Re: forks, derivatives, other distros - what are you thinking/doing
This *is* going to be Yet Another Systemd Thread.
This *is* debian-related and on-topic, since it's about what its users think they should do next about (not|using) the distro and how.
This has a purpose, i hope? 'Cos you can sit and wait if you tihnk it's gonna change the next release of Debian. But what the heck, i haven't dumped my bit into the systemd pool yet.
> 1. What are your issues, reasons for doing so - general and/or specific?
I like(d) Debian for its philosophy, apt, stability and the huge repository, to name a few. I'm not a distro-hopper, i've been using Debian for over a decade.
Big hair ball is bad: systemd is replacing too many things too monolithicaly; i don't like the idea of having a lot of important services in one PID. Logs becoming binary files?! Ew. A sane approach would be to improve the bits that neet improving, like the (sadly named) uselessd, openrc, etc. This has been a good approach for ages and i don't believe in changing it just because "it's so last tuesday". So why not putting some effort into changing those components instead?
Choice: i couldn't care less about systemd if it weren't for a slight problem: i won't have a choice anymore. I dislike gnome and kde, i have fluxbox and a ton of others; i dislike emacs, i have vim; etc. I dislike systemd, now what? I'm left with no system.
Dictatorial decisions: why such a change? Did users have a say? If so, where and how? If not, why not? Obviously this goes beyong emacs vs vim - the worst blind are the ones who refuse to see this, yet whoever decided this doesn't seem to give a shit. I see too many Big Companies backing this up and it bothers me. I thought Debian would be imune to this kind of politics. What about all the stuff that will depend on systemd? And why so much effort on preparing the arrival of systemd and little to none on making sure software works with or without it (like -nox stuff)? Debian lost its philosophy and it's a shame, it's one of a very few distros i thought would be resillient to this. Why are so many big names being so dismissive about it is also strange.
And the impact it'll have on all the Debian-based distros?
> 2. What are you considering, evaluating, or otherwise thinking about?
I have one server running stable. It's probably not statisticaly significant for those up high. I'll hold on to wheezy for as long as i can, then, dunno. I'd welcome a fork starting with the last wheezy release. How much of Debian is going to be vendor-locked by systemd? How much of upstream wants to change their code for both scenarios? These are interesting questions. I'd learn the innards of apt packaging and chip in. Maybe create the non-systemd branch alongside of non-free :)
Gentoo is not going to change to systemd soon, it seems. You can install binaries instead of compilling and i think the benefits of compiling (once) outweight the drawback of how long it takes.
Talking about compiling i'm considering LFS on a VM, mostly to learn about an attempt at bringing some order into the chaos of the file tree and to dive into source code (basicaly what LFS is for). I only need a very basic system for my server and i already compile the webserver anyway. No other means would give me more control, fun and pain.
Slackware's not moving to systemd anytime soon - apparently -, and it's the oldest running distro. If the future of linux distros is dictatorship, i'd rather have my next distro controlled by a benevolent one. It doesn't have apt, but i could give it a shot. I use it at work but not as intensely.
OpenBSD takes a while to get used to, their list only has RTFM on their vocab and i wouldn't consider it for desktop, but it's on my list.
OpenIndiana, or whichever Solaris fork is out there also is on the list, mostly because of ZFS (ZoL seems stable though) and dtrace.
> 3. What other options/initiatives are you aware of that you've discarded or otherwise are not considering, and why?
Any distro backed-up by a big company, like Red Hat or Canonical. It seems as though as soon as they start to grow too much they get lost and drunk on power and money. Remember that nifty search engine?
You don't have to agree with me. I don't have to bother replying either. By then maybe i'll have unsubscribed, i'll probably be busy with LFS anyway. It'll be interesting to see distrowatch a year from now.
--
Cheers,
Nuno Magalhães
Reply to: