''As a personal note, I do not like this decision. To me LFS is about learning how a system works. Understanding the boot process is a big part of that. systemd is about 1678 "C" files plus many data files. System V is "22" C files plus about 50 short bash scripts and data files. Yes, systemd provides a lot of capabilities, but we will be losing some things I consider important.
Linux is literally 62k C files. The amount of time you'll spend understanding how Linux works will dwarf systemd. At least when studying systemd you will be learning a more modern approach of init systems.
I don't mind the inevitable death of System V. It's an archaic relic of the Linux era.
Going systemd-only is not necessarily a good choice (though I do understand it from a practical point of view). There are other, better alternatives for System V that are smaller and more modular so you still get the Unix "feel" without the absurd complexity of interlinked shell scripts that System V relies on.
I'd like to see OpenRC getting adopted in System V's place. Upstart seems to be dead (outside of ChromeOS) but it would also have sufficed. Alas, I'm not someone with the time or knowledge to maintain these tools for LFS, and unless someone else steps up to do all the hard work, we'll probably see LFS go systemd-only.
That said, there's no reason to go full-fat systemd, of course.
I'd just like to interject for a moment. What you're referring to as Linux, is in fact, Linux plus systemd, or as I've recently taken to calling it, Linux/systemd.
Linux is not an operating system unto itself, but rather another free component of a fully functioning systemd system made useful by the systemd corelibs, systemd daemons, and vital systemd components comprising a full OS as defined by Poettering.
I'm with you on this. SysVinit is better than systemd, but far from perfect. I don't enjoy tediously maintaining all of those symlinks, and prefer the BSD approach myself.
One project on my distro is a new init that will be much, much simpler than SysV, more like BSD and without all the years of cruft.
It's a pity. It's also a step back from valuing the Unix philosophy, which has its merits, especially for those with a "learning the system from scratch" mindset. Sorry, but I have no sympathy for systemd.
>The second reason for dropping System V is that packages like GNOME and soon KDE's Plasma are building in requirements that require capabilities in systemd
Do people who really uses LFS even want GNOME or KDE on their system ?
That's funny, I did LFS a few years ago and specifically chose the systemd version so I could better understand it. I don't think this is a huge deal, I believe the older versions of the document that include SysVinit will still be available for a long time to come, and people who want it will figure out how to muddle through. If at some point in the future things diverge to such a point where that that becomes untenable, someone will step up and document how it is to be accomplished.
SysV init was the overengineered cousin to BSD init and I never liked it. Easily my least favorite of all init systems I've worked with over the last 30 years. On the flip side, daemontools or maybe runit were my favorites. Lots of good options for init/supervision tooling over the years and SysV was not among them.
It's always a little amusing when the Open Source Tea Party bemoans the lack of "the UNIX way" and someone else with actual historical experience (and not misguided nostalgia) brings perspective.
On a related note, X11 was never good and there's a whole chapter in the UNIX-HATERS Handbook explaining why.
> Understanding the boot process is a big part of that. systemd is about 1678 "C" files plus many data files. System V is "22" C files plus about 50 short bash scripts and data files.
Systemd is basically the Windowsfication of Linux. I'm always surprised by the people that champion it who also used to shit on Windows with the registry or whatever.
LFS seems to be for people who are interested in how things work. The systemd proponents come off as people who would question why you would want to to drive a manual transmission and say of course you should choose an automatic or better yet, a robot; self driving car.
It would be interesting to see how those opinions line up with the uses of AI
I have owned many manual cars, but I'm fine with systemd on all my machines. Being a nerd about one thing doesn't require me to be a nerd about other things.
Pedantic but systemd is inspired by MacOS launchd, not by Windows services. It has nothing akin to the registry, which even microsoft admits is a pain on windows.
Oh, and usually people shit on windows for many reasons, but some of the very core features of the OS are robust and the Linux crowd could take a hint. Like, you know, the notion of service at the OS-level and not some random bash script that nohup'd a binary. Oh wait, that's what does Windows, MacOS and Linux with systemd.
I didn't say it was inspired by the registry, I just drew a comparison. In both cases you have a huge binary thing that you have to interact with through secondhand tools rather than directly.
We generally aren't in the habit of using "random" scripts to do anything. They're usually carefully developed to do exactly what we need them to do, precisely, and nothing more. Not the giant pile of buggy ass code and security nightmare that is systemd.
By the way, you don't seem to be aware that you can use any language you want to create startup "scripts" including compiled binaries, if you hate shell scripting so much.
Do you even know any shell script? Serious question. Many 'bash' haters know nothing about the language--starting with calling it 'bash' instead of 'shell script.' There are several different flavors just of shell scripting languages, and bash is only one.
It is not cognitive dissonance to learn from others. The pluggable nature of Linux makes developer lifes harder. They have to write wrappers and abstractions to use base functionality. Having a unified api surface is very attractive.
Windows did something right because you can run very old binaries on a new system. Good luck doing that on Linux.
In the end for most people Linux is not an intellectual exercise in freedom but a tool to get work done and systemd is pretty good at that and is getting better.
And another important point: systemd is still lgpl licensed software. There is literally no legal way for someone to rug pull it. So if it works and brings a benefit it might be a good thing to start to depend on it. Just like we depend on the GNU tools.
All I want is init scripts and X11, but the horizons are shrinking. I've already compromised with systemd, and I don't like it. I see BSD in my future, or at least a linux distro from the list here https://nosystemd.org/ - probably Gentoo. Nothing to stop me, absolutely nothing at all. I just need a few days free to backup/wipe/reinstall/reconfigure/restore_data and I'll be good. Better make that a few weeks. Maybe on my next machine build. It's not easy, but I build machines for long term use.
As for Linux from Scratch - This is something that's been on my radar, but without the part I'm truly interested in (learning more about SysV) then I'm less inclined to bother. I don't buy the reason of Gnome/KDE - isn't LfS all about the basics of the distro than building a fully fledged system? If it's the foundation for the other courses, but it still feels weak that it's so guided by a future GUI requirement for systemd when it's talking about building web servers and the like in a 500Mb or less as the motivation.
> isn't LfS all about the basics of the distro than building a fully fledged system?
Not at all, and I'm not sure why so many people have this perception. It's never been this way. LFS is the tutorial you follow to get a full GNU/Linux system. In LFS, not having a base that can theoretically support the latest desktop environments is considered a fatal bug. It's just a conventional distro but in document format.
This is good for its comprehensiveness and as a way of documenting what existing Linux distros are doing, but it limits its utility as a learning aid as well as for embedded development. Most people interested in barebones Linux should instead start with a kernel and busybox and ignore most GNU packages.
I hate it when a website assumes the language I'm speaking based on my IP. There is no apparent way to change it as well. It's just lazy and hostile design in my opinion.
From a completely technical standpoint, is systemd really better than SysVInit? I ask this question in good faith. I have used both and had no problems with either, although for personal preference, I am more traditional and favor SysVInit.
I was considering forking the base book and maintaining it, as I have kept an eye and occassionally built the project over the years (I use it a lot for package management/bootstrapping/cross compilation experiments), but it appears there already is one: https://lists.linuxfromscratch.org/sympa/arc/lfs-dev/2026-02...
I believe maintaining the base book is the most important part, BLFS has some really good hints but a very significant amount of packages have few differences, collecting these in a separate hints file or similar would help a bit, at least for things that don't hard-depend on systemd like gnome.
Wow this is sad. If any distro keeps the old ways around it should be LFS or Slackware I would think. And maybe Gentoo.
I'm honestly worried about the forces pushing systemd in Linux spoiling the BSD ecosystem. And I'm worried that the BSDs do not have enough people to forge alternatives and will have to go along with the systemdification of everything. sigh
*Note, I ended up on Cachy, which is systemd, so I'm not some pure virtue signaler. I'm a dirty hypocrite :P
"The second reason for dropping System V is that packages like GNOME and soon KDE's Plasma are building in requirements that require capabilities in systemd that are not in System V."
I remember LFS from way back in the day.
What do we all think the overlap between LFS users and Gnome or KDE users is? I think it's pretty small.
> packages like GNOME and soon KDE's Plasma are building in requirements that require capabilities in systemd
So drop them. There are other desktops that are faster, simpler, more stable, and aren't hard-coded to make Linux worse. Has everyone forgotten the design principles that made Linux good in the first place? Tightly coupling your software into other software is simply bad design. At some point you need to eat the cost of a looser abstraction to make your system less fragile, easier to reason about, and more compatible.
The proof in the end that SystemD is a cancer in the Linux ecosystem.
Officially it is just a stack and you can decide to use another one if you don't like it. Unofficially RedHat money ensured that other critical stacks will depend heavily on it so that you can't easily swap without replacing the whole ecosystem.
Man. I'd really rather they did the inverse: drop systemd and only maintain the SysV versions of the materials, even if that means dropping GNOME/etc., because I think understanding the Linux init process is far more important than making any specific desktop environment available.
Sadly is Linux is no longer what is used to be for my generation that cut their teeth having to patch kernels for basic hardware support.
Linux is now effectively systemd/linux, and is attempting to become flatpak/systemd/linux through various corporate sponsored initiatives. The only thing worse, in my eyes, are people who distribute things as docker containers.
The Linux distro as such is becoming an anachronism. There’s no real place to innovate without the inertia of choices made by external projects being enforced on you.
I think it’s a generational change. My generation had Microsoft to contend with, and so sought certain freedoms, but this generation has walled gardens and AI to contend with, so freedom à la Microsoft seems okay and so Linux is being Windows-ified, while Windows itself becomes its own abomination.
Modern mechanical engineers, to this day, learn the thermodynamics of steam engines. Not because they are living in the past, but because they are building foundation knowledge that will permeate everything they'll be doing in the future.
LFS should stick to academic pedagogy, instead of trying to compete in the Linux Distro space.
I had stopped using linux at the start of the systemd takeover (it was not because of systemd).
What I don't understand is how this has happened. I didn't care either way but everybody who did seemed to really fucking hate systemd. Then how come it became the default in so many distributions, with so much opposition from the community?
59 comments
[ 2.9 ms ] story [ 77.6 ms ] thread''As a personal note, I do not like this decision. To me LFS is about learning how a system works. Understanding the boot process is a big part of that. systemd is about 1678 "C" files plus many data files. System V is "22" C files plus about 50 short bash scripts and data files. Yes, systemd provides a lot of capabilities, but we will be losing some things I consider important.
However, the decision needs to be made.''
Going systemd-only is not necessarily a good choice (though I do understand it from a practical point of view). There are other, better alternatives for System V that are smaller and more modular so you still get the Unix "feel" without the absurd complexity of interlinked shell scripts that System V relies on.
I'd like to see OpenRC getting adopted in System V's place. Upstart seems to be dead (outside of ChromeOS) but it would also have sufficed. Alas, I'm not someone with the time or knowledge to maintain these tools for LFS, and unless someone else steps up to do all the hard work, we'll probably see LFS go systemd-only.
That said, there's no reason to go full-fat systemd, of course.
I'd just like to interject for a moment. What you're referring to as Linux, is in fact, Linux plus systemd, or as I've recently taken to calling it, Linux/systemd.
Linux is not an operating system unto itself, but rather another free component of a fully functioning systemd system made useful by the systemd corelibs, systemd daemons, and vital systemd components comprising a full OS as defined by Poettering.
-- https://mastodon.social/@fribbledom/116002799114521341
Learning Linux or Unix from scratch shouldn’t include using crude hacks.
One project on my distro is a new init that will be much, much simpler than SysV, more like BSD and without all the years of cruft.
Do people who really uses LFS even want GNOME or KDE on their system ?
On a related note, X11 was never good and there's a whole chapter in the UNIX-HATERS Handbook explaining why.
When will Wayland earn the label "good"? I don't think it currently qualifies.
1. You're using X11 with hardware that is fantastically newer than anything available at the time the UNIX-HATERS Handbook was written.
2. Every graphics vendor that still supports X11 is shipping workarounds for bugs in Xorg.
I used to have a citation for that second one but it went away when Hector Martin dropped off the face of the Internet.
Systemd is basically the Windowsfication of Linux. I'm always surprised by the people that champion it who also used to shit on Windows with the registry or whatever.
Cognitive dissonance is a hell of a thing.
Oh, and usually people shit on windows for many reasons, but some of the very core features of the OS are robust and the Linux crowd could take a hint. Like, you know, the notion of service at the OS-level and not some random bash script that nohup'd a binary. Oh wait, that's what does Windows, MacOS and Linux with systemd.
By the way, you don't seem to be aware that you can use any language you want to create startup "scripts" including compiled binaries, if you hate shell scripting so much.
Do you even know any shell script? Serious question. Many 'bash' haters know nothing about the language--starting with calling it 'bash' instead of 'shell script.' There are several different flavors just of shell scripting languages, and bash is only one.
Windows did something right because you can run very old binaries on a new system. Good luck doing that on Linux.
In the end for most people Linux is not an intellectual exercise in freedom but a tool to get work done and systemd is pretty good at that and is getting better.
And another important point: systemd is still lgpl licensed software. There is literally no legal way for someone to rug pull it. So if it works and brings a benefit it might be a good thing to start to depend on it. Just like we depend on the GNU tools.
As for Linux from Scratch - This is something that's been on my radar, but without the part I'm truly interested in (learning more about SysV) then I'm less inclined to bother. I don't buy the reason of Gnome/KDE - isn't LfS all about the basics of the distro than building a fully fledged system? If it's the foundation for the other courses, but it still feels weak that it's so guided by a future GUI requirement for systemd when it's talking about building web servers and the like in a 500Mb or less as the motivation.
Not at all, and I'm not sure why so many people have this perception. It's never been this way. LFS is the tutorial you follow to get a full GNU/Linux system. In LFS, not having a base that can theoretically support the latest desktop environments is considered a fatal bug. It's just a conventional distro but in document format.
This is good for its comprehensiveness and as a way of documenting what existing Linux distros are doing, but it limits its utility as a learning aid as well as for embedded development. Most people interested in barebones Linux should instead start with a kernel and busybox and ignore most GNU packages.
I believe maintaining the base book is the most important part, BLFS has some really good hints but a very significant amount of packages have few differences, collecting these in a separate hints file or similar would help a bit, at least for things that don't hard-depend on systemd like gnome.
I'm honestly worried about the forces pushing systemd in Linux spoiling the BSD ecosystem. And I'm worried that the BSDs do not have enough people to forge alternatives and will have to go along with the systemdification of everything. sigh
*Note, I ended up on Cachy, which is systemd, so I'm not some pure virtue signaler. I'm a dirty hypocrite :P
I remember LFS from way back in the day.
What do we all think the overlap between LFS users and Gnome or KDE users is? I think it's pretty small.
So drop them. There are other desktops that are faster, simpler, more stable, and aren't hard-coded to make Linux worse. Has everyone forgotten the design principles that made Linux good in the first place? Tightly coupling your software into other software is simply bad design. At some point you need to eat the cost of a looser abstraction to make your system less fragile, easier to reason about, and more compatible.
Linux is now effectively systemd/linux, and is attempting to become flatpak/systemd/linux through various corporate sponsored initiatives. The only thing worse, in my eyes, are people who distribute things as docker containers.
The Linux distro as such is becoming an anachronism. There’s no real place to innovate without the inertia of choices made by external projects being enforced on you.
I think it’s a generational change. My generation had Microsoft to contend with, and so sought certain freedoms, but this generation has walled gardens and AI to contend with, so freedom à la Microsoft seems okay and so Linux is being Windows-ified, while Windows itself becomes its own abomination.
LFS should stick to academic pedagogy, instead of trying to compete in the Linux Distro space.
What I don't understand is how this has happened. I didn't care either way but everybody who did seemed to really fucking hate systemd. Then how come it became the default in so many distributions, with so much opposition from the community?