Re:Systemd redux: The end of Linux
by egil on Friday November 21, 2014 @02:42AM
(#8640244)
The status quo for such projects is to use Busybox, and I reckon it will continue for projects where systemd etc is too much.
[ Parent
| on HN ]
Re:Systemd redux: The end of Linux
by justincormack on Friday November 21, 2014 @08:02AM
(#8641034)
There are other options, in particular the Musl libc-based distros like Alpine Linux and Sabotage (where you can use busybox but dont have to). They also feel much more like the traditional BSDs - Musl libc and pkgsrc is very close to a BSD...
Systemd is too much, but often busybox is not enough. Plus if everything starts conforming to systemd, busybox will have to become like systemd to stay compatible.
Re:Systemd redux: The end of Linux
by pantalaimon on Friday November 21, 2014 @06:23AM
(#8640788)
I doubt you want to run Gnome on a system where you have to use busybox.
Also if busybox is not enough, a minimal systemd system will still be leaner and faster than the equivalent sysV system.
http://events.linuxfoundation.org/sites/events/files/slides/...
It's not just GNOME; an embedded device that acts as a USB host needs something like udev, and as I recall reading systemd has swallowed the hardware plugging notification systems, or was working on it. The success of systemd in getting single-user kernel patches that only make sense for desktops and systemd is most concerning, though. Linux is great because of its modularity; previously, if init didn't make sense you could switch to openrc, upstart, or others, without having to change the way you do logging, DHCP, hardware plug events, etc.