Jeff Massie (Co-host)
Yeah, yep, and it's one of those where really the only people that need to do that are the original maintainers of those files. Is everything good, yep, okay, then we don't have to do it again. We know it's good, it can be in the tree and it doesn't have to be built again every single time over here and have the hdr test run one more time yeah, yeah, all the fun of uh maintaining a whole bunch of code together.

55:46 - Jonathan Bennett (Host)
It's it's always, uh always challenging and entertaining, all that good stuff and it was.

55:52 - Jeff Massie (Co-host)
It was probably like oh, we got to get it in the window quick, throw it up there and go and yep it.

55:59
It should have probably waited till next revision, but they needed it then and that's how it goes they needed it or wanted it yes, well, I don't know one of the other, but or it could have been. Yeah, we're gonna fix that. And then it kind of got on some there. The sticky note lost its stickiness and fell off the monitor and they forgot to go back and clean it, clean it up, or I don't know.

56:22 - Jonathan Bennett (Host)
Oh, yes, so true, so true. The thedo. This is broken. Don't leave this in production, get blamed. Last edited five years ago yeah, it's now at the bottom of the to-do list.

56:37
Yes, no, it's on the floor behind the desk. Yeah, so that sticky note landed All right. So one last bit of news. This is particularly for us Raspberry Pi fans, and this one surprised me when I read it. It's from Evan Upton, and the official Raspberry Pi PoE Plus injector is now on sale at just $25, which, for what this is, that's actually a pretty good price. I must say, which, for what this is, that's actually a pretty good price. I must say it is not a PoE switch, it is literally just a PoE Plus injector. So if you want to run a PoE Plus device over ethernet, uh, this thing will do it and it will do it off of a regular switch. It lets you inject that power, uh, and it's specifically for the raspberry pi. Uh, there's a um, there's a really interesting little uh quirk here in the timing.

57:37
Um, the raspberry pi 5 doesn't have a PoE plus hat yet. It is missing in action, as they say. We do have one for the Pi 4, which works rather well. You can push a whole bunch of power through it. And of course, the old school PoE, regular PoE hats work for the older Pis as well, but we don't yet have anything official for the Pi 5. And one of the things that he mentions here that I found real fascinating is that the snappily named Raspberry Pi PoE plus hat plus for Raspberry Pi 5, that's the name of this upcoming device, the hat itself it's missing in action. He says this promises to be our smallest, most efficient PD accessory and is in the final stages of development. Having absorbed a lot of Dominic, one of their engineers, dominic's attention and brainpower over the last couple of years. It's been going for a long time Watch this space. So coming soon we've got the new Pi 5 PoE Plus hat. There are some things that I would love to see on the PoE Plus hat for the Pi 5. One of the big ones is maintained access to the 40-pin GVIO headers. I don't know if that's actually ever going to happen, but I would love to see it.

59:02
Anyway, coming soon we now have the official poe injector and you know one of the advantages of that is, uh, you know, cheap. How would it go in this case? It's cheap and cheap or good to pick. Any one is kind of how the poe injectors work, and so you know you can get one on amazon for next to nothing. But beware, it may burn your house down or you can pay a hundred dollars for one that's like you know, got all of the markings and you know you're pretty sure it's not going to catch on fire. I'm I'm reasonably certain that the the unit here from Raspberry Pi is not going to burn your house down and it's only 25 bucks. So they, they kind of, they they kind of hit the sweet spot on that. So, yeah, there, there you go. So you're going to wait for the hat or put the $25 out now.

59:46
I'm probably not going to buy one of these, because I tend to buy network switches to PoE plus support, because I do cameras and I do pies and I do, you know, access points, all kinds of stuff. I love PoE. I think PoE is just the coolest thing. I hate having multiple cords on little IoT devices. It's like just plug it in Ethernet and let it power itself. That way it's so much better.

01:00:11 - Jeff Massie (Co-host)
John has got that 64 switch connection switch probably going.

01:00:16 - Jonathan Bennett (Host)
I've got a pair of 24 port PoE plus switches oh 48 ports. Total of 48,.

01:00:24 - Ken McDonald (Co-host)
Yep port poe plus switches oh 48 ports, total total 48 yep and I've got a 15 year old cat5 cable that I ran throughout my house. I'd probably have to replace it cat5 still works you might not have to.

01:00:35 - Jeff Massie (Co-host)
You would be surprised at the speeds good quality cat5 can do, especially if you're not running it. You know 300 meters or something like that you're. When it's short distances you can really crank up the speed you can.

01:00:49 - Jonathan Bennett (Host)
You can definitely do 2.5 gig over cat 5. I think you can do. You can get away with 5 gig over cat 5 pretty well. Um, I think technically you can't do 10 gig over cat 5 at all, but I'd plug it in and see what happened and the power.

01:01:05 - Ken McDonald (Co-host)
Yeah, yeah, yeah you can, yeah, you can do poe over then maybe it's worth me investing in that injector and then the hat when it comes out yeah, I'm just.

01:01:17 - Jeff Massie (Co-host)
I'm just waiting for when they figure out how to do uh poe over wi-fi well, tesla had a something like that going years ago, but that's uh, probably outside of the scope of the show a little bit.

01:01:31 - Jonathan Bennett (Host)
Somebody. Somebody paged tesla not all right yeah, no, not, not the car company, that's not the one we're talking about right now, talking about the original. Uh. Okay, let's get into some command line tips and uh, ken is going to lead us off and he's got an interesting here. Continues with his pipe wire um series of tips. What, uh? What's the next one in that series, ken?

01:01:57 - Ken McDonald (Co-host)
it's gonna be PW-MIDUMP.

01:02:05 - Jonathan Bennett (Host)
Midi dump.

01:02:06 - Ken McDonald (Co-host)
PW midi dump. It has only three options, and let me go ahead and switch over to my command line here and I've also got my QT pipe wire graph up because I'm going to be showing how you're going to need it to help it. But it has the obligatory dash H or dash dash help, which of course gives you all the options that you can use. There's only three. The other one is the dash dash version, which of course tells you that I am behind with 1.2.4. So I'm definitely looking forward to Ubuntu 25.04 coming out so I can at least move up to Ubuntu 1.2.7. But PWMIDI DUP can be used to and let me get back because I don't want to have to type all that over again For those of y'all listening. The command line says PWMIDI DUP space slash media slash dad slash home data slash public slash audio underscore file slash MIDI slash 12 12daysofchristmasmid. It's a MIDI file, so with MIDI files it will just dump it to the screen. You'll see all the events that are recorded in that MIDI file.

01:03:39 - Jonathan Bennett (Host)
So that's probably a little more useful when you've got a hardware device and you want to connect the hardware device and see what it does over MIDI.

01:03:49 - Ken McDonald (Co-host)
And I'll show you how you do that. To do a hardware device, you just run PWMIDI all by itself. For those of y'all listening, my QTP graph just popped up, a. Let's go over here it's there somewhere. Right here MIDI dump, which is a pipe wire MIDI stream input, and I am going to connect it to that device.

01:04:32 - Jonathan Bennett (Host)
And then go twiddle a knob and see what happens.

01:04:35 - Ken McDonald (Co-host)
Then turn on my Yamaha keyboard, which I probably should have had on before I ran it because it didn't connect.

01:04:48 - Jonathan Bennett (Host)
Yeah, I don't think you're connected to the right device.

01:04:51 - Ken McDonald (Co-host)
So disconnect. And, by the way, I found that if you hold the control key down when you press on one and then the other and then hit control C, that will connect it. And now you're seeing everything running and I'm just going to hit a few random keys here.

01:05:16 - Jonathan Bennett (Host)
Nice.

01:05:21 - Ken McDonald (Co-host)
I can name that tune in two notes. And then you can just exit out of it, after you've tested that, by hitting control C. Nice, simple.

01:05:32 - Jonathan Bennett (Host)
Oh yeah, no, that's super useful, for you know, what I probably will use that for is control surfaces. You know, I've got several control surfaces. I've got knobs and faders to be able to do stuff, and that would be extremely useful to be able to see what each fader is mapped to ahead of time.

01:05:52 - Ken McDonald (Co-host)
And I did a live demonstration without blowing out my audio.

01:05:55 - Jonathan Bennett (Host)
Yeah, it didn't die.

01:05:59 - Jeff Massie (Co-host)
Show isn't over yet.

01:06:01 - Jonathan Bennett (Host)
Yeah well, all right, jeff. What do you have for us? Show isn't over yet. Yeah well, all right.

01:06:05 - Jeff Massie (Co-host)
Jeff, what do you have for us? Ddgr, which is a command line utility designed for searching DuckDuckGo directly from the terminal. While the tool Googler has become incredibly popular among command line users, there was a clear demand in online forums for a similar utility tailored to the privacy-conscious DuckDuckGo search engine. That's where DDGR comes in. Now, it's important to mention that DDGR isn't affiliated with DuckDuckGo in any way, so this is another group decided to do this independent of DuckDuckGo. One of DDGR's highlights is its ability to let you specify the number of search results displayed per page, and this is far more practical than scrolling through dozens of results at once, as you would with a web interface. Plus, its default layout is designed to save space while maintaining excellent readability. Another standout feature is DDGR works seamlessly with the Tor network, so a major advantage for those prioritizing privacy.

01:07:09
For examples of how you can use DDGR, be sure to check out the article linked in the show notes. It dives into features like shell completion, customized searches for Wikipedia or any specific website, file type filtering, quoted text searches, custom color schemes, https, proxy tunneling and much more. Now just keep in mind that DDGR requires Python 3.8 or newer To install it. You simply just use your usual package manager. For example, I was able to install it effortlessly using the command sudo apt install ddgr. Now I won't cover all its features because there's so many, but I recommend giving ddgr a try, or even looking at the link in the show notes. It's compact, it's fast, it's user-friendly tool and I used it some and played with it and it's pretty cool. And since it's fast, it's user-friendly tool and I used it some and played with it and it's pretty cool. And since it's a command line utility, it comes with a handy man page for a guide, so you'll have everything you need to know. So happy command line browsing Nice.

01:08:15 - Jonathan Bennett (Host)
So the ultimate next step of that, by the way, is one tool that you can just say I want you to connect to Google, or I want you to connect to DuckDuckGo, or you know whatever other search engine that you can just say I want you to connect to Google or I want you to connect to DuckDuckGo, or whatever other search engine that you prefer. True, that is the. Let's see what will. We call it Y-A-G-R yet another Google replacement, something like that. All right, I have a very, very quick command line tip for you, and it's just cd space dot. And let's talk about what that does. First off, it changes directory to the current directory, and those of you in the kind of middle amount of Linux command line knowledge are going to scratch your head and think why would I want to change directory into the current directory? And I will tell you if that directory was deleted and recreated out from underneath you. That is why you will want to do it.

01:09:17
I've come across this when test running Firefox nightly on my Fedora machine behind me. Test running Firefox nightly on my Fedora machine behind me. When you first start Firefox from the command line, from within the folder, it will go out and check for updates because it's nightly. It basically has an update every day. The way that it installs that update, apparently, is it completely erases and then re-downloads that folder, and so when you run something like slash Firefox, it installs that and then you either close it or, because it's nightly, let's be honest, what's more likely to happen is you're going to crash the browser and then you go to run slash Firefox again and nothing's there. You run LS, nothing's there. It's like you're nowhere on your file system because the folder got deleted out from under you. So cdspace brings you right back there, it refreshes the list of files and folders and then the world is once again right and you can fire Firefox back up to crash it yet again on whatever thing you were trying to test. So there you go. Now you know how to handle that.

01:10:27 - Ken McDonald (Co-host)
And you've rescued somebody another day.

01:10:28 - Jonathan Bennett (Host)
Yes, yes, myself in this case. Uh, you could also use cd space, dot slash, which was how I was doing it. And then I realized actually found on the internet somebody else in the same problem. They're like, yeah, cd dot, it's like, think about it. Yeah, that probably would work, wouldn't it?

01:10:46 - Ken McDonald (Co-host)
that's the same thing, all right now one quick question what does cd space dot dot do?

01:10:54 - Jonathan Bennett (Host)
that goes a directory up. No, that's the way it works. All right, I'm gonna let each of the guys plug whatever they want to get in the last word on something that they want to. We'll start with jeff. See if he has some poetry for us uh, I do and we'll, we'll see.

01:11:12 - Jeff Massie (Co-host)
If I have. Uh, maybe the next one of the week or two I might have uh a story about trying the kubuntu beta. I might just load it on and just go crazy and say I'm just going to run the beta, we'll see how it goes, but nothing else to go. So a haiku this week? Oh, blue screen of death. Fatal error has occurred. I move palm to face. Have a great week, everybody.

01:11:41 - Jonathan Bennett (Host)
I move palm to face that's great week everybody. I moved palm to face. That's great. All right, Ken.

01:11:48 - Ken McDonald (Co-host)
Well, before you try that Kubuntu beta, may I recommend that you back up. You could either use Clonezilla my favorite or Rescuezilla, but definitely back up before you do try any of these betas we talked about tonight. Yeah, you might want to back up before you do try any of these betas we talked about tonight.

01:12:04 - Jonathan Bennett (Host)
Yeah, you might want to back up before you try Clonezilla too. Yeah and I say that only half jokingly if you have anything really, really important on your hard drive, throw it on a flash drive first, because it is quite possible to trash the machine with Clonezilla. Ask me how I know that.

01:12:23 - Ken McDonald (Co-host)
And what's your favorite TAR command for backing up with?

01:12:29 - Jonathan Bennett (Host)
It's going to be something like tar space dash c-a-v-f the name of the file and then the files that you want inside of it. I think that works. I think that's right.

01:12:43 - Jeff Massie (Co-host)
Yeah.

01:12:44 - Jonathan Bennett (Host)
I just use rsync.

01:12:46 - Jeff Massie (Co-host)
Yeah, that works too.

01:12:47 - Ken McDonald (Co-host)
Go through our command line tip spreadsheet and find where we've demonstrated TAR and determine which one of the ones we've shown would work best for you.

01:12:57 - Jonathan Bennett (Host)
We've probably talked about TAR several times over the years. I had to guess.

01:13:01 - Jeff Massie (Co-host)
Oh yeah, you could write a book on TAR. Somebody probably has.

01:13:05 - Jonathan Bennett (Host)
Oh yeah, you could write a book on tar, and then somebody probably has. Oh, I'm sure I bet there is a Riley animal book just about tar. All right, it is time guys, we're going to wrap it up. If you want to find more of me, there is, of course, hackaday. You can find my weekly security column goes there basically every Friday morning. That's also the new home of floss weekly. We have a lot of fun there. You should go check out our latest episode talking about Fedora 42, the beta and the KDE stuff going on there. I talked with Neil Gampa. It's been a couple of years since we had him on so we brought him back. We've got fun stuff going on there as well.

01:13:39
We appreciate everybody that watches us and those of us that are. Um, we appreciate everybody that watches us and those of us that are are part of club twit. We definitely appreciate it, and if you're not, why not? You need to think about it. It's about the price of a cup of coffee per month. Definitely worth it in a way to support the network and the shows that you love. You should check out club twit, appreciate everybody being here or both live and on the download, and we will see you next week on the untitled linux show.