Rob (2:55)

I found my stories. It just didn't have the drama that I like to bring.

Jonathan (3:02)

I see no downsides here. This is a good thing.

Rob (3:04)

I found a little drama, at least flashbacks of drama.

Ken (3:08)

I found some exciting drama.

Jonathan (3:11)

I like boring stuff. I would prefer to talk about the boring stuff. Things are good. We're fixing bugs, we're introducing new features. Boring.

Rob (3:19)

Yeah.

Jonathan (3:19)

No, you guys got to bring the drama.

Rob (3:21)

You don't need a show for that, I suppose.

Jonathan (3:23)

Well, let's see if we can find some drama in the X Elite, the Snapdragon chips. And Rob, what's going on there?

Rob (3:30)

There's probably, I mean there's probably drama there, but really I'm just here to bring the excitement because I mean it's, it's. It was around this time last year when we really started talking about and getting excited for the Snapdragon. Snapdragon X Elite laptops. They were at that time were coming out soon. They came out in June. Yeah, we still don't have an X Elite laptop powered by Linux out of the box. Tuxedo computers announced last year. They were working on one, but that hasn't materialized yet. So, you know, we're stuck making do with the Windows 11X Elite laptops, making them work for our needs. I haven't had one yet, but for those who do, they're making them work for their needs because I haven't got enough copies donated. But anyway, you know, unfortunately for that, we have had Canonical with Ubuntu. They've been working hard, they've been working on the issue. I don't know how they've been working, but from the distribution side, providing Ubuntu 24.10 developer preview images for the X Elite chips. And so Michael over at Pharonix, he decided finally, you know, it's been almost a year, finally let's purchase the, The Acer Swift 14 and, and Benchmark to see what the current state of things are. He goes through. There's some, some issues here and there. You know, apparently some things aren't fully functional yet. Like, like the power management handling evidently isn't properly wired up yet. And under load the laptop would get incredibly hot to the touch, to the point where the system would turn itself off during the benchmark. Fortunately most of us aren't taxing our systems like that. We're not running benchmarks, we're just using them. But you know, overall he noted his experience over the past week was rather disappointing but he still ran a bunch of tests on it and you know, as expected, it doesn't, it doesn't compete with the most modern processors on most workloads. But I thought it came out pretty well when comparing with, with some of the processors from just a few years ago. And to me that's, that's kind of a win for arm at least a good step forward. So, so the benchmarks were against AMD and Intel obviously and they're ranging back to like the, the 8, the 8th 10th gen Intel AMD Ryzen 7 4700, stuff like that. In most of the benchmarks for the X Elite was, you know, the X Elite was down near the bottom, in the bottom half but typically above many of the older processors on the list. Like, like the 8th and 10th gen intel processors, it was usually above those. And the X Elite actually came out on top on some of the, some of them like dark table 1 and a 7 zip benchmark actually came out on top on some of them. So you know, there's some good things there. But you know, Overall the Snapdragon XL8 within the Acer Swift 14 came out just ahead of the Core i7 1185 G7 Tiger Lake laptop and similar to the AMD Ryzen 7 Pro 5850 Ultra. But obviously you know, the article did not point this out but I think it's obvious. I mean nobody expected it to be the powerhouse like up there on top. I mean it's not Apple silicon but, but obviously as to be expected it is still well behind the very newest amd Strix Point Zen 5 and Intel Lunar Lake generations. But you know, the overall out of the 23 systems the two XL8 devices he tested came out at number 16 and 17. So quite a few of them down there, you know, kind of 3/4 of the way through. But this is ARM, you know, and to me on ARM chip that doesn't come out last against AMD and Intel. I think it's a good sign for the future. I think I would have loved to have seen, you know, I kind of hinted there about it's not Apple. I would have loved to see a comparison against another kind of ARM chip. I would love to see Comparison with the M1 M2 Max with Asahi Lennox on there. I mean sure you'd Be comparing different distributions too. But I think that's fair for this because Ubuntu is what you have for the X Elites, Asahi is what you have for the M1, M2 max. So, you know, I think it's a fair comparison. No different than comparing Windows against Mac os, you know, on comparable hardware. So I think, I think that would have been, you know, even if it's not head to head hardware, I think it would have been cool to see where that fit in there.

Jonathan (8:38)

Yeah, definitely, definitely would have been interesting. Did he do a like performance per watt? Sort of, sort of benchmark in there?

Rob (8:47)

I think that was one of them. He could have because of the power stuff.

Jonathan (8:51)

Makes sense.

Rob (8:51)

Yeah, he did like web browser, like Firefox. It, it Tested pretty well there. Seven zip. There's like 15 pages on there. So there's a lot of things he went through. I, I couldn't even summarize well, besides giving his overall score at the end, there's no way I could even summarize going through all them. I tried to pick out what, what some of the highlights are at the.

Ken (9:17)

Top, but based on this, I would probably pick the framework 16 with the Ryzen 7 in it.

Rob (9:26)

That's what he recommended, I think. Sounds like what he said there. But you know, that's if you're looking for pure power. I, I don't. I guess I like you. Like you said, I don't think he had the. Or like I said, I don't think he had the power per performance comparison there.

Ken (9:44)

I'm looking at power per dollar.

Jonathan (9:48)

Yeah, it's a slightly different metric, but one that is definitely important. I kind of thought I heard that these Windows on arm, the Snapdragon laptops were not actually selling very well. And I went looking for a source on that and of course I'm having trouble finding it other than apparently, Apparently Amazon applies the frequently returned item warning to some of the ARM based Surface laptops. So, you know, make of that what you will.

Rob (10:18)

That's too bad. I hope they're at least doing better than Microsoft's last attempt at arm. I can't remember what that was, but I know there was one several years ago.

Ken (10:30)

I've got two questions. First, would you use this in a server environment? And second, how would you update it?

Rob (10:41)

No, not yet. Not, not these.

Jonathan (10:45)

You could, you could use, you can use ARM in a server environment for sure. Maybe even this particular chip, but not in the laptop form factor. It really doesn't make sense.

Rob (10:55)

Yeah. How would you update it? I think he did Update it to 2504 I thought he said in there and I think it didn't perform quite as well if I remember right.

Ken (11:08)

Was he able to update the firmware on it?

Rob (11:11)

Oh, the firmware, yeah. I don't know if you know tools like FWUPD are are on ARM yet or not, but can you do that on the Raspberry PI? I don't know. You tell me.

Ken (11:25)

Yes they are, you tell us. But about actually got two articles linked about fupd. The first link is to an article that Michael Larabel wrote about lead Linux vendor Firmware service and FUPD developer Richard Hughes where he talks at the Prem Day On Premises Computing Conference in Paris. Richard's talk was targeted at the Linux server space where Linux vendor Firmware service and Firmware FUPD Firmware Update utility have seen a slower rollout. Now, according to Michael, we are seeing some adoption by different server ODMs and supporting Redfish and other plugins relevant for firmware updating in the Linux Data Center. If you want to view Richard's talk you can find it embedded in Michael's article. It was a great approximately 30 minute talk. I'd have to go back and listen to it again if you want to hear every single detail, but I don't think we got time for that. So I'm just going to move on to the second link which is Tamarius Nestor's article on the latest FUPD release, which is version 2.0.9 or if you're in the UK z9. It contains support for more devices, new features, and bug fixes. This release introduces support For Intel Arc BattleMage GPUs, the ability to allow installing multiple database certificate updates at the same time, support for showing what certificate signed the EFI authenticated variable, new documentation about updating the KEK and database, as well as the ability to use readline to look up inputs from user and make it optional. This release also fixes a crash that occurred when installing some Wacom firmware types, a crash that occurred when parsing U events that aren't key equal value, and a parsing issue with the DFU descriptor when not using the LIBUSB library. Starting with this release, FUPD is no longer enumerating non updatable option RAM devices, it no longer exports Redfish backup partitions as devices. I don't think I'd want to be updating those anyways. Includes a resolution for more the HSI failures and no longer allows updating updatable hitting devices with the FUPD tool command Marius article does provide more details and a link to the project's GitHub page if you want to get every single detail.

Jonathan (14:27)

So you asked about FWHUPD and the Raspberry PI, and I did actually just find an article from late last year. FWHUPD version 1.9.22 brought unofficial support for updating the firmware, but in the Raspberry PI 5. So, yes, you can actually do that, which is fairly useful because you don't want to have to. So if you're not gonna run the official Raspberry PI OS firmware or, excuse me, the official Raspberry PI OS distro, then you're going to want to be able to use the normal tools like flopd to be able to update it. So that's a good thing. It's a good thing to be able to run sort of your more vanilla stock Linux distros on the PI. I approve. All right. Speaking of not exactly the Raspberry PI os, but the thing that it's based on, I've got some Debian news and a couple of things. One of the big ones is that they just, they just elected their, their, their new leader. The new leader for Debian. What do they call him? Yep, Debian project leader Andreas Tilly is who was elected. And I've got a link here to his platform. You can read through the things that he said he wanted to do. And then he is now doing some of those things. And there's a couple of them that are very interesting to me. One is that he is joining in with the end of 10 campaign. And that is essentially the campaign that says, hey, Windows 10 is going end of life, by the way, Windows 10 is going end of Life October 14th of this year. That's going to be here before you know it. And there's a campaign where people are saying, why don't we switch to Debian or Linux in general instead of, you know, going to Windows 11? I approve of this message, by the way. So he is kind of thrown his weight behind that. And then one of the other really interesting things here is that they're sort of not revamping, but making some little changes to the Debian packaging system. Particularly what happens when a package goes dormant or it's orphaned. You could say essentially what happens when the maintainer for a given package disappears steps away from the project. What has you. And one of the things that they are looking at doing is adding an intent to NMU or intent to orphan. And it's essentially a 21 day notice period where they say the maintainer for a given package has kind of gone missing in action. MIA and we are going to begin the process of turning this into an orphaned package, which from what I understand, that allows some things like non maintainers to suggest updates to it basically just it changes the way that the package is handled. And so this is interesting that they're having to think about this. It's kind of a sign that they do have some Debian maintainers that are stepping away, getting burnt out or retiring, what have you. All kinds of different things. But yeah, I'm glad to see that some of that they are sort of looking into that. It's a problem that a lot of distros and to some extent even the kernel is starting to face. Like as your maintainers get older and some of them start retiring, do you have people coming in to replace them? Sometimes the answer is yes, and sometimes it's no. But in a lot of these places, it needs to be, it needs to be thought about and dealt with potentially. So interesting things, some changes at Debian, nothing huge, but, you know, a new leader and some of the things that he wants to work on.

Ryan Seacrest (18:12)

Hey, it's Ryan Seacrest for Albertsons and Safeway this spring. Take care of your entire home, including the air you breathe, and save $5 when you buy $25 worth of participating products in store or online. Shop for items like Glade Plug Ins, Airwick Plug Ins, Glade Auto Sprays, Airwick Diffusers, and Glade refills. And save $5 when you spend $25 on participating products. Offer ends May 20th. Restrictions apply. Promotions may vary. Visit albertsons or safeway.com for more details.

Ken (18:41)

And it sounds like he'll make a good leader.

Jonathan (18:45)

Hope so.

Ken (18:46)

I'm wondering if he was the one behind looking at orphaning some of these abandoned packages. In some cases.

Jonathan (18:57)

Yes. Basically what they're doing is they're, they're trying to figure out a way to make, they're trying to put a system in place to formally recognize that these packages are orphaned. I think there's a lot of packages that are kind of de facto abandoned, but they need a process. And it's like we're going to put, we're going to put a statement out and say, in 21 days this is going to officially be considered an orphan package. And that gives, you know, the previous maintainer 21 days to say, no, no, no, I'm still here. I've just, you know, life happened or whatever. But if that 21 days goes by then it can officially get marked as an orphan pack package.

Rob (19:34)

And we're actually talking about a technical leader, not just someone at the top who has been in some nonprofit organization and doesn't even know what they're leading.

Jonathan (19:47)

Well, so interesting that you should mention that. Somebody has that story, right? Yeah, Rob, you've got that story. We'll talk about that story later. Yeah, interesting things going on there before then. You want to talk about servo?

Rob (20:05)

No, I would like to skip to this one actually. Okay, well, you missed my message in the back channel.

Jonathan (20:10)

I did, I did. I'd not gone to look at those. I've got, I've got the little red circle with the two in it in my discord notifications and I figured as soon as I got done I would go check those.

Rob (20:20)

Oh, too late, too late. But you know when we have discussed odd choices GNOME has made for executive directors at the GNOME foundation, you know, the previous one came in with a lot of non profit experience and degrees in English and education. Oh, and she was a professional shaman. Yes, and I did put a link in the show notes her website. Apparently after, after she became the executive director, the website went blank, according to article I saw when I was kind of digging through stuff. But the website is up now, so if you want her shaman services, it's Holly something or other, they are available again it appears. But you know, you know, and maybe, you know, with this kind of experience, maybe with their financial turmoil they just needed someone to help with their funds and well, she lasted a little more than half a year after that there was an interim and they're Richard whatever his name is, I don't remember. But, but you know, some of us here, you know, we've often stated the importance of having technical people at the top of technology companies and yeah, okay, it's, it's not always necessary. You know, maybe they have a different focus like fundraising or whatever. But this new announcement may be just what GNOME needs after a lot of the turmoil they've had. So Stephen Debauld has been named the new executive director of the GNOME Foundation. And unlike the executive directors of the past, he is Canadian and a techie, a free software advocate and a GNOME user. He actually used the stuff since 2002. And I don't know if that last one's good enough, honestly, because I first tried gnome back around 1999, so I probably should have been applying, but yup.

Jonathan (22:34)

He totally should have.

Rob (22:35)

Then again, I mean his background sounds a lot like mine, except for up a few steps, but Anyway, back then I was more of a KDE fan. I wasn't so much a GNOME fan. So I tried it back then. I didn't stick with it like supposedly he did, but some fun history around this guy. He's got, he's got some. Apparently has had some technical chops, at least according to what he put out in his blog. He built a graphical mud. MUD before the term nmorpg was coined. If you don't know what graphical mud is, look it up. It's. It's a. It's a type of game basically. MM or PG multi. I can't remember what that stands for. Look it up. Someone else could say it afterwards. So, so anyway, and then he started his first web development company when he was 15. I was like 20. He did it, it was in 1996, so also a few years before me. So he beat me there. He started a couple more businesses while he was in university. Around the time that he started using Linux and Gnome. He was a coder for a good 10 years until he had a bicycle accident or bicycle crash as he said it states it in his blog, which made it a little difficult for him to code after that because of, I know the screen colors and stuff I think it said.

Jonathan (24:04)

But it sounded like. It sounded like his vision. Sounded like it was a problem with his vision.

Rob (24:08)

Yeah. So, you know, if only he had AI back then as an assistive technology. But anyway, after the crash. So it's coding you can't quite handle anymore. He had a decade of what he says he's recruiting, sales, management startups, fundraising and product consulting, which I don't know, it sounds like a lot of good things when you put those two together. Got some technical stuff. He's got the, you know, all the other things that you might want in a CEO or executive director, sorry, sameish thing. So after all that then he began applying at the GNOME foundation and here we are. But it's always good to see an actual user of the product run a company. You know, it always seems a little off to me when, when the person at the top doesn't even know anything about their actual product. You know, like when, when a CEO at a gym clearly doesn't use the gym. I think it would be a little more inspiring to see them out, out there using the equipment alongside us once in a while and you know, to get a better feel for that. What kind of change is really needed out there? You know, kind of goes for, you know, good home too. All the things like that. You should, should have Some a good. A good knowledge of what you're leading. So this guy might be a really good fit. Let's hope he stays around longer than seven months.

Jonathan (25:41)

Yeah, there are a couple of things that he talks about in his sort of intro letter that I really appreciate. I think are good things. He talks about transparency, which is always good to see in an open source project. And then he also talks about reestablishing the GNOME project's foundations. And that's something we've talked about before as well.

Rob (26:04)

I hope that doesn't mean making it look like Windows again like it used to. That's why I did not like about it back in 2000.

Jonathan (26:10)

I don't think so. So as he says, the foundations are that. The GNOME foundation, pun intended here, it exists to support gnome, to support design and development, to support contributors, like that's it. So, you know, hopefully they can get it. We talked about this with Mozilla too.

Rob (26:27)

Yeah, I was going to say, hopefully they take this stance in the future too.

Jonathan (26:30)

Yeah, yeah. Spend your money on developers, write code, ship good projects, ask for donations. Rinse and repeat. That's your one job. You have one job.

Rob (26:43)

Create a vpn. Oh no, that's good. Not to do.

Ken (26:46)

Resilience.

Jonathan (26:49)

Resilience, yes. Resilience, yes.

Ken (26:54)

Kind of mispronounced something one time just to get a laugh out of Rob.

Jonathan (26:59)

That's what that was. Oh my goodness. All right, so let's talk about the Raspberry PI then. The Raspberry PI Connect.

Ken (27:10)

Now, there is a lot of interest in what Raspberry PI PI did this week. Got several links in the show notes. The first one is by Chris Louder. He's writing actually for Raspberry PI about the latest Raspberry PI Connect information. Chris reports that as of Raspberry PI Connect version 2.5, they are dropping the beta prior to version 2.5. The Connect client software has had devices regularly waking up to make HTTP requests and higher data usage than needed. And Starting with version 2.5, the Connect client now holds a single long lived HTTP connection to a Raspberry PI server. Also starting in 2.5, each individual heartbeat is now compressed before it is sent to the server, making it about 50% smaller compared to previous versions. Chris also provides instructions in his article for updating to the latest version of Raspberry PI Connect only while recommending updating to the new version of Raspberry PI OS since it contains the latest version of Connect. Speaking of the new version of the Raspberry PI os, that's what the remaining links in the show notes are about. They're to articles by Simon Long, Bobby Borisov, Richard Speed, Marius Nester, and our fan favorite Michael Arabelle, all writing about the new version of Raspberry PI os. They all say this is probably the final release of Raspberry PI os. That's sad to hear, isn't it? Which is based on Debian Bookworm, since Debian Trixie should be released this summer. There's your drama, Rob.

Jonathan (29:12)

Drama.

Ken (29:14)

Some of the new Raspberry PI OS improvements include using a modified version of Swaylock for its screen locking, separation of console and desktop auto auto login options under the Raspberry PI Configuration app, a new printers application for managing printers on Raspberry PI os, and improved touchscreen handling under Wayland. Since I have just hit the highlights, I do recommend reading at least one of the articles for more details. My recommendation would be Marius yeah, interesting stuff.

Jonathan (29:56)

It's funny you talked about the whole maybe it was dumb for Firefox to introduce a VPN when they should have been working on Firefox and other things we could observe about GNOME with Raspberry PI. The Raspberry PI Connect one could make the argument that this falls into that same category. But man, everybody that's talked about it that I've seen that actually has tried it and used it, has had glowing reviews. And you kind of think about it and it's like, well, having remote access into a little Linux box is sort of a useful thing. So maybe it just squeaks by into that category of sort of makes sense and wasn't a dumb thing to do.

Rob (30:41)

Yeah. I mean, not that we're talking about it today, but you could. Somebody could say the same thing about POPOs with System 76. Focus on your hardware and not make this.

Jonathan (30:52)

But that seems to have turned out okay for them.

Rob (30:55)

That worked out. I think it's worked out well and I'm.

Ken (30:58)

It's actually worked out well for Fedora as well.

Jonathan (31:03)

Well, yeah, because Fedora now ships the as spin with the new Cosmic Desktop. Yeah, well, I mean you could. You could even make that same argument with System 76 about Cosmic Desktop. You should have just stuck to your distro. You shouldn't try to spin up a desktop too. Jury's still out on how well that one's going to work, but so far so good.

Rob (31:23)

We will see some people are good at multitasking and others they just got to focus.

Jonathan (31:31)

I guess that's true of some company like that. That is true of companies as well.

Rob (31:35)

I mean, imagine where Amazon would be if they just focused on selling books, huh?

Jonathan (31:40)

Yeah, true that. The world might be all my money by now.

Rob (31:45)

The world might be a better place, but Amazon would probably be gone.

Jonathan (31:48)

Yes, exactly. Exactly.

Rob (31:51)

Does it ever feel like you're a marketing professional just speaking into the boy? Well, with LinkedIn ads, you can know you're reaching the right decision makers. You can even target buyers by job title, industry, company seniority skills. Wait, did I say job title yet? Get started today and see how you can avoid the void and reach the right buyers with LinkedIn ads. We'll even give you a $100 credit on your next campaign. Get started at LinkedIn.com results terms and conditions apply.

Ryan Seacrest (32:21)

Hey, it's Ryan Seacrest for Albertsons and Safeway this spring. Take care of your entire home, including the air you breathe, and save $5 when you buy $25 worth of participating products in store or online. Shop for items like Glade Plugins, Airwick Plugins, Glade Auto Sprays, Airwick Diffusers, and Glade refills and save $5 when you spend $25 on participating products. Offer ends May 20. Restrictions apply. Promotions may vary. Visit albertsons or safeway.com for more details.

Jonathan (32:50)

All right, so let's talk about some other distro that's doing some other crazy things and that is Ubuntu with 2510. And the Internet has kind of been abuzz because Ubuntu has announced that they are going to use a rust based sudo alternative Sudo RS by default along with some other core utils in Ubuntu 25.10. And that's super interesting. I have seen people decry this because Sudo RS is untested, which may be fair. I kind of have to think though that the only way that you really get it to be a tested tool is by making it default in some distro. So they kind of have a chicken and egg problem here and they opted to go this direction for it. Yeah, I don't see this as being a huge problem, particularly because 2510 is not an LTS. Right. So it is kind of the Ubuntu release where they have permission to break stuff. So yeah, I'm for it. I think it's going to be cool. I think it's a neat idea to be a little daring and see what happens if you want to be a little daring. They actually already have ISOs where they are building 2510 now. It is very early in the process. So these are going to be potentially very rough. Don't run it as your main desktop install, but if you have extra hardware kicking around or you want to put it on a VM, you can go play with 2510 now. It is due to actually release on October 9th and of course it's got all kinds of fun stuff in it like the Rust core utils, but also it's going to have, you know, a brand brand new kernel, probably the 6.17 kernel, GNOME 49 Mesa 25.1. It's going to be, it's going to be a pretty neat release. 25:10 will be, will be cool and you can go test drive it now.

Rob (34:51)

I don't know if we've talked about it or if it was on another show Floss Weekly but, or something like that. But I know somebody had pointed out like, I feel like this was a discussion Floss Weekly and somebody said that it doesn't really make sense for Ubuntu to switch to the Rust core utils but that like sudo rs, the Rust based sudo does make a lot more sense.

Jonathan (35:19)

We may have done that. You know we had the guys from the Russ core utils on the show back a few months ago.

Rob (35:24)

It wasn't them, it was somebody after that that that had their opinion on Russ. It was, it was kind of a less positive opinion or along the facts of they're not going to Was it Lond.

Jonathan (35:41)

No, no. London is against this. He's against it all together. It's a terrible idea.

Ken (35:46)

That's why from what Rob was saying, I was wondering if he was talking about listening to London.

Jonathan (35:51)

Surely, surely Rob doesn't listen to or read Lunduke.

Rob (35:55)

This was the, I think it was the Alex C. The Incas guy.

Jonathan (36:00)

Oh, okay. Yeah, that makes sense. That yeah, yeah, we, I, I have a tendency on that show to tease everybody about whether they're porting their software to Rust or not. So that brings up a lot of interesting conversations.

Ken (36:14)

Rust core utils, that project's been around for a while and actually gotten quite far because I remember covering using it as a command line tip. I'd have to look back to see what episode a while ago.

Jonathan (36:27)

Yeah, so there's like three different core utils projects where people are either the original implementation or people are reimplementing them in different languages and they actually they work together quite a bit. They do things like share a test suite to try to keep them all bug for bug compatible as it were. Except for you know, the few cases where like they intentionally do things Differently, There are just a few of those.

Rob (36:54)

I think what the Angus guy said was that like core utils, you know, it's been around a long time, it's very stable, very tested. And then when you re implement something in Rust that even though sure, you're, I mean you're already something that's supposedly very secure, you're reimplementing rust, you're getting rid of the memory safe stuff, but you're potentially introducing a lot of other bugs along the way that you have to figure out.

Jonathan (37:21)

Right, right. And so like that's the thing to keep in mind. The, the Rust language, it gives you guarantees about memory safety, it does not give you guarantees about any other type of safety. Right. And logic, program logic being the big one. Yes. And memory safety is not necessarily even the most exploited type of vulnerability. Right. And then the other side of this is, on a Linux system, when you start thinking about like the security boundaries, where are those going to be? Well, generally your core utils are not security boundaries unless they're running with like a set uid, whereas sudo is definitely a security boundary. And so having the memory safety there does sort of make sense. But then of course you have that same conundrum of it is newer code and there are other bugs that are not memory safety bugs.

Rob (38:17)

Yep.

Ken (38:18)

But that may have been introduced.

Jonathan (38:21)

Right, right.

Rob (38:21)

With coreutils you're also working on a whole bunch of different ones, so maybe less focus. Whereas like the rust based. The rust based sudo may be more focused where you can, you know, focus on clean up those bugs quicker.

Jonathan (38:37)

Yeah.

Ken (38:38)

The one thing you're going to have to watch with 2510 with rust based core utils is some of your bash scripts breaking really well, stop and think about it. If, depending on how they implement it, if it doesn't have a SIM link to the old name and. Or it doesn't use some of the flags that the old one used.

Rob (39:07)

I'm sure it will.

Jonathan (39:08)

So they're going to, I'm quite confident that they're going to symlink them with all the old names and the Rust core utils are to the point now where they support basically all of the flags. And like there are a couple of cases where they don't do exactly the same thing, but those are things like.

Ken (39:30)

Edge cases.

Jonathan (39:31)

No, it's not even edge cases, it's intentional decisions. Like one of the questions was in the CP command how it doesn't give you any feedback. And so the Rust guys were like, we could put a progress meter in There in the Rust version, it's intentional decisions like that. And I can't remember which, which, which way that particular one went, but is.

Ken (39:54)

That a flag or by default? Do you remember?

Jonathan (39:57)

I don't. I think it. I think it would have been by default, but, you know, it's. It's things like that, that the intentional decisions where they can look at the behavior of the old tools and try to find places where it's very unlikely to break things. But we can also make the user experience. Experience a lot better for people. So other than that, though, they have a huge test suite, like thousands and thousands of individual, individual tests where they go in and they say, let's make sure that the Rust core utils have exactly the same behavior as the old school query tools do. I don't have much fear about using Rust core Utils in Ubuntu 2510. I think that's going to be a nothing burger. Nobody's going to notice it for the vast majority of.

Rob (40:41)

Yeah, I don't think you have anything to worry about, Jonathan, in Ubuntu 25.10. I mean, mostly because you probably won't be running it, but yeah, I have.

Jonathan (40:51)

No worries about it. I will sleep well at night with Ubuntu 25.10. Not on my machines.

Rob (40:56)

And honestly, I mean, I use Ubuntu on things like on my servers, but I use lts, so I won't probably be using it.

Ken (41:03)

You're not going to download the Daily build and run it in a vm?

Jonathan (41:07)

I might throw it on a VM just to say that I have, but I. It's just not. It's not going to be all that exciting.

Rob (41:15)

That's a Jeff thing.

Jonathan (41:16)

If, if they break something and there's like a proof of concept of, you know, here's how you see this horrible breakage. Okay, that's exciting, but it's just. It's just gonna work for the vast majority of this stuff. It's just gonna work and it's gonna be boring because it works.

Rob (41:31)

You don't like the drama, but you sure like you're excited about somebody breaking something. I got it.

Ken (41:37)

Yes.

Jonathan (41:38)

I don't like drama, but I sure like security breakage.

Ken (41:41)

It's fun, especially to write about.

Jonathan (41:44)

Exactly.

Rob (41:45)

You know, they can. That can lead to a lot of drama.

Jonathan (41:48)

Oh, that's true. That's true. All right, we want to talk about Servo now.

Rob (41:53)

I'll talk about servo now. So, you know, it's been said before on this, on this podcast on others that I've listened to that at this point, you know, this far into the HTML and CSS and everything else on the development of that it'd be nearly impossible to develop a web browser or web browsing engine from scratch. And indeed I'm sure it is a tough job. But we've brought up the rust based server servo browser engine before and I think we all pretty much agreed or thought that of it as is just a hobby web browser, you know, something that somebody's kind of doing just for the fun of it that really didn't have much of a feature in the consumer web browser space but the improvements they have been making. My turn to prove that thought wrong. The server web browser has been able to render basic sites for four years. I, I've tried it. I didn't try this update I was going to and I don't know, I forgot but I, I did try. I have tried in the past and you know it's kind of wonky in some sites a lot of basic sites went just fine. You know they've been able to do the basic sites though for years. More complex sites, the more popular websites like like Gmail have been a problem. Well not anymore. The server web browser can now render Gmail and Google Chat correctly and if you do that then you know, what else, what else can it do? Or what else can it do? Those are some of the more complex things out there. I really wanted to try this before the show and try some other sites but we'll see. But I'm thinking in the next few years we may see some serious web browsers start to consider the servo engine. And although it is debatable by some, it may bring a slightly more secure browsing future to a computer near you. And you know those secure browser it's not going to stop you. It's not going to secure you from everything. You know, it's not going to. It's not going to keep you from being scanned by those phishing emails. But it may help with the memory safe issues that some programs have. In other news for the server browser they another thing that they have, this could be a bit of drama. They've been considering a policy that would allow some AI based contributions via select AI tools like GitHub Co Pilot. But based on feedback the developers have now decided to keep the the ban in place. And based on the feedback I've been reading online, this doesn't necessarily seem like a very popular stance. You know, like one commenter saying quote a bit Sad about the blanket AI stuff. AI has been crucial for me in keeping program or yeah keeping programming despite my rsi. So AI could be an assistive technology to some. So I don't know the blanket AI statement is. Is a good one, you know. And another he said it is ironic that a project aiming to be modern is taking advice from people stuck in the past. Though others point out, you know, there are obvious flaws with with code being completely written in AI copyright issues, refractoring that could cause a mess when when running a diff a differential on code to see what's changed because it could refractor and do it all different. So even though you really only change a small piece it could just refractor redo it all different. But. But then it also, you know, lots of them point out with these issues they can be solved rather than completely banning the use of AI. But along with that others also point out that it's likely, you know, if you're not completely refractoring it, you're using it as a tool to just kind of help you with maybe a line or some piece here and there, not. Not doing all your coding for you. It's likely going to be difficult to enforce unless. Unless it's, you know, completely obvious like that. But either way, good progress on the web engine code. No progress to changing a debateable policy. No progress meaning this is status quo what they said before, it's still banned so same as before. So to kind of move forward in some ways and debatable on the other.

Jonathan (46:55)

Yeah, I in reading while you were talking something that surprised me. I didn't realize this servo was a Mozilla project up until like 2020 and then Mozilla dropped it and the servo people continued working on it on their own time and it has continued on. I was thinking that Servo was under a different license, but it's actually under the Mozilla public license which is compatible in most ways with the gpl. So it is a copy left license which is interesting when it comes to browsers. The other browser that's being made from scratch, Ladybird, it is a more permissive license and it actually has some commercial interest in it because of that. There are some things that companies can do using it embedded that they can't easily do with a copyleft license. So interesting stuff there.

Rob (47:49)

I should compare those head by head sometime.

Jonathan (47:52)

Lady Bird to Servo. Yeah, yeah, they have a lot in common as they are sort of the two reasonably successful ground up browsers that are in progress right now.

Rob (48:05)

Yeah, ground up.

Jonathan (48:09)

Built from the ground up. Not torn into little itty bitty pieces.

Rob (48:14)

If Mozilla loses their Google funding, they could be ground up too.

Jonathan (48:19)

Oh man, the world is going to be so. It's going to be so different if some of that stuff happens. Oh my goodness.

Rob (48:25)

Yeah.

Jonathan (48:26)

If Google is sourced, forced into selling Chrome and then Firefox loses their fund, oh my good. Everybody will be using Internet Explorer again.

Rob (48:35)

Yeah, I think there are some articles.

Ken (48:37)

What version?

Jonathan (48:38)

I don't know.

Rob (48:40)

I think there's some art. I think there are some articles about that this week about, you know, how hurt they would be if, if they lost that funding. But really it's. It's nothing we haven't talked about already. So it really wasn't worth bringing another story about that again.

Jonathan (48:56)

Yeah, yeah. Interesting, interesting stuff though. Yeah. So the actual big story here was that Gmail now works in servo. Do you have a servo install?

Rob (49:07)

I have tested in the past, but I don't have it installed right now.

Jonathan (49:12)

How difficult is it to install? Is it just pretty much an APT or a DNF away?

Rob (49:17)

I think it was, it wasn't too hard. I don't remember.

Ken (49:21)

It's not a flat pack away.

Jonathan (49:24)

I mean, if you're really into that sort of thing, I guess I don't.

Rob (49:28)

Remember it being difficult. I remember just doing it on a whim and testing it out.

Jonathan (49:35)

What do they actually call it? What do they actually call it? It's not. There is no servo package. At least not in this machine. Lib, Rust, Servo. Whole bunch of stuff about reservoir. Nope, not available on APT on this machine. Interesting, huh?

Rob (50:02)

Oh, you could.

Jonathan (50:06)

I mean, there's a download button.

Rob (50:09)

Yeah, there's a TAR gz.

Ken (50:14)

Whatever you do, don't log into your bank with her servo just yet, Jonathan.

Jonathan (50:20)

Why?

Ken (50:22)

Well, because they recommend that on their download page.

Jonathan (50:29)

You're right, they do. I. I guess the idea there is that they don't have the guarantees yet about one website not being able to do something nasty. The whole. No, no remote code execution and all of that. It's not quite as hardened as some of the other browsers are. All right, fair enough.

Ken (50:53)

But it looks like it's a TAR that you would be downloading and probably building.

Jonathan (50:58)

Yeah, I grabbed the tar. TAR gz, but. Oh no, there's a binary in there. There's a binary in there. All right, all right, all right.

Ken (51:07)

In the tar.

Jonathan (51:08)

Yeah. Which is weird. A tarball with a binary in it. What's the world coming to just don't know about this.

Ken (51:17)

That sounds like something you'd do for Windows.

Rob (51:22)

Yes, that's because he downloaded from the Windows link the top one there.

Jonathan (51:26)

No.

Ken (51:28)

No.

Jonathan (51:29)

Goodness.

Ken (51:30)

But then it would have been a zip file.

Jonathan (51:32)

Yes, it would have been the zip file.

Rob (51:34)

How can advertising on TikTok help jumpstart sales for your small business? Thanks to TikTok ads, I was able.

Ken (51:40)

To open up a business with my.

Jonathan (51:41)

Childhood friend, get a warehouse, and even hire employees.

Rob (51:44)

My name is Julian and I am one of the founders of the Snacks Lab. We are an exotic snack company. We had over $100,000 in sales from.

Ken (51:51)

Our TikTok ads in the first month.

Rob (51:53)

So our orders went from five a.

Ken (51:54)

Day to over 250 orders a day.

Rob (51:57)

You definitely have to use TikTok ads.

Ken (51:59)

And you gotta start now.

Rob (52:00)

Head over to get started.TikTok.com TikTok ads.

Ryan Seacrest (52:04)

Hey, it's Ryan Seacrest for Albertsons and Safeway this spring. Take care of your entire home, including the air you breathe, and save $5 when you buy $25 worth of participating products in store or online. Shop for items like Glade Plug Ins, Airwick Plugins, Glade Auto Sprays, Airwick Diffusers, and Glade refills, and save $5 when you spend 25 on participating products. Offer ends May 20. Restrictions apply. Promotions may vary. Visit albertsons or safeway.com for more details.

Jonathan (52:34)

Ken you want to talk about Caliber?

Ken (52:35)

I always want to talk about Caliber.

Jonathan (52:37)

It's one of your favorites, isn't?

Ken (52:39)

Is my favorite. And thankfully this week Marius wrote about COVID Goyal releasing a maintenance update to, as I've been saying, my favorite ebook management software, Caliber. This latest release, version 8.4, provides a workaround with a QT Wayland bug that prevented the menu with the layout button from showing. That's a little button, if any of you are familiar with Caliber, that lets you change the layout of your screen. You can have it where you sit up. So you have like the covers of your books showing above the columns that list the data about the different books you have. And you also had a regression that broke the Ebook viewers show book in main Caliber link that was fixed. There's an issue where some links weren't processed correctly for very large epub files that contained many internal HTML files and the ability to handle links to missing internal files with an error pop up now. In addition, caliber 8.4 improves tag merging in the metadata review Dialog, updates the eBooks.com plugin in the Get Books component I've used that occasionally for getting some books, though I actually have to admit my personal preference is Humble Bundle and fixes the Next and previous links in News Download. This release also prevents recursion when creating the base temporary folder if something on the system deletes the created temporary folder. As always, I recommend reading Maria's article for details I have left out Now. Next I would like to share some of my favorite Caliber plugins I'm going to post. Put them into the show notes for those of y' all that are listening later so you can actually go look. But the first one that I have is going to be View Manager. This plugin allows me to manage the views that I use, and I can using it, I can quickly and easily switch between different views of my libraries. Yes, I have more than one library. That's the Another advantage of Caliber is you can have multiple libraries. One that I use for books that I lend from my local Library to the OverDrive system. Another is one that I had created over the years for my paper books. So depending on what I'm doing, I may be it allows me to change the view I have depending on the library I'm in or what I'm doing at the moment. Another plugin that I like is Extract ISBN. I use it to try to find the ISBN for a book using the text within the ebooks format. It makes it easier when you're doing a search for other metadata for it. And of course, as I mentioned, overdrive, there is an overdrive plugin that I can use to search my lending libraries for borrowable ebooks and audiobooks that correspond to selected books in that lending library that I create. I'll just put an empty entry in and use that to search and then I can find out whether or not that book's at the library and then I can check it out, read it, and return it. Saves me the time of having to do it manually through their web page.

Jonathan (56:38)

Yeah, that's actually really cool that it's got that built into. I didn't realize that a lot of local libraries are starting to do that. The whole ebook borrowing thing.

Ken (56:47)

Yep. Then two others. One, if you ever used goodreads.com is real handy. That plugin allows me to sync my Caliber library with my goodreads.com account.

Jonathan (57:02)

Cool.

Ken (57:02)

And with the lending library that I've got sometimes and the paper library sometimes, I'm having to manage the series that it's got I've got set up. So there's a Manage series link that allows me to quickly create and edit my series column values. But the show notes have been updated so you can see all these with links to their descriptions and how to add them to a caliber.

Jonathan (57:32)

You know what we really need to make that really useful? We need a Linux tablet that has an E Ink display on it. That would be cool.

Rob (57:41)

The Pine tab.

Jonathan (57:44)

I suppose that's true.

Rob (57:46)

Or. No, it's not the Pine tab.

Jonathan (57:48)

It's like the Pine Book or something. Yeah, yeah, Pine made one. But is there any software that actually runs on it? I mean, I think that's sometimes the problem with Pine stuff.

Rob (57:58)

I think it just runs Ubuntu. So anything that's Ubuntu and ARM I think you can get to run on it. But how? Well, it would be on an Ink display.

Jonathan (58:09)

Pine note. They call it a pine note.

Ken (58:12)

I'm sure you could take a Raspberry PI and set up an E Ink display with it.

Jonathan (58:19)

I mean, you could, but that'd be kind of like carrying a brick around. That would not be a small or sleek to.

Rob (58:25)

Here's you do you get the display and then you get the Raspberry PI and you have a wire going to. And you just clip it on like a belt clip. So you're just holding on to the lightweight E Ink display.

Ken (58:39)

It'd be no heavier than the old CD players we used to carry around. Yeah, definitely lighter than the old boomboxes.

Rob (58:48)

Yeah. You know, 80s boom box on your shoulder.

Jonathan (58:52)

Yep, yep. Oh my goodness.

Rob (58:55)

Blurry.

Ken (58:56)

You watched Fast Times at Ridgemont High, didn't you?

Jonathan (59:00)

So let's save ourselves from all of this and talk about 1989. Wait a second. That sort of fits. Anyway, there's a couple of things in here that's super interesting and several things actually I didn't know about. One is, and this is kind of the lead of this and that is that the Linux kernel is going to stop supporting the i46 Intel CPU, which, yes, did. First release 36 years ago. It released 36 years ago. But they only stopped selling it 18 years ago, which means that they sold that thing for 18 years. And you might think, why in the world. Well, really it's because it was used in a whole bunch of industrial control. The 6502 is still being sold today. The old processor that things like the Commodore 64 used because industrial controls and even some medical things still use those. But Torvalds and Linux finally going to pull the plug on the 486 in the kernel. And that is interesting. And it sort of Led me down this rabbit hole of asking, well like okay, if you actually have one of those, what are you going to do? And there is of course the idea of using the long term support kernels from Linux and those will be around and support it for a while. That led me down this rabbit hole of the Cipher kernel maintenance project and this I had no idea existed. It is the super long term support kernels and from what I can tell they pick essentially one, something like one particular release from each of the like major versions. So they have a single five dot release that they're supporting in this SLTS from the cip. The civil. Civil. Oh I had it. It's something, something civil, something, something. I don't remember anyway, and the idea is it's for devices that are going to have a really, really long lifetime. And you know, they're talking, you talk about, you know, Linux distros, Red Hat for example, may only maintain something for seven years, 10 years at the most, maybe up to 13 years. Well, these SLTS kernels, they're talking about trying to maintain them for 25, possibly all the way up to 50 years. So you know, their next one they're going to try to maintain till I think 4.4 is one that they've talked about. And so they're going to try to maintain that all the way up through like 2060. Ridiculously long times from now. But you do occasionally find installs on set hardware, whether it be a mainframe or an industrial controller. You know, in some places things like the computer that makes the trains run, the train track run the computers in things like your stoplights. These things will be there for a very, very long time and you do kind of want to be able to push security updates through them. So I was super duper interested to hear about this. The CIP kernels and it's not something I was ever familiar with. So we've talked about the kind of, the decline of the LTS supported kernels. It is, it is really. And there's, there's like three guys behind this. It sounds like they're hoping for more developers in the future. But super interesting to me that they have this goal of supporting kernels for possibly up to 50 years.

Rob (63:11)

Is the 4.4.4, was that the oldest one or did they go back further?

Jonathan (63:18)

I that may be the oldest one.

Rob (63:22)

Love to get like a one dot and a two dot.

Jonathan (63:24)

No, no, no, no, nothing. Nothing back that far. I actually found this through. There's a WIK article on supported kernel versions. I'll See if I can find that and drop that in there as well.

Ken (63:37)

I remember covering slts sometime last year too.

Jonathan (63:46)

Interesting. Yeah, I didn't remember hearing about it at all.

Ken (63:53)

The other thing, I think I touched on it when we started talking about the. Some of the distros going to a 10 years support.

Rob (64:04)

Yeah, you're talking about.

Jonathan (64:06)

This is way beyond that though.

Ken (64:07)

Yeah, 25. Possibly 50.

Jonathan (64:11)

Yeah, possibly up to 50 years. Man, that's a long time to try to keep code run. What are computers going to look like.

Rob (64:16)

In 50 years, you know though?

Ken (64:19)

Well, the ones that are going to be running these kernel zones are going to be looking like the ones we've got right now.

Jonathan (64:25)

Yes, but still, I mean, think about 50 years ago in 1975, what did computers look like? And in some ways, yes, very similar, but in other ways radically different.

Ken (64:35)

You want me to show you around 1983?

Jonathan (64:40)

Do you have a 1983 computer boxed.

Ken (64:43)

Up at the moment?

Jonathan (64:44)

What machine is it? Let's see.

Ken (64:46)

RE 800XL.

Jonathan (64:48)

Oh, very cool. That's a neat era of computing.

Rob (64:53)

But yes, you know, even problem is.

Ken (64:55)

When I got back to the States, the power supply, when I tried putting it, hooking it up, it had trouble converting back to 60 from 50.

Jonathan (65:07)

Yeah.

Ken (65:08)

For 11 years.

Jonathan (65:09)

Yeah. It. It might need recapped and some other things.

Rob (65:13)

You know, if you're running something that old, I mean you could just run the old kernel on and not ever upgrade it. The reason upgrade it is so you have a secure kernel, right? No, no disagreement or.

Jonathan (65:26)

That's basically the reason.

Rob (65:27)

Yeah. The problem you're still going to run into is every other piece of software on there is going to be out of date and potentially insecure.

Jonathan (65:38)

It's true.

Ken (65:39)

But.

Rob (65:43)

You.

Ken (65:44)

Okay, no, I'm just trying to think what software would you be running on it that you needed? Because that's going to be the reason you're running that old in a kernel, isn't it?

Jonathan (65:57)

Yeah. So like for instance, let's just say we've got this old piece of hard, this old computer, it's running some piece of hardware, you've got the kernel on it. Well, you've also got systemd or SYSV in it, depending on how old it is. You may have your tooling in a Python script. Well then you've got an install of Python that can in and of itself have vulnerabilities and then you've got the Python script to keep up to date. So like, Rob has a point. There's multiple moving parts here to think.

Ken (66:25)

About it's more likely going to be an embedded system.

Jonathan (66:30)

Yes, like likely, but not for sure.

Ken (66:34)

And hopefully not connected to the Internet.

Rob (66:37)

Well then who cares if the kernel's up to date then?

Jonathan (66:43)

Well, there are laws that now kick in and have things to say about that and don't necessarily care whether it's connected to the Internet or not. There are potentially reasons too. I'm thinking of the old NES exploits, NES and SNES exploits that people have found that those are not connected to the Internet. But in some cases you can actually, you know, with, with your thumbs. People have practiced enough to be fast enough to actually exploit the NES and SNES manually without using emulators. Have you not seen. Oh, you need to look into this. I'll, I'll, I'll go find the link.

Ken (67:24)

Videos about it aren't there.

Jonathan (67:26)

Yeah, there's a. I think. Was it Seth bling was the YouTuber that did this? It's like in Super Mario Brothers. No, Super Mario Brothers World. You can actually glitch the game out and he's done it by hand. You can glitch the game out and warp to the end credits and beat the game from the first level. And you're literally, you're exploiting the game. You're overwriting a piece of memory in RAM that's not supposed to be overwritten and you set things up to where you actually jump execution into code that you load into ram. It's the coolest thing.

Rob (68:01)

But if you have whatever old piece of computer you have that you want to update the kernel on, if you have that kind of access to it, it's exploited anyway, it doesn't matter.

Jonathan (68:16)

I would just say there's a lot of nuance depending upon where a project or where a piece of hardware is, what it's doing and how much like user interaction it gets, whether it's public facing at all, even if it's not connected to the Internet, if people can poke at it, there, there may be reasons that you want to, that you want to keep it updated even. Even if just because of things like the 2038 bug.

Rob (68:43)

There you go. There's one.

Jonathan (68:44)

There's, there's a really good one. You, you might want to update your kernel because of the 2038 bug.

Rob (68:49)

There you go. You got it?

Jonathan (68:50)

I got it. All right, let's move into some tips. We're going to let Rob take special variables. Do they write this robust?

Rob (69:03)

So anyway, last time I was here I provided a tip and I demonstrated the results with the special variable dollar sign, question mark. To show the exit status, you know, is either 0 or 1. To show whether or not the, the command previous command was successful or failed. And I said I would follow up with an explanation of some of the special variables. Now I almost forgot about, I started, started writing away to my next more utils tip and realized I was gonna bring something to you guys and I, I almost broke my promise, but here we are. So for those watching, I brought up a little, I made a little script to just quickly demonstrate this. So the, the special variables I have to talk about is like dollar sign, question mark which shows the exit status of the last command. I already showed you that last week. Really? Then you have like dollar sign zero that's going to show the file name of the script that's running in. Dollar sign one is going to show the first argument, dollar sign two the second, and three, four, five and so on. And then dollar sign, hashtag, pound symbol, number sign, whatever you want to call it. You know, dollar sign, tic tac, toe box that will show the number of arguments passed. And then dollar sign, dollar sign is going to show the process id. So for those looking, I have a little script. It just echoes each of them out and it's going to show the results. So what I'm going to do here is go back to the command line and I am going to run my test script. So that is test sh space arg1 here. So that's my first argument. Arg2 here is my second argument. If I run that, it's going to spit it out. So Exit status is 0. Successful file name of the script, it tells me what the file name is I just ran. And then it has the first argument pass, the second argument pass. Number of arguments passed is two. The process ID. So you know, that's the process ID that showed up in, in the, you know, P.S. you know, if I did another one, it's going to tell me, you know, if I add an argument there, it's going to tell me there's three arguments passed. But I don't have in the script anywhere to spit out that third one. Or let's say I go down to one, it's going to say one argument pass. The second argument it's blank because there's no second argument. So these special arguments are something you can use to make a script. And so you could use those arguments to find the name of the script. Use not arguments, special variables. You know, special variables to do things like that. You could pass arguments to your script if you want it to Be able to do different things depending on what you do to it. So basically all the, you know, dash, dash help or dash H or dash whatever you want, you know, all those things that you do with other commands, you can pass them to your script and work with them in various ways with these special arguments. I can see you didn't see that last one because it was below my screen, but that was one right past one.

Jonathan (72:33)

Very cool. Yeah, super, super useful to write your. Write your own tools. Yep, yep, Nice. All right, Ken, you got some more pipewire stuff?

Ken (72:46)

Yes, I do, and this is going to be one that would be dangerous if I were doing it live. So I took screenshots while I was in Tumbleweed.

Jonathan (73:02)

There you go.

Ken (73:03)

For two reasons, and I'll go over the both of those reasons in a minute, but this week I'm covering how you can manage the modules used by pipewire. Sounds dangerous right there, doesn't it?

Jonathan (73:16)

Yeah.

Ken (73:18)

Now, I have linked the module section of the pipewire documentation in the show notes for those adventurous ones. And let me go ahead and bring up my terminal so those watching can follow along here. I've already typed PwC just went in for interactive, so I've got my pipewire zero command prompt. Now here is where I'll type in the commands load module followed by the name of the module in question, the one I'm going to demonstrate. I didn't need to add any arguments for, thankfully. Otherwise I'd have probably spent more time playing around with the arguments than actually preparing. But, and as you'll for those of y' all watching, you'll notice that this particular version of PipeWire is version 1.4.2. There's a reason for that going to be when we get to demonstrating the unload module. But moving to the second screenshot, the command is load desk module. Or if you're doing it on the bash command line, it's pwcli followed by dash dash load module, then the name of the module that you're going to do the module I'm going to be demonstrating loading is lib pipewire module adapter. It's just a generic one, doesn't need any arguments, thankfully. And as you see it, when I hit enter it, it came back with 1 equals module 22. This is a module variable that's used when you go to unload it. The nice thing about doing it from the PWCLI's interactive option is you can load it multiple times. I just went ahead and went through two other screenshots showing where I loaded a second, third time, each time it incremented to 2 equals at module colon 24 and 3 equals at module colon 26. Now I'm going to demonstrate unloading. Unfortunately, pw.cli's versions before 1.4 will give you an error message command. Unload module not yet implemented. You just tried it?

Jonathan (76:11)

No, I just find that funny. That's awkward that it was not implemented yet.

Ken (76:17)

Yeah. So when I was trying it on Ubuntu with 1.2.4, it was giving me that command. I'm going, I know I did this. Then I remembered it was in Tumbleweed, but I just put the unload dash module with a space and then the dash fee for three to unload. The third time I loaded it and it returned that. Then I went and reloaded it again and it reloaded it right back into the. Against that third variable. So that's how you can load and unload modules from the PW CLI's prompt. And that way you can go in and play with some of the arguments for some of the modules, like the libpipewire module, FFADO driver. You might find that one interesting, Jonathan, if you haven't already been playing with that one.

Jonathan (77:16)

I was just thinking about that.

Ken (77:20)

And another one that I'm playing with is the library pipewire dash module, dash filter, dash chain. It's got a clamp plugin that you can use that might help with. If I can get it set up to limit hearing any background noise through my mic when I'm not talking. Clamp it. So it's in between a minimum, a maximum volume level.

Jonathan (77:51)

Very cool.

Ken (77:58)

Have you gotten your.

Jonathan (78:04)

Have I gotten My what?

Ken (78:05)

FireWire device working with PipeWire.

Jonathan (78:08)

I've not. The last time I tried to do it, their FireWire implementation was very buggy and had problems. So. And I've not had time to fiddle with it since then. I've got a USB interface, a reasonably nice USB audio interface that I'm using. So, yeah, all right. I have a tip. And this is because I was GMING this week. I was running a Pathfinder game and we were in a module and the module had a player handout. It's like, oh, well, I would like to give my players this handout. And it was, you know, it was nice. It was just a couple of pages of the module that I needed against it them. And I went, I bet there's a way to just re. Sort of remix a PDF and say, I want just these pages. It's like, okay, I know I can do this. I Can print to PDF and do it that way. And so, you know, it'll give me the print dialogue and tell it I just wanted to print these pages and it'll give me a new PDF. Like, but that's kind of lame. Why should I have to do the whole print thing? I'm sure there's another way to do this. So went to looking and found PDF jam. Now there are multiple programs that will let you do this. PDF jam is the one that I use. It seems to be about the simplest and it's just you give it these arguments, you give it the input PDF and then you give it your page arguments and you can repeat those two repeatedly. So like if you want to actually a few pages out of this PDF and then a few pages out of this PDF, you can totally do that. So it's like PDF jam, input 120 to 22, input 2 PDF 5 to 6. And then you give it a dash O and your output and it'll take those pages, jam them all together and gives it to you in the output. Super useful for being able to give my players the specific handout that they were supposed to get. Give it to them digitally, of course, because we're spread Hither and Yon and don't actually get together to play, but was extremely useful. And so if you find yourself needing to remix a PDF, combine PDFs, split PDFs, PDF jam, it is one of the tools to turn to. I got a kick out of it.

Rob (80:11)

Where'd you say you were split?

Jonathan (80:15)

Hither and yon. Are you. Have you. Have you never heard Hither and yon?

Rob (80:22)

No.

Jonathan (80:23)

Oh my goodness.

Rob (80:23)

I have. Must be a Southern thing.

Jonathan (80:26)

Yeah, it probably is. I have some very southern roots and sometimes those come out Hither and yon.

Ken (80:34)

My southern roots are probably further south than yours.

Jonathan (80:37)

Very possible.

Ken (80:38)

I was born and raised in Florida.

Jonathan (80:40)

Florida is not south. Florida is not south. Florida is just Florida.

Ken (80:45)

It depends the part of last century.

Rob (80:49)

Directionally, it's south, just not culturally or right.

Jonathan (80:53)

Florida does not have Southern culture. Florida has Florida culture.

Rob (80:57)

Ken is the Florida man, sorry.

Ken (81:02)

Transplanted to Oklahoma. All right, you don't want to know about the route I took around the world to get it.

Jonathan (81:12)

Oh, I'm sure, I'm sure. Let's let the guys plug whatever they want to get any ending thoughts in if they want. We'll let Rob go first. Rob, what you got?

Rob (81:22)

Nothing special. Just a way for you all fans out there to come and connect with me. And you can do that by going to robertp Campbell.com and on that page near the top you could find links to my LinkedIn, my Twitter, my Blue sky, my Mastodon, and a place to donate a coffee to me or more come and connect.

Jonathan (81:49)

Yeah, pretty cool. All right, and again, let me unmute first. You're here. I can hear you.

Ken (81:58)

I said that right after I hit the unmuted, but I've got a link in the show. Notes to Article Updated article actually about the modular Cerebro cluster board that actually supports the Raspberry PI CM4 and CM5 as well as the Jetson. The Raxta CM5 modular Subaru clusterboard supports has a lot of things in it that it supports, but it's as being a cluster board you can have multiple ones. So check it out. And Jonathan's wife. I didn't mean to have him buy it.

Jonathan (82:41)

No, I don't think I'm going to buy this one. I'm fairly satisfied with my Turing PI 2 for doing this sort of thing, but it does always fascinate me, the ability to put multiple computers in one little space. It's just cool something about it. All right, thank you guys. Appreciate it. I believe that's pretty much the show. We appreciate everybody being here. If you want to find more of me, there is of course Hackaday. That is where you can find Floss Weekly as well as my security column which goes lives on Friday mornings. Go and check that out. Always a lot of fun. And if you're not part of Club Twit, well, we've got the QR code right up there. Scan the code, join the club, support the network that you love. We appreciate that as well. Thank you everybody for being here. Those that watch and listen, those that catch us live and on the download and we will see you next week on the Untitled Linux Show.

Rob (83:35)

Stay on top of tech trends without the time sink Twitt TV's short form podcasts are built for busy leaders like you, delivering essential insights in minutes. Hands On Mac and Hands On Windows provide quick tips for Mac and PC, while Hands On Tech quickly addresses common tech challenges to keep your operations running smoothly. If your conference room needs an upgrade, Home Theater Geeks explores the best screen and sound systems. And if you like watching the shows, join Club Twit to get full video access, ad free versions and more. Get Technology that matters on YouTube. Your schedule. Download our short format shows now at TWiT TV or your favorite podcast player.