Home > Mandriva, Mandrake, Lycoris > Mandriva Fork Announced by Former Employees

Mandriva Fork Announced by Former Employees

  Submitted by gireesh Mandriva, Mandrake, Lycoris 61 Comments

It’s been a troubling couple of weeks for Mandriva, but I decided not to report on it since I found it hard to unravel the events leading up to all this. Now, though, the story has come to its (logical) conclusion: now that most of Mandriva’s employees have been laid off, they came together and forked Mandriva. Enter Mageia.

At the danger of getting things wrong, here’s a short recap of what’s been going on. Edge-IT, a company acquired by Mandriva in 2004, liquidated not too long ago, and most of its employees have either been laid off or left on their own accord, meaning most people working on the distribution are now gone.

Mandriva’s plan is to move development on the desktop side of things to emerging countries, while Europe would focus on server development, since Linux is strong on servers over here. This idea didn’t set well with the people working on Mandriva’s desktop distribution, nor does it sit well with the community.

As a result, the only possible solution they saw was to fork Mandriva Linux, and create a new distribution. “Forking an existing open source project is never an easy decision to make, and forking Mandriva Linux is a huge task,” they state in the announcement, “It was not an impulsive decision. We all spoke a lot before: former employees, Cooker contributors and users’ communities. We collected opinions and reactions in the past weeks as we needed to get some kind of global agreement and to gather, before going ahead. We believe a fork is the best solution and we have decided to create a new distribution: Mageia.”

Mageia will be a community-developed and backed project, independent from any company’s whims. For now, a temporary board of community members is laying the ground work, setting up a not-for-profit organisation, and so on. After their first year, the board will be elected regularly by the community.

An interesting development, surely, and taking Mandriva’s utter incompetence over the past decade into account, it’s surprising it even took them this long to see the writing on the wall.

About The Author

Thom Holwerda

Follow me on Mastodon @[email protected]

61 Comments

  1. 2010-09-19 9:46 pm
    WereCatf

    This was already on Slashdot a few days ago and I commented there too: I am saddened by these news. I’ve used Mandriva for years now and while I’ve tried several other distros I’ve always come back. Why? Because while Mandriva has its own share of quirks and rough corners the corners just don’t seem quite as sharp and lethal as in other distros, ie. it just works generally more smoothly.

    But now with the core devs gone Mandriva is more-or-less dead on its feet and I have to find another distro for me. And I have no idea what to migrate to :/ Mageia won’t be useable for a good while still and who’s to say that it’ll even be as good as Mandriva was in the first place?

    • 2010-09-19 9:56 pm
      lordepox

      Mandrake -> Mandriva -> Mageia. What’s next? This distro has brought a lot to the Linux community over the years, including a good portion of proper package management as we know it today. I used it back in the Mandrake days, and it was way ahead of it’s time then, and stable as a rock. Visual package management was unheard of, but Mandrake had it. I think later distributions have a lot of homage to pay to this distro. Good luck.

      Edited 2010-09-19 21:57 UTC

      • 2010-09-19 10:36 pm
        JrezIN

        don’t forget Conectiva too.

      • 2010-09-19 11:34 pm
        gnemmi

        … proper package management …

        That was something that Mandrake inherited from Conectiva, among many other things like a real, working, svn backboned distro development set of tools that made a sensible distribution development enviroment.

        • 2010-09-20 1:29 am
          djame

          … proper package management …

          That was something that Mandrake inherited from Conectiva, among many other things like a real, working, svn backboned distro development set of tools that made a sensible distribution development enviroment.

          At first they used cvs as everyone else, didn’t they ? Maybe svn was used by connectiva, but one of the first things that comes to mind is the metapackaging system (task-* ), as long as urpmi is concerned, it existed long before the connectiva buy out. And talking about urpmi, people are always praising apt-get but urpmi is seriously cooler, you can even install package in // on mutliple machines.

          sigh, I’m so going to miss Mandriva..

          • 2010-09-20 3:05 am
            gnemmi

            Maybe svn was used by connectiva,

            No “maye” in there .. Conectiva was using svn long

            before Mandriva … and even before svn reached 1.0.

            but one of the first things that comes to mind is the metapackaging system (task-* )

            task-* metapackages were a Conectiva reality long before they made it into Mandriva .. actually metapackages were a part of a whole packaging scheme created around the idea of package “granularity” .. which was a Conectiva feature that slowly but steadily started to grow into Mandriva.

            I’m so going to miss Mandriva..

            And so will I.

  2. 2010-09-19 10:40 pm
    djame

    Mandriva is just the best distro out there… Just try to set up an internet sharing connection from ethernet to firewire (and I use that to connect my mac on this server to backup, firewire network are just way faster, anyway..) on ubuntu and you’ll see what I mean….

    on another notice, the original poster mentionned the incompetence of mandriva heads but what incompetence ? on generating less reviews, yes certainly.

    Just compare the number of reviews any bsd distro or ubuntu get on osnews and just realize the number of reviews mandriva 2010 got, it’s almost ridiculous. As long as Mandriva closed its US office a couple of years ago, its memento was gone for some reason. And it’s the same for Opensuse anyway, they got far way less reviews, notes and buzz than Ubuntu.

    So if Gael Duval ever reads that : go, get your company back and make it works even if that means unfair batlle with giant pocket billionairs.

    Djamé

    • 2010-09-19 10:53 pm
      Thom Holwerda

      but what incompetence ? on generating less reviews, yes certainly.

      How about wholly incompetent at running a business?

      • 2010-09-19 11:04 pm
        djame

        but what incompetence ? on generating less reviews, yes certainly.

        How about wholly incompetent at running a business?

        You mean how about bashing them for failing where redhat, suse, xendros, linwin, gentoo and I can’t remember who else failed as well ?

        Come on, you know well that without the unlimited funding brought to ubuntu, there will be on the list too.

        The problem is not on on how to build the best distro out there but how to build a successfull business model where “customer”‘s alpha and omega seems to be “I want the best, I want it now and for free as in speech and as in beer. Please, I’m hurry otherwise I’ll get your distro on bittorent and that’s it”

        So speaking of the incompetence of people who just tried everything as everyone else did is a bit unfair to me. it’s like telling to people who can’t fly that they suck big time because they can’t do as well as superman or as the next billionaire who got some wings greffed on his back.

        Edited 2010-09-19 23:06 UTC

        • 2010-09-20 1:19 am
          sultanqasim

          Overall I agree with you comment, but I won’t call RedHat a failure. They are a profitable company who build a solid product and earn hundreds of millions every year.

          • 2010-09-20 1:22 am
            djame

            Overall I agree with you comment, but I won’t call RedHat a failure. They are a profitable company who build a solid product and earn hundreds of millions every year.

            Sorry, I was talking about the desktop part of their activity. Now it’s almost completely community driven in the Fedora project.

        • 2010-09-20 2:11 pm
          google_ninja

          so your argument is that since the business plan was impossible, the pursuit of the impossible shows competence in the management?

          Part of being a good entrepreneur is knowing when to stop wasting your (and your investors) time and money, and move on to something that makes more sense.

          • 2010-09-20 2:24 pm
            djame

            [q]so your argument is that since the business plan was impossible, the pursuit of the impossible shows competence in the management? /q]

            no, I’ve never said that. I said that failing was not a sign of incompetence, therefore bashing them for that is just pointless.

          • 2010-09-20 6:22 pm
            spiderman

            so your argument is that since the business plan was impossible, the pursuit of the impossible shows competence in the management?

            Part of being a good entrepreneur is knowing when to stop wasting your (and your investors) time and money, and move on to something that makes more sense.

            Except Mandriva is not a waste of money.

            Maybe they didn’t find a good business model, maybe it does not exist but Mandriva is definitely not a waste of money. They’ve moved the desktop forward and that is a great achievement. Making money is a necessary step to achieve their goals but their goal is not to make money. Maybe they didn’t make enough money but the money they made has moved the dektop forward. Even if they can’t make it any further they still had success.

            “sense” != “money”

        • 2010-09-20 3:24 pm
          Soulbender

          for failing where redhat, suse, xendros, linwin, gentoo and I can’t remember who else failed as well ?

          Redhat and SUSE is still around and gentoo isn’t a for-profit operation. As for Xandros and Linwin, well yeah,I guess they weren’t good enough businessmen either.

          The problem is not on on how to build the best distro out there but how to build a successfull business model

          So you’re saying that Mandriva’s management is competent even though it didn’t manage to create a sustainable business model? Hmmm….

          So speaking of the incompetence of people who just tried everything as everyone else did is a bit unfair to me.

          If everyone else has already failed maybe it’s not the right thing to try. Again, that’s what competent management is supposed to be able to figure out.

          Edit: Ok, you’re trying to say that failure != incompetence. Sure, but if you fail for an extended period of time with everything you try incompetence might be part of the problem.

          Edited 2010-09-20 15:28 UTC

          • 2010-09-20 4:42 pm
            djame

            please if you’re quoting me use full sentences otherwise people will think you’re trying to have a point by extracting things out of their context.

      • 2010-09-20 9:22 am
        danieldk

        Yes they were. I remember being at a trade show (HCC Dagen for Dutch people) ~2000. I was there working at a stand of a Dutch software vendor/distributor. Mandrake flew in some person to sell their product at the same stand. The person was obviously overpaid, knew nothing about their product, and was more concerned about making foreign phone calls (which were expensive in the days).

        The suits were brought into Mandrake to early, and they seemed to be mostly concerned with burning cash and doing suit-stuff. They didn’t care about the Mandrake community, or too much about the technical side of the distribution.

        Red Hat on the other hand, has been governed by enthusiasts for a long time. Sure, they were business people, but they loved the company and the product. And it shows. E.g., when they buy out a company, they are usually quick to opensource the ‘intellectual property’ and build a community around it. They know that open source software forms the genes of the company, and act according to that.

        • 2010-09-20 9:46 am
          jbauer

          Red Hat on the other hand, has been governed by enthusiasts for a long time. Sure, they were business people, but they loved the company and the product. And it shows. E.g., when they buy out a company, they are usually quick to opensource the ‘intellectual property’ and build a community around it. They know that open source software forms the genes of the company, and act according to that.

          Yeah, and Red Hat also figured out a long time ago that you can’t sell Linux to Average Joe. One would think Mandrake and others should have taken the hint by now.

          Edited 2010-09-20 09:47 UTC

  3. 2010-09-19 11:04 pm
    nt_jerkface

    At some point there will be more Linux distros than users.

    • 2010-09-20 12:39 am
      lemur2

      At some point there will be more Linux distros than users.

      Oh dear, you aren’t on about that again are you?

      Here, do yourself a favour and do some reading:

      http://broadcast.oreilly.com/2010/09/debunking-the-1-myth.html

      If we do the math we find that due to netbooks alone Linux captured nearly 6% of the desktop market in 2009.

      Microsoft CEO Steve Ballmer. Using a slide to visualize OS market share Ballmer had Linux desktop market share as a slightly larger slice of the pie than MacOS.

      Rather than simply believe whatever rubbish you are spoon-fed, educate yourself a little beyond what the media-marketing-powers-that-be want you to believe.

      BTW: If you are having trouble deciding on a Linux distribution to use, just choose one upon which others base themselves. There are really only four choices: Debian, RedHat, OpenSuSe and Slackware.

      ReHat is primarily for servers (use CentOS if you don’t need paid support). SuSe is being sold. Slackware isn’t all that user-friendly, and it is getting a bit long in the tooth now.

      That leaves debian/Ubuntu. Debian is more stable in its “stable” long-term-support version (currently, this is Lenny), but less user-friendly and has less of a following, so online help isn’t as prevalent. Use the MEPIS distribution if you opt for Debian, to help a bit in the user-friendliness department.

      http://distrowatch.com/table.php?distribution=mepis

      The best choice for a Linux desktop at this time is thus, probably, Ubuntu LTS (aka Ubuntu 10.04, or Ubuntu Lucid Lynx). This distribution comes in KDE, GNOME, XFCE and LXDE flavours, going from the most powerful desktop down to the most lightweight. Choose one depending on the relative capability (or lack of age, if you will) of your hardware. If you have high-end or relatively recent hardware, and/or you are very familiar with Windows, then KDE will probably represent the best choice of desktop, otherwise choose GNOME unless your hardware is relatively old or under-powered, in which case the LXDE variant is probably your best bet (“Lubuntu” is the name of the Ubuntu variant you want for LXDE).

      http://distrowatch.com/table.php?distribution=lubuntu

      Edited 2010-09-20 00:46 UTC

      • 2010-09-20 12:49 am
        WereCatf

        Oh dear, you aren’t on about that again are you?

        Here, do yourself a favour and do some reading:

  • 2010-09-20 2:26 am
    lemur2

    I don’t think he was referring to the amount of people using Linux but instead to the huge amount of distros there are. Atleast it sounds like that to my ear.

    nt_jerkface:

    At some point there will be more Linux distros than users.

    Linux distros ~= a few hundred

    Linux (desktop) users (according to Ballmer): slightly > Mac Osx users ~= 7% of desktop users ~= (rough estimate) 35 million users.

    35 million >>> a few hundred.

    QED.

    Edited 2010-09-20 02:33 UTC

    • 2010-09-20 4:04 am
      flynn

      Linux distros ~= a few hundred

      Linux (desktop) users (according to Ballmer): slightly > Mac Osx users ~= 7% of desktop users ~= (rough estimate) 35 million users.

      35 million >>> a few hundred.

      QED.

      Stop taking everything so literally. His comment quite clearly meant that the last thing we need is an yet another linux distro. The numbers themselves are irrelevant, learn to hyperbole.

    • 2010-09-20 4:35 am
      lemur2

      Linux distros ~= a few hundred Linux (desktop) users (according to Ballmer): slightly > Mac Osx users ~= 7% of desktop users ~= (rough estimate) 35 million users. 35 million >>> a few hundred. QED.

      Stop taking everything so literally. His comment quite clearly meant that the last thing we need is an yet another linux distro. The numbers themselves are irrelevant, learn to hyperbole.

      Who are you to tell me what to do? I know what his comment was meant to imply, and I refute it utterly.

      Linux distro’s are intended to cover different use cases, in order to make it un-necessary for users to have to identify their own set of packages with perhaps unfamiliar names.

      Some examples:

      For security (digital forensics, if you will):

      http://distrowatch.com/?newsid=06276

      Specialist distribution for use as a firewall:

      http://distrowatch.com/?newsid=06280

      Japanese language:

      http://distrowatch.com/?newsid=06273

      Boot from portable media:

      http://distrowatch.com/table.php?distribution=puppy

      Home entertainment/TV:

      http://distrowatch.com/table.php?distribution=mythbuntu

      Mathematics and statistics:

      http://distrowatch.com/table.php?distribution=scientific

      System rescue / data recovery:

      http://distrowatch.com/table.php?distribution=systemrescue

      Education:

      http://distrowatch.com/table.php?distribution=edubuntu

      Audio, video & graphics:

      http://distrowatch.com/table.php?distribution=ubuntustudio

      These are all based on the same few base distributions, and they are simply easy-to-install variations meant for different specific purposes.

      Why shouldn’t people have the choice of these distributions, and more, if they are found to be useful?

      Edited 2010-09-20 04:37 UTC

    • 2010-09-20 8:53 am
      WereCatf

      — buttload of links omitted —

      These are all based on the same few base distributions, and they are simply easy-to-install variations meant for different specific purposes.

      Why shouldn’t people have the choice of these distributions, and more, if they are found to be useful?

      Well, there’s for example a dozen distros all aimed for desktop use. Ie. they all cater to the same thing, yet there still exists so many of them. And worse, a billion *buntu versions.

      I personally think it’d be better to have the base distro only, ie. Ubuntu, and copy the meta-package system from Mandriva; you’d just install Ubuntu, then install task-education and POOF! You’d have all the needed education apps installed for example. Then to polish the thing you could just include a menu-item which then presents the available metapackages, explains them and their meaning, and allows you to easily install them with one click. This’d allow for more unity as there’d only be one Ubuntu, there wouldn’t be a need to maintain so many different spin-offs and the devs could instead focus on their respective meta-package, and it’d be simpler for end-users. Well, atleast I think that’d be a good route to take.

    • 2010-09-20 1:26 pm
      devnet

      So what you’re saying is…have a small and stable base distribution. Then have task-whatever to install KDE or Gnome or e17 or LXDE, etc.?

      Boy, wouldn’t it be great if this existed and had the Mandriva Control Center?

      Wait a second…it does exist ;) Unity Linux is Mandriva based and a core distribution aiming to be with ‘just enough graphics’ on first boot. It uses Smart for package management and then you just install task-whateveryouwant to get where you need to go. The devs there are ready for the fork as well and won’t skip a beat. http://unity-linux.org

    • 2010-09-20 1:12 pm
      flynn

      Who are you to tell me what to do? I know what his comment was meant to imply, and I refute it utterly.

      You did not refute anything, you posted a bunch of numbers in response to a post that was never about numbers to begin with. Again, learn to hyperbole.

    • 2010-09-20 1:53 pm
      lemur2

      Who are you to tell me what to do? I know what his comment was meant to imply, and I refute it utterly.

      You did not refute anything, you posted a bunch of numbers in response to a post that was never about numbers to begin with. Again, learn to hyperbole.

      You need to get off your high horse. The OP was indeed using hyperbole to try to make a point, but nevertheless the underlying point was wrong. Trying to be clever with wild exaggeration (hyperbole) shouldn’t be allowed to become a vehicle via which it becomes possible to create an invalid impression.

      Learn about the many techniques used to present false argumentation, and associated logical fallacies such as these (elements of which are present in the original post):

      Style over substance fallacy

      Bare assertion fallacy

      Ad nauseam

      Judgmental language

      Appeal to tradition

      and finally,

      Proof by intimidation (which, BTW, is also what you are trying to do).

      PS: I looked up the meaning of “refute” in the phrase “I refute it utterly”. It transpires that the phrase does not mean what I had intended. I should have said “I dispute it utterly”, so my apologies there. Mea culpa.

      Edited 2010-09-20 13:59 UTC

    • 2010-09-20 4:04 pm
      BallmerKnowsBest

      The OP was indeed using hyperbole to try to make a point, but nevertheless the underlying point was wrong.

      Are you suggesting that nt_jerkface didn’t actually perform an exhaustive series of tests and statistical analyses before arriving at his conclusion?!?!? Gasp, say it ain’t so!

      Learn about the many techniques used to present false argumentation, and associated logical fallacies such as these (elements of which are present in the original post):

      Style over substance fallacy

      Bare assertion fallacy

      Ad nauseam

      Judgmental language

      Appeal to tradition

      and finally,

      Proof by intimidation

      You missed one:

      Use the words “shill” or “astroturfer” as if they were synonyms for “someone I don’t agree with”. Or there’s the “lite” version of the same technique, claiming that anyone who disagrees with you must be “brainwashed by the powers-that-be/propaganda” or “a slave to ‘the man'” or whatever other small-change cliche you prefer.

    • 2010-09-20 3:42 pm
      Bounty

      Who are you to tell me what to do? I know what his comment was meant to imply, and I refute it utterly.

      Linux distro’s are intended to cover different use cases, in order to make it un-necessary for users to have to identify their own set of packages with perhaps unfamiliar names.

      Ahhh, the old “I reject your reality and substitute my own” argument. Is it really better if you have to learn the name of the distro with your app, rather than the name of the app? What if you need to do math and multimedia? Do you dual boot?

      What are the different use cases this fork came about for? Those who need an established floundering general purpose distro stay with Mandriva, and those who want a floundering general purpose startup go with Megia? Nice.

      I think you would have been better of with a simple “I disagree.”

    • 2010-09-20 3:51 pm
      jbauer

      Ahhh, the old “I reject your reality and substitute my own” argument. Is it really better if you have to learn the name of the distro with your app, rather than the name of the app? What if you need to do math and multimedia? Do you dual boot?

      What are the different use cases this fork came about for? Those who need an established floundering general purpose distro stay with Mandriva, and those who want a floundering general purpose startup go with Megia? Nice.

      I think you would have been better of with a simple “I disagree.”

      Yep. Distros and forks are made just because no one can prevent them, it’s the nature of the open source beast. That doesn’t sound nearly as planned and intellectual as “made to cover every possible use case”, though ;)

    • 2010-09-20 3:49 pm
      jabbotts

      What’s this Ubuntu fork of Debian? The last thing we need is another distro.. oh.. wait.. turns out there was value in yet another distro.

      Mandriva is floundering, what’s wrong with the technology people who produced Mandriva producing a replacement for it?

      Complaining about too many distros is like complaining about to many toothburshes to choose from at the store. A healthy market has choice and in a healthy market products fail and are replaced by competing products.

    • 2010-09-20 4:04 pm
      Bounty

      What’s this Ubuntu fork of Debian? The last thing we need is another distro.. oh.. wait.. turns out there was value in yet another distro.

      Mandriva is floundering, what’s wrong with the technology people who produced Mandriva producing a replacement for it?

      Complaining about too many distros is like complaining about to many toothburshes to choose from at the store. A healthy market has choice and in a healthy market products fail and are replaced by competing products.

      Now this is a response on it that I like. +1

    • 2010-09-20 5:46 pm
      nt_jerkface

      Complaining about too many distros is like complaining about to many toothburshes to choose from at the store. A healthy market has choice and in a healthy market products fail and are replaced by competing products.

      Distros are not toothbrushes, they are software projects that require skilled contributers for which there is a limited supply.

      It doesn’t make sense to have so many of them working on distros whose differences could be encapsulated within themes and custom packages. I realize this is the nature of open source but that doesn’t mean that Linux users need to be supportive of all these general purpose distros.

    • 2010-09-20 9:24 pm
      jabbotts

      So it’s about developer resources. Why allow developers to choose where they put there time and effort when you can just dictate that they all work on your pet project then? We’ll just decree that Red Hat and any other distribution maintainer merge into one megacorp too then? We better tell the people behind Backtrack that they can’t produce a superbe security distribution anymore because it’s too specialized; it’s not an assimilated part of “MegaLinux; the one true path”.

      Do we also decide that all those engineers wasting effort for different car companies need to consolidate and give us one standard car assembly to not quite fit everyone’s different needs? How do we decide which product categories are allowed to have multiple models from multiple competitors and which are not?

      See Linux Distributions are all different products from different companies. Red Hat and Ubuntu have very different target purposes though they happen to be inter-operable and assembled from similar commodity parts. “Linux” is not all one single product unless you really mean just the kernel alone.

      The nice thing in all this is that resources are still shared where it counts. Red Hat and Canonical both contribute code back into the kernel and various other projects. We have a healthy market full of distributions to choose from while having competitors contribute to each other’s base products. In terms of development effort, your concern, the workload is already shared.

      Based on the license, one can’t really remove the freedom to create a new distribution eiehr

    • 2010-09-20 10:18 pm
      WereCatf

      Do we also decide that all those engineers wasting effort for different car companies need to consolidate and give us one standard car assembly to not quite fit everyone’s different needs?

      A car and a software collection are two quite different things: one is an actual physical entity and thus is limited by the physical world, and the other one is a collection of software modifiable whenever and wherever to suit the needs. A car can’t just be made to suit 2 passenger and then magically upsized to a truck as needed. Software can.

      In the case of f.ex. a security-oriented distro aimed to be booted from a USB stick it does indeed make sense to have it as a whole separate distro. But for example something aimed at education needs is almost always a desktop distro with additional software anyway, so why can’t it just be made an installable metapackage instead of a complete distro?

      It would make more sense to be a little bit more selective and instead make it easy to install additional capabilities to an already existing foundation.

      Just think about it.. a user downloads and installs Ubuntu, spends some time learning it, customizing the desktop to his or her tastes and what not, but then realizes the need for educational and audio creation software. What does (s)he do? A) Install task-education and task-audio-studio or B) Google around, browse tediously through all the available software and try to figure which ones do what and if they are worth installing.

    • 2010-09-21 12:13 pm
      jabbotts

      A distribution is a product assembled from many commodity parts; kernel, boot loader, parts of the userland, parts of the GUI, applications. A car is a product assembled from many commodity parts; engine, starter, trany, chassis, interior. Both are complex products which compete against similar products assembled from nearly the same parts.

      Backtrack is well suited to doing security related work. I can get many of the same parts from other distributions but backtrack has been assembled from those commodity parts with specific intentions. I can just boot Backtrack and get to work or I can take a distro with different target purposes and bend it as close as I can to what backtrack does. In reverse, I could take Backtrack and try to bend it into a more general purpose multi-user distribution but again, it’s not easy because it’s not the intended use. I can sqeeze six friends into a two door jetta or we can take one of the other friend’s cars which comfortably seats six people. Just like cars, distributions have different design goals and limitations.

      The fact that it is not a physical good is a benefit. It takes me a good long while to go test drive five prospective car purchases.

      Your last point is not really relevant as it applies to pretty much every OS out there. I install Windows and discover that I also want education and audio creation software. Not finding it in the add/remove software area, I have to go searching. What education software do I go looking for? Do I use soundforge, adobe products, Windows Movie Editor? Even within a single constrictive platform family, I’m having to shop a little to choose my education and audio software along with considering what Windows distribution it will run on (winXP? Vista, Win7?). Really, user needs that suggest a general purpose distribution instead of a specific one is part of the choice and narrows down the candidate list. Heck, maybe you stick with your installed Ubuntu and keep Edubuntu and Audiobuntu liveCD beside the machine. (I actually have a Qimo VM for that “educational” need to be honest.. works great and I don’t have to dirty my personal setup)

      In the end, no single distribution can do it all. For all of Microsoft’s efforts, no distribution of Windows can be everything to everyone. osX is slick for what it does but it doesn’t do everything either. Debian is one of the most general purpose distribution but it won’t include closed binaries like Mandriva does nor will it include the subtle tweaks and rebuilds that Backtrack does. A Buick and a Civic are designed for different purposes just as distributions are.

      For some distributions it is about world domination. For others, it’s about making a distribution bent to the maintainers whims. Who’s design goal is “correct”? Which assembly of commodity parts assimilates into the other since one must abandon it’s design goals to comply? code sharing can already happen where it’s important underneath; developer efforts are already shared even among competing products.

      And, that’s really my point. In every other healthy product category, we have choices to make. Even if the choice is “thinking is hard, I’ll just do what my friends do” and that’s fine, abdicating choice is a choice and one that does not limit those who do want to make it. Every other healthy product category has items which differ in subtle ways due to target customer and uses. Does your toothbrush have the tung scrubber or the reachy-bendy neck on it? Electric or manual? Disposable use or long term appliance? Yet, when it comes to software, everybody looses there minds; “oh no.. there’s choice.. there’s too much choice.. all you software developers stop causing choice and focus your effort on my individual self entitled demands.. just stop.. stop giving me choices!”

      Now, where I agree fully is within the distribution; the completed product layer of the supply chain. Canonical had to make a choice and say “if your not Ubuntu, Kubuntu or Xubuntu, your not our distro. Create whatever remixes you want but you can’t claim they are works of Canonical.” Too many forks claiming to be the same distro. Android is in a similar situation. Google produces it and the hardware vendors fracture the hell out of it with flavor of the week customizations and a lack of source flowing back to the core distro they all claim to be using. Different jailbreaking methods for different vendor/hardware/os versions and combination.. that’s a mess. It’s all “Android” so why don’t I just get the same rootsh package from the repository regardless of hardware I happen to be running Android on? Different apps from the central repository that only install on specific device/software combination? Maybe I’ve been spoiled by Palm and Maemo where everything based on Palm or Maemo runs what claims to be Palm or Maemo third party apps. I get that the hardware folk think closed binary drivers are some kind of benefit and need to provide there own hardware specific parts; stick with a central Google Android distribution and have the vendors provide a device specific driver binary pack. When you flash the current Google distro you also have the option to include the device specific mini-firmware; one consolidated “Android” without loosing the hardware specific kernel modules.

      Within the scope of a distribution, I absolutely agree that one needs to keep the product consolidated. A distro is a product from a single entity. It is the compilation of programs that is the shipping product. It is a single group of package maintainers working under a single set of project goals.

    • 2010-09-21 1:56 pm
      WereCatf

      Your last point is not really relevant as it applies to pretty much every OS out there.

      Of course it is relevant: why not surpass the other OSes instead of just being contempt with being equal? I’m just saying it’d be great to have a single menu entry that by clicking on it you are presented with various metapackages, all with clear, conscise names and description and NO extra-fluff (ie. ability to install non-metapackages or browse repos or such, there’s a separate package manager for that..) Even better if it worked the same on all distros but I suppose that’s a pipe-dream ;)

      In the end, no single distribution can do it all.

      Of course not and I didn’t even claim so. I’m just saying it’d make more sense to think about the requirements and find out whether just installing a metapackage on a basic desktop installation would compromise those requirements or not; if it didn’t then just instead provide a metapackage, and if it did only then roll a separate spin-off. Like f.ex. a lightweight, security management – and repair – oriented system aimed to be booted from removable media just wouldn’t be doable with basic desktop installation. Nor would very very light-weight ones aimed for very old computers, like f.ex. Puppy and Damn Small Linux. However, f.ex. Edubuntu is more-or-less the same as basic Ubuntu + education metapackage. Just.. use some care in deciding whether to roll yet another spin-off or instead provide a meta-package.

      This is all however just my opinion, feel free to disagree, but I’d be interested in hearing specific reasons then ;)

    • 2010-09-21 3:07 pm
      jabbotts

      It sounds more like themeing at a distro level instead of individual apps or DE which I can’t really find fault with. In a sense, that is what closely related child distros do. Actually, Backtrack4 is now an Ubuntu fork if I understand correctly.

      I should also clarify that my grief is with the knee jerk reaction resulting in “psh.. they should all just be one Linux”. I am personally a fork snob that sticks to the most original unless there is specific reason not too. Red Hat gave me reason to use the Mandriva fork, Debian gives me no reason to use Canonical’s fork. I don’t promote blind forking of projects when one can work with it to accomplish goals. At the same time, I also realize there there are valid reasons to fork. There are valid reasons for separate entities producing separate distributions. That variety is an essential reason that he platform has evolved and produced as much as it has so far. Mandriva has a good base distro that the company is running into the ground; fork them.. brilliant. If Oracle kills MySQL, you can bet the MySQL fork will get more attention unless everyone jumps to postgresql.

      I guess the real question is; when does one create a metapackage and when does one create a specialized child distribution. Do the goals of the specialized child distribution conflict with the parent and is there a different parent distribution the child could work within rather than fork from?

    • 2010-09-20 3:06 pm
      Soulbender

      Dude, it was a joke about the high number of Linux distros.

      Jesus, stop being so uptight. What are you, an accountant?

    • 2010-09-21 3:20 pm
      aesiamun

      Nope, just a pendant ;)

  • 2010-09-20 6:35 am
    nt_jerkface

    Why do the most ardent Linux supporters seem to lack a sense of humor?

    Here, do yourself a favour and do some reading:

    http://broadcast.oreilly.com/2010/09/debunking-the-1-myth.html

    Already did:

    http://www.binplay.com/2010/09/yes-linux-really-has-1.html

    BTW: If you are having trouble deciding on a Linux distribution to use, just choose one upon which others base themselves. There are really only four choices: Debian, RedHat, OpenSuSe and Slackware.

    Again with the condescending attitude. I already have a favorite distro, thanks. It’s called Clonezilla.

    http://clonezilla.org/

    For future reference you should probably assume that anyone who posts at a website called OSNEWS has tried a few Linux distros.

    The best choice for a Linux desktop at this time is

    I don’t think Linux is ready for the desktop. I think it needs to bake a few more years before being pushed on the public. It’s fine for servers and cell phones but on the desktop it still has too many rough edges.

    Edited 2010-09-20 06:39 UTC

    • 2010-09-20 7:09 am
      lemur2

      The best choice for a Linux desktop at this time is probably

      I don’t think Linux is ready for the desktop. I think it needs to bake a few more years before being pushed on the public. It’s fine for servers and cell phones but on the desktop it still has too many rough edges.

      Immaterial if you are trying to claim that Linux isn’t ready for the desktop … since about 7% of desktop users disagree with you.

      So, for those who have not fallen for propoganda attempts at disparaging Linux, and who are aware of what contemproray Linux on the desktop can do and its realised quality, functionality, stability, performance, security and tremendous value-for-money, what is the best distribution?

      That very much depends on what role you have in mind. People who are familiar with Linux already, and who have a definitive role in mind … they do not need any such recommendation about which is the best Linux distribution for their purpose. They will already know.

      So the question of “what is the best distribution?” for Linux on the desktop is typically asked by people who are new to Linux.

      The assumption then would be that these people are after an easy-to-install Linux desktop that covers “typical” desktop usage patterns and which would be relatively easy to pick up for a person used to using Windows.

      My suggestion, based on that assumption, would be for such people to try MEPIS or Kubuntu 10.04. If such a user were to install either of those, they would have a great experience, typically with no problems, and enjoy great software that did everything they wanted of it admirably well.

      Mandriva would have been a reasonable suggestion also, but it appears to be running into some difficulty lately, as evidenced by this very thread topic, and so it is probably not such a good suggestion any longer.

      Edited 2010-09-20 07:13 UTC

      • 2010-09-20 7:25 am
        nt_jerkface

        So the question of “what is the best distribution?” for Linux on the desktop is typically asked by people who are new to Linux.

        I have seen someone ask this question in a tech forum only to have a dozen people argue over the answer. That lack of consolidation is not going to inspire confidence.

        Maybe you think having over a hundred general purpose distros is neat and that it doesn’t matter if thousands of people are spending their time performing redundant package maintenance.

        Ballmer probably thinks all that divided effort is pretty neat too.

      • 2010-09-20 4:01 pm
        BallmerKnowsBest

        So the question of “what is the best distribution?” for Linux on the desktop is typically asked by people who are new to Linux.

        I have seen someone ask this question in a tech forum only to have a dozen people argue over the answer. That lack of consolidation is not going to inspire confidence.

        But does provide for an easy cop-out anytime someone criticizes Linux: “you’re using the wrong distro/you should have used $pet_distro instead” (the standard Plan B when “blame the user” fails).

        Maybe you think having over a hundred general purpose distros is neat and that it doesn’t matter if thousands of people are spending their time performing redundant package maintenance.

        Ballmer probably thinks all that divided effort is pretty neat too.

        And I have it on good authority that he knows best.

    • 2010-09-20 9:13 am
      marafaka

      nt_jerkface: “I don’t think Linux is ready for the desktop.”

      You are right, because Linux has no user interface. That’s what distros are there for.

  • 2010-09-20 2:24 pm
    google_ninja

    re the 1% article, it was a great mix of common sense and downright stupidity.

    He is right that the numbers don’t add up. That’s because that hitslink site is about what is actually being used, not what is sold or bought.

    The over counting of windows is a new one to me, I would imagine windows dramatically under counted — most corporate/governament networks are only visible to the outside world as a single IP addresses, and most corporate/governament networks are 100% MS.

    Using sales numbers doesn’t really mean anything. At my first job, I remember needing a few crappy machines for something. An IT guy brought me into a basement room filled with probably about 300 several years old NT4 machines. If you looked at sales stats, those all would have been considered windows machines. In this case, I have many friends who have netbooks, but they all installed either xp, windows 7, or osx on them. Hitslink would catch that, netbook sales numbers wont.

    At the end of the day, the hitslink numbers should not be taken as gospel, but they are by far the most reliable actual usage numbers we can get.

  • 2010-09-20 2:59 pm
    Soulbender

    I guess poking fun at Linux isn’t very popular. I thought it was pretty funny but alas, I am not allowed to mod you up.

  • 2010-09-19 11:29 pm
    cmost

    As usual, a large corporation takes over a Linux distro and then screws it up. While I’m not a Mandriva user, I am proud to say I got my start with Linux using Mandrake Linux. It was, at the time, the easiest Linux distro to setup and use. Nothing about that has changed today and Mandriva is as innovating as ever. I for one am happy to see that it will live on in the form of Mageia Linux. What would really be the icing on this cake is if Gaël Duval joined the developers of Mageia. Mr. Duval was always a force and champion of Mandriva and the distro suffered after he was essentially forced out several years ago.

    • 2010-09-20 6:34 am
      lezard

      From what I read, he’s not really popular within former employees. I guess when you’re supposed to lead community and you don’t do anything for that, your popularity is taken away.

  • 2010-09-20 12:06 am
    msahagian

    I’m not sure why many people are seeing this as a bad thing. In my mind a fork might actually swing this distro back in the right direction. While I haven’t used it personally since it was, if I recall, Mandrake 6, it seems to me like letting the core developers and community push it where the company would not might be just what it needs.

    • 2010-09-20 12:36 am
      WereCatf

      I’m not sure why many people are seeing this as a bad thing. In my mind a fork might actually swing this distro back in the right direction.

      Several reasons, really. One being that we have to migrate away from our existing Mandriva installations, which means having to reinstall the whole thing, all the apps and whatnot you may have there. Another that we’ll have to use some other distro in the meantime while waiting for Mageia. Oh, and how about the stabilization time: I’ll take atleast a year for Mageia to stabilize, and given how incredibly huge repositories there are for Mandriva it’ll take a good long time for all that to appear for Mageia.

      With all kinds of hurdles ahead of it, one of them being exactly the sheer amount of software they’ll need on their repos, it’ll take atleast 2 years before I can consider using it. That’s the issue, really: Mandriva was usable now, Mageia isn’t.

      • 2010-09-20 6:21 am
        UltraZelda64

        I don’t know… maybe it’s just my opinion, but I think clearing a machine and starting over is actually a good thing. Cleans up garbage that’s accumulated over time. If you have a separate /home partition, this is made even better; keep /home, but nuke the rest. I’m used to the Debian/Ubuntu way of package management, but I’m sure it applies to Mandriva as well: I can usually install re-install the typical programs and set up the system-wide configuration files in a tiny fraction of the time it takes to set up Windows machine. No endless updates/reboot sequence that’s associated with a Windows machine.

        I agree that it will take time to adjust and that the new distro will take some time to stabilize, but there are a couple of things to say about that. First, if what you have right now works and you don’t need to have the latest and (supposedly) greatest versions of software, then why would you need to upgrade right away anyway? If you do want/need the latest, then yeah, you’ll probably have to switch temporarily, but again–with a separate /home partition it’s no big deal. And last, unless you run a business with dozens of machines to upgrade, this goes back to the first point: what’s wrong with a good wipe for once?

        • 2010-09-20 8:58 am
          WereCatf

          First, if what you have right now works and you don’t need to have the latest and (supposedly) greatest versions of software, then why would you need to upgrade right away anyway?

          For security fixes you need something that is stable and keeps updated regularly. And it doesn’t seem likely that Mandriva will be able to deliver that for long now.

          And last, unless you run a business with dozens of machines to upgrade, this goes back to the first point: what’s wrong with a good wipe for once?

          For some it’s fun to try out stuff and reinstall every now and then, but personally I’d rather just keep what’s working. Linux does not really accumulate unneeded cruft so there is really no benefit to wiping it out. The only “cruft” I have is in /tmp, /var/log and whatever is left by the package manager, and I don’t quite consider log-files unnecessary cruft.

          So, reinstalling would benefit me…nothing. But just take time and effort.

        • 2010-09-20 8:59 am
          Yagami

          yeah … people love rewrites and cleaning and what not….

          just dont release a 1.0 version with a bug or with a missing app … OR THEY WILL KILL YOU, INSULT YOU, AND NEVER FORGIVE YOU

      • 2010-09-20 3:22 pm
        josi

        One being that we have to migrate away from our existing Mandriva installations,

        No, I don’t think you need to. If I have understood correctly, it will just be an upgrade like it usually is when upgrading to a new version.

        The future Mageia distro is still being developed in *Mandriva* Cooker.

        Don’t make your up conclusions before the Mageia team have published more detailed plans.

    • 2010-09-20 5:50 pm
      bryanv

      Only if the employees really are that stupid.

        • 2010-09-21 5:45 am
          spiderman

          It does, but does it make sense to do that?

          If Mandriva was the exact copy of another distro but released a little later, would you run it? Why not get rid of the lag and get the real thing? What is the point in running Mandriva when it is lagging behind another distro? And what is the point for Mandriva to copy another distro and call it Mandriva?

  • 2010-09-20 9:11 pm
    MadRat

    When Mandrake Linux (soon to be Mandriva) bought out Lycoris and killed the BEST user-friendly desktop distribution out there it really saddened me for Linux in general. Here was a brilliant work of art snuffed with the stroke of a pen. Damn them for killing Lycoris.