[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ale] vi vs emacs?

On Mon, 19 Feb 2018 14:41:43 +0000
"Lightner, Jeffrey via Ale" <ale at ale.org> wrote:

> I figure if folks are going to go on ad nauseum about init vs systemd
> again 

False dilemma alert and misleading terminology alert:

"Init" is a generic term for, generally speaking but open to
modification, the first process run by the kernel plus the system to
configure and run all the daemons. Systemd is an init (plus some other
stuff), so this might as well say "systemd vs systemd".

Before the invention of Upstart, >90% of all Linux machines used
sysvinit for their init system, so some people just referred to it as
"init". Some still do, but today such reference is misleading.

Assuming by "init" you meant "sysvinit", the system with /etc/inittab
and tweaked-comment shellscripts in /etc/rc.d/init.d that can be
symlinked in runlevels such as /etc/rc.d/rc5.d, "init vs systemd"
implies a false choice that precludes runit, s6, Epoch, busybox init,
as well as some excellent but less popular inits.

This matters. The justification of creating and force-feeding systemd
always boiled down to "it's better than sysvinit or Upstart!" Even if I
believed that assertion, so what: I'm stronger than Stephen Hawking
and smarter than Mike Tyson.

Use of "init" to mean something to compare to systemd, or comparing
sysvinit to systemd, are both fallacies used and useful in justifying
systemd's existence and, shall we say, "marketing methods". Let's stick
to the real costs and benefits of each init system.