Why is Trisquel based on Ubuntu, instead of Debian or Devuan? Can't the developers rebase it again to Debian or Devuan?
- Inicie sesión o regístrese para enviar comentarios
Hello everyone, I have a very serious question to ask that just baffles me. Why is Trisquel, a 100% free as in freedom GNU/Linux distributuion, based on Ubuntu, a distribution that is known to contain lots and lots of non-free software? Ubuntu is infested with proprietary software (such as snaps), proprietary firmware, opt-out telemetry, bloatware, high system requirements, and is literary ditching GNU itself and moving to the Rust-based Coreutils.
And so I wonder, why do the developers make it harder for themselves, especially since Ubuntu seems to be shooting itself in the foot each year? Trisquel was once based on Debian, so why can't the developers rebase to Debian again, or even better, to Devuan? Why is a 100% free as in freedom distro dependent on Canonical and it's awful decisions? To me it seems that Trisquel does not need to be based on Ubuntu at all, and Trisquel already uses many parts from Debian, such as the netinstall option or some other packages.
Rebasing to Debian or Devuan would not only make Trisquel more independent, or make it easier for developers to actually develop and work on it, but it would also make the support cycle much longer. If Trisquel 12 was based on Debian or Devuan instead of Ubuntu 24.04 (which is now 2 years old), then instead of getting support until 2029, we could get support until 2031, because Debian 13's support ends that year. That is 5 years of support instead of 3 years.
Trisquel really should rebase to Debian/Devuan. Im sure that the developers could make it as convenient on Debian as on Ubuntu. A few years ago we atleast had gNewSense, but now even that is gone. Please, tell me, why is Trisquel based on Ubuntu, the Devil of GNU/Linux? We really don't need this dependence. To any developer that is reading this in the forum, please at least consider making a Debian spin like Linux Mint does with LMDE. Thanks.
Hi Gnulag,
This question comes up here on the forum all the time, quite frequently, actually))
Here’s my personal take on the matter.
1. There’s no point in switching, as there’s already a distribution like that => https://gnuinos.org/
Genuen is a libre spin of Devuan GNU/Linux (a fork of Debian without systemd), allowing users to get control over their computer and ensuring Init Freedom.
2. We don’t have the financial or human resources to do it.
3. There’s no desire to change anything, as everything works fine as it is))
I completely agree with you; Trisquel should have moved away from Ubuntu ages ago,
but Debian isn’t the best option either. In my view, the best candidate is Devuan (NoSystemD), but why repeat what already exists? So things are just carrying on as they are...
GNULag, I replied to your comment of this
and sam-d16, if Trisquel switched to a devuan fork, it would be more than a one man team and thus be so much better then Genuen.
That much i will say.
As for reasons the 2nd reason is valid though I suppose for financial reasons.
However aitor is doing genuen with few human resources to my knowledge.
Possibly because he is using devuan's repo for it.
I am a translator!
I can understand that you don't like the idea of "depending on Canonical" (and neither do I), but to me, what matters most is to get a 100% free system that is easily usable for everyone.
You are saying it would make it easier for developers to derive Trisquel from Debian, but did you try yourself? Personally, I would rely on the Trisquel developers to judge by themselves, they may find it easier to start from Ubuntu in order to get a 100% free and user-friendly system for everyone.
I agree. There are good reasons why projects are based on Ubuntu. Linux Mint is still based on Ubuntu even though they also have a version based on Debian. If there was no obvious reasons to be based on Ubuntu, they would have stopped long ago.
In my experience, Ubuntu has broader hardware compatibility and fine tuning than Debian, and so Trisquel benefits from that after removing non-free parts. Also, some packages in Debian are left behind compared to the equivalent Ubuntu packages. When the situation is the opposite, it is not blocking, Trisquel can use the Debian package as already mentioned:
This is a long-term choice made by project developers, users can always change to a different system, there is plenty to choose from.
>"Also, some packages in Debian are left behind compared to the equivalent Ubuntu packages."
While this is true with Ubuntu, the same probably wouldn't apply to Trisquel, which is often released 1-2 years after the Ubuntu LTS it is based on. In fact, some Debian packages will definitely be newer than some Trisquel packages, depending on the timing of releases.
Also, Debian has a very extensive backports repo that a developer could pick and choose through, if all they wanted was some newer packages.
>"Linux Mint is still based on Ubuntu even though they also have a version based on Debian."
Linux Mint, while popular, has been a mess in recent years, often due to their battles in trying to get rid of stuff that Ubuntu crams into their distro. I don't think that's a good comparison - you wouldn't probably want to create the "free version of Linux Mint", unless you just wanted to create an enormous amount of work for yourself.
>"Ubuntu has broader hardware compatibility"
That's almost all going to be based on the kernel. And it's not always true either - the Linux-libre kernels from jxself can handle newer Intel integrated graphics that the Trisquel kernels are unable to handle.
I also never understood, why Ubuntu is used as base. I don't know if it would be easier to switch to Devuan, but the base for the next Trisquel release is already two years old and there is still no final release ready. Two years of development is a very long time in my opinion.
Debian and Devuan are both user un-friendly to a greater or lesser degree. They are both server distributions at heart, with desktop users being a fairly low priority. I'm assuming that the reason Trisquel switched to an Ubuntu base in the past is because at one point in time Ubuntu was known for giving a user friendly setup (which isn't quite as much the case any longer - Ubuntu is now also mainly a big commercial server distro with desktop users a lower priority).
If you wanted to re-base on a user friendly Debian-based distro that also gave you an alternative to systemd, then you would base it on MX Linux. That's really the only option that checks all those boxes. And MX:
a) is incredibly easy to remove non-free firmware from (unlike Ubuntu)
b) has special tools that make it easy for users to configure their systems (not present in Ubuntu)
c) has special tools that make it easy for users to rescue their broken systems (not present in Ubuntu)
d) has special tools that make it easy for users to backup their systems (not present in Ubuntu)
e) has special tools that add persistence and other useful features to live USBs (not present in Ubuntu)
f) offers the user the choice of sysvinit or systemd at boot time (impossible with Ubuntu)
g) runs the Linux-libre kernel very well
h) after removing the non-free firmware, you can make a free software respin with just the click of a few buttons (not present in Ubuntu)
A person could whip up a respin of MX with no obvious non-free packages in a couple of hours. Unfortunately, that would only be a bare starting point for Trisquel, which makes many of its own fully libre packages such as Abrowser and Icedove and many others, and uses its own libre software repositories. But yes, it could conceivably be done, and it would almost certainly be a better, more user friendly base than modern Ubuntu.
On the other hand, it would be a LOT of work to fully transition. Whether you went with Debian, Devuan, or something like MX, you're talking about bringing over many many thousands of packages into new Trisquel repositories. Not an easy task at all. Probably a better idea would be for someone to just start hacking away at MX and making free software respins, and making a new free software project out of it, rather than requiring Trisquel to change its base mid-stream. If the MX free software project ever got enough momentum, you would have some options in terms of Trisquel taking it over, or the MX free software project becoming an FSF approved distro on its own.
Hi Andy,
Thank you very much for your feedback. It’s always interesting and useful to read your posts.
It really is very important that there is full interaction and maximum cooperation between developers and users; otherwise, Trisquel will turn into an OBSOLETE AND UNPOPULAR distribution like ParabolaOS and HyperbolaOS!
These two distributions are a complete disgrace in terms of social interaction with users!
The ParabolaOS and HyperbolaOS forums are 200% the worst forums of all free distributions!
I would rank the forums of free distributions in terms of social interaction between developers and users as follows:
1. PureOS
2. TrisquelOS
3. Guix
4. ParabolaOS
5. HyperbolaOS
I don’t think anyone is demanding anything from the developers; every time I read a user’s question, I see a request and a WISH to ditch SystemD and ditch Ubuntu – and that’s been the case for 10 years, and it’s the right path!
But we all understand that developers’ capabilities are limited by both financial and human factors, so we’re grateful for what we have right now. However, I think we need to start addressing this issue gradually; perhaps we could hold a brainstorming session here with real participants (no bots).
Anyone who knows even a little about code will obviously reject the bloated SystemD code, etc.
==> https://nosystemd.org/ <==
Well, I think we can think about something new without getting angry at Parabola and Hyperbola. I've interacted with both sets of developers a bit, and I appreciate their efforts very much. Obviously Trisquel has the more active forum, so you get into a lot more polite and enjoyable conversation here, but that's not a knock against anyone else.
As far as systemd, my main concern with it is that it's hard for me to control, and to keep down to a reduced size. I've taught myself how to hack systemd a bit over the years, but the systemd developers don't really give you much information to help a user to reduce the size and scope and resources required by systemd. I'm happier when I can use something simpler, smaller, and quicker like sysvinit personally.
But, all that takes a backseat to libre software. If I have to use systemd to use libre software, or I have to learn the Guix system, or I have to tolerate a non-friendly forum, I'll do those things because I still want to use libre software.
So does RHEL. Doesn't make it enjoyable to try to use as a libre software distribution though.
Just a day later after I posted this, Ubuntu made a blog about adopting and integrating AI called: "The future of AI in Ubuntu". Proves my point even more.