from the or-at-least-not-be-bothered-either-way dept.
[Ed's Comment: Not wishing to ignite yet another flame war regarding the adoption of systemd, I hesitated before publishing this story. However, although it is not an formal survey, it might still reflect the views of the greater linux user community rather than those who frequent this particular site. There is no need to restate the arguments seen over the last few weeks - they are well known and understood - but the survey might have a point.]
http://q5sys.sh has recenlty conducted a survey finding many Linux users may be in favour of systemd:
First off lets keep one thing in mind, this was not a professional survey. As such the results need to be taken as nothing more than the opinions of the 4755 individuals who responded. While the survey responses show that 47% of the respondents are in favor of systemd, that does not mean that 47% of the overall linux community is in favor of systemd. The actual value may be higher or lower. This is simply a small capture of our overall community.
Although the author questions the results could this be an indication that we're really seeing a vocal minority who don't want systemd while the silent majority either do or simply don't care? Poll results and the original blog post.
(Score: 4, Funny) by e_armadillo on Thursday November 20 2014, @09:29PM
Hint, Hint . . . but then again the poll we have right now isn't *that* old yet
"How are we gonna get out of here?" ... "We'll dig our way out!" ... "No, no, dig UP stupid!"
(Score: 2) by Subsentient on Friday November 21 2014, @01:06AM
Not sure that's a bad idea, actually. I'd like to see that.
"It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti
(Score: -1) by Anonymous Coward on Friday November 21 2014, @06:07AM
The problem is that systemd trolls are as pernicious as the old SCO "pay your $699 licensing fee you cock-smoking teabaggers" trolls, and likely to be similarly motivated.
Most of them probably haven't even used Linux, and would vote purely to disrupt the poll.
(Score: 1, Funny) by Anonymous Coward on Friday November 21 2014, @07:47AM
That's a nice constructive post right there, not trolly at all, no siree.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:42PM
At least SCO never tried to force systemd on us.
(Score: 5, Insightful) by LoRdTAW on Thursday November 20 2014, @09:36PM
Where was this poll posted? I certainly never heard of its existence. And did it really need a who would win in a four-way fight question? All in all I can say we can safely ignore this poll.
And one last thing: the majority of use cases reported are desktop users. And the main point of systemd is improve the desktop.
(Score: 5, Insightful) by emg on Thursday November 20 2014, @09:40PM
"And the main point of systemd is improve the desktop."
How?
I mean that seriously: what's it supposed to do to improve the desktop? The only benefit I've seen claimed for the desktop is shorter boot time, and my desktop machine already spends more time in the BIOS than it does booting the operating system. Those two seconds saved out of fifteen or twenty are irrelevant to me.
Systemd is supposed to magically make rainbows and unicorns, but I've never seen any real explanation of how or why.
(Score: 4, Interesting) by Arik on Thursday November 20 2014, @10:18PM
For the moment the main effect I can see is to make the freedesktop.org stuff less relevant to anyone that cares about portability.
If laughter is the best medicine, who are the best doctors?
(Score: 1, Insightful) by Anonymous Coward on Friday November 21 2014, @06:44AM
init scripts were crap, but systemd sure doesn't look like it'll improve stuff. Nor does it look like things will really get better. Lennart has bad taste in design- binary log files, monolithic tightly coupled code etc. It's like a chef with bad taste. The guy will continue producing stuff like that and think it's good.
(Score: 2) by Arik on Friday November 21 2014, @07:01PM
I like the analogy. I see too many people saying things that, even if not meant that way, can certainly be portrayed as personal attacks on the man, and I do not want to come off that way myself. He's obviously a very smart and talented man. Perhaps too arrogant and sure of himself though.
Like a chef that cooks an absolutely stunning meal, that clearly very few could even begin to rival, but... his main dish is roast pork and the event he is catering is called "Passover."
Oy vey.
If laughter is the best medicine, who are the best doctors?
(Score: 1) by jmorris on Friday November 21 2014, @07:21PM
Yup. Which is why I always say that if he hates UNIX (he says he does) he should contribute to ReactOS. They would likely welcome the help. But it isn't a big enough project for his ego. He will instead singlehandedly 'fix' Linux, curing it of it's UNIX taint.
Systemd is -not- about the init, it aspires to and is quickly realizing, it is nothing less than a brand new 'OS in userspace.' Replacing all of the device drivers in Linux is simply not possible, but as Google showed you can just take the kernel as a huge hardware abstraction layer and build almost any OS atop it. Pottering learned from that.
Sooner or later Windows itself will sit atop it as the burden of maintaining device driver support becomes more expensive than the alternative of just letting Linus and his merry band deal with it. Eventually *BSD might even run the Linux kernel with a BSD userspace for the same reason. (Apple doesn't support the universe of hardware that would drive that decision process, they can stay on BSD.)
Polls are unreliable on this since the future isn't going to be determined by votes. It will be determined by level of intensity, 90% could vote that they dislike systemd but RedHat will not care and unless feet vote, unless developers migrate, unless donors step up with server space for forked UNIX like Linux variants to reappear they are right not to. Given sufficient intensity, forking will occur if only 10% do not want to abandon The UNIX Way.
Given almost 40% voting to retain UNIX ways in the recent Debian voting, my money is on forks. I'd like to see forks and have BOTH forks survive. The refugees are out there and they must have a home so let them take a fork and build their Windows/Mac clone atop a Linux kernel. And be happy, productive and put good code into repositories. And let us UNIX folk return to building our own world, taking ideas from the PotteringOS side when we like them and they taking our ideas and code when they like the design.
(Score: 2) by Arik on Friday November 21 2014, @08:19PM
He insists that GDM has to pull in X and all of GNOME to do a simple login, because that's all required for his screen reader. So, blind people don't get access to a primary shell? Talk about second-class citizens. I think I would have made a screen reader with a lot fewer requirements, but then again I don't hate blind people.
If laughter is the best medicine, who are the best doctors?
(Score: 5, Informative) by Pav on Thursday November 20 2014, @10:38PM
It doesn't help with boot times [distrowatch.com], at least not on the default install of both Debian and Arch... for GUI or non-GUI.
(Score: 5, Interesting) by Foobar Bazbot on Friday November 21 2014, @12:12AM
It depends on hardware as well as configuration, but for me, it did: substantially improve boot times in Arch. That's on an Eee 900A netbook, which had such short battery life I initially liked systemd for making shutdown/boot a practical alternative to suspend/resume.
As I looked into systemd, I wasn't philosophically happy with a number of aspects, but I'd still have tolerated that version of it on that machine, solely for the boot time reduction. However, the project's further expansion, following the same sort of wrongheaded design principles, leaves me firmly in the anti-systemd camp.
(Score: 0) by Anonymous Coward on Monday November 24 2014, @05:01PM
"...substantially improve boot times..."
How is this beneficial for servers where boot times should not occur? Also, how important is it to shave a negligible amount of time during boot?
(Score: 2) by Foobar Bazbot on Tuesday November 25 2014, @01:28AM
How is this beneficial for servers where boot times should not occur?
Did you miss where I said I found it useful on a netbook? (FYI, netbooks aren't servers.) And where I said I'd have accepted pre-feature-creep systemd on that netbook, implying that I wouldn't have accepted it in other use cases?
Also, how important is it to shave a negligible amount of time during boot?
Since, in my case, the time saved was not negligible, that question makes no sense.
(Score: 3, Informative) by MrNemesis on Friday November 21 2014, @12:52AM
systemd boots so fast that your stopwatch will gawp on in amazement and its heart will beat faster at witnessing the beauty of such a vision of heaven unfolding before its eyes. That's why boot times seem so similar between it and sysV init.
In all seriousness, from my tinkering with it, there is... some difference in certain scenarios. I've got a new build HTPC running jessie from an SSD that I've tried with-and-without systemd, and systemd goes from grub through daemons, XFCE and XBMC fully loaded twice as fast as sysV; 2s instead of 4. The CPU is the limiting factor there but the time difference is barely noticeable even if you're looking for it. On spinning rust systems where you're likely to be IO bound the difference is much smaller or non-existent. For heavier-duty server loads (i.e. more time spent waiting for IO and network responses) they're for all intents and purposes identical. But besides the point IMHO since POST on any of my mobos takes at least 10s before the OS can even start loading.
"To paraphrase Nietzsche, I have looked into the abyss and been sick in it."
(Score: 2) by tonyPick on Friday November 21 2014, @06:08AM
To be fair, I suspect it might help a bit more for boot times if you're running virtualised servers where HW/POST time isn't a factor, and particularly if you're running a *lot* of virtualised servers. The case also known as "Red Hat's probable customers"). In fact there's probably quite a win there if you're shutting down/restarting/migrating a lot in those environments.
(Of course, that's pretty much irrelevant as far as the rest of us are concerned)
(Score: 2) by TheGratefulNet on Friday November 21 2014, @03:27AM
today, I did an apt-get update of my rasp-pi board (I had pointed to jessie. probably not a good idea) and it borked the bootup process. sits there counting down 1:30 and then hangs. damn! glad I have a backup.
so far, I see no benefit to systemd and only down-sides. this 'upgrade' costed me time and now I have to blow away my sdcard and redo it all from backups.
I'm also going to be avoiding jessie until its known to work (at least on the pi).
I guess I don't see what was wrong with old init. it worked, it was understood and it was lightweight.
linux guys are starting to piss me off. they take functioning stuff and feel like they have to break it. damn.
"It is now safe to switch off your computer."
(Score: 0) by Anonymous Coward on Friday November 21 2014, @03:39AM
You're yet another victim who has had to waste hours dealing with a problem that never would have happened if Debian had done the right thing and shunned systemd.
I don't care how much time this will "save" the Debian package maintainers. It has already wasted so much time of users whose computers have become infected with systemd that the maintainers' time savings will never even begin to approach how much time users have wasted dealing with systemd-induced problems.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @03:49PM
I don't care how much time this will "save" the Debian package maintainers. It has already wasted so much time of users whose computers have become infected with systemd that the maintainers' time savings will never even begin to approach how much time users have wasted dealing with systemd-induced problems.
Exactly. It doesn't help that there are 40,000 packages (and less maintainers) but millions of users.
(Score: 4, Interesting) by Hairyfeet on Friday November 21 2014, @05:10AM
Follow the money and you'll see what the story REALLY is, and its all about "cloud computing" which RH is pushing. Are you planning to run a cloud computing datacenter? No? Then its not gonna help you and will in fact add a big fat single point of failure. For those that want to learn more this blog [blogspot.com] is a good place to start.
And before somebody goes "Ur teh windows guy u no get to talk" I have ALWAYS supported the USERS over the suits, I supported the users when Balmer was trying to shove Win 8 down the user's throats, I supported the users when Pulse was shoved out there in an alpha state, i was and always will support the users over the suits because hey, if you listen to the users you'll find their wants and needs are rather sensible.
Besides anybody that doesn't smell a tuna factory sized fish at the way things have been going on, what with stability focused distros like Debian jumping on when even the supporters admit critical parts aren't complete, with users being banned and discussions being erased...WTF? And wadda ya know, this critical subsystem required by every distro, including those focused on privacy and security, which replaces a part that even nitpickers like me honestly had no real complaints with, is being pushed like its the most important thing since the kernel.....by a company that gets more than 85% of their revenue from the 3 letter agencies. Look it up, DoD,NSA,CIA,FBI, if the US government quit buying RH licenses tomorrow the company would be dead in the water. And to add to all this the stomping of any opposition with bans, erasing threads, and treating everyone who says they are opposed as some sort of troll or idiot? Yeah after snowden I'm sorry but I think everyone with a brain should be worried about putting a critical subsystem like that in the hands of a company practically owned by the US gov. Its just common sense.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 2) by FatPhil on Friday November 21 2014, @08:43AM
Ur teh windows guy - why ur OS not taken into consideration? This pole only considered how popular systemd woz - which is about 1% favouring it. 90+% of people prefer windows. If popularity is an important metric for the systemd supporters, then they have lost.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 0) by Anonymous Coward on Friday November 21 2014, @09:44AM
So you stopped reading at "windows" and missed his points entirely.
Bravo, bravo!
(Score: 2) by Hairyfeet on Friday November 21 2014, @04:46PM
If you wanna take my OS into consideration? Fine its a perfect example of how voting with your wallet empowers users more than the so called "meritocracy" of Linux. After all when Balmer tried to ignore the will of the users with Windows 8 the users refused to buy, OEMs stopped pushing it for Win 7, and Balmer ended up replaced by Nadella who has given the users EXACTLY what they asked for as Windows 10? Its just Win 7 with improvements to speed and ease of use. Hell I've got it running on a circa 2009 AMD netbook and even with such a weak APU and every driver running in compatibility mode its faster in every way and even has hardware acceleration.
Compare this to the supposedly more user oriented Linux, the users as you pointed out have said loud and clear "We don't want this!" only to be told they were idiots, have their voices silenced with secret votes, users banned, and threads erased, and the heads of the major distros saying loud and clear "We don't give a single fuck what you want, you can take it or leave". And wadda ya know, many are doing just that and leaving. There are many tutorials on how to migrate to BSD popping up all over the place and some have even said they are taking my advice and running Windows 10.
IMHO the mess with systemd just proves what I've been saying for years, that Linux is a server OS controlled by corporate interests. If you are not running a fortune 500 datacenter? Your voice is worthless and ignored by the PTB. At least with Windows we can use the power of the wallet to change the direction, it looks like nothing the users say or do will keep them from getting system'd right up the butt.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 2) by LoRdTAW on Friday November 21 2014, @08:59PM
They did what everyone wanted: nuke that atrocity of a start screen, and make metro apps run in a damn window with a full screen option. 99% of the problems were solved right there.
There were no AMD APU's made in 2009. The Bobcat series did not debut until 2011. Hell the Core i7 was still very new in 2009. Perhaps you were speaking of the first Bobcat APU's? That was 2011. A netbook most likely has a cheap, SLOOOOOW 5400RPM disk and low memory, maybe 2GB.
I had Windows Vista running satisfactorily on an AMD x2 w/2GB RAM as long as you turned off all of the stupid fucking services that ground the hard disk on a near constant basis (system restore, RAC, ReadyBoost, Superfetch, Windows Search, and Windows Defender.) Windows 7 fixed those services and they no longer trashed your system. 7 runs fine without tweaks on an old P4 with 2GB RAM and can run office apps and a few others without much effort. Using a fast disk with more RAM can make an old PC feel like a new one. Vista damaged their reputation not because of a bad underlying architecture but because of shit services that ate up I/O and CPU for no reason.
(Score: 2) by Hairyfeet on Saturday November 22 2014, @01:14AM
My bad, got the presentation and the release date mixed up, it was presented in 2009 and released in jan 2011. I grabbed one of the first units, the 1215B [asus.com] which is still fucking great to this day. I slapped in 8GB of RAM and it makes a hell of a mini workstation, even does 1080P over HDMI for those times I have a support call with a customer that has lousy eyesight. they really need to bring back the netbooks, when they were sub $300 I couldn't keep 'em in stock, I ended up having to buy up refurbs as the stocks ran out.
And I'm sorry but I ran 8.1 and it was STILL deep fried ass, as Metro crap was still too tightly ingrained. As I ran it it was like playing pop goes the weasel, you never knew when something Metro'd would be slammed in your face. With Windows 10 all that crap is gone, the only Metro is a few live tiles on the task panel and a simplified control panel that you have the OPTION of using as a quick shortcut to most common control panel uses, kinda like "sort by category" on Win 7. I can't wait until its released, if the rumors are true and its free or less than $30 a pop I'l be grabbing copies left and right.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 2) by LoRdTAW on Saturday November 22 2014, @07:22PM
10 looks to be the ticket. I believe it still has Metro as the weather and news apps looks and acts like a metro apps but in a window. And lets be honest, its not like they couldn't window metro apps, they just wanted to play the "me too" game and force them full screen to act like they knew about mobile.
The price is 30 bucks or less? Shit, i'm all over that. I spent $300 on Win 7 Ultimate and spent $130 on numerous copies of OEM Win 7 pro at work. Now that 7 is ending mainstream and going into extended support, OEM and retail versions will be hard to find. I only hope 10 comes out sooner rather than later. I wont touch 8, even my manager refuses to let 8 in the building. The only experience I had with it was testing in a VM and deleting it the same day.
(Score: 2) by Hairyfeet on Sunday November 23 2014, @12:52PM
You didn't hear? here is the scuttlebutt straight from the mouth of Mary Jo Foley who usually has the inside track at Redmond.
According to her the IMHO frankly awesome new CEO Nadella is tired of the XP holdouts and is fearful that 7 will become another XP so he wants to nip that shit in the bud. Since so few of their sales are from upgrades he is debating either making Win 10 free for everybody or making it free for everybody who bought Win 8/8.1 as an apology and making it a flat $30 for the Home version for everybody else. If the rumor is true you'll be able to upgrade at any time to a higher tier if you need to OR, get ready for this, you'll be able to just buy any of the higher features ala carte. So if you just want bitlocker or domain support? Just buy it! Its not as solid as the pricing rumor but there is talk the higher tiers will likely see deep cuts as well, with pro being $75 and Ultimate $100 but there is serious talk of just getting rid of Ultimate and just having Home and Pro.
And I'm running the latest on my netbook and while they do have a few of the simplified Metro apps like the Music and News apps along with the Windows Store again it appears its all gonna be OPTIONAL so you can just run what you want which frankly is what pissed everybody off about Win 8, lack of options. You can even switch to Metro if you want which IMHO is fucking awesome as the Metro UI is great for HTPCs as a 10 foot UI so with Win 10 we can have the Win 7 UI for work and play and Metro for HTPCs, best of both worlds.
But Foley is usually right on the money and if it turns out to be free or just $30? I'll be all over that shit. I'll probably keep 1 system on Win 7 simply for Windows DVD Maker, I have yet to find anything that makes nice DVDs simpler than DVD Maker and on my home system I'll probably make it a dual boot for a year like I did with XP and 7 before I nuke 7, just in case I find an old program that don't like Win 10. And the rumor is it goes RTM summer 2015 so that the OEMs can have it for the big back to school rush so we should be able to just glide from 7 to 10 without any worries. I know it let me keep ALL my programs I had installed under 7, and I do mean all, even the drivers and lower level stuff like codecs were switched over to Win 10 without me having to do jack shit! That's right, you happy with the programs you have installed on Win 7? You can KEEP EM ALL, they will be right there in all programs after you upgrade! Honestly I have NEVER in all my years seen a more easy upgrade, it was less than 40 minutes from sticking in the flash to surfing the web in Windows 10, fucking awesome.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 2) by BasilBrush on Friday November 21 2014, @07:24PM
"Well, let's say you can shave 10 seconds off of the boot time. Multiply that by five million users and thats 50 million seconds, every single day. Over a year, that's probably dozens of lifetimes. So if you make it boot ten seconds faster, you've saved a dozen lives. That's really worth it, don't you think?"
-- Steve Jobs
Hurrah! Quoting works now!
(Score: 2) by LoRdTAW on Friday November 21 2014, @07:38PM
I never said systemd WILL improve the desktop. My point was a majority of the responders to this so-called poll appeared to be desktop users. And supposedly systemd will make desktops better. How? I dont know. And I don't know because systemd doesn't appear to fix anything. It only breaks POSIX which is what Unix is and what Poettering apparently knows nothing about.
And I am going to go out on a limb here and say that I bet there are plenty of clueless Linux desktop users out there. The other day I was reading about speeding up the boot time of the raspberry pi. Everyone kept spouting the same nonsense: "Use Arch instead of Debian cus it has systemd which makes it boot faster!". Then you had me too posts about how systemd made their pi boot twice as fast. Meanwhile they completely ignored the whole bit about how both distros are completely different and have different init loads to handle. I mean seriously, there are some dumb people out there.
And those are the kinds of people who will vote in favor of systemd, people who think they know shit but don't know shit.
(Score: 2) by pe1rxq on Thursday November 20 2014, @09:45PM
Appearantly I am also not part of this 'community'... But even then with 22% vs 47% both are pretty sizeable minorities.
As for myself, I don't really like the ideas behind systemd, but I never understood why people liked sysv either.
I just hope Slackware can hold out with just a few shims and keep a sane init system.
(Score: 1, Troll) by maxwell demon on Thursday November 20 2014, @10:06PM
If you post here, you are of course part of this community. However you may be in the majority here with your opinion while being in the majority elsewhere (for example, I'd expect the favouring votes would be much more pronounced at Red Hat).
Anyway, I'm pretty sure those who don't care either way are vastly underrepresented because many of them wouldn't care enough to participate at the poll.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 1) by maxwell demon on Thursday November 20 2014, @10:08PM
Err ... the first "majority" should, of course, have been "minority".
Reminder to self: Don't forget to proofread.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 0) by Anonymous Coward on Thursday November 20 2014, @10:47PM
People don't like sysvinit. But when the alternative is something as fucked up as systemd, sysvinit looks damn good.
(Score: 5, Insightful) by frojack on Thursday November 20 2014, @10:14PM
Where was this poll posted? I certainly never heard of its existence.
No, you wouldn't have, and neither did I.
Unless you happen to be on a site where the issue was actively discussed, you just don't run into these things.
Polls are put up, the troops are alerted, mass voting begins, with the usual mass-stuffing of ballots, favored group is all in ahead of time. When it looks like the voting is slipping to the opposition, the voting is closed.
Having just converted (fresh install) from an older release to a new one with systemd I see absolutely no improvement, stability, speed, or lower memory utilization. No obvious degradation either, until I want to do something like starting services or sucn
I do a lot more digging through man pages to find out how to do simple things. Mostly my distro wires the old commands into systemd , but not all of them.
So I see no benefit what ever. Simply a new learning curve.
No, you are mistaken. I've always had this sig.
(Score: 2) by emg on Thursday November 20 2014, @10:20PM
True. Someone asked me today how to run a script at startup, and I immediately said to just call it from /etc/rc.local.
Except we then discovered their new OS has systemd, and /etc/rc.local has a comment saying something along the lines of 'don't even think of calling stuff in here, because systemd makes things so much easier with unicorns and rainbows!'.
So we've no idea how to do something as simple as running a script at startup any more.
(Score: 2) by Konomi on Thursday November 20 2014, @11:00PM
You can redirect systemd users to this: How can I make a script start during the boot process? [archlinux.org] seems pretty strait forward.
(Score: 3, Informative) by Anonymous Coward on Thursday November 20 2014, @11:10PM
That's far more complex than adding just a line or two to /etc/rc.local. It's a huge step backward.
(Score: 2) by FatPhil on Friday November 21 2014, @08:53AM
Why doesn't it work? Is it command line parameters ("/path/to/script.sh") being passed to the program ("/bin/sh") that don't work? Do I conclude that the program may not take any command line parameters? The power of unix-alikes is the power of the command line, the power to parameterise things. If programs may simply run, be launched, and only do one thing, they're no better than icons on a desktop which you click to activate - they're nothing but a "do it" button. What if I want the program started with more verbose debug logging enabled? Can I not add the "-v" switch to the execstart line?
This sounds like a massive step backwards in user-friendliness (for those users who wish to be in control of their systems, rather than just click "do it" buttons).
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:17PM
I think I have figured it out. If you have a .sh to launch on init with parms then you create a second .sh to call the .sh required with the parms needed. Then add a service in systemd/system to that intermediate .sh
(Score: 0) by Anonymous Coward on Friday November 21 2014, @09:52PM
RTFM you noobs...
(Score: 0) by Anonymous Coward on Thursday November 20 2014, @11:01PM
Jesus Christ. Are you going to do the right thing and switch to FreeBSD?
(Score: 0) by Anonymous Coward on Friday November 21 2014, @08:02AM
Jesus Christ will never switch sides. Suffering it is.
(Score: 2) by opinionated_science on Friday November 21 2014, @02:14PM
As I have mentioned before, opensuse had "init" wrappers to systemd for a few years now.
My major problem with systemd is the opaque logging, and slightly cryptic configuration.
But I don't complain because I am not not writing code - I'm a user.
A lot of people who are complaining about systemd dont't seem to realise what a mess the init.d stuff is. Is systemd the answer? I don't know but it has definitely speeded up boot times, on my desktop and laptop.
Furthermore, if you read the roadmap for systemd, there are some very cool things that can be done with it, that CANNOT be done with init.d scripts.
If it doesn't work for you, get together and write a wrapper script so that you can use the old init.d system That would probably be the best compromise.
I don't know if you feel it , but there is a war between corporate ownership and FOSS culture. We need the best tools for GNU/linux to prosper.
If you don't like how systemd is doing something , please submit a patch, or convince someone else to do so....
(Score: 1) by jmorris on Friday November 21 2014, @07:48PM
But I don't complain because I am not not writing code - I'm a user.
I am writing code. Two days ago I ran smack into systemd. Hear and despair:
I am interfacing an RFID reader via serial. Damned thing would not work. Open calls would throw errors unless I was root. Permisisons were right, selinux is long discarded so WTF? Google a bit and get reminded of the rules controlling tty type devices. The first user to open one owns it, that user can multiple open but not others, even in the right group. Ok, that clue lead, via fuser, to discover that gpsd was somehow installed and that it was claiming the freshly connected USB-Serial adapter. Fine. 'service gpsd stop' (Running Fedora 20) and it translated that to a twice as long invocation of systemctl... that at least was decent enough to warn about the sodomy that was coming. "Warning: systemd controls your horizontal, systemd controls your vertical. Systemd reserves the right to automatically restart this service on demand via socket activation." Ok, 'chkconfig gpsd off' should work. Same thing, another translation to a steaming pile of systemdctl and no difference. Systemd knows best.
And sure enough some damned thing keeps on fondling that gpsd socket. Why? Who the hell knows anymore, a modern Linux desktop is as inscrutable as Windows. But fondle it does and damned if systemd doesn't keep right on launching gpsd. So finally I pulled out the big gun and just did 'rpm -e gpsd' and for now systemd didn't reply with a "I'm sorry Dave, I'm afraid I can't let you to do that." How long until package management gets sucked into systemd? (Is Debian ready to be rpm based?)
So after a few hours lost to Freedesktop.org and PotteringOS foolishness I'm back to troubleshooting why my code doesn't talk to the reader... but that is my problem since the port is now open and sending/receiving bits. (Confirmed by null modem to a nearby server's spare serial port.)
(Score: -1, Flamebait) by Anonymous Coward on Thursday November 20 2014, @09:39PM
Fuck you, "janrinok".
(Score: 0, Informative) by Anonymous Coward on Thursday November 20 2014, @09:46PM
Yeah, must be some kinda kraut solidarity or something.
(Score: 3, Insightful) by emg on Thursday November 20 2014, @09:42PM
How many of the people polled actually know enough about init systems for their opinion to matter?
If you'd asked people about Pulseaudio, I'm sure most would have been in favour of its unicorn and rainbows solution to Linux's sucky audio... but, a few months later, they were posting on the Internet asking how to get rid of it because it was such a disaster for several years afterwards.
(Score: 3, Insightful) by frojack on Thursday November 20 2014, @10:20PM
If your computer boots, you are usually satisfied.
If you can get your work done, you're probably ok with it.
If it improves your life, things run faster, its more secure, easier to understand, goes less berserk, you are a happy camper.
So far, NONE of the promises have been fulfilled. Nobody's life is easier due to systemd.
But by the same token, none of the disasters that was pulseaudio have shown up either. Just a steep learning curve for zero benefit.
People new to linux will just learn it like we learned sysvinit, and they will know nothing different.
No, you are mistaken. I've always had this sig.
(Score: 1) by fritsd on Thursday November 20 2014, @10:46PM
"If your computer boots, you are usually satisfied."
Maybe it's my pessimistic outlook, but I have a related but quite different requirement as well:
"If my or my customer's computer *doesn't* boot, and I can debug and fix it, I am usually satisfied"
System administration is easy as long as everything goes to plan. It's when it all goes foobar that you need your toolkit to fix it. And that is when system complexity bites you.
(Actually, that seems to match your expression "easier to understand")
(Score: 2) by maxwell demon on Thursday November 20 2014, @11:53PM
If you can fix it, it does boot (it it still doesn't boot you haven't fixed it). Assuming your satisfaction only comes after having fixed it, not after finding that it doesn't boot, the line "if your computer boots, you are usually satisfied" applies to you. It's just that for you, usually the computer boots because you managed to fix it after it initially didn't boot.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2, Informative) by Anonymous Coward on Thursday November 20 2014, @10:50PM
There have been disasters. I had it happen to my Debian system. Systemd got installed, and it no longer booted properly. While searching for solutions, I ran across a lot of Debian bug reports and mailing list postings describing similar disasters affecting other people.
Now my Debian system doesn't boot at all. That's because I replaced it with FreeBSD after this ordeal, rather than trying to fix it. It's the best decision I've made in a long time. I'm done with Linux.
(Score: 1, Troll) by quixote on Friday November 21 2014, @01:32AM
to have an informed opinion on systemd. I'm pretty sure I'm using it (Debian jessie/testing), and did notice that boot times suddenly went down by about 10x when (I think) it was implemented.
Does that mean I'm for it, for some unspecified value of $for? Well, I don't like what I hear about it making the init process more opaque, about it maybe having some kind of roots in Redhat wanting to push us all into clouds, about Poettering of pulseaudio being some kind of central to it. So, yeah, it makes me nervous.
But there's another factor that's actually moving me toward giving it the benefit of the doubt: The devs who do seem to know something about it respond politely to doubters and even try to explain what's going on. And -- the biggest factor -- the anti-systemd-ers who slag off supporters as "social justice warriors," scream like wounded banshees, and even issue death threats.
Enough already. I just wanted to let the screamers know that you've lost one potential vote by what you're doing.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @08:06AM
social justice warriors
hipsters
{insert others here}
Whenever you encounter the “label of badness” of the day you can safely stop reading.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:22PM
It's a blanket term now. "You'll take it and like it", attitudes.
Ironically similar to the windows 8 interface and the user response.
We all know how that turned out.
(Score: 2) by choose another one on Friday November 21 2014, @01:42PM
How many of the people polled actually know enough about init systems for their opinion to matter?
Ah yes, the fallacy of the democracy - we count everyone's vote as equal whatever their level of knowledge and intelligence, which allows us to conclude that any vote is wrong because the voters were stupid (oddly, this is a more typical conclusion of those who lost the vote than those who won it...).
Question is, how would you decide who "knows enough" to vote ? Would it per chance have anything to do with agreeing with your opinions ?
(Score: 1, Interesting) by Anonymous Coward on Thursday November 20 2014, @09:45PM
Any poll that does not measure the strength of people's convictions is going to be misleading.
For example, the oft cited poll [washingtonpost.com] that shows 75% of NRA members support universal background checks as do 91% of Americans in general. What that poll does not reveal is just weakly those people want background checks - it is easy to agree to something that appears to have no consequences and requires no effort. But the minute you ask these people to do something to make universal background checks happen, like call their congressional representatives, nobody can be bothered. They just don't care that much.
Ignoring all the problems with this systemd poll being unscientific (self-selecting sample, etc) even if that part was OK, they still didn't ask the questions that matter like "would you want systemd if it meant X" where X is list of downsides that are part of systemd's design and implementation.
(Score: 4, Insightful) by tibman on Thursday November 20 2014, @10:05PM
It seems like most of the reasons to not use systemd are technical and ideological. If you are neither of these things then you just don't care. An end-user who never reads their logs simply doesn't care what format a log file is in. Trying to make a survey where most of the questions are about things people simply don't care about will end with some wacky results.
SN won't survive on lurkers alone. Write comments.
(Score: 2) by choose another one on Friday November 21 2014, @10:58AM
Actually one of the common reasons I have seen stated is "because of who it is written by" - which is one of the things that is really odd (and disturbing) about the debate.
One person seems to have become the new Microsoft where everything they produce must be bad simply because it is produced by them. I just don't get how that can be a rational conclusion by a technical mind, if there was a rational technical argument then it seems to have been long buried under what appears to be a very personal hatred.
(Score: 2) by tibman on Friday November 21 2014, @03:31PM
Miguel de Icaza is the same way. Though he actually does work for Microsoft (now).
SN won't survive on lurkers alone. Write comments.
(Score: 1) by jmorris on Friday November 21 2014, @08:02PM
By their fruits shall ye know them.
PulseAudio still does not work. My Thinkpad has no end of issues due to it. It still doesn't have a real reason to exist yet, mostly because of GNOME's dependency, almost every system installs it by default. Some see a bad design pattern repeating.
And once you actually LOOK at systemd the reasons to dislike it are easy to find. Even if you think adopting svchost and event logging from Windows is a good idea, the implementation in systemd is substandard. I don't think it is a good idea so wouldn't like it if DJB himself dropped GPL code with his usual mathematical proof of correctness from his Cathedral.
(Score: 2) by efitton on Friday November 21 2014, @11:38AM
You make a good point. That said, self-selecting sample would be the even bigger problem. You can't use statistics of any sort on a self-selected sample. Random samples or it doesn't count. (Ok, matched pair in a few isolated cases but 99+% of the time).
(Score: 4, Insightful) by maxwell demon on Thursday November 20 2014, @10:00PM
The most interesting part would be the correlations. For example, how many of the people who use Linux primarily for web servers are for systemd? Are Debian users more critical of systemd than Fedora users? What is the average age of those in favour of systemd compared to those against it?
Indeed, some of the questions only make sense to have in that context when correlating them with other data (like age, place of living, or education background).
And of course the faceoff question makes no sense at all here.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 3, Interesting) by tynin on Thursday November 20 2014, @10:07PM
I'd also like to see the number on how many servers / workstations the voter deals with. How many hardware / configuration variants. As those two values grow I would expect to see them more against systemd, but I'd interested in seeing the data regardless.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @08:28AM
You'd be wrong. systemd is specifically pushed by server farms and Red Hat for large facilities. It's advertised and argued in favor of the desktop because the server people already want it.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:45PM
The "server people" don't want it. That's why they're in the process of switching their servers over to FreeBSD.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @10:18PM
You're wrong. We do want it. I'm not switching any of my servers to FreeBSD, and I actually adopted systemd early in my custom Ubuntu-based deployments.
(Score: 2) by choose another one on Friday November 21 2014, @02:56PM
Voting by number or compute power of servers might not get the answer you want either - consider the RedHat cloud admins, how many servers do they manage ? They are in favour (see Atomic, OpenShift, GearD - all using systemd as one of the pieces). Also, Amazon EC2, widely believed to be running RHEL as host...
My guess is that you would see a lot of smaller / discrete / on-premise server admins against, and then switching to in-favour for the really big stuff, but that is just a guess. There are definitely some "big" server admins in favour, as above.
(Score: 4, Insightful) by Marand on Thursday November 20 2014, @10:56PM
The most interesting part would be the correlations. For example, how many of the people who use Linux primarily for web servers are for systemd? Are Debian users more critical of systemd than Fedora users? What is the average age of those in favour of systemd compared to those against it?
Indeed, some of the questions only make sense to have in that context when correlating them with other data (like age, place of living, or education background).
I was thinking the same thing. Presenting all the information separately like they did with the images doesn't really tell anything useful other than ~1/4 of the people polled are against systemd, which is more than just the "small vocal minority" some people claim. The ODS file has the context you want, so maybe someone with more patience than me will pull out some better stats.
One thing I did notice, though, is that ArchLinux is disproportionately represented here, at 37% of those polled, with Debian+derivatives (Ubuntu, Mint, etc.) taking 28%. This probably skewed the results, since only 15% of Arch users voted against, with Ubuntu+Debian making 36% of the negatives. This isn't too surprising, considering some of the comments I've seen on SN about how Arch devs have reacted to anyone even questioning the systemd adoption.
Also, apparently 9 people in the poll listed "FreeBSD" as their current Linux distribution. None were in favour of systemd, though three were undecided. Maybe they'll change their mind when they notice BSD isn't a Linux distribution :)
(Score: 3, Interesting) by maxwell demon on Thursday November 20 2014, @11:43PM
Looking at the time ODS file, I notice an interesting point:
The last two fields are missing for the first 76 resp. the first 82 entries, but present on almost all following entries. Since the entries are sorted by date, this gives me the strong suspicion that the survey was altered while it was carried out.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2) by Marand on Friday November 21 2014, @03:54AM
Could be. I saw some gaps but just thought it was unsupplied data, like the options were left blank. It's also possible they scrubbed troll responses out; it's odd seeing any kind of poll data without at least a handful of people writing "PENISPENISPENISPENISPENIS" in every text field.
That user disparity, with over 1/3 of the respondents being from Arch, really bugs me though. Bugged me enough that I started looking up information about user bases, which just makes the results even more suspicious.
I know it's not actual user numbers, but distrowatch puts Debian alone at almost double the hits-per-day of Arch over the last 12 months, with Mint+Ubuntu+Debian having 6x the hits of Arch. Arch is 7th in popularity, behind MInt, Ubuntu, Mageia, OpenSUSE, and Fedora. Yet here it's over a third of the respondents, and over 3x the number of Fedora users.
In Wikipedia's List of Linux distributions [wikipedia.org], almost half the chart is filled with Debian offspring, with Redhat offspring being the next-largest. Arch is barely a footnote, yet in this poll, the combined count of Redhat (11%) and Debian (28%) derived distributions only account for 2% more of the total vote than Arch (37%).
I'd joke about the poll looking like it was run on the Arch forums, but I think it's just ballot stuffing by the Arch user base, which has a reputation for being excessively proud of their distro of choice. Speaking of jokes, here's one:
Q: How do you find the Arch users in a community?
A: You don't have to, they'll tell you about it any chance they can.
(Score: 2) by Pav on Friday November 21 2014, @07:02AM
FreeBSD has its own systemd ie. launchd. There are those that want to see its adoption, but there doesn't seem to have been a real push (yet).
(Score: 0) by Anonymous Coward on Friday November 21 2014, @08:14AM
I'd like "How good is bash/dash for programming?" question somewhere there.
(Score: 3, Insightful) by MrGuy on Thursday November 20 2014, @10:01PM
Linux is getting to the point where a majority of Linux users are not Linux hackers. They are people who care that it works, not how it works. Those people are not up in arms about systemd, not because they strongly favor the other side ("systemd is WAY better and you're a fool if you don't embrace it"), but rather because they don't care ("meh - seems to work, makes nothing I interact with regularly worse and some things maybe better.")
Most users do NOT CARE how elegant the design under the hood is.
The people who care, both pro AND con, will pretty much always from here on out be a vocal minority of the overall USER community. Not just about this issue, but about most issues. Because the subset of the linux user community that are also part of the linux developer community is shrinking as the userbase expands.
This means the statements "over 80% of linux developers are opposed to X!" and "less than 25% of the linux user community are opposed to X!" could simultaneously be true (numbers made up on the spot for illustration purposes only).
The more interesting question, to me, is whether the decision making should be driven by the user community, or the developer community. In cases like systemd, when we're talking about a decision with major architectural implications, personally I listen to the developers. That said, there are a number of decisions (for example, around look and feel) when the opinion of the using community is probably the more important voice.
(Score: 2) by isostatic on Thursday November 20 2014, @11:07PM
But it's clear that the developers want systemd. Nobody is forcing people to develop for systemd, no one is prwventing a fork of Debian to eliminate systemd.
(Score: 4, Insightful) by Anonymous Coward on Thursday November 20 2014, @11:15PM
Debian shouldn't be forked to remove systemd. Debian should have been forked for those who want it to include systemd.
A two-decade-old community of millions of users shouldn't be held hostage, and then have their computers trashed, by the small handful of idiots who are too dumb and inexperienced to realize how fucking awful systemd really is.
(Score: 2) by isostatic on Friday November 21 2014, @09:50PM
But the developers disagree with us. Fortunately the minority of developers can fork and call it "traditional Debian" or something.
Reality is most developers for some inexplicable reason want systemd.
Just like we got fed up with beta and forked and supported SN.
(Score: 1) by Nuke on Friday November 21 2014, @02:15PM
Like another commenter here suggested, when I first heard of a Debian fork in the context of systemd I thought that people were talking about systemd being the fork, as it was so radical. I was astonished when I realised that the non-adoption of systemd was supposed to be the fork.
(Score: 2) by isostatic on Friday November 21 2014, @02:26PM
Does it really matter?
And no, the majority of Debian developers like systemd. Those that don't can leave and develop their own Debian. With blackjack, and hookers.
(Score: 2) by marcello_dl on Friday November 21 2014, @12:17AM
I tried out jessie many months ago, was not satisfied, likely because it was too early (no systemd specific problems, IIRC driver for old ati card was not stable, but no noticeable gains in performance/boot time either), got back to wheezy. No problems because pre systemd wheezy just worked, so, since "I do NOT CARE how elegant the design under the hood is", I don't think that systemd is justified by its making stuff work. It is justified by its making stuff work *in some way*, and time will tell if somebody will get advantage of systemd way or if all the fuss about it was paranoid... oh, I mean, unjustified. Paranoids means "eventually right", lately.
(Score: 2) by turgid on Thursday November 20 2014, @10:06PM
Which "community" would that be? Does it affect Slackware?
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 1, Interesting) by Anonymous Coward on Thursday November 20 2014, @10:29PM
Does it affect Slackware?
Not yet... It will. As slackware does pull from the other packages out there to feed its own ecosystem.
I personally am on the fence. I see the merits of both sides. I also see that the current init system could use some good improvements. More commonality sure would not hurt it any. It seems like we keep reinventing things. For something as 'mundane' as the init system we are spending a lot of time bickering about it. Most of my work is WELL after the system has come up. I can also see why chucking out 20 years of scripts could be a bad idea. Those scripts are battle hardened by years of use and are dead simple to fiddle with.
(Score: 1, Interesting) by Anonymous Coward on Thursday November 20 2014, @10:57PM
Slackware has other problems, many of which are more severe than even the problems that systemd brings.
The first is that it's basically in a comatose state. It's not totally dead, but it's hardly vibrant, either. It's a stagnant distro, with a stagnant community. It's not that it has taken a stance against systemd, it's just that it hasn't even gotten out of the 1990s yet. Systemd is a good 15 years in the future, relative to where Slackware is today.
The second is that it's a royal pain in the ass to use. Again, this is because it's a fossil. Slackware trades off convenience, without getting anything in return. Debian, before it was ruined by systemd, was convenient to use and just as powerful as Slackware. The convenience actually made Debian more powerful than Slackware for a lot of people.
There are others, but I don't want to spend all day on this.
Slackware isn't a solution to the problem of Debian being infected with systemd. FreeBSD is. FreeBSD is basically Debian done right. It has a better kernel, better userland, better licensing, better developers, a better community, and yet it's still modern and usable.
(Score: 2) by pe1rxq on Thursday November 20 2014, @11:11PM
I started using Slackware in the 1990s. (I probably still have a few infomagic CD sets somewhere.)
I still use Slackware today, and it looks nothing like the first versions I used. It is a modern linux system with up to date packages.
The only thing it has left from the 1990s is its sanity.
(Score: 2) by meisterister on Friday November 21 2014, @12:19AM
I have only 2 itty bitty minor problems with FreeBSD, only one of which will affect my adoption:
1. There's no updated version of Java. I actually quite like Java. To me it's how others view Python: in Java it's easy to bang out an algorithm and get things rolling. The only problem is that I also really like Eclipse and Netbeans (and I kinda have to have Java for work). Also, before someone chimes in with "you can compile it". No I can't, which leads me into #2.
2. It defaults to llvm as the C compiler. On my CPUs (all of which are AMD), GCC can curb-stomp llvm and then have a little bit of energy left to do 20,000 jumping jacks while llvm is still figuring out what's going on. Needless to say, I like GCC. I'm used to it, and it does everything I want it to. It's a shame that the version of GCC that can be installed on FreeBSD is from sometime in the last geological age.
Aside from this, FreeBSD has been awesome. I'm going to try out PC-BSD soon because it looks like the the BSD Ubuntu, but better (or at least, how Ubuntu used to be).
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
(Score: 2) by fnj on Friday November 21 2014, @12:39AM
openjdk-7.71.14,1 not "updated" enough for you? How about openjdk8-8.25.17_1? Either one is only a "pkg install" away.
Really? We're talking about gcc-4.8.3_2 here. There's also gcc49-4.9.3.s20141105. Again, either one, and quite a few others, are a "pkg install" away. And they can coexist with clang and with each other.
(Score: 2) by meisterister on Friday November 21 2014, @01:01AM
I just flipping tried this a month ago! It was probably a config problem on my part, I'll try again soon.
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
(Score: 2) by TheGratefulNet on Friday November 21 2014, @03:32AM
is there a freebsd for the raspi or beaglebone? (I'm curious).
my raspi got horribly borked today when I updated it and systemd broke my boot. I'm not happy about that.
"It is now safe to switch off your computer."
(Score: 2) by tempest on Friday November 21 2014, @06:08PM
Raspberri Pi support was added in 10x, but still has some work to be done from what I understand. I haven't heard much about beaglebone support, but the arm relelease branch suggests it is: ftp://ftp.freebsd.org/pub/FreeBSD/releases/arm/armv6/ISO-IMAGES/10.1/ [freebsd.org]
(Score: 1) by linuxrocks123 on Friday November 21 2014, @02:36PM
As a long-time Slackware user, I must take exception to this. The Slackware community is by no means dead or dying. There's a vibrant user help forum at LinuxQuestions.org, where you can find, among more useful things, multiple flamewars about SystemD. The packages are generally quite recent, though the distribution follows a consciously conservative update policy, which might be mistaken for "stagnation". For instance, right now it uses GCC 4.8.3 instead of 4.9.2, and it hasn't switched to Python 3 yet (but it is tracking the Python 2.7 bugfixes).
There is a package auto-updater included, although it's completely optional and, unlike apt-get or rpm, won't force dependency updates on you. There are also 2 or 3 unofficial ones, including slapt-get, which is what I use. And just in the past 2 months or so I came across, through Google, no less than THREE new unofficial Slackware package repositories that seem to have basically come out of nowhere:
http://sotirov-bg.net/slackpack/ [sotirov-bg.net]
http://slackonly.com [slackonly.com]
http://ponce.cc/slackware/ [ponce.cc]
I have no idea where all these repos are coming from all of a sudden. Maybe they're a front for the NSA ;) This is in addition to slacky.eu and slackware.schoepfer.info, which have been around for a while. And then there's slackbuilds.org which is a "run these scripts and make a custom Slackware package just for you" site rather than an actual repository, but still.
As another indication of vibrancy, we've got derivatives: Arch Linux and VectorLinux are the most popular, but there are others. So yeah. Slackware's doing fine.
As far as Slackware switching to SystemD? I don't see it happening, personally, and certainly not in a way where you can't disable it if you want. Slackware's known for customizability, mostly by not being so obtuse that you can't figure out how to dig into the internals and tweak stuff if you have to.
Re SystemD: I really don't think an apocalypse is coming where you can't run any software without needing SystemD as your init. Many people -- including in the Slackware community -- fear this happening, but, well, almost all "Linux" software also runs on the BSDs, and MacOS, and Cygwin. None of those except Linux even have SystemD as an option for their init. It's traditional in the Linux community for software to be portable, across archs, OSes, and system configs. I don't see that changing.
PS this is random, but I just now discovered that pkgsrc -- NetBSD's "ports" repo -- supposedly works on any Linux distro, as well as Cygwin etc. I've never used it, ever, but it looks cool.
(Score: 2) by jasassin on Thursday November 20 2014, @10:57PM
If all distros can use systemd and be universal, less time is needed developing init scripts and nuances in booting/configuration. They could use that time on desktop development. Difference in hours spent on systemd vs separate distribution sysvinit systems could be a considerable factor in systemd's value.
jasassin@gmail.com GPG Key ID: 0xE6462C68A9A3DB5A
(Score: 0) by Anonymous Coward on Thursday November 20 2014, @11:07PM
I don't think it's possible for there ever to be a net win in terms of time saved.
I've lots several hours to systemd trashing my Debian system. I've read enough other Debian bug reports, mailing list postings, and other comments from people who have had their Debian systems ruined by systemd. Collectively, we've probably lost decades, if not centuries, of time to dealing with systemd problems we didn't even want to deal with in the first place.
And those are just people using Debian! Things are going to get even worse once Debian 8 starts getting more widely used. Every other distro that has adopted systemd has caused similar problems for users, so it's not just a Debian problem.
Looking at this globally, systemd has wasted far more time than it will ever save. It doesn't matter if a small handful of developers save a few minutes here and there, when hundreds of thousands, if not millions, of people each waste many many hours each dealing with problems caused by systemd.
(Score: 0) by Anonymous Coward on Thursday November 20 2014, @11:36PM
>when hundreds of thousands, if not millions
I think you mean hundreds, if not thousands.
I'd be amazed if the Debian Install base had more than 20-40k total users, excluding servers since 1 "user" can control hundreds of "servers", etc.
a few of the comments here are indeed correct though, just because the vocal minority don't like something doesn't mean that everyone feels the same way. I know for a fact I'll never take part in such surveys because I'll be you know... actually using linux and not browsing some shitty personal blog with maybe 50 unique visitors daily.
(Score: 2) by Marand on Thursday November 20 2014, @11:25PM
If all distros can use systemd and be universal, less time is needed developing init scripts and nuances in booting/configuration. They could use that time on desktop development. Difference in hours spent on systemd vs separate distribution sysvinit systems could be a considerable factor in systemd's value.
There's a fallacy to that statement, but I'm failing to think of the proper name for it. The problem is that the statement is mistakenly treating all developers identically, assuming that every developer is both sufficiently familiar with, and willing to deal with, any part of the system equally. The guys that deal with system services aren't just going to go "okay, nothing to do here, I guess I'll go add a feature to gedit today instead!" Likewise, it's foolish to assume all the desktop devs are qualified to have their fingers in the init system and other underlying bits. (Incidentally, that know-it-all assumption is what keeps getting Poettering and Sievers cursed out by people, like when Sievers demanded Linus change the kernel to accommodate bad behaviour by systemd)
Of course, it's just as likely that the admin types that deal with services are instead getting frustrated by systemd. I wouldn't place any bets on that making them want to spend their time improving the the desktop.
(Score: 4, Insightful) by Wrong Turn Ahead on Thursday November 20 2014, @11:18PM
If systemd were good, a poll wouldn't be needed. Forcing adoption of systemd also would not be necessary. There would be a growing, natural excitement around systemd, if it were truly good. Instead, we've had several rounds of crapware and poor decisions forced on users; who are then ignored or attacked when they voice their opposition to leaders and developers with a superiority complex. Linux as a whole has suffered for it and Debian's handling of the entire process has been shameful and stinks of dirty politics.
I am an admin and I've been actively planning a migration path with my staff to move our Linux servers over to FreeBSD. After over 20 years, I've grown so disgusted with Linux that I no longer promote it like I once did. If systemd is the road to new users then I simply don't want new users...
(Score: 1, Insightful) by Anonymous Coward on Thursday November 20 2014, @11:26PM
You aren't alone. Lots of us are moving to FreeBSD, too.
Linux has become tainted. Not just Debian, but all of Linux.
It just isn't suitable for serious use any longer. Debian's quality has taken a nosedive since systemd got introduced, and that's just not acceptable.
(Score: 2) by Konomi on Thursday November 20 2014, @11:37PM
I think in this case it might be more of a "If certain people weren't getting so upset about systemd we wouldn't need a...." you get the point. I'm still amused at how the mood for systemd changed so quickly when I first heard about it my thoughts were somewhere along the line of: "Yeah boots faster that's good I guess I leave my PC on all the time though and I can't wait to start fixing the transitional issues". Both were pretty true for me (that's not to say I don't like the features I've gotten from systemd a main one being the ability to quiet the system startup and still get a service status which I wanted in sysvinit since the start of using Linux), others I knew were jumping up and down at how amazing it was. Fast forward today and it seems like everyone hates it suddenly, it makes me wonder how many certain parts of the community are just echo chambers, on both sides. Or if certain individuals just like making a louder noise when they hate something than when they like something.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @05:24AM
If you care about the truth that's it. I like systemd, I currently dislike a couple things about it's implementation, but I agree with it's design rationale and think it's a better way to do things than the old proven ways. Just because something doesn't need fixing doesn't mean it can't be improved upon. So far it's been working fine for me in my workstation, in my laptop, in my admittedly small server deployments (3-4 racks per), and in my ARM dev boards. It may not be up to all of it's promises yet, but for me it is already delivering more than any alternative.
For others it may be different, and I respect that. I don't mind, heck I support people moving to FreeBSD if it serves them better.
However you'll (almost) never see comments of mine, because the very few times I've commented I've gotten downmodded and/or insulted ad-hominem -- never any meaningful discussion, so I just don't bother commenting.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:52PM
Maybe systemd hasn't ruined any of your systems yet, but it has ruined the systems of many others. In the case of Debian users, it hasn't just ruined their systems after performing updates. It has ruined them permanently, because systemd has ruined Debian itself.
Systemd has wronged them in so many ways, and all of this violation is completely unwanted and unnecessary. Nothing else has caused as much harm and pain for so many open source users in such a short amount of time.
That's where the hostility toward you likely comes from. You are supporting something that has harmed a huge number of people in terrible ways. So don't be surprised when you face vitriol from them for supporting systemd.
(Score: 1) by Wrong Turn Ahead on Friday November 21 2014, @09:40AM
This is more than just a few dissenting voices being blindly echoed en masse. I think it's more likely that many people weren't following systemd progress as closely as they are today. I believe that the people who choose to use Linux are also the very same people who prefer to think for themselves and don't just follow whoever is yelling the loudest. Poettering appears to be very smart but he lacks humility and respect for the very community he is trying to change. His attitude is such that he is the developer and he knows what's best for us; anybody who dares to question or criticize him or his projects are met with hostility and rudeness. It was this very same attitude by the Gnome devs that caused so much unnecessary resentment during the Gnome 3 roll out.
Good or bad, the project itself has become toxic to the point that it is now dividing communities and distracting from the larger goal of improving and promoting Linux. It no longer matters if systemd turns out to be the greatest single invention of our time... An improvement to Linux that comes at the expense of shattering the community is not an improvement; it's a loss...
(Score: 0) by Anonymous Coward on Friday November 21 2014, @11:27AM
I believe you greatly overestimate the fraction of the community that is going to leave Linux or is even upset over systemd. Yes they're a lot, but nowhere near the member count. Time will tell. Furthermore if systemd does turn out to be a great invention then it won't be a loss, it will be natural selection (ie. good riddance). Of course that's a big if, and the reverse is also true, but I'm personally on systemd's side of the fence.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:55PM
Firefox is a good example of how people will flee, and they won't come back once they're gone. Firefox used to have 35% or more of the browser market. Then they made bad decisions that ruined Firefox. Users fled to Chrome. Now Firefox is at 10% of the market, and this number is continually dropping.
A lot of people have already moved from Debian to FreeBSD or Slackware. They won't be back to Debian as long as systemd is present, that's for sure.
(Score: 2) by Pav on Thursday November 20 2014, @11:19PM
Like all good disagreements it seems art [imgur.com] has become a weapon. ;) I have no idea who's responsible for this.
(Score: 2) by Konomi on Thursday November 20 2014, @11:28PM
I feel a bit sad when something that is normally left to political satire enters the Linux community, that picture is pretty shameful.
(Score: 3, Informative) by fnj on Friday November 21 2014, @12:51AM
Funny, it looks like brilliant satire to me, and satire has a long and respected history in journalism.
(Score: 2) by RamiK on Friday November 21 2014, @08:40AM
Agreed. Funny stuff right there :)
(* I'm pro-systemd. But the day a big company picks up Plan 9 and puts a browser and drivers there is the day I switch.)
compiling...
(Score: 1) by jmorris on Friday November 21 2014, @08:19PM
Preach! Or more realistically, instead of chasing Microsoft and Apple (systemd and every desktop installed by default) why not chase the good ideas in things like Plan9, BEOS, etc. Actually try to innovate?
(Score: 2) by Pav on Saturday November 22 2014, @04:41AM
...and another one [akamaihd.net].
(Score: 0) by Anonymous Coward on Thursday November 20 2014, @11:19PM
Linux Mint or Ubuntu users, perhaps the "desktop" majority, simply employ their computers as an adjunct to thumb sucking and thumb suckers don't know, don't care, and can't tell the difference about anything or everything.
(Score: 2) by el_oscuro on Thursday November 20 2014, @11:45PM
"This whole thing is wildly inaccurate. Rounding errors, ballot stuffers, dynamic IPs, firewalls. If you're using these numbers to do anything important, you're insane."
SoylentNews is Bacon! [nueskes.com]
(Score: 1) by SlackStone on Friday November 21 2014, @03:31AM
Could it be that there are forces at work that want to see a complex mess evolve, one with potential security bugs? Could it be that these forces also have a marketing and spin team working the media?
(Score: 2) by pixeldyne on Thursday November 20 2014, @11:53PM
I don't care how long the system takes to boot up since they're almost never allowed to reboot or shutdown. I'd switch to freebsd but since they dropped xfs support I've had to learn and put up with systemd (I think it's too complicated). And please, just because .Net is going multiplatform, lets not make powershell mandatory either.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @12:57PM
Do you have a real need for XFS? Unless you're working with existing XFS filesystems, you'll be better off using ZFS. It offers the benefits of XFS, and then so much more.
(Score: 2) by quitte on Friday November 21 2014, @12:02AM
I do care about the future of linux. A lot. I've been using it as my primary OS for almost 20 years now
Unfortunately there is an increasing number of things I just don't like about it. And I'd prefer if there was less resistance to change, On the desktop it's all fucked up. On all Operating Systems. For example I'm fairly certain that it's more than just nostalgia that makes me think that in the CRT days we had less problems with tearing and smoothness.
XGL looked like a nice way towards a modern X11. But instead we got AiGLX. DirectFB was so very impressive and on Matrox cards we had excellent and smooth video output more than a decade ago. Now everything is a mess of extensions and no video is ever smooth.
And init? In my mind there is no question that we need something that is faster and more intelligent than linearly calling a number of scripts. Then sysvinit scripts became worse when Distributions tried to get the boot speed down and add concurrency. So as it is now sysvinit is nowhere near as elegant and simple as it seems. It is a mess. It has to go.
So while I don't know wether systemd is the answer I'm glad that sysvinit is phasing out.
Hopefully there will be less resistance towards wayland. X has to go even more than init does.
(Score: 1) by Synonymous Homonym on Friday November 21 2014, @09:34AM
What does Wayland do that DirectFB/DirectFBGL doesn't?
(Score: 2) by quitte on Friday November 21 2014, @10:48AM
Wayland has a chance for widespread adoption.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @01:08PM
... once it's integrated into systemd, and forced onto your desktop.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @08:34PM
... once it's integrated into systemd, and forced onto your desktop.
...from behind!!!
(Score: 1) by Synonymous Homonym on Friday November 21 2014, @05:47PM
But what does Wayland do that DirectFB/DirectFBGL doesn't?
(Score: 3, Funny) by srobert on Friday November 21 2014, @12:34AM
Polls like this only confirm that people don't know the FACTS about systemd.
I don't want to spread rumors, but it's been well-established that systemd was born in Kenya and shouldn't have been eligible to serve as an init system.
I've heard that systemd causes cancer and is linked to pedophilia in lab rats.
systemd kidnapped the Lindbergh baby.
systemd was the real culprit behind 9/11.
systemd is contributing to delinquency among our nation's youth.
systemd dwells in the darkest depths of Mordor and shares power with no one.
Anyway, that's what I heard.
I really don't care about systemd one way or the other.
Thank God for FreeBSD.
(Score: 0, Troll) by Lagg on Friday November 21 2014, @12:37AM
Even if the poll was perfect it wouldn't matter. It'd likely just get attacked for being impartial or whatever. Maybe even rightly so. If not then it'd be intellectually bankrupt comments like below saying "if systemd was good a poll wouldn't be needed". Because yeah that's how project polls work. This is what happens when you turn what should be technical discourse and merit into politics. That said I'm not too surprised that approximately half of a given community likes something and the other about-half doesn't.
The real educational value here are the comments about it like in this very thread. It shows how disgustingly low people will stoop. I just saw one hell of a passive-aggressive comment saying that essentially the only people who can like systemd are admins that don't care how things work.
First off, fuck you in general. Secondly, such comments fail to realize that sysvinit is a clunky piece of crap that was 90% shell script boilerplate while the other 10% of implementation was never touched. If anything what I'm seeing are people who do know how things work preferring systemd. I've looked at its code and they do some really irritating things in it like variable length arrays in structs and other things they should really know better about. There are also some design issues. But it's actually being fixed. How about you look at the code of bash and init(1) itself and tell me how it looks there since I'm clearly just an admin who doesn't care how things work because I think systemd has potential.
Then there are of course the comments (again, like below) that totally fail to realize that systemd is not one huge daemon. Probably because they never bothered to so much as look at the project dir layout (not that it stops them from calling people technically ignorant, of course). It's a project name encompassing multiple ones with systemd-the-initd at the center. The comments making this assumption sound as stupid as someone saying that Kwin is bad and uses an ebil binary file format because Kate writes a sqlite file (not that it does, this is just off the top of my head) while at the same time calling Kwin "KDE".
Or my personal favorite: The comments implying that people who are partial to systemd are young idiots that love overabstraction. I don't even need to explain how stupid the age argument is particularly when the very same people that do that haven't the slightest clue what traditional unix actually is. The unix philosophy we think of is better attributed to what plan9 tried to do stressing an importance of solid IPC with good process management. Guess what systemd does? Actual unix was a piece of shit and originally not even written in C. If you want to see what the traditional unix looked like, read the unix hater's handbook and then hit yourself in the face with it until you learn to argue like someone with balls.
Considering that it's really easy to see why sysvinit came out of it. Yet apparently people think this is the gold standard which all inits must be held to. Quite the laugh.
This also brings me to the next point. It's just another stupid social argument for people to fall into. Just like I did just now. That's why this stuff doesn't really matter.
http://lagg.me [lagg.me] 🗿
(Score: 3, Insightful) by E_NOENT on Friday November 21 2014, @12:58AM
Nice rant. You make some good points.
However, the opposite of love is not hate, but apathy.
I, for one, no longer really give a crap about systemd, or Linux in general. Eventually the debate will die down as people either move on, or get over it.
I think the people complaining about systemd/redhate/gnome/pulseaudio/xinet/selinux/apparmor/whatever are finally realizing it's time to move on, but are working through their rage issues by complaining.
So, I say, let 'em gripe for now. It's part of the grieving process.
I'm not in the business... I *am* the business.
(Score: 3, Insightful) by emg on Friday November 21 2014, @01:28AM
I think most people who know anything about it would agree that SysV init is a horrible kludge.
We just want to see it replaced with something that's clearly better, rather than a worse kludge. Systemd seems to add a lot of complexity, embrace and extend a lot of basic OS functionality, for little benefit.
(Score: 2) by Lagg on Friday November 21 2014, @01:42AM
The maintainers are indeed adding complexity where it doesn't need to be but they're also removing a lot of it with the other daemons. For example networkd and timedated is much less of a pain in the ass both usage and code wise than dhcpcd and ntpd. Seriously, check out the code and compare. Despite the maintainers being asshats about good practice sometimes the simplicity difference is quite refreshing. I think if these were not under the systemd project umbrella and were their own thing people would be loving it.
But yeah I am getting very sick of always seeing "hai we added this big batch of features and will do cleanups later" but I think that can be helped by people stopping acting like braindamaged children and actually trying to communicate productively with maintainers /or send patches. I've also advocated booting the current maintainers. Even though I don't blame them for giving up trying to listen to people given the stupidity observed even at places like soylent I do blame them for being entitled pissants. Even though I have my own issues with Lennart the other maintainers I downright loathe. Particularly the one who had the gigantic brass balls to tell Linus that the kernel command line "belonged to them".
http://lagg.me [lagg.me] 🗿
(Score: 1) by NotSanguine on Friday November 21 2014, @03:19AM
I think most people who know anything about it would agree that SysV init is a horrible kludge.
As an admin who, over the last 25 years or so, tracked through BSD Init, SysV init and now onto systemd, I was extremely annoyed when I was forced to learn SysV init and find moving to systemd to be almost as annoying.
Back in the days of proprietary Unix, we had no choice as to the init environment or the management tools (smitty was pretty annoying, and don't get me started on sam). You took what they gave you and loved it.
These days, there are a plethora of choices and, while I understand the hate on systemd (especially the lack of a mechanism to choose between SysV init and systemd), for better or worse the folks who develop and maintain the various Linux distributions made a conscious choice to move ahead with systemd as the only path.
We don't have to like it, nor do we have to use it. As many have mentioned, several flavors of BSD are available without systemd or even the SysV init.
What is more, nothing is stopping anyone (which is completely different from the bad old days) from creating their own Linux distribution which uses whatever init system they choose.
I absolutely agree that those who don't want systemd should shout from the rooftops ("I'm mad as hell and I'm not going to take it anymore" anyone?) as they have been doing. If that (as it has been and things don't look good) is unsuccessful, perhaps it's time to move on.
I'm not a big fan of systemd, nor am I very fond of SysV init. I may decide to move to a BSD variant or I may decide to just eat it and stick with my preferred Linux distribution. As I have neither the time nor inclination to fork my own Linux distribution, I can (as it was in the old days) take what they give me and love it, or move to another platform.
Getting all butthurt about design decisions when you aren't even involved in the development of the platform is pretty dumb if you ask me. To those who really want to use Linux and don't want systemd, you have the tools and the access to do something about it. The rest is just bellyaching and hand-waving, IMHO.
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 3, Insightful) by emg on Friday November 21 2014, @05:05AM
Yeah, because everyone has the time and ability to create their own distro.
And, of course, when a significant number of people who use a distro do decide to fork it, the people who were telling them 'if you don't like it, make your own distro' start whining about them being 'splitters'. And the trolls come along shouting that Linux is far too fragmented to ever compete with other operating systems.
(Score: 1) by NotSanguine on Friday November 21 2014, @08:04AM
Yeah, because everyone has the time and ability to create their own distro.
I don't have the time either. Which I said. But I'm not going to throw a tantrum (not saying you are, but with some of these folks that seems to be the best description).
And, of course, when a significant number of people who use a distro do decide to fork it, the people who were telling them 'if you don't like it, make your own distro' start whining about them being 'splitters'.
Which is why I don't pay those folks any mind. Forking is one of the best parts of FOSS. These are tools, not religions.
Except emacs. Everyone should use emacs. :)
And the trolls come along shouting that Linux is far too fragmented to ever compete with other operating systems.
That was never the problem with Linux anyway. It gained acceptance because FSF couldn't get the HURD going quickly enough. If they had, Linus' creation would have probably ended up like Minix and that would have been that. And it found its niche just as HTTP and commercialization of the Internet were taking hold.
It wasn't simple enough for the average end-user, and MS used its market dominance to keep it off of the major manufacturers' builds (with their pay for every machine you sell OEM licensing), so it never took off as a desktop.
And now there are distros that could be decent desktops, but that won't happen on a large scale for some time, if ever.
Which means the trolls are talking out of their asses and it smells that way too. Why pay attention to them either?
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 3, Insightful) by Marand on Friday November 21 2014, @06:06AM
Not worth getting into most of what you said, but a few things stood out and I wanted to comment:
How about you look at the code of bash and init(1) itself and tell me how it looks there since I'm clearly just an admin who doesn't care how things work because I think systemd has potential.
Does anybody actually use bash for init scripts? Debian and its derivatives use dash, which is much smaller than bash, fewer features but safer. I would hope other distros do something similar. Given that dash is 110K and sysv's init is 39K, there's a lot less there to break than in systemd, whose main binary (the one that I believe is the init) alone is over 1MB.
Now, I'm not saying that "well it's got more code so it must suck", but I think it's premature to be pushing everyone to it as an init when it's fairly new, has a lot more code to it, and is just generally less tested.
The comments making this assumption sound as stupid as someone saying that Kwin is bad and uses an ebil binary file format because Kate writes a sqlite file (not that it does, this is just off the top of my head) while at the same time calling Kwin "KDE".
That comparison is only valid if you can run the individual parts of systemd separately without kludges like Debian's systemd-shim. Yeah, it's in a bunch of individual parts, but they aren't intended to be used individually. That's like claiming a jigsaw puzzle isn't a single picture broken into a thousand pieces, it's really a thousand separate pictures. That's only true if the individual parts are useful without the rest.
Considering that it's really easy to see why sysvinit came out of it. Yet apparently people think this is the gold standard which all inits must be held to. Quite the laugh.
I haven't seen many (any?) people arguing that sysvinit is awesome. The argument is generally that it's either simpler and easier to grok, or that it's well-tested in a way that systemd currently isn't, or that it's interchangeable in a way that systemd isn't.
What I mean by interchangeable: I never had to keep a sysv-shim process running just to make dbus or hal happy before. I could change syslog daemons at will and not have to do anything special. For example, I was using a deprecated syslogd instead of Debian's rsyslog for years because I never noticed they changed the default out and init didn't care. I didn't get forced into it, and I could switch out inits the same way. None of that affected me being able to use power management or removable storage, either.
Then systemd shows up and suddenly your syslog and init have to be blessed by Poettering or you can't use power management, removable storage, and some other desktop feature stuff that's worked for years regardless of your init. Is it really a surprise that some people don't consider that an improvement?
The only reason you can use a non-systemd init right now in Debian without being punished for it is because some devs (Ubuntu or Debian, not sure which) made systemd-shim, which is faking what systemd components expects so that you can run the init (and thus syslog) of your choice without the higher-level parts knowing. That shouldn't have been a kludge, it should have been standard behaviour. If that had been the case from the start, there would have been a lot less outcry.
You see, I don't have any particular issues with some of the parts of systemd. I have the logind part of it running (thanks to the systemd-shim package) without having to change my init, and so far I've had no real issues with it. It's just another chunk of userspace crap, akin to having hal or dbus going, and I'm indifferent. If I hadn't had to jump through multiple hoops to get it installed with the shim instead of changing my init abruptly, I wouldn't even have cared about its addition, it would just be another oddball piece of desktop crap that got added in an update.
Likewise, I didn't care about the init side, until updating desktop software started quietly attempting to replace sysv with systemd. The default behaviour for existing sysv users should have been to set up the shim instead. Having more init systems isn't bad, and it could have gradually been brought into mainstream use by making it an installation option or something. I think if it had gone that way, a lot of us wouldn't have minded. It's harder to be pragmatic about something like this when it's shoved in your face and you're told "TOO BAD, GET USED TO IT, IT'S THE FUTURE"
Doesn't matter how awesome it may theoretically be, people are going to be pissed about that, especially the ones that want to approach huge system changes cautiously. And you know what? That's the kind of user that Debian generally appeals to, so this complete change of behaviour and relatively quick adoption of systemd was like poking a hornet's nest with a stick, except the nest was filled with grumpy admins.
(Score: 2) by Lagg on Friday November 21 2014, @11:59PM
Hehe. I see the ones who modded the original post are doing wonders for convincing people I'm wrong about this being twisted into an emotional and political issue. Why discuss technically when you can bury what you don't agree with I suppose. Good job guys.
http://lagg.me [lagg.me] 🗿
(Score: 2) by Marand on Saturday November 22 2014, @09:05AM
Hehe. I see the ones who modded the original post are doing wonders for convincing people I'm wrong about this being twisted into an emotional and political issue. Why discuss technically when you can bury what you don't agree with I suppose. Good job guys.
Dude, what did you expect? You started with "people are just going to attack the poll", followed it with calling commenters "intellectually bankrupt", and then topped it off with a "fuck you".
Just because I, along with a few others, chose to overlook the tone and bad attitude and responded to other parts of your post doesn't mean the people that downmodded you for it were wrong. You can be 100% correct and still earn a flamebait or troll mod by being a dick about it. You earned that moderation, regardless of any good points made, because of how you opened fire right from the start.
(Though I think it should have been flamebait instead of trolling, but that distinction tends to be lost on people)
You see, you make good comments sometimes, but I've noticed before that you're also pretty abrasive with some of your posts, and that's going to get downmods. That doesn't mean you're wrong about something, or that people are trying to bury you for disagreeing. It's just as likely that you came across like a jerk, got downmodded for it, and not enough people came by and modded you back up on the merit of the rest of your post.
If I hadn't commented before you posted, I probably would have downmodded you too, because it read like an angry flamebait rant. I almost ignored it and didn't comment, but you seemed sincere so I decided to go for rational discussion instead and wrote a reply about parts of the rant.
TL;DR: don't let the downmods bug you, it's probably over your tone rather than an attempt to bury dissent.
(Score: 2) by Lagg on Saturday November 22 2014, @01:58PM
True I do have a tendency to be an asshole about things but I'm just sick and tired of the passive aggressive political nonsense. They deserved both the observation of it being intellectually bankrupt and they deserved to be told to fuck off. Because it's childish shit flinging. It gets old, it got old quickly and it gets really old seeing it clustered together in one place.
and let's be honest. If I were to use that exact same tone but in a way that was critical of Lennart or the other maintainers I'd get a +5 interesting. It was modded troll because I don't take sides and tell both parties that they're being stupid. The downmods in themselves don't bother me, but that they do this and yet still have the sizable brass balls to call other people ignorant, groupthinkers or cultists does.
http://lagg.me [lagg.me] 🗿
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @11:12AM
Thank you for making sense...
(Score: -1, Troll) by Anonymous Coward on Friday November 21 2014, @12:52AM
The only people who want systemd are feminists, faggots, and the supporters of feminists and faggots.
Capatchata: ovaries
Says it best.
Lennart is an SJW faggot.
(Score: 2, Insightful) by khallow on Friday November 21 2014, @05:12AM
If you take over an open source project and then force it to be solely dependent on other code you wrote, that strikes me as an abuse of power and manifestation of a serious conflict of interest. I'd be pleased to have interpreted this story incorrectly here, but this strikes me as rather shifty behavior on the part of systemd developers which would explain all on its own the anti-systemd ire.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @05:41AM
Linus doesn't care. He has redhat stock and a fat wife. He's happy. For some reason.
(Score: 1) by jmorris on Friday November 21 2014, @08:32PM
More important is he doesn't have to care about 'Linux' the UNIX like OS anymore since the vast majority of deployed instances of his kernel are under Android. Linux is a device abstraction layer now for other people to build Operating Systems atop. If you don't like PottteringOS build another UNIX like OS using Linux as the base. There are in fact still several distros doing that, Debian, Ubuntu, Fedora, RHEL aren't interested in that though. Diversity is a strength, Linus still wants his 'World Domination' after all.
(Score: 2) by choose another one on Friday November 21 2014, @01:36PM
How the heck do you "take over" an open source project ? I can understand how you do that with commercial software, but a major point of FOSS is _preventing_ the vendor tie / lock-in.
What happened to the pre-pottering maintainers, did he buy them off or bump them off or what ? Surely even if he did, if it mattered to anybody, the pre-pottering versions of those projects are forked and fully maintained elsewhere, right ?
Could it perhaps be (I don't know, just postulating) that "they" volunteered to maintain a project that no one else cared about enough to maintain, and if so whose fault is that result ?
But, perhaps more importantly, to me udev and dbus are not traditional-Unix-way at all, they fit right in with systemd on that basis, especially udev (and devfs and HAL etc.) which look awfully like desktop solutions being put into servers... I don't see why any traditional Unix admin should miss them.
(Score: 1) by khallow on Friday November 21 2014, @02:38PM
How the heck do you "take over" an open source project ? I can understand how you do that with commercial software, but a major point of FOSS is _preventing_ the vendor tie / lock-in.
By developing and managing it. For example, one of the primary developers of udev, Key Sievers is a developer of systemd and an employee of Red Hat like Lennart Poettering. Similarly, D-Bus was a Red Hat project. I guess that's also the other side of the coin. Somebody has to develop this stuff in order for it to be useful. If they want in exchange for that work a possibly gratuitous linkage to systemd, is that really more serious than possibly not getting usable software in the first place?
(Score: 2) by choose another one on Friday November 21 2014, @04:53PM
So, essentially you are saying that the GPs argument boils down to believing it is wrong for someone / some people who creates and maintains a project to integrate it with another project they also create and maintain, and even though they release it for free use under open source licence so no user is tied into any vendor / maintainer, that just isn't good enough, they should not be allowed to integrate projects because ( because what ? I am stumped, is it just "unless I say so" ?).
I don't think I understand the anti arguments at all any more, because that is heading into srsly wtf territory.
(Score: 1) by khallow on Friday November 21 2014, @05:35PM
So, essentially you are saying that the GPs argument boils down to believing it is wrong for someone / some people who creates and maintains a project to integrate it with another project they also create and maintain, and even though they release it for free use under open source licence so no user is tied into any vendor / maintainer, that just isn't good enough, they should not be allowed to integrate projects because ( because what ? I am stumped, is it just "unless I say so" ?).
"Should" is not the same as "allow". Prior to the merge of udev with systemd, one could use udev without having a particular init system. It looks to me like a dick move to increase the prevalence of systemd-based systems while simultaneously breaking functionality in udev. So yes, I think it's wrong to do that. Do I think it shouldn't be allowed as a result? No.
As to whether free use under open license is "good enough" or not, depends on how much effort it takes in the future to work around systemd and its blob of dependencies. There can be negative opportunity cost here compared to a more open approach.
(Score: 1) by fritsd on Saturday November 22 2014, @11:57AM
"So, essentially you are saying that the GPs argument boils down to believing it is wrong for someone / some people who creates and maintains a project to integrate it with another project they also create and maintain, and even though they release it for free use under open source licence so no user is tied into any vendor / maintainer, that just isn't good enough, they should not be allowed to integrate projects because ( because what ? I am stumped, is it just "unless I say so" ?).
I don't think I understand the anti arguments at all any more, because that is heading into srsly wtf territory.
"
Um.. now I'm going to risk coming across as a frothing-at-the-mouth anti-systemd zealot, because of the way you phrased that, but essentially: yes.
Where you said "they should not be allowed to integrate projects" that's of course impossible; they can write whatever code they like.
So the words "should not be allowed" are far too strong. Instead I'd like to water that down to
"It is bad design for them to purposefully integrate these unrelated projects".
The reason I believe it is bad design is, that it makes the combination of projects more difficult to maintain, more complex to understand (because of the new cross-links), and it makes it more difficult to swap one of the component projects with a newer, better version that is not under the same umbrella. How am I going to use syslogd-ng or rsyslogd or [even better future syslogd], if everything under the systemd umbrella is rewritten to only log to libsystemd-journal0?
An example: does your company ever need to print?
cups depends on cups-daemon
cups-daemon depends on libsystemd0 (apparently all those libs are now joined together into libsystemd0 (since version 215); I hadn't noticed that.
cups-daemon also recommends on colord (recommends are often installed automatically)
colord depends on libsystemd0
colord depnds on libpolkit-gobject-1-0
libpolkit-gobject-1-0 depends on libsystemd0
cups-daemon also recommends (not depends) avahi-daemon
avahi-daemon depends on dbus
dbus depends on libsystemd0, do you know why? because it used to depend on libsystemd-journal0, do you know why? Because nifty binary logging, that's why.
I can tell you from my own experience that it's already not trivial to *not* install systemd.
(Score: 1) by johnck on Friday November 21 2014, @05:33AM
No. God no. Are you crazy?
(Score: 0) by Anonymous Coward on Friday November 21 2014, @06:59AM
The ones that don't care about technical stuff like that, should be using Windows.
The silent majority (the actual majority, not the majority of a tiny minority) has spoken and they are using Windows for desktop stuff. And it sure looks like they've made a better choice. Microsoft could run backwards with crap like Vista and Metro and still stay way ahead of Desktop Linux since the Desktop Linux developers keep sabotaging stuff.
And the market shares of OSX, Android have proven that you can't blame Microsoft solely for the dismal share of Desktop Linux.
It's sad really. It would be nice to have something better.
(Score: 2) by isostatic on Friday November 21 2014, @09:15PM
I had to acquire a new device to run some windows only camera configuration. Normally for single purpose things like this that I leave in another country I get a Mac mini, but the software wouldn't run in wine.
So I bought a cheap windows laptop. It came with windows 8, and a mouse. It's horrendous. They keyboard is unusable, the junk ware that's installed took hours to remove.
Now I know that windows 7 msdn install on a vm or rdp from my t410s with Ubuntu isn't too bad, so it must be the junk that you get with widows pcs, and the terrible hardware.
The silent majority wants this, because of the low price. They don't want quality, they want cheap, be that windows, Linux or mac.
(Score: 0) by Anonymous Coward on Friday November 21 2014, @08:08AM
No.
If they did, the systemd proponents wouldn't need to be so vocally against letting systemd stand on its own merits.
When Arch switched, one of the arguments was that nobody was willing to maintain sysvinit-scripts. They spent the next couple of days banning people for volunteering to take over that job.
A few days ago, Debian had a vote on whether to add a requirement not to use dependencies to actively prevent people from running sysvinit. It was voted down. Nobody on the pro systemd site wants you to have a choice, even when the default is systemd (and I do expect most people to stick with the default).
Heck, if systemd was allowed to stand on its own merits, we wouldn't be having this discussion. The whole for/against systemd comes from a lot of us being pissed about attempts to force systemd down our throats, and being forced to switch distros to avoid it. Multiple times in some cases (Arch and Debian).
(Score: 2) by E_NOENT on Friday November 21 2014, @10:12AM
Heck, if systemd was allowed to stand on its own merits, we wouldn't be having this discussion. The whole for/against systemd comes from a lot of us being pissed about attempts to force systemd down our throats, and being forced to switch distros to avoid it. Multiple times in some cases (Arch and Debian).
Hear, hear. I'd like to add that for some of us, switching distros isn't exactly trivial. It's even worse if you've got clients or customers using one of the affected distributions running your commercial software. What happens to people using the init scripts (and everything else !systemd) you wrote last year on a product you shipped? Better hope everything works!
It's a big transition for some. Maybe Red Hat can help by selling you some consulting hours, you know, to help ease the transition.
I'm not in the business... I *am* the business.
(Score: 2) by CoolHand on Friday November 21 2014, @05:36PM
Preach it Anon! That is about 95% the reason I'm against it.... If they have to try so hard to have it take over the Linux world, then there must be some nefarious motives behind it. I don't like that... AT ALL..
Anyone who is capable of getting themselves made President should on no account be allowed to do the job-Douglas Adams
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @07:25AM
all i can say is... suck shit hater-faggots :D