Hello Beautiful community!
I am a student/jr-level IT-guy who has used linux as a daily driver for 2 years now.
I chose Fedora because of it’s similarities to RHEL and RHEL-clone. It was also easy to set up with UEFI and LUKS/LVM, which I somewhat struggled to do on arch. Having wayland, GDM and XDG preconfigured also made starting configuration a lot easier.
When I used arch-based EOS, I usually took the “easy route” when configuring. Instead of using systemd, I just launched stuff on i3-config. Instead of compiling stuff myself I just installed it with aur. Instead of using LUKS or LVM I just had some encrypted directories.
Maybe it was because I was much more experienced when I started with fedora, or maybe it helped to have an already usable system when starting. Either way I feel I learned more using fedora than EOS, even if I heavily modified EOS as well.
However as I am now considering switching, I’d love to hear what experiences people have had with their distributions. Especially Nixos and Debian users, as those are what I’m considering myself.
How much configuration did it take to make the system usable? Are there some limitations with the repositories, distribution or OS in general? And importantly: have you learned something useful while working on your own system?
Did some distribution make you feel you were missing out on something important with your last distribution?
Have you had bad learning experiences with some distro? Have you switched away from distro for the same reason you installed it?
Would you suggest your distro for someone learning linux-admin skills? If you could go back, what distribution would you have used to 1) learning linux the first time 2) working in a jr-level position, still learning basic system administration, 3) when learning to code?
Thank you for your time and comments. I hope this post is general enough to be a worthwhile discussion.
‘Bikeshed issue’ refers to the effect of having a discussion where everyone can easily form an opinion on. Such as choosing a paint colour for a bikeshed that needs to be painted.
I do think this standard, if successful, would trickle down to users outside enterprise settings. Similarly to how Red Hat was/is the force behind Gnome, Wayland, (if I recall correctly) pipewire, and many now ‘universal’ parts of modern linux user-space.
It’s very clear this project aims to be that force in enterprise linux. And if successful, they would determine the direction of development.
And simply put: most people prefer stuff actively developed by a full-time team of software engineers. Some of us don’t, but usually those need to adapt to the new standard, or miss out on software developed assuming such ‘standard’ userspace.
This is why I think it truly is a bikeshed issue. Everyones bike will eventually be in the shed, if the shed gets painted.
I personally am carefully optimistic, as long as the community (you and me, not just our bosses) care enough to contribute. And the organisation makes it easy and accessible.
Of course having meaningful community participation is only the first step. The community can make bad decisions or incoherent decisions, that’s part of having meaningful power.
Lastly I think the organisation knows the reputation of the companies founding it is, on average, not great. So I expect them to truly make their best to engage the community meaningfully and in good faith. Without it, I don’t think they will convince even rocky linux to switch, let alone achieve meaningful compatibility standard of any kind.