Tech Over Tea - Chatting With An Early Slackware Developer | David Cantrell

Episode Date: May 15, 2026

Today we have David Cantrell on the podcast, one of the authors of the Slackware book and one of the early developers from the Slackware project to talk about what things were like back then before Slackware became the oldest surviving Linux and longest running distro.==========Support The Channel==========► Patreon: https://www.patreon.com/brodierobertson► Paypal: https://www.paypal.me/BrodieRobertsonVideo► Amazon USA: https://amzn.to/3d5gykF► Other Methods: https://cointr.ee/brodierobertson==========Guest Links==========Red Hat: https://www.redhat.com/Slackware: http://www.slackware.com/Slackware Book: http://www.slackbook.org/==========Support The Show==========► Patreon: https://www.patreon.com/brodierobertson► Paypal: https://www.paypal.me/BrodieRobertsonVideo► Amazon USA: https://amzn.to/3d5gykF► Other Methods: https://cointr.ee/brodierobertson=========Video Platforms==========🎥 YouTube: https://www.youtube.com/channel/UCBq5p-xOla8xhnrbhu8AIAg=========Audio Release=========🎵 RSS: https://anchor.fm/s/149fd51c/podcast/rss🎵 Apple Podcast:https://podcasts.apple.com/us/podcast/tech-over-tea/id1501727953🎵 Spotify: https://open.spotify.com/show/3IfFpfzlLo7OPsEnl4gbdM🎵 Google Podcast: https://www.google.com/podcasts?feed=aHR0cHM6Ly9hbmNob3IuZm0vcy8xNDlmZDUxYy9wb2RjYXN0L3Jzcw==🎵 Anchor: https://anchor.fm/tech-over-tea==========Social Media==========🎤 Discord:https://discord.gg/PkMRVn9🐦 Twitter: https://twitter.com/TechOverTeaShow📷 Instagram: https://www.instagram.com/techovertea/🌐 Mastodon:https://mastodon.social/web/accounts/1093345==========Credits==========🎨 Channel Art:All my art has was created by Supercozmanhttps://twitter.com/Supercozmanhttps://www.instagram.com/supercozman_draws/DISCLOSURE: Wherever possible I use referral links, which means if you click one of the links in this video or description and make a purchase we may receive a small commission or other compensation.

Good morning, good day, and good evening. I'm as well as your host, Brody Robertson. And today we have someone who's been around in the FOS space for how many years has it been? Oh, gosh, I didn't know I would have to do math. What year? We can skip the math. Yeah, since the 90s, mid-90s. Okay.

So you may know him from his Slackware stuff. now being a red hat, being involved in fedora stuff. Welcome to the show, Dave Cantrell. How's it going? It's going great. Thank you for having me. Absolute pleasure. I wasn't aware of your work. One of my viewers had sent the interview you did with Stewart over on Slackware Linux Arm,

and I thought it was a good episode, and now you're here. Yeah, great, yeah. So for anyone who's unaware of who you are and why, what you do, I guess, do you want to just start all the way back at the start? That seems like the best place to go. Yeah, you know, it's actually, it's really a common question I get, you know, how did I get started in this space? Because I think definitely the stories that people have now about how, you know, their first, you know, usage of open source software or Linux, specifically, is very different than mine. So I enjoy sharing that. So I will say back in the 90s,

so I was in high school in the early 90s and was definitely interested in computers. And the computers that were available to me, also I'll say we didn't have a computer at home until I was in high school. And that was what was common at the time was, you know, your parent would bring home an old computer from work. So you were getting something outdated, but it didn't matter. You had one at home. Most of the modern stuff I was using at the time was at school. And that was dominated by Apple.

And then IBM later came in and sort of inserted PCs from various generations. So that was, you know, my exposure. I had heard of other stuff. I knew that there was, you know, large computers, you know, companies had mainframes things like that but i had never you know seen one or really used it or anything like that um so you know we my friends and i we would get into programming and stuff and we had you know most of the information you would get would be through just talking to people uh going to bookstores and the library and finding books and kind of going from there books back

then were kind of a key source to learn a lot of this stuff initially. Because unlike people at a university or a company, you didn't really have reliable internet connectivity at home. Right. And if you did have an internet connection, what information was there really available at the time? Exactly. Yeah. You know, at that point in time, most of the, I would say technical stuff would be on Usenet, which, you know, dial-up home users weren't really exposed to. And even then, you know, you kind of had to be a technical user to get into that. So in high school, I remember someone had heard about Unix. And then that led to, I already knew about that. But I was like, yeah, but that requires really expensive hardware. We don't have any of that. And then someone said, no, no, no, no,

I meant to say, he said, Linux, because no one knew how to pronounce it. And so it's like, okay, I'm curious, what is this? So we head to the bookstore and we see a lot of books that are talking about this. And we have no idea of where to start. Like what we didn't even know what it was. Roughly what year would have, would this have been in? for me that was 95 96

I think two three years after Slack Blair and Debian started yeah yeah and at that point publishing these releases along with a how to book was kind of the

norm and there was interest among publishers to do that so that that's kind of the you know aside from being able to download it again you know that was a thing that people at universities could do i i wasn't there yet so it was like that you know i'm getting it from a book and the book would you know they were trying to put together these books which is how to get it set up initially so we went through that and i remember uh the three friends that i was i was working with uh we each

spent about a week trying to get our uh lennox set up on our system and uh I was the only one that succeeded, but even with that, I didn't have, I didn't have dial-up support and I didn't have sound support and I didn't have X, but it did boot and I was able to log in. And so, you know, you got to start somewhere. So that was sort of the beginnings of it. And then I went to university and my degree is in computer science. So Unix was the norm at that point. We were doing everything on Solaris, which I had already also been exposed to in high school.

And so I was, you know, and playing around with Linux, you know, there were similarities. So everyone, you know, was starting to put Linux on their computers. I already had that. It was sort of this halfway install. But suddenly I had access to all these people that could help me, you know, how do I get X working? And, you know, you just start sharing that knowledge. And pretty soon I'm just using it full time. And then you start, you start finding, well, I'll say finding bugs or problems or something like that.

And all of the notices, you know, all the documentation, every developer is putting their email address in there. I have email now at the university. So I email people and say, hey, I found this. You know, sometimes I would be able to fix it. Other times it would just be, I want to let you know. And that's how I got to first talk with Pat Vultry. And then ended up meeting him in person at a trade show in Chicago. So in the U.S., well, they may have been worldwide, but we used to have these trade shows called Comdex, later Condex Windows World.

It's just a typical industry trade show. And he was there with. Slackware CD sets and you know, we're like, oh my gosh, this is this is what we're using. So that sort of started things off and we continued talking and that led to my involvement with Slackware. That was around what was what release was that? I think it was release 3.3 maybe. Yeah. According to the Wiki page, you were part of the team as of four. So sometime before that point. Yeah, well, I, so.

So I was around for 3.6, definitely in 3.5 and 3.4, but I don't know that, I know my name wasn't in the announced file, but we were, we were helping. I was going to trade shows along with Patrick at that point while I was still in university. And then that led to a job offer from the publisher of Slackware, Walnut Creek, CD-ROM. Long since gone at this point. But if you were around in the 90s, you know, you were on a modem probably. And downloading lots of data was time-consuming, prone to error, and probably expensive. So it was really common to have these companies that were pressing.

CDs of just really anything you could download. And so Walnut Creek CD-ROM was a company that was doing that. You could buy CDs of fonts and clip art and icons. And then they had Linux distributions. And they also did FreeBSD. They were the official publisher for that. And so that's why I was employed by them is that was kind of like the the home base really for for slackware at that time and yeah then you know was kind of involved

and through i think release uh 7.1 and then that was that was early 2000 so we had the dot com fallout and uh that led to well so walnut creek had already been through a number of acquisitions and mergers and it was kind of a disorganized mess in my opinion we we had a hardware company that either bought us or we bought them or we merged with them i don't quite understand um b sdi did come in and acquire us that was a commercial business selling uh commercial bsd unix and you know they were i look back at that now that was that was definitely a a cash grab with the dot com wave and it just didn't pan out. So we were we were let go because as it was getting parted out,

the engineering, engineering assets, let me see if I can remember this meeting. The engineering assets were going to Wind River, but they only wanted the BSD side of the house. So that left those of us working on Slackware, we were let go. And, you know, that was kind of the end of getting a paycheck to work on Slackware. Though, you know, the distribution obviously is open and it continued.

Pat did ask me to continue working on it, but I was, I was concerned about, you know, now I've, you know, I'm no longer employed. I left university to do this. Now I'm going to worry. Finishing it. Well, after that, I ended up going back to university to finish my degree. So I left like midway through. No regrets with any of the decisions. It's just, well, I guess I wish I could have continued working on Slackware.

But it was probably the right decision to just focus on school and get that finished. and then go from there. But yeah, I know I'm going through the whole story here. Let's see. It's okay. We can jump back at any point. It's totally fine. Yeah, yeah.

But so after I graduated with a degree, I worked for a number of companies before ultimately ending up at Red Hat in 2005. And I've been there ever since. And I work on, I'm in engineering and I work on, well, Red Hat Enterprise Linux, Fedora Linux, and now Send to Us Stream. So basically the operating system products that Red Hat does, that's where I am. I've seen the company grow quite a bit since I've been there. It's definitely got a different field than when I started. But, you know, we're still open source first. And it's really nice that you can like I've I've been able at least to carve out a

career around being able to get paid to work on open source software. You know, like this is a concept that decades and decades ago wouldn't have made sense to any company. And I think that all of the players out there, companies that have supported Linux in particular, but really all open source development has fundamentally changed the computing industry. And that it's kind of fun to be part of that. One thing I didn't expect at this point in time was that, you know, kind of like the number one Linux platform is probably an Android phone. And no one, no one really knows. You know, they, it's Android. It's not Linux. And when I got involved with this, the sort of the dream was, you know, what if what if you didn't have to use a Windows PC? You know, what if Linux was,

was good enough to replace that? We don't really need that anymore. Like it's, it's, it's found its home elsewhere, you know, and it's, I mean, gosh, it's on like TVs and every embedded it. Yeah, it's what you have the Android auto as well. Right. Yeah. Yeah, of course. So like, it's, it's everywhere. And it's running, you know, REL, Red Hat, Enterprise Linux is, is everywhere, it seems. And if it's not that, then it's another Linux distribution. I mean, gone are the days of having a wide range of commercial Unix operating systems and Windows. I mean, sure, those are still out there, but they're not, they're not like at that number one position anymore.

So that's kind of neat. But, yeah, I feel like I'm trailing off here. in my thoughts so good I'm feeding the timeline on Red Hat 2005 you said you joined yes yes

was it Red Hat Linux but still at that point or was Fedora Core it was Fedora Core okay okay yeah so that that split happened I'm gonna get the date wrong

but it was a couple of years before I joined where Red Hat Linux what had been Red Hat Linux basically became Fedora Core. And a new commercial, uh, supported version of, of the distribution was rebranded as actually Red Hat advanced server,

RHAs. Final release of Red Hat Linux was March 31st, 20073. Yeah, that's, that sounds about right. Yeah. And then, yeah, and, and really the code base just kind of for the distribution. just kept continuing. But yeah, so the idea was the business model will be, you know,

we're going to make a commercial product that's based on the open source distribution, that's free to everyone, and we're going to sell support and subscription services. And you know what? Like that, I got to give it to Red Hat for figuring out how to carve out a revenue stream and letting us basically do all of our work in the open. Like I don't like I've never had to keep like code behind closed doors or anything like that. Now I'm sure that we do have products that are like that, but at least working on the distribution, the expectation at Red Hat is you're doing this in the open first.

And I've tried to explain it to people that because of that and it's so ingrained in how we do the OS at Red Hat. we don't really have a way to make and ship something proprietary. Like, we don't even have a process or mechanism for that. We don't know how to do it, which is kind of cool, you know. But, yeah, so the expectation is you're doing everything in the open and, you know, and then we stabilize it and support it forever. We're still supporting, like the promise for REL for businesses was 10 years,

of support, but businesses pay for additional years. So they're still, you know, like, I'm sure, like, REL version 6 and version 5 has users and they're paying us to continue supporting that. And I think that says something to the stability of the platform and the reliability. And that is, that definitely goes to all of the open source developers that, you know, those of us at Red Hat and plus everyone else. Like, I mean, Linux is a really solid OS. You know, it's really reliable. Yeah, I've got the support page open right now.

End of maintenance, we have seven and six. They're in the extended life phase now. That's right. Yeah, that's what we're called. Well, that basically means the type of updates, I think it goes to critical and security. But you have you have to pay for those additional. Right, right. That makes sense. I don't want to sound like a red hat salesperson here.

I guess I'm just explaining how we do things. I try to clarify, you know, we, the stuff that appears in REL is, it's in Fodora years before that. But Fodora itself is not just a test bed or dumping ground for things. Like, we actually do test and make releases and all that kind of stuff. There are vendors out there that also, just like you can get with Ubuntu, you can get systems pre-installed with Fedora. So, you know, that's kind of cool. And just for context, like, how old Rel 6 is, that is from 2010.

Yeah. Yeah, I worked on that one. Yeah, the first rel that I had patches that went into was a version of rel three. I forget which one. So I've been there for three, four, five, six, seven, eight, nine, ten, and we're working on eleven. I don't even know any. It's just all right. But, you know, all of us that work on development stuff at Red Hat, we all use Fedora.

So that's the development platform, which is, you know, I wouldn't want to use REL. I mean, I know that business is a buyer, but I wouldn't want to use that as my daily driver. It's missing a few things I like. Right. It's, what's an enterprise product is not, right? You know, you know, you don't need to say anything that's going to get you in trouble. Yeah, I know, I know. It has a function that is, that is not for the general product, like general purpose user.

It's a business focused operating system. Exactly. Exactly. Yeah. Yeah, I hope they're not listening, you know. I don't want to, don't want a bad mouth the product. but yeah, it's, uh, it seems to work. So, you know, the model does anyway.

So jumping all the way back, when you first got started with Linux, was that first distro you used Slackware? It was. It was. And what hardware were you using at the time? Just to be able to have an idea of like, what kind of things, you know, how much things have improved? really. Yeah, okay. So that was a system that I had put together from sort of cast off parts. It had

and I got to look up the processor real quick. Okay. Hang on just a sack here. Oh, good, oh good. Yeah. The, okay. So, so. I don't know if you remember a microprocessor company called Syrix. Syrix, Syrac. See, I'm 28, so. Sure, yeah, no, I, yeah, yeah. So I'll explain Syracor.

So, okay, okay. But yeah, so Intel processors, X86 processors were, oh, I'll clarify. So I started on X86. I know that there were other architectures that were supported. but that was definitely the easiest one to get going. Slackware was only on X-Cad6.

So I had an X-86 system that was using a Cyrix 486 DLC microprocessor. That was, I think it was 20 or 25 megahertz. It had four megabytes of RAM. I had a 105-mabyte hard drive. I did not have a CD-ROM drive, and I did not have a three-and-a-half-inch floppy drive. I only had a five-and-a-quarter-inch floppy drive. Again, assembled from parts. And I had a Paradise VGA graphics adapter, which could do 16 colors at 640 by 480.

Maybe I'll be Maybe I'll be to find a picture Probably sure you'll find something Paradise Systems That what a crusty board that is It's yeah So this is what we would do We would assemble

You know systems from parts You know and try to try to You know It was And I was doing this because it this was not my my daily driver system um it was one that i wanted to see if i could get lenox so you know i wasn't well i didn't have money to go and purchase a new system this is when computers were still several thousand dollars in the u.s uh new so you're you're getting parts

and people businesses especially like in high school i i was making money by you know, helping small businesses, you know, like here upgrade windows on our computer or help me hook up this printer or, you know, and you would get a little cash here and there. And then you could go and purchase parts like from generic vendors. So it would be usually like it was all import stuff. And it would be like in a office part. building and it would be a windowless door and it would just say like you know the the guys initials you know like uh hl computers or something and you would go in and it would just be all these like just brandless boxes and it would be like oh you need a you need a vGA card there it is you know it's

$20 and you know so it's just cheap stuff but uh it did it did run Linux so i used the term run loosely It was actually faster than Windows on that computer, which you can probably not be surprised by. So Cyrix was a X86 compatible CPU manufacturer. They were fabulous. I think IBM made the processors. Yes. Yes. Yeah.

So they were founded to basically try to make a low-cost. x86 knockoff and there were a number of vendors a number of companies that were doing this because intel processors at that time were quite expensive um but they were reliably performed well um but you know i did not have like four or five hundred dollars to buy a 486 like a genuine 486 so you know i went with the knockoff now the there there were a lot of linux kernel developers that had these sort of low-cost CPUs. So there were plenty of workarounds in the kernel to deal with these sort of misbehaving process.

Right, they were clones, but not perfect clones. Not perfect clones. You know, to make it cheaper, they would say, let's just leave out some instructions. You know, Cyrix was definitely, the big thing they saved money on was just not having an FPU. So, yeah, it was, I mean, it got me to my next computer.

So, you know, but yeah, that's definitely a far cry from the kind of systems that we have these days. But yeah, that's, that was the first system I installed on. Even the cheapest of cheap devices just completely outclass something like that today. Just like a basic phone completely outclasses that. Oh yeah, yeah, absolutely. Yeah, it's and, and, you know, when I talk to people about, you know, like, well, you got it installed on that, but then how long did it take to compile things? And the answer is it took a long time.

You would do a lot of waiting. And in that time, you would either be reading through the books you had or doing something else. Because you had to let the computer finish. And we're now like, I don't know how fast my system can build the kernel now, but it's minutes, if that. And this used to be an all day affair. You know, oh, I got to recompile the kernel. Okay. Well, here we go.

And then you would get the infamous GCC Signal 11 errors, which were due to cheap hardware. It was bad memory. It was so common that there was the famous Sig 11 FAQ, which let's see if that's still online. There's got to be, if it's not, there'll be an archive of it, surely. Oh, yeah. Yeah, it's on the Linux documentation project. I just take them... The Sig 11 problem?

Sig 11 problem, and then there's the Sig 11 FAQ. Ah, yep, yep. So... Question 11, what does that mean? Answer. Signal 11, or officially known as segmentation, officially known

as segmentation fault, means the program acts of implication that was not signed. That's usually a bug in the program, so if you're writing your program, that's usually most likely, because, however, this FAQ will concentrate out the possibility beside that. Okay. How long is this document?

Oh, it's quite long. And it's full of just anecdotal feedback from people. But no, like, like, Sig 11 is real. And, you know, we, we do debug memory problems. But what would happen is you would get this error building the kernel. And you would think like, okay, well, surely someone built the kernel before releasing it. I mean, Linus wouldn't release it if it didn't work. So what am I doing wrong on my end?

So people would get online and they would start asking questions. And pretty soon in those conversations, someone would say, have you read the SIG-11 fact? Because you probably have bad memory or something is wrong with your hardware. And that was really common. It was just really bad components and running GCC building the kernel, building GCC itself, any of those really, really large programs that took a long time to compile would tax systems like I just described and would really push them to their limits. You might not normally see that had you been running DOS or even Windows because they are not

stressing the system like you are under Linux. Yeah. So that's sort of a thing of the past. We don't have to have the FAQ explaining the truth. hardware is terrible. Considering those issues of the past, I can understand why when Linus had

his new system built by Linus Tech Tips, the other Linus, the Canadian Linus, made a big deal about having a full ECC system. I can understand now why that was such a big deal considering that legacy. Yeah, yeah. You know, you can't,

the the and he is definitely stressing his system constantly and you yeah you don't cheap out on the components but you pay more for the the really reliable stuff and you know that's that's also reflected in the cost of enterprise grade hardware that that is on all the time um you know consumer grade hardware which Linux developers you know those of us at home we don't turn our systems off we're stressing we're always doing stuff and we see those consumer products or consumer rated gear, you know, tends to fail faster. You know, it was really a major problem back when we all used hard disks because those were mechanical and they just couldn't, they weren't rated for the hours.

We were, we were giving them. But no one could afford, you know, an enterprise grade disk. So, and there were, they were a workaround solutions for that too. Like I look at, at raid solutions as, like, like dealing with the fact that the hardware will fail. So you might as well prepare for it. And we have, you know, like software rate analytics was done for that reason. I remember Alan Cox, he had, he was describing a setup one time where when he was doing bring up on a new system or really working on something that was barely running the kernel.

And, you know, it's, it's, you're getting colonel panics constantly. he would just NFS mount all of the stuff on the system because he knew it would crash. It's like, I can't also deal with local data loss. So all the data is over the network. That's how he did it, which kind of makes sense. Of course, you have to have networking available on the system at that point. But yeah, once we gain NFS support, you know, that was possible. So you mentioned that, you mentioned that,

I'm going to jump back through your story. I think it's what we're going to do for like half this episode. You mentioned struggling... No, no, totally fine. I like these episodes. These are the good ones where I don't have to talk much. So you mentioned struggling to get X set up. And I think people now sort of take for granted.

I guess the simplicity, not only of setting up X, but setting up everything with their system. And, you know, you can just install a package. And for the most part, it's probably going to be fine. If it's not fine, you're probably missing a package. And even just using X itself, things like X Rander, for example, those didn't exist yet. So if you wanted, my understanding is you wanted to reconfigure your monitor layout, you had to restart the X server. Correct.

Yeah. Yeah. And it was, so, so back then X, so you, you had the driver for the graphics device, which was, early on, we actually had different executable X servers for different types of graphics adapters. Like you had the one for ATI, you had the one for ATI. you had the one for Matrox, you had, you know, and of course there was a generic one. So, but all those eventually got separated out into drivers and were then loaded by, you know, the X server. But the sort of one of the rights of passage, so to speak, was getting X working and configuring your display, which everyone had.

So I described how I would go and buy these parts. They're generic, you know, no brand or anything like that. That's where my monitor came from as well. X did not know about that monitor. It didn't really know about anything except maybe like a dozen monitors that no one had anyway. So you were, you had to write mode lines in your XF86 configuration. fig file. And, you know, the X server will do what you tell it to do. So you would write a mode

lining. You had to get the refresh rate and you had to calculate all of this stuff for your monitor using the information from the manufacturer. And that might not be correct. You know, you may have gotten a manual with the monitor, but it could have been the wrong one. You know, so you would write, put all this info and then fire it up. I hope you got it. right, but some unfortunate people would smoke their monitor. I was about to mention my understanding is you could pop some bad monitors. Yeah, and that was, you know, that would certainly turn people off from using Linux, because now, you know, they've been told to try this, and now it has, you know,

damaged their hardware beyond repair. But it was, you know, it was, it was tough, you know, and it was just one of those things that we had to learn about. Now, you take on a PC, you would take an operating system like Windows. You know, part of what Windows Microsoft would do for that product is they would just have a database of pretty much every manufacturer in existence and have a way to automatically figure that out based on detecting it. And there were no, at that point in time, there were no real standards for getting that info. So they would research it and they would build that into the OS. So that was where that sort of ease of use thing argument would come in where it's like, well, Linux is hard to use because I have to give it all this info.

I was like, well, yeah, right now it is because we don't have all that info. But we, yeah, we got to a better place within Linux, certainly with kernel mode setting. for the display and then the X server being able to, you know, just directly use that. And of course, Wayland has improved things even further for modern displays. Because, you know, we don't really have X was from a time period where we had CRTs. We didn't have LCD displays. And now you've got the ultra high-deaf and, you know, all kinds of other displays that X wasn't really built for. Wayland is taking into account, well, you know, maybe, maybe displays won't always look like this.

You know, we should we should build in a way to to improve that. And I think they've done that. My, I actually, I do still use X. And that's not because I'm anti-Wailant. It's because I haven't, I still use software that does not work in Wayland even through the X layer. And I know it will eventually, but I just don't care. I'm going to use whichever one is the path of least resistance for the software that I use. And that's why I still have X as my graphical setup.

What software do you have that's causing issues? It's some maritime software that I use. Yeah. Okay. Yeah. Yeah, and that's the problem is like it's just really very specific low number of users. And not a lot of those users are developers. So it's one of those things where I like, I fear that if I ever want to have that work on

Wayland, I'm going to be the one that makes it work on Wayland. But so a lot of these specific programs for X that are very, you know, maybe industry specific, relied on stuff that was in X that now in Wayland is sort of a different paradigm. So things like getting input devices, window management, things like that, just the whole paradigm's different in Wayland. So that's one of the limiting factors right now. So, you know, it's not just a matter of changing your include lines and recompiling against Wayland. You have to kind of re-architect your software around a new environment. Actually, you could probably think about it as like,

I have this program on MacOS or Windows and I want it on X. And that's not a simple change. That's a new UI, basically. What desktop do you happen to be using right now? The one I have been rocking for a few years now is XFCE. Okay. So shout out to XFCE. I've tried a lot of the other modern ones.

I tend to do that maybe once a quarter. I should clarify, if anyone from XFCE is listening, please don't take that to say yours is not modern. I know it's modern. I keep it updated. But I really like the XFCD sort of simplicity paradigm and the panel at the bottom. And just it does what I want.

want and it stays out of the way. Sure. That's what I like. I, before that, I was using Nome 2, which we had in, in Fidoa. Wait, so you've been on X, that was a while ago. That was quite a while ago. It was a while ago. Yeah.

But with Nome 3, the change to Nome 3, it was, it wasn't, in my mind, it wasn't really a simple upgrade. It was a total change of how you think about the environment. And I didn't really like that. I found it kind of jarring and I felt I was less productive. So I was just like, I'm not really interested in this. I'll move along. But, you know, it's we have plenty of environments out there and they all do very different things.

And, you know, you can customize them the way you want. But yeah, I guess I guess I have been on XFCE for a while. Just a little bit. Just like, new G2 G2K versions. Yeah, that's true. Yeah, yeah. What do you use?

Right now I'm on Neri, which is, okay. Have you heard of Neri? It's a relatively new one, but it's a relatively new one, but it's not, no.

It's not a tiling window manager. It's a scrolling window manager. So it uses the tiling concept where, you know, the window, you spawn a window, automatically goes into a tile, but you have this infinite horizontal plane that you can work with. So rather than having just your monitor, you can expand out past that point and then slide along

like that plane, basically. Mm-hmm. Okay. Yeah. That sounds similar to just having virtual desktops. Yeah, you have it, yeah, you have the virtual desktops as well, but you can also just have all of those just on a single virtual desktop. Yeah, yeah. Yeah, that's... that's an interesting paradigm. Do you find it

productive or do you find it productive or do you find things getting lost? I have certainly run into issues with things getting lost. You've got to be kind of how would you say it? Very

not, what's the word? I'm blanking on words right now. Planned you've got to really think how you're using virtual desktop to keep things organized basically. Otherwise, you can very easily get things very lost. Yeah. Yeah. That actually reminds me of a feature, well, or an anti-feature, depending on who you talk to. In older versions of X, like when I started and we would have,

I had a I think it was a 14 inch display I'm sorry that's I don't know do you use inches for yeah yeah yeah yeah I yeah I pretty much anything we use inches full just displays

that's it yeah that's what I thought because it's like the same marketing materials yeah yeah yeah yeah so okay so you can you can imagine that one was it was small And I'm guessing on the size. I can't really remember exactly, but it was small.

And what was common at the time for X for Linux users is, okay, so you had 640 by 480 was your viewport, but you would actually configure the display to be 1280 by 1024. And what you would do is you could move all the way around. And so it's kind of it was you had two dimensions to scroll instead of just the horizontal one you're talking about. Some people loved that. And I could not work within that sort of. I was like, no, I don't want it. And so that's like, that's where the like that virtual desktop came, where you would have a page. And it's like you could just, you know, you have four, four, eight or something like that. That was a little easier to wrap my head around than having like,

thinking about the display being physically larger in front of me, but my viewport was small. But yeah, people would do that and it was really weird to see them use their systems. There actually have been some modern re-implementation of that that have also added some tooling in to sort of make it somewhat more productive, adding things like a window overview so you can zoom out and then easily more easily find the windows. I can see that. It's not a model I can kind of get behind, but I like the, really I like the people who are experimenting with different models, even if it's a thing that nobody really cares about.

I like the people are trying it out, see if it works. If it does it, hey, move on, do something else. Absolutely, yeah, yeah, no, I find this kind of stuff, you know, it's always interesting. Like, I don't spend a lot of time configuring my, my desktop, but I'm always curious if maybe there's one that I would like more. So, you know, a couple times a year, I probably fire up all the different ones that I can think of. And eventually I get tired of doing that. And I'll say, like, okay, well, I looked at all of these. And I'm always curious what KDE is doing.

I remember when that was announced initially. The cool desk of environment. Right, right. And, you know, I didn't know anything about cute or Qutee, as some people say. I say both just to annoy everyone. I, so I tend to say cute now to someone who doesn't know that's how it's pronounced. and then I'll say QT to to like KDE developers

or like QT developers who are you know they're going to get annoyed I don't know it's just like it's kind of a bit of fun and they're like you know what you're out there I'm like yeah I'm just messing with you but yeah I remember when that came out and I remember the screenshots I want to say it was

you can correct the date I'm going to guess 1997 initial release was 98. Oh, okay. But there was an announcement 96. So could have been somewhere in that period. Maybe you saw the beta or something.

Yeah, yeah. And I thought the screenshots were really, really cool. Because at that point, you know, we had, we had FWM and sort of all of the fan outs from that. And, you know, you can have some really, really cool looking desktops. There used to be a website called themes.org, which was collecting all of these FBWM themes

and later other window managers. And that went away. But that's how we would have these really cool setups like you would see in a movie or whatever. And people would spend days setting up the background graphics and patches to terminal programs. stuff and it, but you could get like a really, really slick setup. And then of course, you would want people to see it, but it was before any of us had laptops. So, you know, you're like,

hey, come over to my house and look at my desktop. But now you can do that on a laptop and you get, you know, the person at the coffee shop or the airport sees it. They're like, whoa, what are you running, you know, which I always think is kind of funny. What I like doing, especially going through an airport is I like to take my laptop and just for the travel day, I will run a theme that makes it look like Windows 95. Okay.

I was not sure where you're going to go with that one. I thought you're going to be like, oh, I like running something that looks like a hacker from. That's going to and still have a TSA. Well, yeah, no, no, that's, well, I do have a story about that, but the, I

just always, I, if I'm in an airport, I, I, I, want, I want, you know, someone to actually see them. We're on an airplane. Like, like, just do a double take and think, is this person really using Windows 95 this year? Like, and then they're curious what you're doing, you know. And it's, I don't know, it's just kind of a fun little gag.

But I've done that a few times. I've definitely got some weird looks. Regarding the, the KD1 thing, they're actually, this is what I love about open source. there is a developer who decided I'm going to go and revive KDE1 He decided to just go and fork the project Yeah

Port it over the Q2 because Q2 Because Q2 has a couple of nice additions to it Like some scroll wheel support for example So what happens when you deal with 90s code basically Sort of a lot of basic Things are missing Yeah And he wants to eventually,

he wants to eventually basically do a rewrite of the back end and port it over to Wayland as well. That would be, you know, that would be really cool to see. Send me the URL for that, if there is one or wherever. Yeah, I think he's got a refo. Let me see. Okay, yeah, I'll send, I'll email your link afterwards. You can get it set up right now,

with the development preview in Debian if you want to. Actually, I can send in the chat here, can't I? Yes. How do I get the chat here? I don't even know. There it is. That's the chat. Okay.

There you go. Oh, Libra Next.com. Okay. That would be fun to see. It's very buggy right now because it's in the 90s code. Yeah. Oh, I so remember this. Oh my gosh.

That is, I definitely remember because when KDE came out, you know, at that point in time, anything that was new that was coming out and you were a Linux user, you were downloading source code and compiling it. So my friends and I, you know, we're like, let's build KDE, but it was all in C++, which G++ took even longer to crank through all the code on our. slow systems. But yeah, we spent, I think it was like a week building all of the source packages for KDE1. But it ran. It looks really slick. I thought it was pretty cool. But then Noam was announced after that because there are gear apart for each other. Yeah. And what I remember from that

is it was the pushback on using QT. Yeah. At the time it had the proprietary license. Yeah, it was, which, yeah, I get, you know, that's, that's, you know, and they, and they, because GMP was already out since the G and GTK stands for GMP, though it probably stands for GTK toolkit now, but yeah, yeah, I think it is just GTK at this point. Yeah, but that's where it originated from. And they said, okay, well, we're going to do a desktop environment and we're going to use these GPL libraries as the base, which yeah, you know, it's kind of cool. And I mean, I think they're both really nice environments. I don't use them on a daily basis. So I don't know the, you know,

the warts and stuff that people run into from time time. But they seem to both have like, from my viewpoint, roughly similar sized user bases that, are that are like deliberately, I'm using this because I want to not, yeah, yeah, not, I'm using it because I installed a distribution and this was the default interface. Like I'm not, I'm not counting those, those users. Right.

A good example of that, just with Fedora, right, is, uh, last release, released before, whatever release it was, um, promoting Fedora KDE to be alongside Fodora workstation where now they're both right there on the home pay. You don't have to like dig through the spins to find it. Yeah. And, and, you know, that's nice. We had made that decision a while back. And also for, I guess for the audience, because we didn't mention it at the beginning.

I am on the Fedora Engineering Steering Committee. I'm also on the Fedora Council. I have been involved with Fedora for a very long time. We have the current stable release is Fedora Linux 43. And the first one that I had a hand in was Fedora Core, I think, three or four. So I've seen a lot of Fedora. But yeah, there was a push years back to say, you know, we want to promote, you know, kind of a primary experience for users.

That's not to say that we were going to remove other desktop environments. But feedback we were getting from people that wanted to use Fedora who were not technical users, you know, they're not going to jump in and start building software and submitting patches. They just want to use a computer. They were confused. They're like, why is it asking me? I don't know what this means. Do you want Nome or KDE or LXDE or XFCE?

I don't know what any of this is. Why is we like, and that's fair. So we rethought, like, how do we present these? But now you skip ahead and on certain hardware, MacBooks, anything that Asahi runs on, KDE tends to be a little farther ahead in terms of support on that hardware, the Gnome does. Now, I haven't tried it in like a year, so that could be totally different. But I know that was a big push because people were coming. end of the project, joining the community, they would have a MacBook with one of the Apple,

M whatever processors, and they want to run Linux. And so that's Asahi Linux, which is part of Fedora now. And they would be surprised. They're like, I can't, it's only giving me KDE. Are we like really Fedora? Or are we, you know, what's going on here? So I think it was okay what we did. promoting that because, you know, there's, there's no reason that people shouldn't use it.

So it was, now, the other ones, like, we haven't received really any request to, say, promote XFCE to that level. But I don't think those users are really in the same category as, am I on Nome or KDE? I think it's, you know, somehow it's different. Well, I think specifically with XSC, I think first point of order is getting their Wayland version ready, right? Yep. They are working on that. That is something that is coming.

But if we're talking about, like, what is, and Fonora's always been in this distro, which jumps ahead and does things, argue, I would argue a little bit too early in some cases. When Fedora enabled Whalen by default, I want to say it was, oh, 2013, some really early year. Yeah. No, I don't disagree. Sometimes this stuff is too early. But yeah, sorry, I interrupted. No, no, no, that's good. It's good.

With a distra that sort of has that legacy of wanting to push forward modern technology, making a distro which making a like a desktop that only has the X11 version ready put alongside like Gnome and KDE. It doesn't really fit with sort of what the distro is trying to do at this point. Absolutely. Yeah. And it it adds a lot to the testing matrix and sort of, you know, what has to be ready to actually, you know, call it the release.

And so yeah, the work the work builds up that way. Right. If you're going to have an X version, then X has to be a blocking, a blocking problem. And with Gnome and KDE dropping the X11 versions, you know, I've spoken to a number of Fodora people that do not want to deal with X11 at this point. I'm very happy to just not have that be a blocking problem anymore. Yeah, yeah, no. And it's, so definitely Whalen by default was not without controversy or, pushback, but I will say that doing that infidora does allow us to really take those new technologies and iron them out. So maybe you can default in whatever release, and then that that bakes for several releases, and then it's really just solid. And we've done that for a number of

things. And a lot of upstream developers on these things, on various projects are also working on Fedora. So it's kind of a natural fit to do it there, I think. I was a little bit early. It was 2016 that it was made default for Gnome. That's not right. Yeah, Fedora 21, so that would have been 2014, something like that. It was introduced.

Yeah, okay. Yeah, that sounds right. I can't remember. Yeah, this is just what I'm seeing in the AI overview that shows up on the search, and it could be wrong. I don't know. That sounds somewhat right, though. Yes, yeah, agreed. So you've been jumping around.

I've been jumping around. This is great. I brought you on to talk about sort of Slackware and Liddix history, and then we've been talking about Fedora for the bar. I was like 20 minutes. Yeah, I know. Yeah. No, I like it. It's fun.

You got a lot to say. I do. Yeah. Excuse me. If, you know, if this, if this ends up being not usable, I'm happy to do a more organized session. But yeah. So, yeah, I know you had questions about Slackware. And I know, I think the last thing we talked about was X, which took us down that.

Yeah. Yeah, somehow we got there. So when you got started, my understanding is Slackware didn't yet have a web, like, a web presence beside like the, the FTP server. Correct. Yeah. So sort of one thing that, and I don't remember the exact conversation with Patrick, but we were interested, a couple of us at, university, we're interested in providing, we're like, well, you know, a lot of projects have a website, we can put one together for the project. It's kind of weird. That's even a thought, right? Because now that's just, that's just the default thing you do. If you have, even, even it's not like a complete product, it's just make a website, buy a URL, just throw it up there.

Yeah, yeah. And we, we put together a website. He liked it and wanted to go forth with it. So we, you know, that's, that's how Slackware got its, its web presence was, uh, us sort of saying, hey, can we do this? And, and then we did. Um, and we, because he already had Slackware. com. Hmm. He had the domain name. Uh, and then we, you know, we set up a server and, uh, put it on there. And, uh, yeah, we, we did several iterations of the site. Um, everything. was, well, I will say that that was Slackware.com working on that website was the last time I really did any web thing. Not really a web guy myself. I'm not. I'm not, but it sort of brought me into the Slackware fold where I could start working on other stuff. But we did, the site was, you

know, it was HTML written by hand. And later we used PHP, which was new at that point, PHP version 2. And we could do a little more dynamic content. All of the graphics were done by us in the Gemp, what few or what little graphics there are. And then the biggest thing I think we added in, 2000 was the online discussion forums that we put on the website.

And those got to be really popular because at that point, we didn't have, we didn't have like a discussion mailing list for the project. What we monitored mostly were Usenet postings, which was really just a subset of users and not everyone knew how to get into that. And then there were also IRC channels, which we didn't really go into those. Those were just community members. So we're like, you know, it would be nice to have a discussion forum because, you know,

people are asking us the same question. Why don't we put it here? And we found this guy had written an open source sort of discussion board software called Forum, P-H-P-H-O-R-U-M. And it was written in PHP. So it was a nice thing that we could integrate into the site. And at one point, the author of that software came out to California for something. I wanted to meet us.

He's like, yeah, you're one of the biggest users of my software. And, you know, just wanted to thank you. And we would provide him feedback and patches here and there. I wrote an import tool when he re-architecteducted a bunch of stuff. and we had to extract all of the discussion like threads and then re-import them into a new database because it was not compatible. And so I wrote, I wrote a tool to do that. And I was like, hey, you know, someone else might want this. And he said, no, probably not, but I'll take it anyway. Yeah, he was, he was nice. And yeah, it was, it kept running after all of us left. But I don't,

I don't think it's on the website anymore. I don't think it. Yeah, I couldn't see a link to it. It seemed like it is gone. Regarding forum, their latest version was from six years ago. So it seems like the project is dead at this point. Yeah, that would be my guess. Yeah, started in 1998.

I feel like this is software that served a purpose at the time. but it's no longer really I wouldn't I don't think I would use it now to set up a discussion board and not not because it's bad just because there's not really a need for that. Sure. Yeah, yeah, yeah. We've got other other solutions. And also most like it's so easy now to get an open source project like from the ground up, you know, you could go to GitHub and you suddenly have a way for people.

people to have discussions. You have ways for people to send you patches. You have ways for people to open issues. You have a wiki where you can add user contributed documents. You know, all that stuff that that large projects have, but took time to set up. And the vast majority back in the 90s, the vast majority of open source, you know, scratch and itch type software that you would find would just be dumped on an FTP site. You know, and because that's, It's all the developer. It's like I want to put it out there, but I'm not going to make this huge project.

So you would go digging and FTP sites to find stuff. And now I think GitHub, Codeberg, certainly, and GitLab maybe to a lesser extent, they have these nice environments for that. And the idea is not really new. Like I don't know if you remember SourceForge. So that came out of a company that had started as VA research, which made Linux workstations like custom built.

And then that became VA Linux. And then they created sourceforge.net, which was an early sort of thing like GitHub. You know, this is before get existed. And I don't know. If anyone listening to this or watching this, if they remember that, you know, early on,

the Linux kernel was not in version control. It was just released as G-Zipped TAR files. Or you could download a big patch and apply it to the source tree you already downloaded. But patches, new things were submitted by email. Mm-hmm. And that continued for a very long time, even when many other projects were using CVS. That was really the only thing that we had.

At some point, we got subversion when CVS, people were getting frustrated. I can't remember some of the other. There were like a dozen that just sort of all blasted out around the same time that were going to quote unquote fix CVS, but none of them really addressed the fundamental problem of meeting the ability to have distributed version control. Yeah. And Linux first went into a product called BitKeeper, which was commercial. And that was, in my opinion, that was not a good move, but it did introduce developers to the concept of

distributed version control. But then with, I mean, I mean, you know, the story, how we got get out of that the need to, you know, we're working this way. We need a tool because what the license was going. I can't remember the license was going away or the company folded or something like that. I did a video on this a long time ago.

Yeah, yeah. Story. Oh, they had, yeah. right. They had given um they'd given kernel developers free access to it and then they were switching

they're switching their license or something uh I don't yeah I I I I'd it says the history on the Wikipedia page someone was to go exactly exactly what it was there there was some change like that which basically forced the creation of guests. Oh yeah, also there was some like drama where they didn't want people that were using

BitKeeper to go work on Mercurial. Yep, there was that and a number of developers had had reverse engineer the Right, yeah, yeah. And then

was that who was that that did that? Well, I remember very famously that happened and then the CEO of Bitmover, the company that made BitKeeper revoked that Curl developer right, like that license

like revoked his right to use the tool. Andrew Tridgell. Yeah, Tridge. That's who I thought it was. Right. I had to forget, this is the Australian guy. Of course, the Australian guy.

Why would you be an Australian guy? We show up everywhere just doing random. We're a tiny little island. No one lives here, but we're involved in everything. Yeah, you know, it's funny. We were talking about all the desktop environments. We never once mentioned enlightenment, which, you know, is still going. But, yeah, the, yeah, so I don't know why I got on.

Get an oh yeah source for those talking about yes yes. Yeah, we were talking about websites and now it's like like projects are it's so easy to just get all that stuff sort of immediately when you set up a project But that wasn't always the case source forge was created to make that easier for people But it its implementation was rough I mean even from day one it was like I am not moving a project here and and and then at one point. Let's see as in recent years, I say recent, it may have probably past 10 years, but as it was being passed around by different companies that they were just buying the property, one of the owners at some point, they were taking software that was on source for download, and then they were like adding like, I mean, malware. It was like, it was like they're adding.

like here's some additional shareware that we're going to put in there um and yeah it's it was it was garbage but uh i mean that was just the the unraveling of it all uh that that happened you know after it was long since past the point of relevance i think um but yeah it was it was big for a while just because there wasn't anything like it and then then gethub showed up and you know Things like GitHub and GitLab and Codeburg, none of those things would exist without Get, so they even, you know, owe their success really to yet another thing that came out of Linux. The reason I brought up the site is you briefly mentioned at some point, you didn't have an email until you got to university. And that's another thing which sort of is just taken for granted as well.

this easy access to email provide. You know, you don't have to go set up your own email. So if you can just like, you can just sign up for a thing and just have email now. Exactly. Yeah, that was not the case. Email came with your, um, ISP and, and sometimes like they wanted to charge extra for, or email.

Um, so you, you're like, well, I don't need it. So I'm not, I'm not going to pay for that. And also it was, you know, you would change ISPs, at least in the U.S. You would change ISPs frequently because the prices were all over the place. So you were always moving to whoever's the cheapest one this month. So you didn't want to keep changing an email address. So you just, you know, no one ever really bolted onto that.

You still see some outliers. Like obviously we had America online, which was its own sort of walled garden kind of thing. Funny that we're kind of moving back towards that. if we haven't already. You mentioned America Online. Do you know they're actually still operating as an ISP? Yes, I do know that, which is wild to me. And people still use AOL.com email address.

Yep, yep. And anytime I see one of those or I see like Yahoo.com, my first question is what company is running that? Because it cannot possibly be those companies are still running. I want to say Microsoft owns Yahoo at this point. I have no idea. Who owns Yahoo? Nope, some private equity firm.

90% ownership by Apollo Global Management and then which required and stay can come. And Verizon owns the other 10%. Oh, that's right. Yeah. Yeah, I remember when that all went down and they kind of split it up. But, but so, so, so it's probably Verizon that's running the email. Yeah, yeah.

But yeah, it's just, it is funny to see people using. Yeah. And I'm like, I, and then I, I feel the same way about a hotmail address. Yeah, how me, yeah. And I want to ask these people, I'm like, how do you possibly deal with that for email? Like, you can't get anything but spam on that address, right? Like, I mean, you know. Oh, haunt mails, but.

I think it was the Microsoft owned. Yeah, yeah, yeah. Okay, they merge that into Outlook. Yeah, okay, okay. Oh, you know what? I do remember that now. Hotmail. That was the one also where Microsoft acquired Hotmail and Hotmail was running on free BSD. I think it was free BSD. It may have been another BSD, but it was all open source software was the back in and they they tried to move it all to Windows and it failed so they moved it back to the open source stuff I always thought that was kind of a funny story so when you got started in Slackway

obviously this was well pre-Git what was the development like back then was there a source control system of any sort was it just throwing patches out Yeah. So Slackware, we didn't have any version control system. We would maintain the distribution trees and kind of work in there. And then we used R-Sync a whole lot to move things around. And that's that's that was you would know what the master tree is. And you just it was sort of, you know, And that's really the same way that commercial software was handled before any kind of version control.

You would have in companies, you would have a release engineer, basically, which was the person who was collecting all the correct copies of code and patches and putting it together in the release. That was all done by hand. But yeah, we didn't have anything. We just, we would keep patches. So Slackware would, you know, gather the upstream source and then create patches against those releases. And that's what you see in the source tree and Slackware. And then at that point in time, there were various build scripts that were used for each package. I think they now.

try to make those a little more consistent. So every package, you know, but the packages are built by a script that kind of drives the building and packaging process. And, you know, there was a lot of, there was a lot of internal knowledge about how the packaging was done. And that was, it was not all like in the scripts at that point. So you kind of had to know we had notebook. that we would write down the steps and what to do. It's like, okay, well, after you build this, you've got to copy these two files over here and compress this and delete this

because it's a temp file and stuff like that. And you know, kind of the early days of Linux distributions, everyone was kind of figuring out how to, you know, how to build and package software for distribution. And the idea of software packages that we are inundated with in the world of Linux, that's not new. That's from commercial Unix.

That's how the parts of the OS were built and distributed to customers. So it was in a way that idea was like a natural fit for a Linux distribution because you were assembling it from various projects. So you would have a GCC package and you would have this package and all that kind of stuff. And so it was a nice way to organize it on a per project. level that you were incorporating code from. But, you know, like, Debian went through various iterations before they arrived at the format they used now, which I think the current Deb format was largely originated by Bruce Perrin's,

I think. Maybe I could have. I'm not too sure on the, the format myself. But the, well, the Debian packages is pretty easy. it's an AR file. Well, I mean like the, where the origin of what came from. Yeah, yeah, go on. Yeah, but a Debian package contains, it's an AR file.

They may support other payload formats now, but it originated as an AR file, which if you don't know AR, that is the archive command in the tool chain that you use to create static libraries. So an AR file puts all the object files there. So a Debian package is an AR file that contains two tarballs. One is data.tar.pick your favorite compression format and then control dot tar dot compression format. And so the data package is, you know, that gets extracted to the root file system. And then control contains the metadata and the scripts.

And, you know, that format has been really flexible for them for a very long time. If you look at an RPM, which I spend a lot of time working on RPM-related tools and libraries, and the RPM format is also one that has been with us for a long time, but is far more complicated than a deviant package. There's a lot of header metadata. Yeah. And Slackware, of course, just uses TAR files. And those are extracted to the root file system as root, so permissions are correct.

And, you know, that's a fine use of the TAR format. There is metadata in a Slackware package. Like I think there's a package description in there as well as a post-install. script. I don't think Slackware, at least when I was involved, did not have like pre-install scripts or anything like that. It was just a single script that ran after the package was extracted, which in many cases, that's really all you need. And there was not a description. I think that got added later. But the biggest complaint, when I was at Slackware, the biggest complaint that we got from packages is the lack of dependency metadata. So you would install,

because there's none of that there and the tools don't gather it from anywhere else. But people would install a package and then try to run it and I would say, you know, missing some shared library or whatever. And then they would say, I don't know what package that said because it's telling me it's missing lib whatever. So so yeah, I mean, I get that that's confusing, but with Slackware just install all the packages, then you have everything. You've got it all there and it doesn't really matter.

It certainly doesn't matter now with storage space is much larger than we had back in the day. Yeah, I have, what do I have in my, actually don't even remember what I have in my system. Let's do an LSBLK. I have 2 terabytes 500 gigs there

another 2 200 gigs there another 2 20s about like 4 and a half terabytes in my system yeah yeah you know I mean like so what are we what are we talking about like

saving here like why are people like what's a full install of any of these distributions now it's it's you know Well, even with some of the larger ones, like Ubuntu ISO is 6.5, no, sorry, 5.8 gigabytes. 5.8 gigabytes is the ISO. Well, that's a lot of stuff. It's a lot of stuff. But even so, it's 5.8 gigabytes.

That's true. Yeah. Yeah. Yeah. It, I don't know. I don't miss having that conversation with Slackware users. you know, why isn't there a dependency information? You know, it was common. I'm sure it still comes up, actually, with them, but I don't know. It was like, how many times can I tell the same thing that people, but whatever. The other systems have their own problems, too, with RPM, you know,

we've got, you know, we have dependency. information in there, but it was a long time before we got the overlay tool that's now called DNF that is basically like apt to depackage where, you know, you say you want to install this package. It'll tell you all the other ones it needs as well, which is really that that's the complete story there. You know, it's one thing to embed dependency information in the package. And then it takes another tool to actually use it and do something useful with it. So that's what I think like app did for Debian and now DNF does for RPM. So, yeah.

So Slackware has the title of being the oldest distribution still maintained. Obviously, it's not the first district. It was based on SLS. And there are like three or four other things before Slackware. Yep. Yeah. So Slackware is. and Debian started about a month apart from each other.

And obviously Debian now is this you know, it was the basis for Ubuntu and Ubuntu is this behemoth as well. Debian is this, you know, it's this massive server distro and Slack was obviously still

around but it's kind of just there just being maintained. like just sitting there like what do you think sort of you were there earlier like what do you think sort of happened with that like shift between the two

where debby and has become this like critical infrastructure and slackware is still around but and it has its fans but it's it's never really had that same level of adoption yeah I well you know I don't I don't think anything's going to happen to Slackware like, you know, if that's what you're asking. No, it's more like, you know, people, it's been around so long, there's always going to be someone who cares to keep it, to keep it going, right? Sure, yeah, yeah. Well, I know, like, we, there are actually a lot of commercial installations of Slackware, believe it or not,

and we would we had a lot of commercial users but we weren't out there really pursuing it right canonical does or like red hat does or susa does and i i think that that pat really wasn't that that that wasn't the thing that that drove him to make slackware and he was not really interested in becoming that sort of business focused, you know, we're going to go out and, you know, do whatever we need to do to secure these contracts or whatever. It was more like, look, this is the thing we're making. Right.

If it works for you, I'm happy to have you use it and we'll, you know, help you with it. But, you know, it's my project. I put it together, you know, and this is what I'm doing. And I think that's probably why you haven't seen it sort of really at the same level as say Debian, given that they were around the same time. Like Debian definitely, like Ian Murdoch, who created it, he definitely did go down the business path of, you know, finding the, the, the, the consumer sort of or the commercial success paths.

I mean, he founded progeny systems later on to sell commercial support for Debian. He brought people into the Debian fold that, you know, promoted its use in commercial settings, you know, and really built sort of that that community around it to, it to keep it open and self-governing and things like that, which has given it this sort of inertia that I think has allowed these other things to build off of Debian and rely on it, knowing that that's there. And we never really had anything set up like that for Slackware. We didn't have like a community of elected members, you know, making decisions and things like that. It wasn't, it was never set up for that.

And I don't know if it's right or wrong. I mean, it's just, I think, different approaches to it. There have been a lot of commercial attempts at Debian. And like, you know, I mentioned project. Before that, you had Corel. The Corel Corporation did Corell Linux. And that's actually why projects like Wine gained a lot of functionality is Corel was investing in the Windows compatibility layer. It's also why KDE got, I don't know if you, you.

remember on on well certainly it appeared on windows 95 but later versions would have the network neighborhood icon and you could open it up and it would show you other windows computers right yeah there's okay so that network neighborhood browsing um kdee gained that and that was contributed by krell um a later known as my network places my network places yeah and you know that that that that that browse capability, support for that. So it worked in the desktop. Karell did that. Of course, Krell Linux didn't really go anywhere. Progeny folded.

And then the next one to come along was canonical. And I was like, when I heard there was another commercial Debian that was coming out, I was like, really, guys? You're going to try this again? And then it was, and I, whatever the guy's name, Shuttleworth was sending, sending out CDs. You just would ask for him. And it was like America Online in the 90s where they were sending out CDs and floppy disks everyone. And it was just, but there was no, there was no cost to it.

And I was just like, okay, this is, this is interesting. but what are you doing to make money? Like how, like what, like you can give it away and have a bunch of users, but who's paying for this? And I remember he was talking, I think it was a DebConf where people were asking them like, like, what's revenue stream here? And he just said like, you don't have to worry about canonical running out of money. And it was basically he was saying like, I'm just super rich.

and we'll always fund it. And I think someone said like, yeah, but that's not a business plan. It was basically the Silicon Valley business model for anyone else was doing it. It's just burn money until you have market share. Right, right. And I mean, and then once you have market share, then you can abuse the users and start extracting money from them. But I was just like, this, I mean, okay, fine. If this is what they want to do,

cool, but it sounds like he just wants to do this project and surround himself with fans. And okay, that's cool. If I was a billionaire, maybe I would do that too. But I don't know. I guess out of that, there is a variant of Debian that people like. But I, and I don't really know how closely. Ubuntu works with Debian upstream. I know that they,

you know, they're based on Debian. Yeah, but I don't know which direction contributions flow, really. Hmm. So, yeah,

I don't know. And then there's Sousa, which is, you know, kind of its own thing out here that has, has had peaks in valleys and, and as the, uh,

industry has evolved, never really gaining a huge amount of market share in comparison to the other distributions, but always sort of steady state. I, you know, like Susan initially was, it was going to be a German translation of Slackware that was only going to be sold in Europe. And then they kind of reinvented it. They hired a guy, Florian LaRoche, who had made Jurex, I think. And they kind of rebranded that as Susa Linux. And then it started appearing in the U.S.

But it was, it was a different distribution entirely. And I remember using or looking at Susa in what version was it? I remember it was an insane number of CDs. It was like six CDs of stuff. And it was just an enormous amount of software. And the installation process was so tedious. It was like one.

I had so many details. And then I got on the system. And this is just me, you know, having used Slackware, having worked on that, now at Red Hat, familiar with those systems. And Susa just really felt sort of like kind of clumsy at that time. It may be very different now. But I was just like, this system feels like a mess. It was, yeah, I didn't really.

I don't know. And I don't know. Are you using Sousa? I'm on Achelelinic right now. Okay. Yeah. Yeah, it was, it was different. But I don't know. I don't think that Slackware really had an interest in becoming a big, big commercial thing. They were, you know, Pat was more interested in creating something that he liked to use that that other people would use. And it has found lots of commercial use.

It's just not since there's no, there's nowhere to talk about that. Right. There's no, you know, we have the Red Hat Summit trade show. You know, there's nothing like that for Slackware, which those kinds of things can be talked about.

But it's really cool, actually, that it's active still. And there are a lot of users, you know. And it's quite interesting to me now in 2026, I still get questions about my time working on Slackware. So it definitely is a distribution that has an impact on new users. It's people just coming into the open source world,

into the Linux world, and they find Slackware. I think it's still true that Slackware is probably one of the only distributions that's going to make you really become comfortable with the parts that make the OS, you know, you really need to understand it. Whereas a system like Fedora or Ubuntu or Red Hat Enterprise Linux, we spend a lot of time really sort of papering over all of that kind of stuff to make a uniform experience. I mean, and there's good and bad, you know, I see there's like Arch Linux. Yeah, yeah, I'm on Audra now. Yeah, Arch, yeah, which I think is another one that that is a good system to, you know,

if you really want to learn how a Linux system is put together, you know, Arch is a good system for that. It used to be Gen 2 was one that people would tell you to go. But I don't, I don't think that one does as good of a job really. Because there's a lot, I think there's been a lot added to Gen 2 to kind of make it uniform and gin tooy so that it's not you're you're learning the Gen 2 way to do something as opposed to you know what they provide a lot of um gen like the the gen 2 package manager it provides its own specific way to configure the build options and things like that right yeah and that's that is specific to gen 2 that's not really really a general Linux thing, which I think, you know, if you want to learn, you know,

kind of what really goes on to put a system together, Slackware is a good, good place to start. Just a bit of a side tangent with the whole Debian and SlackWestle. People obviously know the like the naming of the Debian releases. I think all of people don't like know why. And a big part of that was early on with the type. with Pixar where you had the early mailing list hosted at Pixar like that's how it's so even from that like earliest point of the distro it already had some sort of

not like it's funded by Pixar it's like already like a foot in the door in a corporate space yeah totally yeah and that was I'm pretty sure that was Bruce Perrin's when he was working at Pixar and that's where those names uh originally originated, which, yeah, you know, and definitely, you know, he was, he was certainly using it there and, and putting it to use at Pixar, which is, you know, that, that's, that's, yeah, one more leg up there for, uh, for, for, for, for, for debby. So I, I had to get to it eventually. Um, you are technically a published author because you're on the Slack Web book. yes, yes, I am, yes, that that's right.

You mentioned obviously earlier that sort of books to inform people how to use Linux for a really common thing back then. I'm assuming that's sort of the reason why that existed. Yeah, it's, I mean, you didn't have another, you didn't have a place to start, I guess. That was, you know, probably just to help you get going.

Because if you're, you know, you're starting from nothing. Yeah, I mean, the books, like you, well, okay, so you can have like different kinds of books, right? You can have like reference materials. So if you're familiar with the subject, but if you're just starting out with it, you, you almost need like a guide. Like, you know, I'm going to take you through this stuff. So, yeah, that was, that was kind of the objective. with that Slackware book that we wrote as was, you know, let's not just put a book together that that's all the man pages, you know, like, like a new user, they don't know what the LS command is,

but new users know that there are files and how can I find my files? You know, so like that, that was kind of the approach we wanted to take with it. But it was replacing a book that we had been including with Slackware that wasn't even written for Slackware. It was written more of as a generic Linux book. But I think it was being included with like in the box set that you could buy from one at Creek C-Rom because there wasn't another book. And it was roughly the appropriate.

it weight and thickness to go in the product so that people felt like they were getting their money's worth. You know, like, because it wasn't, it wasn't like super thin, you know, but it also wasn't the one of those huge books where most people know that 90% of that's going to be just recycled garbage. Right. It was that sweet spot. So they wanted us to have a book that was, that was like that, but that was actually geared

towards. slackware and usable, you know. So yeah, that, that, I'm technically a published author. I mean, three other people, but. That's true. Yeah. So Chris was the, the main, the main author on that. And then Logan and I contributed other sections to it.

And we, we had to write that at the time, the, the, uh, sort of the up-and-coming, I guess, documentation, type-setting system, whatever that was the new hotness at the time was SGML. So the book was all marked up with that. And SGML had one major lacking thing. There was no way to generate an index. So we had to so the so I'll say okay so the Walnut Creek was also having the free BSD team do a book as well and so we were asking them you know what do you and they're like oh you need to use SGML tools and all okay great and then we got to that like you know we need to make an index now and Walnut Creek C Cibron was like you got to make an

index. But there's not there's not way to kick one out. And they're like, you have to have an index. And so we went and we asked the free BSD guys like, oh, yeah, what did you guys use to do an index? And I'm like, oh, we just didn't do one. And you, you open the back of the book and there wasn't an index there. So I was like, cool, that's great. So I think, I think we had a conversation with them. It's like, well, the free BSD guys didn't do an index. So do we have to? And they're like, Yeah, you have to. So, okay. So then Logan and Chris, well, I think we all worked on it.

But I forget who, we wrote some pearl code to generate an index. And it went through many iterations. And then we had put like, I remember we put things in the book so that we could like test to make sure that it was working right, things that would show up. And yeah, oh, God, that was some late nights. But yeah, we got the index generated. Seemed to work. Of course, now you mention SGML and everyone's like, what?

But, yeah. Chris later on ended up redoing it all. in LaTeck. A far more sensible choice. Yeah. And, you know, looking back at it now, I'm not sure why we went. Well, I'm pretty sure we went with SGML because that's what the company was kind of saying, like, use this. But, I mean, we had, we had Latak and Slackware, you know, that, in fact, that was one reason that a lot of people would go to Slackware.

They were, they were in education or whatever. and they needed to work with Latec. And they're like, you're, you have it included. And it's all, it's all ready to go. So I'm, I'm going to use this. But yeah, we should have used that. We should have, we should have pushed back when I, when I look back at it now. We can probably just keep going for a long time.

But one more thing I want to do before we sort of end off is that, is there, is there some, like, highlight from your long, long history doing this that you, you want to mention that maybe we haven't talked about just yet. Oh, gosh. Well, does it have to be Slackware specific? No. Red Hat's fine. Whatever's interest you.

Well, okay. So one story that I like is a good friend of mine used to work at NASA at Johnson Space Center. And this was after my time at Slackware. by now I was working at Red Hat and he he worked at NASA for seven years and he worked on he worked on a couple of teams but the first one that he worked on was the team that was responsible for maintaining the treadmill that's on the international space station and and there have been several iterations of the treadmill and the one that he was working on I forget what they called it it doesn't matter now the The treadmill is a very complicated piece of equipment on the ISS. It has to float freely in space.

It has to record vital statistics from the astronauts using it. And it also has to work. If it does not function and they can't get it repaired in a certain number of days, then they have to evac. because it's critical to prevent severe muscle atrophy. Right. So they have to work out on that thing a lot. And the thing has had daily maintenance, weekly maintenance, monthly maintenance.

Okay. Well, he contacted me one day. He's like, hey, man, I'm working on this, like, really, like, problem that's just ragging my brain here. I can't figure it out. And I'll add that he is an electrical engineer. And he's a software guy out of necessity, but that's kind of not what he focuses on primarily. So he was asking me a software question. He said, yeah, I'm working on the treadmill.

And we have these data cards that come back. The astronauts bring them back. And they each have their own data card. And it records data for each time they use the treadmill. And so, you know, it'll tell it's the vital stats. But then it's also like how long they used it, you know, what. or did any of the sensors not work, all that kind of stuff. And so they would get these cards back and they would,

they would dump the records. And they came back from some flight and all of the cards were corrupted. They couldn't read anything. So they're like, we're kind of panicking. Like, like, it's the same with the backup treadmill. Like they're all, like, broken. And I don't know what's going on. And so we went back and forth.

And I'll set this up for you. So the ISS, you know, is pretty old at this point. And the equipment that was going on it was built well before it launched. And it was, the technology was certified, you know, to fly to space. And so my point here is you can't just take an iPad and take it to space. There is a whole process that NASA has to go through to clear things for flight. So the, I mentioned memory cards. the treadmill was not USB.

The memory cards were PCMCAIA memory cards. So PCMCAA is a 16-bit bus that used to go in laptops for expansion cards. You could also have memory cards, and they had four megabyte memory cards that the astronauts would use. They would bring it back, and they would dump them. What was happening is at this point in time, the ISS and this trance, and this trance, treadmill had been in use for a very long time. Missions were getting longer. And I worked with him and I said, okay, well, what does a data record look like?

And he's telling me. So I wrote like a header file for that. And I said, okay, well, can here, try this. Dump, you know, and what we found out is that records were there. But what was happening is that the treadmill, when it got to the end of the card, it just loop back around and kept writing it up. So on these long missions, it was just going back to the top and then right. And I said, okay, so you're going to have, to dump this data, you're going to have, you know, you can have like two corrupted records or two incomplete records. But the rest of the

ones are there. It's just we just have to seek until we find the beginning of a new record. And he's like, okay, I think I get it. And I said, you know what? Here, I know what to do. And so I wrote a small command line program that would take one of the raw data files from the card and kind of recover every record that it could find. So, you know, let's see, it's supposed to have, you know, 500 records and it would throw out the two corrupted ones and, you know, you would have 498. So I slapped, you know, LGPL on there and said, here you go, told him how to run it. And he's like, yeah, this is great. Yeah, I'm able to read all the cards. Thanks.

This is super helpful. And I kind of explained what was going on because he wanted to be able to explain it to his team. And I was like, yeah, cool. That's neat. And months went by. And I asked him about that. He's like, oh, yeah. So fun story about that.

we developed a new procedure before astronauts return or actually after astronauts use the treadmill, they are to take their memory card, float over to the return spacecraft, which has a think pad in it, insert the memory card, and run your program to dump all the records to a text file and then clear the card for the next use. okay and that became part of the official use the treadmill procedure i was like wait hold up are you saying i have software in space i was like yeah you have software in space and he's like but don't get too excited because that treadmill is going to be retired and then we have a new one coming out i was like you know it's still kind of cool so i told people about it well then skip ahead to when that treadmill's

retired. I'm sharing this story in some bar and he's there. And I get to this point. I'm like, actually, come to think of it, whatever happened to the old treadmill and the and the think pad with my software on it. And he's like, oh, yeah, that was deorbited. And I was like, wait, what? He's like, yeah, when we sent the new one up, they bundled all the old treadmill stuff together, including the think pad with your program on it, bundle it up. And then they push it out. the ISS and it or it goes it falls back to earth and it burns up in the atmosphere and another friend of mine was like that is the coolest story to like the end of software that I've ever heard so I wrote an open source program it was used in space and then it was burned up in the atmosphere

once it was done I can't think of a better story than that you know No. That's a pretty high bar to clear. Yeah. It was pretty cool. But yeah, that's my NASA claim to fame. Yes, I've had code in space and burned up in the atmosphere, deliberately. Okay. Sorry if that was kind of long. No, no, that's totally fine.

Totally fine. We need you to have a Wikipedia page now. Mark Scheneworth, he's been to space, but he has not had his software launch back to Earth. Yeah, exactly, right? Yeah, feel free to start one. Probably bad for him to start your own. Probably.

Probably. Yeah. But I'm sure I have plenty more stories, but that's one of my, That's one of my favorites. Just because it's so common for us to write software that we open source and then we move on from. And years go by and people will ask you questions about stuff. And you're like, well, it's really cool that you're using it.

But man, I don't even use it anymore. I don't know. I don't know. You want to take it over? And then they they don't. So having a closure to an open source project is really kind of cool. And it involves space. Yeah.

That was, that yet, that was actually, again, I've not heard a story cooler than that. I don't know how I'm ever going to have that one topped. Well, yeah, I, uh, yeah, I'm sure I have, have more, but. I need to find somebody wrote something for the Mars rover. Maybe that'll top it. Yeah, right. Yeah. well um that's as good a place as they need to end it okay well well thank you and i i hope this is useful

this was uh you know very fun thanks for having me absolutely pleasure um yeah if you ever want to come back on and do another one talk about either maybe more focused on red hat or something like that uh yeah i would i'd love to do something and it sounds like you're a lot more stories to talk about anyway, so. Yeah, you know, you're around in this long enough. You have, you have a lot of stories. I'm now thinking of some I didn't even get to for Slackware. So I don't know, maybe we do a part two at some point.

Okay. You have more questions. Yeah, yeah. No, it was a lot of fun to talk. I enjoyed this. It was a really good episode. And I hope the viewers enjoyed it too. Yes.

I hope so. I'll do my sign off and then we can end it unless you, Do you have any, like, places you want to direct people to or nothing? No, I don't. I'm not, I'm not promoting anything. Fair enough. Thank you. Check out Fedora.

Yeah, do that. Yeah, get fedora.org. Yes. I guess if you're in, if, you know, this is your time to do your red hat marketing piece you really wanted to. Go buy red hat license. Do that. Tell your employer. Yeah.

Tell your employer. license. Yeah. Okay, I'll do my outro and then we'll sign off. Okay. Okay. My main channel is Brodie Robertson. I do Linux videos there six-ish days a week. Sometimes I stream as well.

I've got the gaming channel, Brodion Games. Right now I'm playing through Shenmoo 2 and Metal Gear Solid. Maybe, I don't know, when this comes out, maybe I'm done with one of those. You'll see. And I've got, if you're watching the video version of this, you find the audio version on basically every podcast platform on Spotify. we have video as well if you like Spotify video for some reason. I don't know why.

We have an RSS feed. So if you like that, grab the RSS feed. And the video is on YouTube as well. I'll give you the final word. How do you want to sign us off? Oh. I never tell anyone they're doing it. Oh, wow.

Let's see. I think, you know, keep exploring open source and keep submitting poll requests to your favorite projects. Okay. Yeah, good way to end it.