Wednesday, 24 December 2014

Is Devuan really a good idea?

Posted by Gary Newell  |  at  10:08 20 comments

I was idly looking through the press releases and news stories when I came across this article which talks about the new Debian fork called Devuan.

Devuan is a complete fork of the Debian system minus systemd.

I know that there are lots of people who aren't happy with the inclusion of systemd as part of the next release of Debian but to make such a radical decision to clone the entire thing and start your own project could be deemed overkill.

I can understand a single developer or a handful of developers taking a Debian or Ubuntu base and then creating a new distribution with a specific purpose in mind. I actually think smaller distributions are a good thing because they come up with and implement ideas that might not reach the light of day in one of the base distributions.

Many people are of the opinion however that it is better to pool resources and have just a few distributions where everybody works together to make those distributions as good as they can possibly be.

I am not sure that the Devuan fork is a good thing. In a democracy, we vote for people to make decisions on our behalf and if we don't like the decisions they make we either vote them out again or we move to another country that is more aligned to our way of thinking. In Linux terms this would mean if we don't like Debian then we would switch to another distribution.

In the UK at the moment there is a rail project being implemented which aims to reduce journey times between the north and the south, called HS2. It has greatly divided opinions as to whether it is or isn't a good idea.

People who live in the areas where the new infrastructure is being implemented aren't particularly happy. Where they used to overlook rolling green fields as far as the eye could see they now have heavy machinery and rail tracks. They are obviously against the idea.

People who live in the north and who regularly visit London will see the rail link as a good idea and will be happy that it is being implemented.

There are of course another group of people. The unaffected. These people don't care if HS2 is or is not implemented. They can see the argument from both sides but are largely uninterested.

The one thing that nobody is thinking of doing is creating an exact clone of the United Kingdom and moving it slightly to the left.

If a few developers working on their own Linux distribution is seen as counterproductive then what effect does it have when a whole group of developers take a mammoth distribution such as Debian and decide to fork it?

The people who are against systemd may well go over to Devuan but at what cost? Is productivity going to be as good, is it going to move forward as a distribution at the same pace as Debian would have had they stuck with it?

The people who are for systemd now have less people working on the distribution but are partly responsible for the split.

The people that lose out the most are those who don't care either way about systemd. There are now two projects instead of one with less people working on each.

What do you think about the new Devuan project? Is it a good idea or should the people involved look for a comprimise?








About the Author

Gary Newell started the Everyday Linux User blog in 2010 and has written reviews on dozens of different Linux based operating systems. He has also written a number of tutorials.

Get Updates

Subscribe to our e-mail newsletter to receive updates.

Share This Post

Related posts

20 comments:

  1. I'm going to remain neutral on the subject of init systems, but I would suggest that a project is better for losing contributors whose involvement is conditional. Debian will always attract developers, not least because so many distributions are based on debian.

    ReplyDelete
  2. My understanding was that Ubuntu was started with Mark Shuttleworth approaching a number of Debian developers and getting them onboard his own project - effectively taking an mount of talent away from Debian. Even today, large swathes of people develop for Ubuntu instead of Debian, with Ubuntu mechanics in mind - Debian is an afterthought.

    My end point being - is this really a loss? The persons who were unhappy with Debian's direction were probably looking at Slackware or *BSD as alternatives to jump to - so they were in a sense lost from the outset - but if the compatibility in other areas remain, their contributions to Devuan could still be meaningful to Debian, right?

    Alas it is all highly speculative now. The Devuan group's size is probably not stable, and their ability to pull this off is yet to be proven. With Debian the size it is, I doubt this will be in any way cataclysmic to their ability to progress, and old hands leaving on such large operations generally means fresh blood has a chance to take the lead, and the mothership will plough onwards...

    ReplyDelete
  3. To be fair when the peasants in England in the 17th century did not like the decisions being made by the government of the time they did pick up sticks move to other areas of the world such as America and Australia and rebuild a country the way they want. As a result we now have a more diverse world with strong and in many respects differing cultures throughout. The only reason nobody does this now is because there is nowhere left to build new country's.

    I don't think there is anything inherently wrong with new ideas and differing directions. What you lose in the short term you gain in the long term. Whatever your views on systemmd I think the creation of Devuan has every chance of one day being a stable much loved part of the linux community. Diversity is one of the strongest parts of the linux world in my opinion as it means that what doesnt work can die off while what does can live on and split again itself to start the whole process again. This is one of Linux greatest strengths over proprietary organisations such as Microsoft. After all if the Windows 8 debarcle had happened in the Linux world everyone would have had other distros or even a new fork to move off with.

    I say long live the diversity of the Linux community.

    ReplyDelete
  4. To be fair when the peasants in England in the 17th century did not like the decisions being made by the government of the time they did pick up sticks move to other areas of the world such as America and Australia and rebuild a country the way they want. As a result we now have a more diverse world with strong and in many respects differing cultures throughout. The only reason nobody does this now is because there is nowhere left to build new country's.

    I don't think there is anything inherently wrong with new ideas and differing directions. What you lose in the short term you gain in the long term. Whatever your views on systemmd I think the creation of Devuan has every chance of one day being a stable much loved part of the linux community. Diversity is one of the strongest parts of the linux world in my opinion as it means that what doesnt work can die off while what does can live on and split again itself to start the whole process again. This is one of Linux greatest strengths over proprietary organisations such as Microsoft. After all if the Windows 8 debarcle had happened in the Linux world everyone would have had other distros or even a new fork to move off with.

    I say long live the diversity of the Linux community.

    ReplyDelete
  5. Please pardon me if I'm wrong, but it looks like your premise is based from a users's perspective. Have you tried looking at the situation from a developer's perspective? When I was first introduced to linux, there seemed to be a common vision of making linux so great, that it would displace Windows to a large degree. That vibe seemed to be everywhere. Perhaps I'm wrong, but currently, I don't see any common linux vision.

    As I understand from a several linux articles, most developers are unpaid volunteers. If a developer had joined a project because they were in agreement with the projects goals and vision, and over time the vision or the goals change so that they no longer were in agreement with it, or at least, didn't agree with it as much, and if it seemed that things were never going to revert to the previous vision..what motivation would they as an unpaid volunteer have to stay with that project?

    While I currently do not contribute to any projects open source or otherwise, I'm sure I would find it very difficult to continue volunteering time toward something I didn't agree with. Although it quite possible my viewpoint of this specific situation is overly simplistic and underinformed.

    Maybe recreating the entire distribution is overly ambitious, and maybe one or both projects will now fail. Or...maybe just maybe there's enough vision on both sides to cause them to flourish in their own way. Only time will tell.

    ReplyDelete
  6. The compromise is the best choice, of course. And we in devuan were asking for it before to fork, thanks to the Ian GR in debian. But sadly, there was no compromise acceptable so we are forced to fork.

    ReplyDelete
  7. Sure, we shouldn't have forked LibreOffice or MariaDB for those simple issues. And while OOXML was chosen in a democratic process, we shouldn't have insisted on ODF. It's so silly.. I think we should change all FOSS licenses, so forks will become illegal and all development effort will be concentrated on the real thing. That will stop stupid things like Android right in it's tracks.

    ReplyDelete
  8. Every time a major distro switches to systemd there has been outrage with loss of users/devs. Gentoo and Slackware don't use systemd by default but neither is as accessible as something like Debian. The main thing here is that some people feel very particular about their init, and you don't want devs working on a distro they don't like using anymore.

    If these devs feel this strongly about their init then their options were not:
    "Fork Debian or stay with Debian."
    Their options were:
    "Fork Debian, work on a non-systemd distro, or start one."

    If that is the case, I'm glad they forked Debian.

    ReplyDelete
  9. (This comment was emailed to me by somebody who would like to remain anonymous)

    The option to switch to another distro is less and less of a possibility as
    systemd creeps everywhere and try very hard to gobble other functionality to
    force itself upon distros and users.

    As systemd gets into every major linux distros, I was counting on debian not
    to fall for this suspicious trend.The only other viable alternative would be
    to abandon GNU/Linux for BSD. I expected debian to offer the option of choosing
    to use systemd for those who chooses so, not to make it the default init. Kind
    of like gentoo did.

    Then devuan came to be and now I have hope again, for this is not about
    systemd, it is about providing freedom of choice and empowering the user.
    Something that debian is expected to take a strong stance for and I feel
    debian has failed to uphold its engagement of the debian social contract.

    Hopefully debian will get the message[1] and there will be a merge back of
    devuan into debian to restore freedom of choice. But until them, after 14
    years, I'm leaving debian for the closest thing without the systemd nnonsense
    forced upon by people who don't listen to the users they're supposed to cater
    to.

    ReplyDelete
  10. As far as systemd goes. Even though it's an "empirical" bit of kit, I think it can be dismantled and controlled by those who are willing to do the job. The whole premise of Linux is "freedom" and to have systemd come in and take over' a slew of processes, is a bit disconcerting. But, I also don't mind nor am i offended by the forking of Debian, if there were no forking...there wouldn’t be Linux Mint....or a host of other distros that are "based upon" Ubuntu.....Fedora.....Red Hat......Debian.....Gentoo....etc. I myself will still use Debian....CEntOS and Fedora along with Ubuntu...Linux Mint...PC-BSD, and openSuSE. If there are people who want to step outside of the box and create something new from a base of a "standard" distro...then by all means they should. There have been some stellar distros that started in just this fashion. Some are still around and others have been retired....such as Fuduntu (a brilliant distro!!) PearOS (Apple's nemesis!) and Snow Linux (a great OS for tutoring others about Linux!). So let the Devuan crowd continue its endeavours....and those who wish to support them can continue on, and those who want to continue using and supporting their tried & true distros, can also go on doing this. That's the real "strength" of Linux and the Open Source community, the freedom to do as you wish, and the freedom to support and use what you like.

    ReplyDelete
  11. I notice that, before Devuan, many people's reaction to complaints about systemd was more or less "Show us the code or shut up". So that's what they're doing. Result? Reaction has shifted to "Stop showing us the code and shut up".
    The non-contradictory thing here is the "shut up" part. If I were one of those bothered by systemd, I would by now be paying little attention to critics who seem less interested in the merits of the case than in just finding some way, any way, to get me to give in and quit rocking the boat and disturbing their sleep.

    ReplyDelete
  12. Is Devuan really a good idea? is the GPL good idea?

    ReplyDelete
  13. Is Devuan really a good idea? yes is the GPL good idea ....

    ReplyDelete
  14. Devuan is a great idea for those doing the fork. They may have needs that SystemD can't provide them. Also if it separates the political bull that's been flying about then so be it. Debian can move on and those who don't want SystemD but a more simpler init system for the machines can move forward as well.

    One size does not fit all as Microsoft found out with it's Metro interface. Lots of people jumped ship when Gnome found this out with the 3.x front end stuff and XFCE and such got more popular. Also Ubuntu found this out to some extent when they went full on Unity and Mint took off. Debian is finding this out now with SystemD.

    ReplyDelete
  15. Getting back the old SYSV init is enough for me to strongly consider devuan. I, for one, hate having systemd shoved down my throat, without having an alternative. Systemd is much less documented, much more complex than SYSV init, and it caused me to reboot much more often.

    ReplyDelete
  16. So much inaccurate, misinformed... where to start.

    The fork isn't about the default init system. It is about the attitude problems that were exposed during the debates. Systemd isn't just the new default for Debian, it has been made clear that the systemd developers have no intention of coexisting with other systems, that systemd is THE future and no other systems will be supported. The non-Linux (Debian/kFreeBSD and Debian/HURD) versions of Debian are thus on death watch.

    More importantly it is not (only) about the init, systemd is already replacing most the userspace of an 'OS' that isn't the kernel below it or glibc and the higher libraries above it. They also are showing every sign of merging it all with GNOME to leverage each other into total domination of desktop environments. It was GNOME adding a hard requirement for systemd that forced it onto every distro, it would be coming full circle for systemd to require GNOME and thus seize the entire OS.

    Systemd and it's cluster of tech is totally alien to a UNIX person and it is set to get worse. Read their writings, their vision is more like Android with 'rom images' instead of operating systems like we know them, with programs mutated into 'apps' vended from appstores and hosted on read only OS images where the only use of yum/apt is at the distro vendor who puts together the image for distribution.

    ReplyDelete
  17. Distros aren't democracies from an end-user perspective. The people who maintain the code and distribution have the final say. Users may be solicited for comments. But at the end of the day, those who will do the actual work decide what that work is. And in Debian's case, the initial decision seemed to be for systemd. Not so much for it's utterly compelling and superior technical merits, assuming (depending on who you listen to) that there actually are any. But more because it's something totally new and different. Which is to say less boring for the maintainers.

    As far as a compromise is concerned, that's not an option. There are two very different philosophies and attitudes, combined with some very real technical considerations, that make compromise impossible here. It has to be one or the other. To do both simultaneously would require more effort and coordination than would be practical for Debian. It would amount to an internal fork of their own distribution.

    If there's any truly democratic thing to be found in FOSS/Linux, it's the nuclear option to fork something. Forking isn't a bad thing. It's a prime example of democracy at work.

    ReplyDelete
  18. I'm a long time user of Debian - ever since Sarge. I don't think that as a user I've ever been aware of issues with the init system. As an end user what differences would I notice if either was used ?

    David

    ReplyDelete
  19. First off, i'm not against systemd. I think that is technically good (I use it everyday), but maybe no in Lenard way. That's a large debate.

    You cannot compare "democracy" with open-source. You are wrong about the community, as it gets more devs on every fork, not the opposite. When a project gets a fork gets more kudos from the devs that forked the project. It's not negative to fork but positive: it's open, it's public and that will always be good to people.
    Cannot compare a train that is done by government with a distro that is done by popular devs. Some can do a demo against on it, but last one decides if train will be implemented. On open source, you, the user, decide on your software. Do you prefer to go with train or bus?

    You choose your everyday software and its consequences.

    Regards.

    ReplyDelete
  20. Having been a person who has looked at the systemd code base and also tried to work with it to get some real work done. I have to side with the Devuan people. It shouldn't be included in the Debian distribution. First of all its a bad design. And when we are talking about a core service and consider what its replacing I'd have to say that this was the wrong choice. SystemD over complicates things. Its designed mostly for desktop machines. Meaning the 90% of the servers out there who deploy linux are SOL. Thats is called not taking your user base into consideration. As such perhaps Devuan should replace Debian. And Debian and Ubuntu who have made the bad decision of going this route deserve to suffer that consequence. That is MYHO.

    ReplyDelete

Feel free to comment on any of the blog posts. Please try to be constructive.

Offensive messages will be removed as will blatant adverts for misleading products and sites.

Thanks for visiting my blog

    Popular Posts

    Total Pageviews

    Subscribe

    Enter your email address:

    Delivered by FeedBurner

    Popular This Month

    What are other people buying?

    Ubuntu Buy
    openSUSE Buy
    Manjaro Buy
    Zorin Buy
    PCLinuxOS Buy
    gNewSense Buy
    Edubuntu Buy
    Fedora Buy
    Uberstudent Buy
    Linux Lite Buy
    Lubuntu Buy
    Xubuntu Buy

    Followers

    Feedburner Followers

Blogger templates. Proudly Powered by Blogger.


back to top Google