It's much like the days before Soylent News ie. like an anthill has been kicked; chaos, with uncoordinated swarming over the problem space. To summarize what was discussed in the last 18 hours [1]: there seem to be systemd-less Jessie ISOs already available e.g. Exe GNU/Linux and Refracta. Someone tested systemd boot speed and discovered there's no difference, at least on default installs. There were offers of server space, and a discovering of important dependencies eg. GIMP depending on libpulse0 -> libsystemd-id128-0, discussion of GUI environments without dependency issues e.g. LXDE/LXQt, automatic package building infrastructure e.g. USE="-pulse -systemd". For those considering a move to FreeBSD there seems to be an Apple-sponsored systemd-alike being pushed by some. There was also some dreaming about the intermediate to long term e.g. if people really want the features of a launchd or systemd (or upstart or...) there could be an opportunity for an end to the SysV/BSD init split, but of course that would require years of work, consensus building and helicopter cat-herding — maybe it isn't even impossible. The #debianfork chat channel is on Freenode.
[1] Story was submitted at: 2014-11-20 03:12:49 UTC.
(Score: 2, Insightful) by Anonymous Coward on Saturday November 22 2014, @04:45PM
What matters is what happens to the project 12-24 months down the road. Do they still have critical mass, or will it just be a couple guys doing all the work? Or just copying patches from the parent project.
(Score: 2) by cykros on Sunday November 23 2014, @03:24PM
Eh, Slackware is still at the core a couple of guys, and 22 years later, it's still going strong. The bigger question is, which two guys?
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @04:46PM
Launchd in the FreeBSD world isn't comparable to systemd in the Linux world.
Launchd was never taken very seriously in the FreeBSD world. The efforts involving it and FreeBSD were just research. This is good, because it allows ideas to be tested out. There's nothing wrong with doing that. Launchd on FreeBSD was shown to not be worth it, and the efforts naturally died out.
The problem in the Debian world is that a small cabal of developers took systemd seriously, and decided to unilaterally push systemd into the systems of millions of unwilling victims. They caused massive harm, to the point of destroying what was once the most important and vibrant Linux distribution and community around.
(Score: 2) by meisterister on Saturday November 22 2014, @05:03PM
Oh thank god...
You wouldn't believe how much of a relief this is. It's nice to know that at least some developers have sense!
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
(Score: 2) by wantkitteh on Saturday November 22 2014, @06:09PM
I rather like launchd, I've used it under Mac OS X for a while now. IIUC, Apple developed it to bring the functionality of a number of different triggered launch daemons (eg: init, rc, cron, inetd) into a single unified launcher system. And then they stopped - no attack of the creeping feature creature, just a few bug fixes and a couple patches, that's it. Launchd runs quite happily in parallel with most (if not all - I think it completely replaced SystemStarter) of the daemons it was designed to supersede - given that this accommodation has resulted in duplicated functionality, I can understand how a lot of people wouldn't think too much of it, especially in the BSD community.
(Score: 2) by fnj on Saturday November 22 2014, @08:04PM
Summary of response by the grown-ups to launchd: do not need, do not want.
(Score: 2) by Pav on Saturday November 22 2014, @10:58PM
In any case there seem to be some big plans [slideshare.net] for the next ten years eg. switching from text to XML everywhere, launchd-like functionality and much more.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @12:23AM
Bear in mind that's the opinion of *one* individual and a former Apple employee.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @12:46AM
Dear god did i find it annoying to have to learn xml to just get a small config file so i can keyboard kill my xserver on gentoo.
I was jumping for joy when they switched back to plain text files. What took 3 lines in plain text took 4x as much in xml.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:28AM
Those "developers" happened to be SJW's and feminists.
Forward thinking style people.
Sadly if you're not pro SJW you're not wanted in the fork either.
(Score: 2) by doublerot13 on Saturday November 22 2014, @04:52PM
The one strength of the OSS community that endures is fragmentation. This also means that one ring to rule them all projects like systemd can never work.
I've been wondering how long or if at all this simple consequence of the linux world will dawn on Debian[and everyone else].
(Score: 1, Insightful) by Anonymous Coward on Saturday November 22 2014, @04:53PM
I think that forking Debian to remove systemd is a horrible idea. Doing that just lets the systemd freaks win.
It's better to move to FreeBSD. Or if you really, really need to stick with Linux for some reason, go with Slackware.
That way Debian can eventually die off. That's what should happen to an open source project that chooses to infect itself with systemd. Natural selection should be allowed to take place, and rotten projects like Debian will eventually be culled when the most important users have moved elsewhere.
Face it, Debian is dead. Any distro that's derived from Debian is also inherently dead.
Move to FreeBSD. It gives the same benefits of Debian, but with a much better overall experience.
(Score: 3, Insightful) by doublerot13 on Saturday November 22 2014, @05:00PM
I'll admit for the first time in a long time it is being considered as an option.[FreeBSD]
Thanks to things like everything-in-a-browser or virtualization the host OS doesn't have to do much but be stable and get out of the way. FreeBSD does this well.
(Score: 2) by zeigerpuppy on Saturday November 22 2014, @05:05PM
Debian still has significant advantages over FreeBSD.
For instance FreeBSD does not have a mature Xen system. When I needed the combination of Xen and ZFS, I looked at all options, Debian was the only distro that had mature support for both.
I don't think Debian will die so easily, I do hope that the exodus eventually makes them reconsider their decision regarding systemd.
Otherwise, a fork will be the only way. At this point it would probably not be too difficult as the bulk of packages are not dependent on systemd. I agree, however, that if there's no policy change this may become harder to maintain in the future.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @05:22PM
What are you going to do with your Debian infrastructure now that the Debian project has been infected by systemd?
On one hand, it would be a bad idea to upgrade to Jessie once it's available, because it's clear at this point that systemd is not suitable for use. The stability of your systems will be at risk.
On the other hand, if you don't upgrade your systems to Jessie, you'll eventually start missing out on critical security fixes. If the recent bash and OpenSSL disasters have shown us anything, it's that even the most established and widely used open source software can be very vulnerable to severe exploits.
(Score: 3, Insightful) by zeigerpuppy on Saturday November 22 2014, @06:01PM
Luckily there's until at least mid 2016 until security updates stop for wheezy, hopefully by then this whole mess will be sorted.
If Xen has better support in FreeBSD by then, it may be a reasonable direction to go with my infrastructure, at least.
(Score: 1) by macson_g on Saturday November 22 2014, @07:28PM
"infected"? really?
(Score: 3, Insightful) by Anonymous Coward on Saturday November 22 2014, @07:33PM
I think "infected" is a good word for what systemd has done to Debian. When I updated my Debian system and systemd was installed, and then it stopped booting properly, it was equivalent to a Windows system getting infected with malware.
(Score: 1) by Horse With Stripes on Saturday November 22 2014, @09:32PM
"infected"? really?
You would prefer infectd? Or would that be infectedd?
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @11:01PM
"You would prefer infectd? Or would that be infectedd?"
You would prefer infucktd? Or would that be infucktedd?
TFTFY
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @10:16AM
Hi Lennart!! How's it going?
(Score: 2) by sjames on Saturday November 22 2014, @09:14PM
I actually tested that to make sure I would have a stable place to go while the forks work out.
Using the beta installer in expert mode on a vm, I set it up w/ btrfs, shelled out, created an '@' subvoulme and mounted that subvol to /target, then did a normal install. I de-selected GNOME and selected xfce and Mate instead.
Just before the bootloader install, I shelled out again and chrooted into /target. Then apt-get install sysvinit sysvinit-core. Exit and complete the install.
That leaves pieces of systemd in the install but doesn't run them. I got a nice and fast boot-up into a proper system.
That probably breaks pulseaudio, but I don't use it anyway.
(Score: 2) by fnj on Saturday November 22 2014, @08:15PM
Xen is a loser for losers. It had its day and blew it. I was into it and had some server VMs set up. RHEL5 was clearly pushing it and it was clearly The Way. Then RHEL6 dropped it like a hot potato suddenly and pushed KVM on us. I was pretty torqued. But I adapted and learned KVM and started using it. And found it to be far superior.
KVM has been ported to FreeBSD. Xen never will be.
If I was virtualizing only FreeBSD on FreeBSD I would use Bhyve.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @11:30PM
Yes! My decision is clearly the correct one, and nobody else could ever have any different needs from myself.
I also use the correct operating system, unlike many of you fools out there, and listen to the proper music. Anything else is just not right.
While I'm at it, the right foods are the ones I like, and parties in line with my political beliefs are the right ones to vote for.
Anything less is simply silliness!
If you examine my statements above, you will find that they are true. I will hear no contradictions, for there can't be any that can counter my opinions.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:31AM
caeogen is that you? You sound like a pompus douchebag. Must be caeogen (#debianfork op)
(Score: 2) by meisterister on Saturday November 22 2014, @05:08PM
I would like to politely disagree. If Debian is forked, the base concept of Debian will stick around (as far as I can tell, it's supposed to be a fast, free, open server OS that cares about its users... ha!), but Debian itself will still rot and die. Eventually people will just know the names of Exe and Refracta, and Debian will be a distant memory.
To be honest, the only reason I use Debian/Ubuntu/Mint is because of its package system, and I'd rather not see apt die because some asshats decided to screw over an awesome OS.
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @05:14PM
That may be the case, but the problem then is that this true core of Debian, without systemd, is still very susceptible to future infection by systemd.
This is where the BSDs have a big advantage. They have a number of things, from not using the Linux kernel to having intelligent developers who wouldn't put up with hipster idiocy like systemd, that will make it damn near impossible for systemd to infect them like it has infected so many Linux distros.
That's not say that the BSDs couldn't get infected. They very well could! But their natural immunity will surely help prevent infection.
(Score: 2) by Thexalon on Saturday November 22 2014, @05:44PM
I view it this way: Organizations rise and fall, companies come and go, but the code lives on as long as somebody, somewhere, has the source code.
For example, there's basically no risk of apt and dpkg being completely destroyed forever: I'm one of lots of people that have recent source tarballs and don't intend to delete them until I get newer source tarballs. So if Debian crashed and burned completely, all I'd need to do is set up a torrent on a server somewhere, and say "Hey everybody, come and get dpkg and apt-get right here". At worst, any further improvements or bugfixes would grind to a halt because developers were working on other things.
Plus, you don't think that a systemd-less Debian fork wouldn't use apt? It's a great tool, it's free (as in both speech and beer), what they'd be forking from already has it, and there'd be no reason to replace it.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.
(Score: 3, Insightful) by novak on Saturday November 22 2014, @08:55PM
Sorry, but no.
First:
Face it, Debian is dead. Any distro that's derived from Debian is also inherently dead.
No, actually. First of all, systemd probably makes things easier for companies, by unifying much of userspace. Anyhow, lots of bad ideas don't die. Windows is a _terrible_ idea but it's far from dead. systemd is just one more horrific idea that is not going to fail because it has too much support from people who have the wrong priorities.
Second, linux itself does not depend on systemd, but it's important to keep things like eudev around so that a distro without systemd works as well as one with systemd. I don't plan to abandon linux just because of systemd, especially for embedded systems. But as long as I can keep it working well on the desktop, I plan to. I already used "niche" distros which don't use systemd, so this hasn't hit me as hard as some.
I fully support the fork. I hope that some of the distros depending on debian for upstream switch to the fork to prevent linux from becoming as much of a mono-culture.
novak
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:34AM
Debian is run by social justice warriors and feminists.
I hope it's dead. They kicked out anyone not of their stripe years ago.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @09:07PM
If one is going to mention Linux alternatives, please don't limit the list to just Slackware. There is also Gentoo Linux which is still intent on providing a choice of init systems. Gentoo does not exclude systemd but it also does not mandate systemd. In fact, with Gentoo, systemd is, as it should be, just another option.
Gentoo should be the example for all other Linux distros.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @01:16AM
Supporting, and not explicitly excluding, systemd means that Gentoo is half-way into the grave.
In Gentoo's case, all they need to do is pull a Debian and focus on systemd, and then you're fucked.
In Slackware's case, systemd would first need to actually get itself integrated, before it could be used as the default. This actually reduces the risk of systemd infection significantly.
Unless Gentoo completely removes systemd as an option, we just can't consider Gentoo to be a viable option.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:24AM
Please do not spread FUD.
Gentoo has singlehandedly developed EUDEV as a replacement for the systemd dependent UDEV. This act underscores the commitment of Gentoo to bring their users choice. EUDEV is a courageous contribution by Gentoo to the Linux community.
Debian could have adopted EUDEV and followed the same path, but they, and most other distros, chose the cowardly and lazy way out of the dilemma. As a consequence, Debian and the others deserve little respect and much condemnation.
Gentoo demonstrates what simple creativity can accomplish.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @01:12PM
The only ones responsible for any fear, uncertainty and doubt in this situation are Gentoo. They need to come out and say, "No, we will never support systemd going forward." and then they have to remove any existing support for it.
I don't care about udev and eudev and whatever else they might be working on. Until they eliminate systemd, and put safeguards in place to ensure it's never integrated with Gentoo, there will be fear, there will be uncertainty, and there will be doubt.
(Score: 2) by novak on Sunday November 23 2014, @05:03AM
Gentoo is very nearly LFS. Any system you imagine is possible, simply compile the required packages. Of course systemd is an option- I'm not sure how anything would not be an option in such a setup.
As to the comments along the line of "12% of gentoo users think systemd should be the default and that's 12% too many," either way it's a matter of seconds to switch. It's basically a moot point in gentoo.
novak
(Score: 3, Informative) by rleigh on Saturday November 22 2014, @10:35PM
I'm a Debian developer and user of about 16 years. I've spent a lot of time this year getting started and increasingly involved with FreeBSD. I moved some servers over to FreeBSD 10.0-RELEASE after it came out to test the water. Now running for 9 months on ZFS with some jails, including Debian kFreeBSD jails. Now testing out PC-BSD 10.1 after its release a few days back. It's pretty slick, and gives me a perfectly functional KDE desktop. If it works out well, it will probably replace my last bare metal Debian unstable installation if it's as effective for development, and I see no reason why it shouldn't be.
After investing many hundreds of man hours into Debian, as well as the associated emotional investment, it hasn't been easy to contemplate switching. But I've been deeply unhappy with things for over 18 months now. The direction was set and in the absence of drastic action, the reality is it's rapidly becoming a system I don't like. The quality is already going downhill; these systemd folks simply don't have the same concern for testing and corner cases we did with sysvinit. If you're a corner case you're doing things wrong and must change! We never had that attitude. With sysvinit/initscripts we support all sorts of crazy setups, even if not officially blessed. If they broke we considered it a bug; historical baggage is not something you casually drop and burn users who had a system installed a decade ago. You can plan a migration to a new way, or you carry on supporting it. You never drop it outright. Debian has ~20 years of that legacy, most of which will be dropped on the floor in the name of modernity. Maybe it will be "cleaner", but it's also destroying a good part of what makes Debian Debian. Theoretical perfection shouldn't trump compatibility with our past.
Anyway, if I'm not going to be using it, I won't be contributing to it. Altruism got me started with this, and kept me going for a long time, but it can only go so far. When I look back at the reasons I started using Debian all those years ago, the current system isn't what I want. systemd is not the only reason, it's just the straw that broke the camel's back. The general trends in the development of gnome, freedesktop stuff, are all IMO taking us backward, not forward. These modern systems are less usable than what we had years back. The decline has been slow, but it's been going on for a long while now. If I was to blame it on one major thing, I'd say GNOME/GTK failed entirely to encourage a healthy ecosystem of third-party developers. Primarily due to being technologically backward and not caring about developer's needs or applying needed patches. I should know, I was once such a third-party developer and the experience was miserable. If you look at the development of e.g. gnome at the moment, it's primarily focussed upon trivial look and feel tweaks and the desktop. There is almost no care for people actually using this stuff to develop applications, and I think this has led to the current insular and inward-looking environment. They are by no means the only offenders, but I think they are the most deserving of criticism for trashing GTK with version 3, making it unusable for third-party development outside gnome. The original gnome was a platform for development of applications.
I'll say this about FreeBSD: in some aspects it's more primitive than Debian. It's not as slick. But it's perfectly functional, and in many ways reminds me of what Debian was like a decade back. In a good way. The current evolution of the pkg tool to install stuff from a repo rather than building everything from ports makes it just as convienient to administer as a Debian system. I'm now in the last stages of porting my software to BSD, adding BSD-specific functionality such as ZFS and jail support.
Given the choice, I'd certainly prefer to stick with Debian. However, that choice was taken out of our hands by the people foisting such terrible changes upon us.
Regards,
Roger
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @01:06AM
Don't let these mercenary carpetbaggers to run you out. Help the fork. Many of us who have been merely users are willing to contribute towards that effort. Debian is too good to abandon.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:36AM
Seconded. When they edged out Ted Walther it was clear that something new and rotten was in debian.
Time to rebirth the phoenix.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:38AM
The one thing linux has that BSD doesn't is the P_X/G_security patch to the kernel.
That's what keeps me on linux.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:44AM
Wheezy SysV is awsome. I've told all my friends NOT to update to Jessie (it's a bitch), stay with wheezy forever.
(It even has parallel boot up)
Maybe fork from wheezy. Most of the software is done - feature complete, no need to update other than security patches to daemons.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @10:11AM
I can't say I understand what the big deal is. If you want to use sysvinit, then you can. The summary references Gimp requiring a systemd library, but that library doesn't depend on systemd being used as the init system.
Debian seems plenty happy to continue supporting sysvinit. If the problem is that packages have dependencies on systemd libraries, then it seems like it would be much more productive to fix those dependencies in Debian or upstream than to fork Debian.
(Score: 1) by jmorris on Saturday November 22 2014, @05:21PM
Ok, somebody point me at the Exe GNU/Linux based on testing, all I see them talk about on the homepage is being based on current stable. Refracta for now is just an alt book media that is trying to use the Debian repos. They seem to know that isn't going to work long term but it gets the action started and somebody had to start.
Now we need to revert GNOME3 and put MATE in as default desktop so it will look pretty much like last stable Debian. Just make a clean sweep of the cruft and start going forward from there.
Hope both the systemd and one or more non-systemd forks thrive. If Debian dies the windows refugees will simply swarm elsewhere, bringing their defective ideas. The problem is they vastly outnumber the UNIX folk in the Free Software movement.
(Score: 1, Insightful) by Anonymous Coward on Saturday November 22 2014, @05:47PM
The BSDs, especially OpenBSD, have a built-in defense against systemd. The core OpenBSD and DragonFly BSD developers would never put up with total bullshit like systemd, and they're also quite exclusive, which further helps to keep hipsters and systemd deviants out of the picture. FreeBSD and NetBSD are maybe more susceptible, but I don't think they'd give in anywhere as easily as Debian did.
The inclusivity of Debian has obviously been a huge problem for it. What was meant to help it thrive has instead caused it to fight for its very survival.
Every open source project needs to take a hard look at what Debian has gone through. The obvious thing to do now is to put up a "hipster firewall". Hipsters and those who haven't taken a strong stand against systemd need to be blocked from participating in any existing open source project. It's clear at this point that if these sorts of people get involved in an open source project, that open source project will end up destroyed. GNOME, Firefox and now Debian are proof of this.
(Score: 2, Insightful) by jmorris on Saturday November 22 2014, @06:26PM
FreeBSD is going to do what Apple wants. Not sure I want to hitch my wagon to the fail that is the destiny in that direction now that reality is reasserting itself. OpenBSD is small, very small, which is good for a server but bad for a general workstation. And all of the BSD forks are very spotty in hardware support, which is where the Linux kernel shines. Put BSD userland atop a Linux kernel and it would get really interesting. But for me, I like a properly package managed distribution. I just want it to remain UNIX, not a port of WindowsNT internals to the Linux kernel.
As for your 'hipsters' arguments, I don't buy that mindset. It isn't so much 'hipsters' as non-UNIX thinking. It is a 'culture problem' though. Specifically it is younger people who accept the base concepts of Win/Mac implicitly and code with those in mind. We made the mistake of allowing unlimited immigration into our world while making no effort to acculturate them to The UNIX Way before handing them commit access to the core cultural artifacts of our unique culture. It is clear that in Debian land, UNIX folk are now a minority and the new 'majority of the ascendant' no longer plan to bother even paying lipservice to what we want.
And as for GNOME, just no. You could not be more wrong. They didn't get invaded by hipsters, look at the key people behind GNOME3, they are the same old people. We did have ample warning from the early GNOME2 days they were not the sort of stable people who should have been in charge of such an important project and they finally went off the rails for good. But that is OK, Free Software can handle it, even though they knew GNOME2 would be the most important competitor to GNOME3 and carefully took steps to ensure co-existance would not be possible the MATE team has proven them wrong and is fast making it a quality traditional desktop for those wanting one. Same is going to happen with Debian, forking is what ensures a project can't go too far off the rails.
(Score: 2) by fnj on Saturday November 22 2014, @08:24PM
Utter bullshit. FreeBSD has nothing whatsoever to do with Apple. Apple just took an ancient version of FreeBSD a long time ago and used PARTS of it together with a bunch of stuff, including the kernel, that is completely different, and created their own thing. It is frozen in time with the ancient pieces. Whatever Apple does with it is their own business. No foreign directions have ever come back to infect FreeBSD, or ever will.
(Score: 2) by meisterister on Saturday November 22 2014, @07:07PM
I'd say that NetBSD may actually have even more resistance to that way of thinking. Basically the entire point of that OS is to be portable to every computer system imaginable, and godawful kludges like SystemD would fail hard in that kind of environment.
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
(Score: 2) by choose another one on Sunday November 23 2014, @02:42PM
The BSDs, especially OpenBSD, have a built-in defense against systemd.
The BSDs in general have a built in anti-systemd defence in that they don't have cgroups, on which systemd depends.
They _don't_ have a built in defnece against things _like_ systemd - otherwise launchd would not exist, and it does.
OpenBSD is different, it has an individual built in defence against everything (begins with T).
(Score: 2) by morgauxo on Saturday November 22 2014, @06:54PM
I wish the Windows refugees would go to ReactOS. They could actually do some good there!
(Score: 2) by HiThere on Saturday November 22 2014, @11:44PM
If you check their site you'll find one, but I don't know how maintained it is...but that's not important.
An OS doesn't NEED to follow the path taken by testing or sid, but it's a big advantage to have a stable place to start from. There's absolutely nothing wrong (and a lot right) with starting off from the "stable" branch, and being careful about what you add. First you need to ensure that all trademarks are handled properly. (If you're keeping the same licenses, then Debian stable is a relatively safe place to start from...you just need to keep the security patches current.) Then you change the artwork so that you're recognizable. THEN you start figuring your path of development...and there are good reasons to NOT pick the same path Debian choose. After all, if you liked that there wold have been no reason to fork.
But to build a community of support, you need to attract people. So you need something that's new and different. So far they list 5 people. That's not enough to do much more than maintain security...and that only because they started from a stable place and they can use a lot of the same patches that Debian uses.
Is it viable? I see no way to tell at this juncture. In the optimum scenario they would at some point merge back together with Debian as a separate project that maintained a separate installer, etc. (Just what "etc" is depends on how systemd develops...but it's beginning to look extensive.) I don't expect this, not after the results of the recent vote. But the head is a professor emeritus, and he may well be able to talk some of his college's active professors into having some of their graduate students in the Computer Sciences department support the development. It's not obvious that it would work, but that's the kind of scenario that BSD Unix came from. OTOH, I don't know what college he's emeritus from, so this might not be practical.
Whatever, remember that Ubuntu was nearly identical to Debian at first, and the first Mandrake Linux was nearly identical to the Red Hat of the day. Changes, if they're to be successful, need to start off small.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:49AM
Wheezy is a fine base to start with.
Build your house on rock, not shifting sands.
Then you can add additional packages that are not in debian and/or update existing packages IF there is NO regression or creeping dependancies.
Fork ALL of userspace. The only way out of systemd (and other new nontraditional nix) mess. Debian makes it easy as a repo clone to start.
(Score: 1) by fritsd on Saturday November 22 2014, @05:37PM
"The #debianfork chat channel is on Freenode."
That's great. Um.. So, how do I subscribe myself to this, and lurk or join in the discussion??
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @05:53PM
Do you have an an IRC client installed?
(Score: 3, Informative) by Aiwendil on Saturday November 22 2014, @06:08PM
Quick intro to IRC:
1) Install an irc-client (irssi is very popular in linux (it is the repo of most dists), there are several irc-clients (some even in java) and _lots_ of software actually supports the irc-protocol so a quick search for ircclient operatingsystem (ie: ircclient linux) will turn up quite a few)
2) Pick a nickname (no spaces, at most 15 chars) ( /nick Foobar )
3) in the irc-client connect to Freenode ( /connect irc.freenode.net )
4) join the channel #debianfork ( /join #debianfork )
Quick intro to common commands in ircclients:
/join #channel - joins a channel
/part #channel [reason] - leave a channel with "reason" as "quit-message")
/query nick - open a "private window" with someone
/server server - disconnect from currently connected to servers and connect to a server
/connect server - connect to a server, and remain connected to other servers (if currently connected to any)
/disconnect server [reason]- disconnect from server (possibly with a reason).
/quit [reason] - close the irc-client (with a reason if still connected)
/nick Nickname - set the nickname you want to use, this will fail if it is taken on the network you are connected to.
/help - will bring up the help for you client in most clients.
In most text-based clients alt+N (alt+2 and such) will jump to the window numbered N (window 2 in case of Alt+2).
Tab-completion for nicknames in channels are supported in most clients as well.
It is recommended that you read the documentation for the irc-client you chose (at least the usage-section) so you learn how to set up defaults and such.
People on irc are usually quite helpful if you are polite and do not demand help but rather ask for it, oh, and the tolerance for spamming is _very_ low on irc (due to necessity) so be patient.
(Score: 1) by fritsd on Saturday November 22 2014, @06:40PM
OK it worked thanks.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:51AM
You really need one on oftc.net too.
Debian has channels on both.
(Score: 2) by ticho on Saturday November 22 2014, @05:38PM
I'm just staying away from as much of this as possible for as long as possible, and will wait how things turn out. As long as there's Gentoo, I always have a rather simple and effective exit strategy, but for now, a systemd-less Jessie still works for me as a primary desktop.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @05:42PM
What are you going to do when Gentoo eventually switches to systemd? Use Slackware? Hell, you'd be better off using a distro with systemd at that point!
This is why you need to consider moving to one of the BSDs today. They're shaping up to be the only viable way to use a UNIX-like system in the near and distant future.
(Score: 2) by ticho on Saturday November 22 2014, @06:10PM
I will evaluate my options when (if) that happens. So much can change until then.
(Score: 1) by curunir_wolf on Saturday November 22 2014, @07:13PM
Yea, I really don't think that's going to be an issue. See this thread [gentoo.org] on systemd, and note the votes (only 12% want systemd as the default). That's not likely to change any time soon, but Gentoo, in principle, would never require use of systemd. There will always be a way to use it, and a way to avoid it. USE="-systemd".
I am a crackpot
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @07:18PM
That's still 12% too many.
(Score: 2) by CoolHand on Saturday November 22 2014, @09:54PM
Hmm, your post made me think of this:
Benjamin Franklin : “People willing to trade their freedom for temporary security deserve neither and will lose both”
--------------------------------
Maybe not directly applicable, but I think the principle is close. If you're willing to sit on the sidelines letting other people fight for your freedom, you may end up with no freedom at all...
Anyone who is capable of getting themselves made President should on no account be allowed to do the job-Douglas Adams
(Score: 2) by ticho on Sunday November 23 2014, @02:53AM
Yes, you are entirely correct, and I fully realize that, and I might regret it one day. However, the whole situation leaves a bitter taste in my mouth, thanks to all the extremists (from both pro- and anti- systemd crowds) spewing their bile on various internet forums. Therefore I try to distance myself from it, and just keep on hacking on whatever few small projects I feel like scratching an itch on, and not feeling dirty.
(Score: -1, Troll) by Anonymous Coward on Saturday November 22 2014, @06:28PM
I see. So that means once debianfork is released, it'll be much like the days after Soylent News: a no-name project populated by a tiny handful of cranks, incapable of raising a pittance to scrape by with their yelling-at-clouds community while the vast majority keep chugging along, getting work done without the drama.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @07:17PM
Öh comeon, we need some drama.
Otherwise, take all the sound out of space movies and the reality is a bit boring.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @07:22PM
You're thinking of Pipedot.
(Score: 1) by curunir_wolf on Saturday November 22 2014, @08:12PM
It's much like the days before Soylent News
I see. So that means once debianfork is released, it'll be much like the days after Soylent News: a no-name project populated by a tiny handful of cranks, incapable of raising a pittance to scrape by with their yelling-at-clouds community while the vast majority keep chugging along, getting work done without the drama.
Yea, I don't think so. I've seen the "vision" of the systemd developers, to wit:
What we are working on:
Verifiable OS images
All the way to the firmware
Boot Loading
No, thanks. I'd prefer to own my computing devices, not lease them and require permission from some manufacturer on what I can run. There are already enough systems like that out there.
I am a crackpot
(Score: 2) by kaszz on Saturday November 22 2014, @10:20PM
Ah, you mean like smartphones where the signed boot sequence ensures people don't run any norty software that lack blessing by the mass surveillance community? ;-)
(Score: 2) by Phoenix666 on Saturday November 22 2014, @06:47PM
I don't mean to spin out one small part of the human experience into a general statement. The systemd controversy is not the sum total of human endeavor. But, it does rather suggest a productive model for collective endeavor that surpasses the original, centralized one. Humans commit errors all the time. Thanks to group dynamics, some errors can continue into implementation despite individual objections.
So the ability to fork important projects evades such collective failures. That's a big deal. I focused my entire career in social science on looking for such examples 10 years ago, and found none. I am not a marquee name and my findings would certainly carry no authority, but I was at one of the top 3 social science institutes in the world (University of Chicago) and I can assure the community that there was nothing then or since behind closed doors that would have investigated these sorts of endeavors in detail. Most, frankly, have focused on the aftermath of the failures of the labor movement.
Freedom of thought and endeavor are, I believe, a core component of human survival. I hope that amid the purely technical discussions of the merits of systemd that debate participants would be mindful that they are also charting something greater than that specific technical discussion. How they talk is as important is what they talk about.
Washington DC delenda est.
(Score: 2) by kaszz on Saturday November 22 2014, @10:16PM
Nerds are quite successful because they spend their resources on facts not social courtesy. So the difference is how failures are handled.
(Score: -1, Offtopic) by poutine on Saturday November 22 2014, @07:19PM
I'm sorry, it's a little ridiculous to compare some no-name site's struggles with something that affects a large portion of enterprise companies. Keep the circlejerk editorial out of articles, you're not as important as you think you are soylentnews.
(Score: 0) by Anonymous Coward on Saturday November 22 2014, @07:39PM
And why should anyone care what a nobody like you think?
(Score: 2) by CoolHand on Saturday November 22 2014, @09:48PM
I've moved to PC-BSD on my work desktop since this started (replacing Siduction), however, I've also been re-investigating OpenIndiana and Dyson, both based on Illumos (OpenSolaris). They both seem workable. I had thought OpenIndiana dead a while back because there updates were so slow, but it actually seems to be getting healthier. Dyson is really cool because it recompiles Debian packages, and a pretty good package repository because of this. I'm just afraid how systemd (and possibly shrinkage in the number of packages if a split) could affect it in the future...
Anyone who is capable of getting themselves made President should on no account be allowed to do the job-Douglas Adams
(Score: 1) by lentilla on Saturday November 22 2014, @09:54PM
I sincerely hope that any fork is truly a scion of the original tree. Debian is more than just a collection of .debs and an awesome package manager - there must be a commitment to stability and strict adherence to the Debian Free Software Guidelines. Debian's longevity is a direct result of all the above. That's why many projects are based on Debian rather than the other way around.
Frankly, I'd prefer it if the project was simply called "Debian2" or "DebianMinus". Import ("fork") the entire ecosystem so that we could simply point our /etc/apt/sources.list at the new project and be done with it.
Just like a tree, these forks are occurring because the trunk picked up a life-threatening malady. With time, it may indeed turn out to be a fortuitous mutation. It saddens me to see all this energy diverted into a fork but it is a sensible strategy for survival.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:54AM
seconded.
(Score: 1) by hopp on Sunday November 23 2014, @04:51AM
Thirded.
Funny I came to debian in the late 90's form FreeBSD and I may return there if there is no viable systemd free debian-ish distro.
(Score: 1) by anti-NAT on Saturday November 22 2014, @11:25PM
As Linux user since 1991, and a systemd user since changing to Fedora a couple of years ago (having previously used Slackware, Redhat, Debian, Ubuntu and Arch Linux), I've been amazed at the amount of entitlement expressed by those objecting to systemd. There seems to be a view that just by using something that you've been given for free, you are somehow returning so much value to the person who created it that you are entitled to tell them what to do and how to do it.
People who are creating and giving this work away for free are getting something from other people who choose to use it - social validation of their work. However the return they get is completely disproportional to the amount of time and effort they've put into it. Consequently they retain the choice as to what to do, how to do it, or even whether or not do continue to do it at all. If nobody uses their application it probably won't bother them too much - they created the application to scratch their itch, so that is the fundamental definition of success - does it work for them. If somebody else benefits from it, that's great. But if nobody else does, it doesn't really matter.
The joke of the recent Debian Re-Proposal to make the whole Debian distribution not depend on any particular init system is that it was attempting to force more work onto those who develop various applications and systems, by those who won't be doing or volunteering to do any of this additional work at all. It is one of the best examples I now know of the cliché, "all care, no responsibility."
The people who do the work make the choices. If you disagree with Debian using systemd, put the effort into forking it. If you can't be bothered putting any effort into forking it, then you should accept what you're given (for free).
(Score: 1) by rleigh on Sunday November 23 2014, @12:24AM
It's not about entitlement. systemd has the same (DFSG-free) licensing as all of the rest of the software on the system. However, all of the rest of the software on the system is optional, and it's up to me, the admin and user, if I wish to make use of it. systemd is being forced upon us whether we want it or not, taking away my choice.
When we talk about software freedom, we have traditionally focussed entirely upon licensing because that's what mattered. However, software freedom is ultimately about empowering the end user, and while free licensing is a big part of that it's not the only part. If a freely-licensed software package artificially restricts what I can do with the system though its policies, complex dependencies and forced bunding of lots of interdependent components, then my freedom to replace individual bits to meet my needs has been reduced. This is where systemd is taking us, and is why I and many others object to it.
You talk of forcing work onto people. Let me tell you that asking developers to retain initscripts for their packages is far less work than supporting systemd. Firstly they already exist since they have been maintained by all packages for the last two decades. Secondly because systemd itself forces developers to do a lot of work to unbreak their software to work with systemd. I've had several "bugs" for my software, all of which were due to systemd affecting how fairly fundamental stuff works (like basic POSIX system calls for crying out loud). There were each huge breaks with the past (and all other UNIX and Linux systems) which are each individually a massive failure on the part of the systemd developers since each is a hard break with backward compatibility. This is software which had been working perfectly for a decade! You just don't break things like that in an operating system. But it's clear that quality software engineering is not part of the programme here.
I see in a lot of these discussions (paraphrased) "if you don't like it, fork it", as you have mentioned above. This is an appalling attitude. It's bandied about without real understanding of its implications or meaning. It's not reasonably possible to fork an entire distribution due to the maintenance overhead. But more than that, free software has always been about mutual cooperation and forking is the antithesis of that. It's a consequence of developing things without cooperation. It should only happen in the most extreme of circumstances when no agreement over direction can be made. It's a sign of failure. If it were not for systemd driving a wedge through the entire community, we wouldn't even need to consider such action.
(Score: 1) by anti-NAT on Sunday November 23 2014, @01:59AM
However, all of the rest of the software on the system is optional, and it's up to me, the admin and user, if I wish to make use of it.
So what makes a distribution? I think it is a set of packages, file system layouts and the corresponding glue to make it into a functional computer that can run applications. If all of the rest of the software on the system is optional then you really haven't got a distribution - or, alternatively, what actual distribution you're running shouldn't matter to you, as much as apparently running the Debian distribution actually does.
systemd is being forced upon us whether we want it or not, taking away my choice.
Tell me when you specifically chose to use a distribution that used Sysv init scripts? Or was that choice made for you, without any of your input at all, because it was inherent in your choice of the Debian (or some other) distribution?
It isn't possible to create a distribution without making choices about what software to install by default, and how to glue it all together. Debian in this regard is probably more democratic than any of the others, because in some cases these decisions are put to a vote. If you've chosen not to participate in that vote, and then disagree with the outcome, then you've only got yourself to blame.
You had a choice to participate in the decision to change to SystemD.
You have a choice to stay with the existing non-SystemD version of Debian you're currently using. You also have a choice to ignore any security issues that arise with it, or a choice to backport security patches to those packages.
You have a choice to find a different distribution if you don't like the direction Debian is going.
You have a choice to participate in a fork of Debian.
Seems to me you've had or have a lot of choice.
It's bandied about without real understanding of its implications or meaning.
Perhaps you missed the part about me using Linux since 1991? I spent the whole of the 1990s as a system admin. I fully appreciate the amount of work involved in a fork.
Those that are complaining about this should instead redirect their energy to a fork. Talk is cheap, which is why the volume of complaining about this is excessive.
It's not reasonably possible to fork an entire distribution due to the maintenance overhead.
Another example of "all care, no responsibility.". You don't want wear the cost of maintenance overhead of a fork but you expect application developers or packagers to wear additional maintenance costs by having to develop init scripts for the variety of init systems, and you want to force that on them. You're good at complaining about choice apparently being taken away from you, but quite happy to take it away from somebody else.
But more than that, free software has always been about mutual cooperation and forking is the antithesis of that.
You're quite wrong about mutual cooperation being what free software is all about. Forking is fundamentally what is all about - if you don't like what is happening, you've got the source code so you can do something else with it. The freedom to fork is the whole point of open source.
It should only happen in the most extreme of circumstances when no agreement over direction can be made. It's a sign of failure.
No, it's no more than a sign of disagreement. If you believe variation is a sign of failure, how do you explain the multiple free Unix's, multiple Linux distributions, multiple boot loaders, multiple c libraries, multiple graphics libraries, multiple graphics environments, multiple command shells etc. that both continue to be used and continue to be maintained? Is Linux a failure because there is more than one Linux distribution?
The Free/open source software process is Darwinian. It is survival of the fittest, specifically fittest for purpose. Debian will survive even if nobody is using it other than the package and distribution maintainers. If they're happy with SystemD then that is all the the people who have to be happy with SystemD for Debian to survive.
(Score: 2) by choose another one on Monday November 24 2014, @09:18AM
+1 dead right wish I'd written that.
I keep wondering about all these people being forced to use certain software, forced to accept design decisions that they disagree with made by software vendors/maintainers, having their choice taken away... seems like they need a visionary leader to stand up and proclaim a manifesto for freedom.
Oh, wait, someone already did that 30yrs ago, and when I looked up the manifesto the phrase is right there as I recalled it from the 80s:
Users will no longer be at the mercy of one programmer or company which owns the sources and is in sole position to make changes.
It's funny though, I can't find the bit in the manifesto where it says if you bitch and whine and threaten loud enough, then maintainers shall do your bidding. Nor can I find the bit where the users are supposed to summon rms and he appears in fairy wings and a pink tutu, waves his magic beard and grants the user a new maintainer. No, I found this:
a user who needs changes in the system will always be free to make them himself, or hire any available programmer or company to make them for him
So that is what free software _is_ - users are always free to quit bitching and start fixing. If you thought it meant free as in beer, then it's worth every penny you paid for it.
Oh, and while we're looking up the history and what things mean, I wish the "it's not good because it's not Unix" people would read up on what the GNU in GNU/Linux actually means.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @12:41AM
Or move to a better system with like minded developers.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @01:19AM
It's not "entitlement" to want Debian to work perfectly fine, like it has worked without systemd for the past 20 years!
(Score: 1) by anti-NAT on Sunday November 23 2014, @02:09AM
Prove that Debian won't work perfectly fine with systemd. If you don't trust the Debian distribution maintainers to make wise decisions about something so critical, then you most definitely should find a different distribution.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:24AM
The debian devs are all SJWs and feminists now. They kicked out all the libertarian or conservative minded devs years ago (example: Ted Walther). That should never be forgotten.
They're riding off the coattails of the greats that are no longer with debian.
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @01:27PM
nice try at proving debian won't work with systemd... cos debian devs are feminists?
dude... i want what you're smoking, cos it must be some funky shit
(Score: 0) by Anonymous Coward on Sunday November 23 2014, @02:26AM
Yea the people who "do the work". Do you mean packagers or programmers?
Oh you mean packagers. Fuck them.
--
Opensource game rejected from Debian for authors social beliefs
An opensource casino video game was recently posted to
the debian bug tracker as a request for packaging, as
is the standard method for pursuing such things in debian.
The bug was quickly closed, tagged as "won't fix"
The reason given by one of the debian developers
alluded to the authors conservative views and his
advocacy of them:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=770314 [debian.org]
The piece of software in question is licensed
under the GPL and is one of the only of it's
kind for linux (ascii-art console slot machine software)
Is professing progressive politics now a hard requirement
for being allowed to contribute to opensource?
(Score: 2) by Arik on Sunday November 23 2014, @04:10AM
If they do it the systemd way, I will use something else.
That's not entitlement, that's standards.
If laughter is the best medicine, who are the best doctors?
(Score: -1, Flamebait) by Anonymous Coward on Sunday November 23 2014, @01:23AM
I offered my years of programming and system admin experience.
I was banned, reason: not pro-women's rights.
Only SJWs and pro-feminists are allowed to help the fork.
I hate all these techi fucks. They are all feminists.
Every single one of the.
Every FUCKING one.
(Score: 2) by Pav on Sunday November 23 2014, @04:17AM
After long periods of silence with a few updates eg. experiments with packages compiled by hand etc... there will be something unambitious/humble to try (hopefully soon). From IRC:
[nextime] anyway, if anyone want to help putting hands on things, now is the right moment to write to vua@debianfork.org listing what you can do/offer to the project
[nextime] we are starting to build a debian repo ( with dak or mini-dak ), we are searching someone that can draw a temporary logo, we need to put online a initial web site for the new distro and make a debian-installer of jessie using sysvinit and pinning our repo