[ale] systemd bad. Very bad.

[James Sumners]

On Thu, Jun 29, 2017 at 4:11 PM, Jeff Hubbs <jhubbslist at att.net> wrote:

> There's a certain issue of passing/failing the "ho-ho test:" if systemd
> were intrinsically, deeply, intransigently flawed, why would it be getting
> implemented across distributions that have completely independent
> management bodies? It's even supported as an alternative init system (one
> of five) for Gentoo (where I live my life).


Really? We're going to go over this again? This DNS issue is a prime
example of the problem. systemd's tentacles stretch far and wide; it is
gobbling up every major function possible. This has led to very deep
dependencies on systemd that are difficult to extricate. Therefore, if
distro maintainers don't have the resources to keep surgically removing
systemd from everything they want in their distro, it's just easier to let
the devil in.

Want another example? https://github.com/tmux/tmux/issues/428 <== systemd
maintainer asking tmux to write in explicit support for systemd. Why in the
hell does an init system need deep integration into tmux?

Here's another example of the reach, and subsequent _un_improvement, of
systemd --
https://lists.freedesktop.org/archives/systemd-devel/2017-February/038343.html
. Its DHCP component can't renew a lease; most people consider that an
important part of the spec.

For me, the only good thing to come out of systemd is Void Linux (
voidlinux.eu), the hilarity that is https://twitter.com/systemdsucks , and
DistroWatch adding a "not systemd" init filter to their advanced search
filters (added by popular request).

I suffer through the tragedy that is systemd for work. It'll never touch
one of my personal systems.

-- 
James Sumners
http://james.sumners.info/ (technical profile)
http://jrfom.com/ (personal site)
http://haplo.bandcamp.com/ (music)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ale.org/pipermail/ale/attachments/20170629/339a08d4/attachment.html>