Just realised that this didn't go to the list. Apologies for the
omission - and for the length of it.

-------- Forwarded Message --------
From: Peter Duffy <peter@???>
To: sawbona@???
Subject: Re: [DNG] Yeah… everything
Date: 27/06/26 13:56:25

On Sat, 2026-06-27 at 09:15 -0300, sawbona@??? wrote:
> Hello:
>
> On 27 Jun 2026 at 11:08, Peter Duffy wrote:
>
> > ... trying to visualise the thinking which probably drove the
> > systemd development process.
> Indeed ...
>
> *Probably* is the key word within what you have taken the trouble to
> write. ie: a seemingly sound explanation to the reasons behind the
> creation of the absolute clusterfuck that systemd is. 
>
> The motivatiions you enumerate and describe could well be correct,
> but cannot be looked at or analysed within a context less void which
> is, I'm afraid, what you are doing.
>
> Needless to say, I write this with all due respect.
>
> In a reply to a post by Martin S. I described how, not being a
> developer, programmer or experienced Linux user quickly came to
> understand what systemd actually was and was about.
>
> That being so after many years of having to deal with the MS registry
> which, in my opinion, is nothing but an MS sanctioned virus at the
> heart of its operating systems, from W95 onwards.
>
> You may have read it.
> If not:
> https://lists.dyne.org/lurker/message/20260621.132524.4c0f261e.en.html
>
> ---
> TL;DR:
> systemd is nothing but a tool developed by L. Poettering to slowly
> but steadily infiltrate, control and eventually crush the Linux
> ecosystem as we know it.
>
> This is being achieved via one of the arguably most important
> distributions (Debian) and will eventually result in its becoming a
> faint memory of what it once was.
> ie: just collateral damage
> ---
>
> By attempting to justify, understand or reason about the existence of
> systemd we are only ignoring its real purpose.
>
> And we do so at our peril.
>
> While accepting the possibility of being wrong, I also wrote:
>
> " I sincerely hope to be wrong, but I foresee (a turn of phrase if
> you will) that the time will come when one will not be able to boot a
> PC (or whatever replaces it) if some form of systemd is not present
> somewhere in the hardware."
>
> As always, time will tell if the Linux ecosystem was up to scratch.
>
>
> Best,
>
> A.

I agree with what you say. I suppose that what I'm trying to figure out
is whether that was the intention from the start, or whether other
parties saw the potential in systemd to further their objectives and
started to drive the development of it in a certain direction. And I'm
not sure: everything that I've read by LP seems to suggest that he
considers that he's clever enough to know what's best for everyone. But
it's always seemed to me that to understand the enemy is one step
towards defeating it.

I keep remembering my first encounter with systemd. I was supporting
about 350 linux developers and their workstations, all running CentOS
6. I'd been asked to try CentOS 7 with a view to upgrading all the
workstations to it. I installed it on a sacrificial box. The first
thing I found was that our custom workstation deployment system didn't
work any more - and that the reason was that this thing called systemd
had redefined or removed a number of commands that it used. I then
tried to remove systemd, and found that I couldn't. I googled about
this, and found that lots of other people were trying to remove
systemd, and were all being told: "You can't! This is the way linux is
going to be from now on! Get used to it! Or naff off and install BSD."
I then tried to shut the box down: it displayed a bureaucratic message
like "Asserting shutdown state" and then hung. Eventually I powered it
off and rebooted it. When it came up, it complained about ungraceful
shutdown and ran filesystem checks. I reported back to my boss that we
should stick with CentOS 6, which we did. Eventually we moved to ubuntu
14, the last pre-systemd release. At that point, I changed to a new
job, and eventually persuaded the new employer to use devuan.

So even at that point there were at least 3 things to hate about
systemd: (1) it broke a deployment system on which we'd expended a lot
of time and effort, and which worked really well; (2) it was apparently
so buggy that it couldn't shut down the system properly; (3) the
unbelievable fact that there was nothing in the install to indicate
that systemd was being installed automatically, and was not removeable:
the arrogance of the people responsible for this still brings on the
red mist even now.

Side-slip to an alternative reality: again, I'm installing CentOS 7.
This time, a screen appears saying something like: "A new init system
called systemd has been developed, with the following features: [blah
blah blah]. Do you want to keep the existing init system (default) or
try systemd?" and an option box and a button to continue. Also, for
those who opted to try systemd, it could later be swapped for another
init system. It would have been a win-win: those who liked systemd
could use it, and those who didn't wouldn't have to. Most of the
hatemail against the systemd developers would never have happened - so
that they could have focused on making systemd better. But of course,
for that to have happened, the systemd developers themselves would have
had to be different :(  

Sorry for the length of that. I'm sure that everyone on here could
recount equally traumatic or worse experiences.