i did that too
it is... bloated, but it's still the only DE complete enough that i can recommend it to anyone

for example i plugged in a smartphone and KDE recognized it and i could click to open it in dolphin and use normally
while usually how id get pics from that phone would be to use gmtp, dl a pic, gmtp would crash then id restart it and repeat
(the fuse based mtp thing disconnected a lot more)


from what i see KDE will never go full GNOME, so it's fine

I like the way MySQL handled this. On startup, mysqld attempts socket activation. On non-systemd systems, socket activation fails, which mysqld treats as an indication that it's on a traditional Linux system, and it starts up normally.

As for the rest of the uses mentioned in the KDE article, KDE did not have to use systemd to implement those features. The developers chose to use systemd, but could have chosen other ways to do the same thing.

Therefore, XFCE developers have a choice, too. On systems that offer systemd, they could use systemd's services, or they could use the existing traditional solutions which are right there next to systemd, but not (yet) part of it.

I am a software developer, and if it were up to me, I'd choose the solution that jfw on all systems, and not tie my software to only working on systems which support the nonstandard solution.

I think the systemd evangelists understand this, which is one reason we're seeing such an aggressive push to put systemd on all platforms, and a widely-proclaimed narrative of systemd's inevitable ubiquity. As long as it's just one solution among others (which is exactly what it is), people have less of a reason to use it.

The danger is that developers will hear "systemd is the only way to change the system clock!" or "systemd is the only way to restart processes when they crash!" and believe them, and not even consider other ways to do these things.

as developer, do you know situations where resources you have defines what will be shipped and therefore the choice is limited?

you mean Debian Ubuntu Suse Arch ... they all have been manipulated by a small group of fanatics and the decision makers of this distributions are a bit limited in understanding the facts?

and that developers believe simplified half true rumors would be something new?

Yes, it is way ahead of other desktop environments in many areas, but KDE is also permanently stuck in a never-ending rut of bugs, incomplete implementations and outright poor implementations that whatever it offers on the one hand it takes away on the other. If I tried to put this on an enterprise workstation I would be laughed out the door no more than 5 minutes later. Far too many bugs, far too many decorations and choices getting in your way, some necessary UI features barely noticeable in the system tray and others (tooltips) stuck in your face when you least want them. A mail client that is sometimes light years ahead of the rest, and at other times light years behind.

We're talking about a desktop environment that was released in the mid-nineties. There is no longer any excuse for it not to have the refinement and polish it needs to be considered a serious contender for professional and business use. But it still doesn't have that refinement and polish all-round. Here and there, undoubtedly, but all-round? Not even close. I am not joking when I tell you I would consider Windows 95 before KDE for a customer desktop. It took Microsoft two years to get from Windows for Workgroups 3.11 to the ground-breaking Windows 95, and they nailed it first time. KDE still hasn't arrived, mainly because nobody knows exactly what they want. Lack of clarity and definition means a never-ending vista of unfulfilled potential; with programming as easy as it is nowadays, all the tools are there to keep going, but the destination keeps receding into the distant future. Nobody knows where it is.

And we all know the team will never be happy to sit still and refine KDE until it reaches a level of maturity fit for professional use. They would much rather bring it to a good-enough state and then completely disrupt everything once more to move on to a new desktop paradigm. The imposition of a wildly out-of-control systemd promises to make this permanent beta state of affairs the default for at least a decade to come. I have no intention whatsoever of hanging around until they all finally get there. The further "ahead" they go the more I return to computing as it was in the 60s, 70s and 80s, when the real geniuses were around. Have any of these johnny-come-lately devs come up with a UI as good as Fvwm, an editor as good as Emacs, a programming language as good as Lisp? I don't think so. Billions poured into Linux and FOSS by Google, Red Hat, the European Union and all the rest, and we still don't have a single genius from today's generation worth remembering as long as McCarthy, Stallman, Nation and others from their generation will be remembered and revered?

Something not quite right, is there?

8 members found this post helpful.

Pipe down Chachi.
Even a whiff of hostility from you isn't tolerated.
I needn't remind you of the last time you failed to reign it in a bit.
So don't.

There has been quite a lot of posts in this thread basically saying that it was boring, and that "enough is enough". I can't help finding this a bit of a paradox.

Why do you read it? why do you comment in it?

- The thread is very well identified - "...mass exodus...", you know what you are in. Given the number of posts, I bet you know what to expect

- It is very easy to avoid - just don't read it

- the few hundreds of kilobytes of the thread content are not going to change Linuxquestions.org economics!

- the only one who could legitimately complain is poor unSpawn, forced to read all posts to limit abuse!

Actually I would argue that this thread is a perfect honeypot for systemd discussions and flamefest amateurs. And I like to read it too!

P.S. I don't want to poke on Didier here, he just happens to be the second contributor to the thread (by the way, thanks for the per thread contributor statistics tip!

3 members found this post helpful.

I actually stopped using KDE after 4.x. It's just a bad design.

For me THAT ^^ is the ultimate question and the reason why there are so many pages here on systemd. Still, after many.many hours of research and outright asking of proponents here and elsewhere on several forums, I have yet to get a straight answer on that.

Exactly what, aside from the possibility of faster boot times, does systemd offer as a benefit that even strikes anywhere close to a breakeven point for it's costs and risks? I've seen a video in which one of the devs of SuSe stated simply that "systemd is the future" without qualification for which applications, without one hint of why that is let alone should be in any application.

I actually can see considerable value in CoreOS for server farms but it seems to me that even there that can be accomplished largely just by cgroups and other containers and does not require systemd.

So please someone tell me what is the "raison d'etre" for systemd? cuz I come up goose-eggs.

2 members found this post helpful.

Though it has been brought up and responded to a few times I imagine we are skirting close to becoming OT with this, with the only possibility for being On Topic being some comparison to implementation and maintenance of systemd having any similarities to that for KDE. But until it is designated I cannot not respond.

I am so sick of people, especially since it is almost always from someone who doesn't use it or at best "gave it a week" making broad brush statements about KDE as "bad design" or especially the Numero Uno - "Bloated". These sweeping generalizations are useless and therefore amount to unproductive FUD. A User can turn off "The Dreaded 3" and whatever else they'd like to trim out as long as the tradeoff suits them. Polls show KDE still highly regarded by many so it's design can hardly be just swept away and dismissed as "bad".

Gezley, in post #1294 actually nailed it by posting real and specific issues that the KDE team does seem to foist on us - the total lack of some overseeing vision of what to keep and what to toss out, leaving it in a permanent beta state (although I can deal with the choices and don't find the DE particularly buggy). It is in grave danger of becoming nothing to nobody by trying to be all things for everybody.

I do hope they get their act together and soon for professional use and that they manage to avoid being locked into systemd since although they have a lot I have grown accustomed to and fond of I have zero doubts I can and will leave them behind should they make the choice to be systemd-only.

1 members found this post helpful.

To me this is the problem of systemd, outside the fact it's an octopus.

I'm certain I, or of the systemd opponents could get a round with just systemd-init, systemd-logind, and systemd-udev just peachy. This would be nearly the same as using uselessd and/or s6 with udev and logind/consolekit mixed it.

But it willfully compounds an unfixable single issue common on any operating system by making on single part of the services mandatory rather than optional, namely, systemd-journald.

You don't offer the fastest booting init and then add in a mandatory component that can slam the formula 1 race car into a concrete block wall at 200 mph.

Manipulated, yes. By a small group of fanatics, no!
Manipulated by very large and well monied corporate interests.

Paid off? Possibly, in some cases.
Pressured in various ways such as peer, professional, career, simple ego - absolutely.

A bit limited in understanding?
Not if they paid any attention at all to the ample discussions in various fora!
But it isn't about understanding - it is about money, profit, power and politics...
That is what makes the discussions such as this thread so "unreasonable"...

It isn't about reason, or understanding, and that is what makes it all so maddening to us reasonable and understanding little users!

Linux has hit the big time now, and the bullies want to control the ball.
The crowd seems to love the excitement, but that does not make it a fair contest, nor does it mean that the bullies are the better players.

6 members found this post helpful.

sometimes the anti systemd fanatics are the best advertisement for systemd.

honestly, I care about which compiler to use but not about which init system i have.
but the more I read the more it seems that technical solid opinion comes from the sytemd people, while most anti sytemd people are driven emotional and argue with incompetent hysteric crying.
(I do not mean ReaperX7 who brings solutions and the view others that have at least some technical background)

That some people say fan boy and are not able to reflect their own fanatic is delicious
shall I really take a person serious who has proven to not have the flexibility to use a search engine to find out how to boot a systemd machine into runlevel 3 but 'ended up going around in circles and writing instructions down and in the end walked away.'
who declares windows 95 to a better alternative than a Linux system with KDE and sytemd and who obviously wishes time has stopped 20 years ago??

shall I really take persons serious who still reduce systemd to faster boot time

and people complaining about KDE.

bloated KDE, why, just because you are not able to read a little documentation does not mean that KDE is bloated.
changing, ...Software without commercial background is as it is, the developers decide what makes fun.
here you want company driven stability, but systemd is not ok.
unstable, KDE works perfectly fine for a lot of people so possible the stability factor is in front of the screen.
also here on LQ KDE is desktop environment of the year, and this is why I think that a non functional KDE leeds to more exodus than a working systemd where most people would not even notice that it is here.

1 members found this post helpful.

I just came across these slides (thanks Michael Larabel for the pointer) from a conference from Samuel Thibault at FOSDEM.

For those who suffer of the tl;dr syndrome, here is a quote from Michael's article:Incidentally, Samuel works also in the field of accessibility for Debian and has helped me to include brltty in the Slint installers for Slackware version 14.1 in answering questions on the brltty mailing list.

With systemd we have a very strict policy: we want to gently push the distros to standardize on the same components for the base system. That means that if you use ply and systemd together you get the best features but you can still use them independently too. It's loosely coupled, but we do want to get people to use this combination and no other by offering them the best support for this combination.

-Lennart Poettering

1 members found this post helpful.

+1 on that. That is exactly why I decided to move my MLED project from KDE back to Xfce.

1 members found this post helpful.