Left out another major chunk: the conflict that arose over the eventual systemd adoption. To me that conflict altered the concept of ''What is Debian'' permanently, for better or worse (depending on who you listen to).

Regardless of whether one likes or dislikes systemd itself, I think that unfortunate debacle can only be seen as causing harm to the entire Debian project.

The politics of it certainly generated a lot of distrust and resentment among the users and contributors. The project's reputation was undoubtedly hurt.

Perhaps most importantly where the technological impacts. It's one thing when a user can generally ignore the politics surrounding a Linux distro, and the software still does what it needs to do. It's another matter when one routine update after another causes their computer(s) to no longer boot, among other serious problems, all thanks to systemd. Users definitely notice incidents like that, and it decreases, or even eliminates, their trust.

So much hard-earned and invaluable goodwill was unnecessarily lost during and after that period of time.

If any good did arise from that situation, it was that more people became aware of the BSDs, or tried them again if they'd used them in the past. FreeBSD and OpenBSD saved users who needed the reliability and trustworthiness that Debian used to offer, before systemd negatively affected the quality of Debian.

> before systemd negatively affected the quality of Debian

Is there any publication quantifying this?

I followed the whole debacle with interest, and my personal experience with my servers was the exact opposite: adopting systemd improved reliability and made administration significantly easier. It's sad that this was politicized by a small part of the community, but the end result was worth it.

Yeah, I agree.

Systemd is incredibly controversial among a niche group of people who have strong opinions about how init & core system functionality should work, and then there’s an outer ring of people who focus on one or two problems with some relatively minor problems that Systemd caused for which there are viable workarounds. Like how Systemd terminates processes that belong to your session when you log out.

I remember writing SysV style init scripts or rc.d / BSD style init scripts. It was awful. You had all these copy-pasted shell scripts with various gaps in functionality depending on who wrote them. Getting a service to run in Systemd feels like heaven by comparison. I don’t even care about, like, Docker.

I think the reports of problems (like background processes getting termed on logout, how any security problems in Systemd tends to be severe by nature) were just so numerous compared to the reports of the benefits (like the boot time improvements and the massive improvements running daemons). It was some bad decisions and a lot of bad PR, but the overall impact IMO is very positive.

Have you looked at OpenBSD?

Here's the /etc/rc.d/sshd:

  #!/bin/ksh
  #
  # $OpenBSD: sshd,v 1.7 2022/08/29 19:14:25 ajacoutot Exp $

  daemon="/usr/sbin/sshd"

  . /etc/rc.d/rc.subr

  pexp="sshd: ${daemon}${daemon_flags:+ ${daemon_flags}} \[listener\].*"

  rc_configtest() {
          ${daemon} ${daemon_flags} -t
  }

  rc_cmd $1

Pretty much any service/daemon is similar. You define a few things and you're done.

A declarative unit file is still more readable for me.

Why in the world korn shell?

I stumbled upon an auditor with programming skills, who did all his scripting in Korn shell. To me it was like he was from another planet.

It's the standard shell on OpenBSD.

Here is the systemd unit for comparison:

  [Unit]
  Description=OpenBSD Secure Shell server per-connection daemon
  After=auditd.service

  [Service]
  EnvironmentFile=/etc/default/ssh
  ExecStart=/usr/sbin/sshd -i $SSHD_OPTS
  StandardInput=socket

I leave it to the reader to decide which of those two is easier to understand and maintain.

Debatable, but I was mainly reacting to "... rc.d / BSD style init scripts. It was awful. You had all these copy-pasted shell scripts with various gaps in functionality depending on who wrote them."

Simply not true in (modern) BSDs, well at least OpenBSD; I'm not familiar with the others.

Yeah, maybe it’s not inconsistent any more. But meanwhile, systemd has gotten really nice, even a good version of rc.d seems awful by comparison.

I had the same feeling when Apple came out with launchd in 2005. It felt like such a massive improvement over the existing state of things. Systemd also feels like a massive improvement.

When I was trying to resolve the numerous problems that systemd was causing for me on multiple computers that Debian had previously worked perfectly fine on, I certainly ran across a lot of bug reports, mailing list postings, forum postings, IRC logs, blog articles, and other online communications from people who were also having problems with systemd.

Beyond that, I've heard of enough problems involving systemd from my Debian-using colleagues and acquaintances, too.

I don't know if it's been formally studied in any way, but it was clear to me that I definitely wasn't alone in experiencing problems involving systemd.

The widespread negative sentiment that exists toward systemd, including from well beyond the Debian community, didn't just come out of nowhere.

From what I can see, it was generated thanks to a lot of people directly experiencing a lot of unnecessary problems caused by systemd.

Sigh.

I’m pretty convinced it was net positive in the long term. If Debian hadn’t gone with systemd, then where would it be today? Stuck in shell script hell?

We would have been forced to bite the bullet and implement a packaging extension that allowed debs to describe what to start at boot time, when, and how. The compatability layer would have allowed any init system to be used.

Instead we welded the systemd engine into the chassis and pray that when it comes to replacing it we aren’t the ones on the hook.

> The compatability layer would have allowed any init system to be used.

I don’t see how such a compatibility layer would work in a way that doesn’t suck horribly.

Yeah, it was annoying for a bit of time, but not anymore! I haven't had systemd installed in quite some time. sysvinit works without a hitch on bullseye and bookworm.

added: Debian's (to me) about (among other things) technical superiority, a robust packaging system, as well as user freedom and choice. It's super easy to not use systemd these days, "what is debian" didn't change, there was just a slight delay in reality catching up to principles. :)

For anyone who hasn't yet seen it, there was an interesting adaptation of the Debian systemd discussion into an Ace Attorney-style format: <https://aaonline.fr/search.php?search&criteria[sequenceId-is...>

Your comment makes it sound like volunteers have a great degree of freedom and that's not the case because, obviously, in these organisations you will be shown the door if you don't do what others tell you to do.

From the description, the Debian organization seems to be an anarchist one. A bunch of people, not coerced to be there, have created a diffuse rotating democracy for making decisions. Self-sufficiency is key to the organization, that emerges from thoughtful usage of resources.

Why would you consider an organization with a constitution anarchist?

Contrary to popular beliefs, anarchists have no problem with constitutions or laws or governments. Anarchy is ANti-hierARCHY, which in the extreme case extends to not accepting the hierarchy of the State (i.e. an organization with a license to exert force over all others in society). But,

1. Many anarchist movements do not demand this much change. They only demand removal of specific forms of hierarchies they think most problematic. The Occupy movement for instance was demanding the curtailment of the political power of the 1% over the 99%. Its always better to think of political movements as directions of evolution in political space, rather than specific destinations.

2. Then how do anarchists propose that laws/constitutions be imposed? By consensus and discussion. By making sure everyone is on board. Or by temporarily giving someone conflict resolving power (as in the Debian case). Plenty of societies and organizations operate this way, and work fine. Read The Dawn of Everything for some historical examples. See a region in Syria [1] as a modern example.

[1] https://en.wikipedia.org/wiki/Autonomous_Administration_of_N...

> From the description, the Debian organization seems to be an anarchist one.

Then it mislead you. Anarchist organisations aren't typically characterised by large, long and complex sets of policies, constantly evolving, that are strongly policed. Often by bots.

This is the reaction from one software engineer that stumbled into Debian infrastructure: https://lists.debian.org/debian-devel/2023/09/msg00334.html

To quote one part of that email:

    I've been maintaining free software for 30 years so I've got a lot of experience with a lot of different tools, and I've rarely encountered anything that is as comprehensive and well-documented as all this stuff is.

This style of organisation is characteristic found in engineering organisations try to deliver high quality products, not anarchist organisations.

And while it's a flat(ish) style hierarchy, it has leaders (the DPL), a judiciary (the technical committee) and even behaviour police (whoever polices the conduct - it is policed).