[snassar]

Cool. Nice to have in contrast.

My personal feeling is this: systemd might be a software blackhole that eventually eat and destroy everything it comes close to and that's OK. If it is so bad then other developers can create a new or modified system that keeps the good stuff and eliminates the bad stuff, while working with other software projects to integrate better.

If we're stuck with systemd, is is better than others worse than some and it isn't like it is the first highly opinionated software project to grace us. A new ecosystem will get developed, wikis written, books sold, training courses packaged. Wise developers of other software will whisper in dark corners about the evil that stalks the land in the form of bad code and design of systemd, and crazed developers will haunt Slashdot, Hacker News, and LWN holding up the hand-lettered signs to make sure we understand how terrible Lennartitis is.

If something new comes along (more modular! less Lennart! doesn't insult the tender heart of tmux developers) developers will fight it out and we can go back to the 5 init systems that we're currently using.

[fridsun]

The problem is there are better things before systemd, namely daemontools (and its derivs runit, s6, ...). It wasn't popular, presumably because people didn't care that much.

OTOH, Nosh offers a tool to auto-convert service files into equivalent daemontools shell scripts: http://homepage.ntlworld.com./jonathan.deboynepollard/Softwa...

Few seems to care enough either.

History of modern init FYI: http://blog.darknedgy.net/technology/2015/09/05/0/

[digi_owl]

Best i can tell, init as more than what started processes at boot, and stopped them at shutdown/reboot, only really became "important" as people used laptops as oversized mobile phones.

And as the concept of containers and *aaS/cloud took hold, some of the same things that benefited said laptops was found to benefit containers.

The reason systemd happened over any of the others though was that perhaps the main dev had a penchant for NIH solutions, and worked for the 800 pound gorilla of the Linux ecosystem...