Quoting Craig Sanders ([email protected]):

[init systems:]

> the majority had no say in it, and probably aren't capable of switching
> to something else if systemd doesn't meet their needs.

They can follow recipes, though.  There's a continuum from people who
can fully maintain software through people confident patching and
building software, to those who can manage a few local packages and some 
third-party package repos for particular purposes, to those fearful of
or unable to handle doing _anything_ non-default.  All but that last
category _can_ follow a recipe.  (If they're willing, and not just heady
with the recently trendy appeal of being outraged on the Internet.)

I wrote a recipe telling all about how to make current Debian 8 'Jessie'
use either OpenRC or your choice of the other packaged init systems
_partly_ to make a point to the Devuan crowd -- that the people loudly
claiming Debian is now systemd-captive are mistaken, and are indulging
gratuitous melodrama.  (This was not entirely well received.)

And, of course (equally), I wrote said recipe to assist those seeking such
a recipe.

I fully concur with Russell that most Debian users really neither know
nor care about options among init systems -- for the simple reason that
most *ix users, at all times, seldom even think of such things.  (But I
would assert that those of us who value reliable system, deterministic
operation and security do.)

> so it's OK to break promises because some (other) people said some mean
> things somewhere along the line?
> 
> right.
> 
> i think what actually happened is that they knowingly lied just to get their
> preferred option approved, and actually had no intention of enabling or even
> allowing continued support of anything except systemd.

Quite possibly, but this needs to be seen in proper context.  IIRC, it's
things like 'daemon package [foo] has an unfixed bug in its SysVInit 
script'.  (Invented theoretical example, because I can't be arsed to
find a real one.)  In which case, guys, your hands aren't broken:  Make
whatever's the obvious fix.  If you're a Debian developer, do a NMU.  If
you're not, put it up in a local repo.  Either your patch will get
merged or perhaps your superior outside package will embarrass the DD
into doing it right _or_ your package will become the standard version.

I've been pretty lazy in the past in allowing bloat and questionable
architectural decisions to enter my systems via Debian package
dependencies, but I finally got my attention drawn to the problem -- 
and it wasn't systemd but rather the whole stack of rather ridiculous
desktop-computing glue creating a dependency hairball:  systemd, udev
(now a wholly owned subsidiary of systemd), PolKit (policykit-1),
udisk2, packagekit, network-manager, systemd-logind, D-Bus.  

IMO, those bits of CADT-ware don't belong on my systems, particularly
servers, and mine is the only opinion that ultimately matters locally,
so I set about getting rid of them.  No polemics required:  It's merely
a technical task, and not actually a particularly difficult one.  As I
showed on my Web page. 

Further, if Debian package maintainers make stupid decisions over time,
I can correct those.  A .deb isn't exactly difficult to understand,
dissect, change, and rebuild, after all.  So, noisy politicking isn't
required, and IMO is almost always a rather poor way to get real-world
objectives achieved.

(Speaking of politicking, I note with amusement the spin-aspiring
Subject header.  Russell, I'll bet that was yours, right?  Here, let me
help:  https://en.wikipedia.org/wiki/Begging_the_question )


> By inclination, i'm in the anything-but-systemd camp because systemd is
> the only one that's actively hostile to other software that it sees as
> competing with it (now or in future).

    At least some in the Debian community are particularly annoyed by
    the systemd team's unwillingness to take patches for portability 
    to kernels beyond Linux. That led Adam Borowski to jokingly dismiss 
    OpenRC because it lacks "a hostile upstream". 

https://lwn.net/Articles/511726/

_______________________________________________
luv-main mailing list
[email protected]
https://lists.luv.asn.au/cgi-bin/mailman/listinfo/luv-main

Reply via email to