Systemd Brand
brand.systemd.io
brand.systemd.io
That's an interesting comment, since the green [ OK ] was in use for bootup messages long before systemd was a thing. Nothing about it says "systemd" to me.
Gentoo also did it from the get go, although it's not SysV.
Except you don't know really know that. It is objectively true that such usage was around for a long time before systemd in some major distros. I was seeing it at least 10 years ago, and I still see it on my systems that don't use systemd.
> the green [ OK ], is what many people already identify systemd with
is true for people like me that were not familiar with those other systems.
For example, most probably use Ubuntu (and probably only Ubuntu), and as far as I know it didn't display those [ OK ] until after it adopted systemd. So, at least there's a very large portion of users that never saw those [ OK ] until systemd came along.
If I remember right, it was the same case with Archlinux, which is where I was first exposed to systemd.
Imagine, the biggest news in your distro is the migration to systemd and the first and most obvious change you notice is that the boot process has all these neat [ OK ]s instead of random text. I think it's to be expected that people would identify systemd with that.
I like systemd precisely because the components are nicely integrated with each other. It seems like what we would design after learning the lessons of the last 15-20 years of how to manage a linux/unix machine.
I would assume more programmer types would like that the redundant "intelligence" that was placed in every init script is now gone, and replaced by a common core that allows you to declaratively say what's different about your daemon (or mount, timer, etc). I've also read that systemd has smart defaults, which is a great thing overall over init scripts.
I don't know if this is true, but I would also guess that it makes learning for newer people easier, since you have a similar paradigm and set of tools for your services, mounts, timers, etc.
It's consistent, which is great if you don't want to learn 20 different ways of doing things.
But then again, I also love the ideas from MS PowerShell and wish we had something like that in the *nix world, so we could pipe objects around rather than having to work with text and use awk, sed, cut, print, and all those things to keep legacy workflows around from decades ago. So maybe I'm just weird :)
I seems to me that it is quite useful and easy to use.
Really, that's it. There's technical reasons for really advanced users, but for average users systemd works just fine, and it's easy to use.
In my experience, it's been a pleasure to use.
That you've had a wonderful experience with it doesn't mean that others don't have legitimate issues with it.
The distribution maintainers have spoken; systemd is a vastly superior to any alternative out there. It's no contest, really. It makes building and maintaining complex and dynamic system setups laughably easy. So easy that just about every distribution maintainer threw their arms up in the air and said "Fuck that constantly breaking, inconsistent, unportable, unmaintainable init script shit that's SystemV and move to some thing sane."
I've yet to work with a system engineer that dislikes systemd. Has Not Yet Happened.
People complaining about systemd is like people complaining that these newfangled LED lights are not producing enough heat.
That's not adequate in itself. If we were talking about a database that only lost data if your last name started with "Z", then JWZ would probably be quite vocal about it. Even if it worked fine for you and me, that would be a serious problem. If systemd (or anything else) had a major design flaw, that would be important even if most people don't get bitten by it.
> I've yet to work with a system engineer that dislikes systemd. Has Not Yet Happened.
I've heard from plenty who dislike it. Plenty who love it, too, to be sure! But there's definitely a non-negligible set of sysadmins who don't care for it.
> People complaining about systemd is like people complaining that these newfangled LED lights are not producing enough heat.
More like people complaining that every now and then it catches their loft on fire, but so rarely that everyone else thinks it's peachy.
Yes, obviously.
> The distribution maintainers have spoken; systemd is a vastly superior to any alternative out there.
This seems like a logical leap. Sure, distros have spoken -- but that doesn't speak to whether or not systemd is generally superior. It only speaks to it being acceptable for distros.
> People complaining about systemd is like people complaining that these newfangled LED lights are not producing enough heat.
I don't see that parallel at all. But I guess it doesn't matter.
What I do know is that I've used systemd a lot, and it certainly has its benefits. But it also has its drawbacks. Personally, the tradeoff is not acceptable to me.
This is why systemd has spurred me to stop using Linux in favor of BSD.
I have a lot of Unix machines, so longevity is important to me. I'm switching to BSD now because it seems to me that the sooner I get that done, the less painful it will be to do. It's a huge task as it is!
As an example GDM/GNOME should not depend on systemd components and they should accept patches to support alternatives.
(shrug)
Hater's will hate, I guess?
SystemD also fails quietly or with useless errors and makes problems hard to diagnose - so a bug not directly caused by SystemD looks like it comes from SystemD because SystemD slime is on everything - is extremely poorly-documented, extremely over-engineered, and subject to rapid change due to the moodiness and instability of the OverPoetter.
Reasons to dislike systemd as a professional sys admin?
https://github.com/systemd/systemd/issues/2402 - brick EFI based systems!
https://twitter.com/bitfield/status/736868148348063744?lang=... -- Systemd broke your very mature stable application like screen and tmux?
https://nvd.nist.gov/vuln/detail/CVE-2018-15688 - Root a box on bootup with poison DHCP offers
I can keep going, but how are you doing on my questions?
Well that seemed needlessly contentious. What a shame.
Like I said. Needlessly contentious.
A handful of improvements came tied to a bunch of new constraints and limitations that were not mandatory to get those improvements. This annoys the people who deal in those less visible areas and got the short end of the stick.
A large chunk of what systemd does could have been implemented without also sprawling it's tentacles and cross-dependencies into everything. Had they rolled out that featureset first and given the rest of the ecosystem some time to evolve in response adoption would have gone much smoother.
If you don't think that's just insane... well I'm just getting old I guess.
Perhaps you should learn more about a project before launching into criticism?
Friendly reminder that they managed to infiltrate all distros by creating a hard dependency from GNOME to systemd, which they could only pull off because both projects belong to the same corporation.
> hard dependency from GNOME to systemd
> both projects belong to the same corporation
It is honestly impressive how you've managed to fit three different factually incorrect statements in a single sentence. Kudos.
GNOME does not run at all without systemd.
Even the biggest Linux distros, like Gentoo, struggle to create a patchset to modify GNOME so it does not require systemd; every new version of GNOME takes months to reach Gentoo stable. Most distros have bent the knee and switched to systemd because they had no other option, as GNOME is the most used Linux desktop so they kinda call the shots.
systemd is part of freedesktop.org (https://www.freedesktop.org/wiki/Software/systemd/) > freedesktop.org is a completely volunteer organisation with no corporate backing or funding stream. We are a member project of Software in the Public Interest, Inc., for the purposes of holding assets.
meanwhile gnome has it's own foundation at https://www.gnome.org/foundation/ > The GNOME Foundation is a non-profit organization that furthers the goals of the GNOME Project, helping it to create a free software computing platform for the general public that is designed to be elegant, efficient, and easy to use.
I'm gonna guess this is the information Tobias Bernard (the designer) provided them after creating the logo. This is basic design/UI stuff in my mind.
I do find it a little funny they don't use 'systemd green' for the hyperlinks on the page, and instead use a different, similar color.
I've seen small-ish FOSS projects having long debates on uppercase vs lowercase in logos, libre fonts, etc. And why not, even if we don't always agree with the end result, art does tend to help build a community :)
It's identifiable and helps give a friendlier face to new contributors and outsiders.
This isn’t your grandfather’s open source, no sir.
(Edit: check the #k8sordie tag on Twitter)
not tested or proofed, must provide own logo_mark_beast.ppm of the logo
You can like it or hate it, it's design. It's bound to be subjective. But why is it "a waste of graphic design"?
On a side note, this kind of unhelpful/non-constructive feedback isn't welcome on HN.
If so please look in the mirror. If not serious and just jokingly, please consider how you got to a place where you thought that was a rational and decent thing to do.
A stop job is running for Session c2 of user ... (9min 30s)
"I'm choosing euthanasia etd 1pm." @12:35 PM · Oct 4, 2016
From Wikipedia: Pieter Hintjens (3 December 1962 – 4 October 2016) was a Belgian software developer
It's his choice, and I support it if that's what he really wanted, but it's still really creepy to see.
So his choice was that or further suffering to the same end result.
It was what he really wanted.
I fondly remember the July 2005 party we had at his Molenbeek warehouse after defeating the EU software patents directive. His 4am African music and dance demonstration. Memories...