The fact that you think this is a good idea tells me a lot.

Richard obviously he doesn't understand basic security measures like using Firewalls, access control lists, and port management to know that ports, if proper security is in place, will be automatically blocked if a daemon is stopped without warning.

Then again does systemd even have a feature to setup and manage a proper Stateful Packet Inspection and Filtering Firewall through IPTables that automatically blocks incoming and outboard ports if a target daemon service is no longer in use?

I do not believe it does...

I'm always looking at alternatives to Linux. Just in case, if the FreeDesktop guys manage to ruin the ecosystem.

It's completely unimportant, which init your distribution uses, if everything stays stable, compatible, interoperable and non-proprietary. But if the latter is not the case and the Linux/Unix ecosystem is wrecked by Red Hat and/or Canonical, distro-hopping is pointless anyway.

Slackware is the final Linux distribution I will use. If Slackware has to adopt 0pointer or a display server war ruins the desktop experience, I won't hop to another Linux distribution, which will pop up for sure, or start building one from scratch or anything. Instead my decades-long journey with Linux itself will end right there.

All the same reason I've taken to learning FreeBSD. Someone's going eventually add the unwanted onion to Linux and that'll be it.

Using LFS has taught me a lot of what Linux needs and doesn't need to be or become, but if and when the day comes, I'm ready for FreeBSD. At least the FreeBSD guys have a sense of sanity when it comes to their operating system.

BTDT. FreeBSD heavily depends on the ecosystem (it shares with Linux), too. So if the latter becomes an unstable mess of mediocre incompatible concepts, there is almost nothing FreeBSD can do about it. Having a nice well-designed operating system under a broken pile of junk just doesn't change much.

The early Linux years were mainly about copying and re-implementing well-thought Unix concepts into free software. That worked well, because the template was rather good. Now people are trying to invent their own "better" concepts and fail miserably, because doing a good system/framework design requires a lot more than just programming knowledge and most people trying to do this, are just not qualified for this job.

So this is not about 0pointerd, it's just the focal point. I have no use for an amateur-designed Red-Hat-like non-Unix. If that is the future of Linux, I'm out.

linux should never try to adopt to windows/mac os.. in fact is windows since years trying to adopt some things from linux (and fail). i think if i remember right that was the core of the systemd thinking by his creator.. easy windowslike linux.. thats a no go..

ill like to say, if i would have the money, i would buy microsoft out just to let it die than..

well that all typed - i agree in all points with jtsn.

btw. i also think the work "systemd" should be blacklisted in the slackware part of this board. :D

So, get yourself a copy of The Linux Programming Interface, ignore everything it says about POSIX compatibility and hack away your amazing Linux software. It’s quite relieving!

– Lennart Poettering @ fosdem.org


...

So how are systemd’s goals achieved? Basically, by throwing well-proven Unix paradigms out the window and clearly admitting it.

– Lennart Poettering @ fosdem.org

What I actually suggested in that interview was not so much that the BSDs should adopt the Linux APIs, but instead that people should just forget about the BSDs. Full stop.

- Lennart Poettering


It is not too hard to see who and what he is targeting. He is against old school unix/Linux. Therefore he will do everything in his power to change/inhibit/rebel against what is and has been standard.

Systemd will happen if the linux community lets it happen.

If they let this mad-hatter be at the controls... who knows if linux will even look unix-like in the future.

Boy that is a terrible thought... I wonder what Lennart's equivalent of an .exe is.

Probably... .sysd

All I miss in SysVInit is ability to tell init to stop respawning of certain processes marked by "respawn" action. Sometimes I do need to stop execution of some vital daemons(for maintenance purposes) but init will respawn them in no more then 10 seconds. For this purposes I have complex workaround now, but I want get rid of it asap.

Auto-respawn shouldn't even be enabled by default. It's existence as a default through systemd is actually a security hole in the design of Linux and modularity design. No daemon should ever auto-respawn if it is shutdown due to security issues, or is purposely crashed from an outside threat.

I think eventually, worst-case-scenario, the *BSDs and UNICES are going to have to start making their own versions of Linux based software, and either take over development of projects like xorg if and when it's abandoned for Wayland, or direct forks of projects specifically for BSD and UNICES.

If it divides and splits the community apart then so be it. Maybe BSD should fork their own stuff out and redevelop stuff specifically for BSD and UNIX. If they could get HP, Oracle, Illumos, and other UNIX core projects onboard, they could effectively zero-out Linux entirely and leave it to rot after the kool-aid works it's poisonous magic.