Well it seems no matter how loudly we here in the Linux blogosphere try to hum a happy tune or discuss cheerful FOSS matters, we just can’t seem to drown out the shouts and screams coming from those standing too close to the Systemd Inferno.
Stand back, people! It’s dangerous!
The embers, of course, had been hot for some time already before the blaze flared sky-high a few months ago. Now, the conflagration appears to be completely out of control.
Need proof? Two words: Debian fork.
That’s right: Debian, the granddaddy of Linux distributions and embodiment of everything so many FOSS fans hold dear, may be forked, and it’s apparently all because of Systemd.
A more upsetting development would be hard to conceive.
‘Roll Up Your Sleeves’
“Debian today is haunted by the tendency to betray its own mandate, a base principle of the Free Software movement: put the user’s rights first,” explained the anonymous developers behind the Debian Fork site. “What is happening now instead is that through a so-called ‘do-ocracy,’ developers and package maintainers are imposing their choices on users.”
Their conclusion: “Roll up your sleeves, we may need to fork Debian.”
Quick as a flash, word traveled to Slashdot, LXer and beyond.
Down at the Linux blogosphere’s Punchy Penguin Saloon, a profound hush fell as soon as the news arrived. Fortunately, it lasted only a fraction of a second.
‘I Say Go for It’
“Freedom of choice implies the freedom to be a complete idiot, and clearly Free Software has its share,” Google+ blogger Kevin O’Brien said.
“I have been skeptical about Systemd, but I have trouble believing there are enough people this crazy to actually pull off a fork of Debian,” O’Brien added. “I predict a year from now we won’t remember what this was all about.”
On the other hand: “I say go for it if you’re that passionate about it,” offered Linux Rants blogger Mike Stone. “This is Linux we’re talking about, after all, and Linux is open source. Anybody should always feel free to do what they want with Linux, as long as they’re willing to share.
“The fact that SysVinit will still be available on standard Debian kind of makes forking it over Systemd seem a little silly, but I’m not going to stand in the way of anybody that wants to fork any FOSS for their own use,” Stone added.
Indeed, “Linux’s strength is also its Achilles’ Heel,” Google+ blogger Rodolfo Saenz opined. ‘In the Linux world, forking is inevitable. It is part of Linux’s evolution.”
‘A Lot of Misinformation’
At the same time, “I think if they were likely to actually fork Debian, they would have just gone and done it rather than throw a massive public temper tantrum,” consultant and Slashdot blogger Gerhard Mack suggested.
“Secondly, I think there is a lot of misinformation out there about what Systemd does and how it works,” Mack added. ‘At the beginning of all of this I was very worried about the stability and security of the systems I maintain after reading the nerd rage on Slashdot, The Register, and sites like Boycott Systemd, so I looked into Systemd for myself.
“What I have discovered is that they seem to be confusing Systemd with things that are bundled with Systemd but run separately using a ‘least privilege needed for the task’ type design,” he explained. “There are things I don’t like, such as the binary logs, but then I can just configure it to run through syslogd as usual and ignore the binary logs.”
Particularly “hilarious,” Mack added, is that people “suggest that only desktops need to boot quickly,” he said. “I have seen some automated systems that load VMs on demand, and they would be much more effective if they booted faster.”
‘I’m Really Confused’
It will be “a sad day if Debian forks over this Systemd thing,” longtime Debian user Robert Pogson told Linux Girl.
“I am one of the haters, I guess,” Pogson said. “I see adopting Systemd as something that kept Jessie’s bug count high for months. I just don’t see the need for it. I’ve read that some desktop users complain that Systemd is all for server users and I’ve read that some server users complain that Systemd is all for desktop users. I’m both and I’m really confused.”
Meanwhile, “do I need to learn a lot about Systemd to use it?” Pogson wondered. “I’m too old to learn too many new tricks. Does it give me any benefits, or is it just a nuisance?
“I see faster booting as a rather small benefit for a lot of nuisance value like binary logs… what’s with that?” he added. “I’ve learned to use grep on current logs to get what I need. Hiding them is just making GNU/Linux more like that other OS. Yuck!”
‘Nonfree Software Is the Real Enemy’
Debian is an organization of roughly a thousand developers, Pogson pointed out.
“They work hard and make the world a better place,” he said. “Forcing them to choose which fork to take is really cruel and unusual punishment for such generous people. If the fork is 50/50, Debian might take years of recruitment to recover. That does no one any good.
“The ‘new’ Debian would be rather weak,” Pogson added. “Would it have the hundreds of mirrors that make Debian wonderful? I doubt that. Debian is a great distro. Disemboweling it out of spite is just wrong. Why can’t we come to some amicable agreement? Why do we have to race at full speed to the edge of a cliff when we don’t know if we can stop?”
Bottom line: “If this civil war gets any worse, I may switch back to Debian Stable/Wheezy, my ‘bomb shelter,’ in the hope that I can wait for peace to break out,” he concluded. “I don’t need the drama. Bill Gates must be laughing at this waste of energy. Nonfree software is the real enemy — not folks building/using Debian GNU/Linux.”
‘It Is What Happens’
It is a sad development, Google+ blogger Gonzalo Velasco C. agreed.
At the same time, “it is what happens in the FLOSS world when you don’t listen to your peers and users and listen to others that have their own (commercial) agenda and ‘suggest’ you use a tool as hungry as Systemd, regardless of its merits and modernism comparing to old sisVinit,” he said.
“There are a lot of technical discussions and arguments out there, and Debian must show it is neither deaf nor blind and re-discuss the issue,” he added.
Red Hat’s Influence
“Do the users wish to be beholden to Red Hat’s corporate roadmap? If the answer is ‘no,’ then a fork is the only choice left open, as it’s pretty plain to see that Debian will go Systemd whether their users like it or not,” SoylentNews blogger hairyfeet said.
“It all comes down to cloud computing, and RH intends to foist its version of SVCHOSTS for Linux onto Debian and Ubuntu,” he added. “The reason why is obvious: it gives them pretty much every major Linux distro, as they are nearly all built on RH, Debian or Ubuntu.”
So, the answer is simple, hairyfeet said: “If you want RH calling the shots, then stay; if not, fork.”
‘Seems Like a Lot of Work’
Of course, there’s nothing to prevent a fork, Google+ blogger Brett Legree pointed out.
“If someone wants to do it, that’s their choice,” he noted.
“Seems like a lot of work, though,” Legree added. “I mean, I figure that most people wouldn’t care either way what init system is being used, and those who do know can probably figure out how to configure Debian (or whatever) to use a different init system. That’s been possible up to now, and I’d expect it will continue to be so.”
Forks are a lot of work to maintain, agreed Chris Travers, a blogger who works on the LedgerSMB project.
“Trust me — I know from experience, as LedgerSMB began life as a fork of SQL-Ledger,” Travers said.
Still, “there are huge differences in philosophy between init scripts and Systemd, and this is an area where there is probably room for a good Unix-like distro to keep the old ways,” Travers said. “There are certainly worse things than forks developing. This being said, I wonder if people who really want Unix should instead switch to the BSDs.”
‘Like Killing Mosquitoes With Shotguns’
The Debian community was not aware of everything the changes in the init system would bring, Google+ blogger Alessandro Ebersol suggested. “They thought it was a non-issue.”
Now that “a large number of Debian sysadmins are not pleased,” however, forking would be “an extreme measure,” he said, “and a last resort. There are still a lot of things that can be done.”
After all, Debian is “the GNU/Linux that runs on anything, in any *nix setup — remember the Debian BSD flavor, and that Debian BSD will have to be accommodated to work with the new init system,” Ebersol pointed out.
“So, I believe all is not lost for Debian, but a fork, right now, is too extreme, like killing mosquitoes with shotguns,” he concluded. “There’s still time and place to make peace and amendments in the Debian community.”
Unix != Linux
There’s a few complaining that decision that took more than a year and based on technical merits and the future direction of Linux is in opposition to the philosophy of Unix. This decision is actually in line with the philosophy of Linux and survival of the fittest.
In the article it mentions embers and conflagration when it’s been more like the slow glacial advance of Debian, with the screams of the few loud malcontents crushed in it path.
From some of the comments mentioned, in your post, I would have to say that the authors hadn’t really read through the DebianFork.org web page and what they are actually working towards.
One of the reasons that I have standardized on Debian is its ability to plug software components together in a way I’ve not really seen repeated on another OS. My typical GUI software environment is made up of: wdm, e16 and Xorg. On most systems it takes a bit of hacking together the appropriate configs and scripts to combine those. Yet with Debian I can change up the recipe and Debian puts it together. In other words I could use gdm, kdm, xdm, … and just by installing it becomes my display manager. Or if I install both it lets me choose my default.
If I want to use XFCE, LXDE, KDE, Gnome, WindowMaker, OpenBox, BlackBox, … as my window manager… once again it just plugs together. I sometimes have several window managers installed and the "environment" selector in any of the display managers I’ve installed automatically updates with the selection of window managers. The menus of the window managers automatically update with the selection of apps I’ve installed.
As a consequence I’ve often referred to Debian as an OS construction kit (without having to compile everything). It seems to me that the solution that is being pushed by the DebianFork.org site is just an extension of that feature, which has always impressed me. Plugable init systems HOORAH!
If they did fork and did stick to the pluggable init system then I would hope that the fork would become the preferred distro. It would give the freedom to all those current derivatives to choose how they want to manage their services. It would also give all of us lowly users a choice in how we want to manage our systems.
You have a need for faster start ups in a virtualized environment? You want to gain the envy of your windoze using friends with your boot times? "apt-get install systemd".
You prefer text logs? You prefer something that just does what you ask of it, without needing to apply the "do it dammit option"? Or you just prefer the openness of scripts in the start up? "apt-get install sysvinit"!
Perhaps it would even open it up to more experimentation and some better solution for "init" will rise to the top. Quite frankly, I feel, that what "DebianFork.org" is trying to get Debian to adopt, is the very definition of "Debian".
I also have to agree that forking Debian is a monumental task. They can count on me to use their fork, if they go that road. Perhaps I’ll volunteer as a maintainer. I just don’t have enough time remaining in my life to jump through all of the hoops of the current Debian organization to get involved.