You are not logged in.
Pages: 1
I agree with points 1 and 2. Maybe what we need is an alliance with the BSD guys so that everybody uses the same non-systemd init system.
However, the problem has deeper ramifications than init alone. The problem is that everything else is starting to depend on systemd: from desktop environments to drivers.
And the world as we knew it becomes narrower and narrower...
Devuan (and Debian) are about stability not shiny new stuff.
Three things:
1.- MX empowers the users to decide. It provides options and systemd is opt-in.
2.- Debian also used to provide options and still does to some extent. One exception being systemd, which is pretty hard to avoid/remove. In that particular respect, who knows why, they have chosen the tyrant way.
3.- It is not at all about "shinny stuff". For different reasons, I need to compile a lot of software for my work like scientific code and machine-learning stuff like Tensorflow. Some of that code you can build in any system, old or new, because the developers use stable OSs such as Debian, or, more often, RHEL or derivatives. However, most of the developers are running more current systems such as Ubuntu, Gentoo or Arch and, very often, their code depends on newer versions of several libraries than the ones you can find in Debian stable. Of course, you can install them locally and link against them anyway or use containers. However, from a practical point of view, it is a lot faster and easier if you can just compile most of the code without having to go down that road (think, for instance, about the dependencies of the dependencies and so on). This essentially forces me to use Ubuntu (Mint Xfce) at work, if I want to be productive.
Linux is not just server administrators and desktop users/distro-hoppers. There is life in between. There is also us scientists, engineers, developers, audio/video creators, designers, architects and other technical professionals. Let's say workstation users, for short. Typically, we need a trade-off between stability and currency.
Traditionally, we have been a minority but, with the advent of so called artificial intelligence our numbers are growing exponentially.
I just wanted to point out that, nowadays, there isn't a current systemd-free Debian-based distro.
So, either:
1.- You stick to MX or Devuan with old software.
2.- You migrate to Debian testing or derivatives (Ubuntu) with systemd.
3.- You move away from Debian entirely.
In fact, I may give Artix a go.
In the meantime, seeing that a few of the Devuan derivatives seem pretty redundant, why don't join efforts and produce a testing-based distro which would really address currently unmet needs?
And what about a working (tested) how-to on how to decontaminate Debian Testing?
Thanks Panopticon. I understand that. I briefly used Aptosid and Siduction in the past.
Obviously, it is not a job for the Devuan core team.
Which is why I was asking about derivatives.
Hello,
I would like to know if any of the several Devuan Derivatives currently offers the option of installing a system based upon Devuan Beowulf / Debian Testing.
I am aware that some people have successfully upgraded their ASCII installation to Beowulf, but it seems a bit of a messy solution. AntiX proposes a similar approach but I tried it and it was a bit of a mess.
I also know that there is a Beowulf installer but, currently, it seems to have issues as well.
Whereas for Debian stable we have excellent systemd-free alternatives such as Devuan or MX Linux, it would be very good to have one with more up-to-date software. Specifically, my problem with Devuan and MX is that often I have troubles compiling software that depends on more current versions of libraries.
Best,
Miro
Pages: 1