Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
Linux Software

Linux Foundation's Desktop Linux Survey Results 172

DeviceGuru writes "While the Linux Foundation's third annual desktop Linux survey doesn't officially end until November 30th, the number of daily respondents have shrunk to a trickle and the Foundation is working on analyzing the results. They now have up an early look at the raw data. For starters, almost 20,000 self-selected users filled out this year's survey compared to fewer than 10,000 in 2006's survey. Not surprisingly, the Ubuntu family of Linuxes is the most popular among organizations, at 54.1 percent. This was followed by the Red Hat family — RHEL (Red Hat Enterprise Linux/Fedora/CentOS) — with 50.2 percent. The Novell SUSE group — SLED (SUSE Linux Enterprise Desktop) and openSUSE — came in third, with 35.2 percent."
This discussion has been archived. No new comments can be posted.

Linux Foundation's Desktop Linux Survey Results

Comments Filter:
  • No Debian? (Score:4, Interesting)

    by cerberusss ( 660701 ) on Thursday November 22, 2007 @03:40PM (#21448375) Journal
    Both my current and previous employer has supplied me with a Debian desktop. No Ubuntu so far...
    • Re:No Debian? (Score:4, Insightful)

      by yog ( 19073 ) * on Thursday November 22, 2007 @03:46PM (#21448417) Homepage Journal
      Ubuntu is based on Debian so you could argue that Ubuntu has gotten Debian out to the masses. My home workstations have progressed from Redhat to Fedora to Suse to Ubuntu and I feel that they are all fine distributions with their particular strengths, but Ubuntu definitely wins on the plug-and-play aspects. I put it on a Dell laptop and except for having to manually download and configure ndiswrapper to handle wireless networking, it practically required no technical knowledge. The most recent release in fact does away with the ndiswrapper step, I believe. It's not surprising that Ubuntu wins. I hope that the other distributors learn from the success of Ubuntu and make their next releases "just work", thus undercutting one of Microsoft's main arguments against Linux.
      • Re: (Score:2, Insightful)

        by Nazlfrag ( 1035012 )
        How the hell Microsoft gets to play the 'just works' card is beyond me. Unless, of course, by 'just' they mean 'barely', like 'it works, but only just.' Apart from the overly heavy handed authentication that breaks on trivial hardware changes, the godforsaken registry from hell and lets not forget their latest innovation in insanity, you can [Accept] windows 'barely works' or live in [Denial].
    • by cyphercell ( 843398 ) on Thursday November 22, 2007 @03:46PM (#21448423) Homepage Journal

      [sarcasm]It's okay, Debian's in the Ubuntu family of Linuxes [/sarcasm]

    • Re:No Debian? (Score:4, Informative)

      by Firefalcon ( 7323 ) on Thursday November 22, 2007 @04:06PM (#21448569) Journal
      From the linked article:

      "Debian (22.2 percent)"

      So looking good... :-)
    • That performance or control over the OS isn't what drives adoption, but instead, it is bloat. Ubuntu could be claimed to be less bloated than RHEL and SUSE (both of which drip with bloat), but overall, I was surprised that among corporate offices and IT places that do Linux, not many are really using Gentoo or LFS or some such OS with a higher degree of control over what goes into the final installs, etc.

      Oh well, guess its best to be among the few, than among the many.
      • Re: (Score:3, Insightful)

        Yet MS products which give you just about 0% of control are still dominant so It is not surprising that most are Ubuntu, Debian or SUSE based because those give you better hardware detection, plus, Gentoo, LFS, Source Mage, and Arch Linux despite being great distros, lack commercial support that you can get from Red Hat, SUSE and Ubuntu. Also, the fact that you have to rebuild every update from scratch is a real pain on Gentoo, despite it being great for a home user, having 1-2 hours of 100% CPU usage in a
        • Re: (Score:3, Informative)

          by mechsoph ( 716782 )

          Also, the fact that you have to rebuild every update from scratch is a real pain on Gentoo, despite it being great for a home user, having 1-2 hours of 100% CPU usage in a business means that 1-2 hours employees can't work.

          An institutional Gentoo installation would probably have one or two compile/test machines to produce packages, then just install the binary packages on all the production machines. At least that's how Purdue's CS department seems to do it.

    • The survey's your run of the mill online questionnaire. So it's a self-selecting sample (head over to reddit for lots of foaming posts how some mainstream media page took down a poll after Ron Paul got 80% of the votes. The vast leftright-wing conspiracy at work...)

      The only thing it measures is hype, and the most interesting fact you can get from the data is that debian users got better things to do than fill out online-surveys. =)

      Of course this doesn't mean that the data doesn't correlate with reality, i

  • by chrb ( 1083577 ) on Thursday November 22, 2007 @03:42PM (#21448387)
    www.linuxfoundation.org appears to be some kind of domain search squatter.
  • by Confessed Geek ( 514779 ) on Thursday November 22, 2007 @03:44PM (#21448411)
    Update the link in the original front page post.

    http://www.linuxfoundation.org/ [linuxfoundation.org] is NOT http://www.linux-foundation.org/ [linux-foundation.org]

    The first is just a traffic collector page.

    The Linux Foundation mentioned in the story is at
    http://www.linux-foundation.org/ [linux-foundation.org]

    Thats where you will find the article/survey.

     
  • "Family"? (Score:5, Insightful)

    by Bo'Bob'O ( 95398 ) on Thursday November 22, 2007 @03:51PM (#21448457)
    Family? I guess that make sense. Ubuntu of the Debian Order, Linux Class, UNIX Phylum. I guess that would make the Genus the particular type (server/home), and the species it's version number.
  • %139.5 (Score:1, Interesting)

    by Anonymous Coward
    Well, I could have believed %100 Since this survey was filled out by linux users, but %139.5 ?!!!
    Am I the only one who sees a problem with the math here?
    • Re:%139.5 (Score:5, Informative)

      by J0nne ( 924579 ) on Thursday November 22, 2007 @04:09PM (#21448593)
      If you'd RTFA, you'd have read that you could pick multiple distro's. The question was 'which Linux distributions do you run in your organisation', and apparently lots of organisations run several different distro's, instead of standardising on one.
    • Maybe some people or organizations use more than one version?
    • Re: (Score:3, Insightful)

      by grcumb ( 781340 )

      Well, I could have believed %100 Since this survey was filled out by linux users, but %139.5 ?!!!
      Am I the only one who sees a problem with the math here?

      Yes. If you bothered to RTFA:

      "Yes, that does add up to more than 100 percent. It would seem that groups using Linux in the office have not standardized on a particular distribution, or even a distribution family."

      Linux users are - amazingly - capable of using more than one OS at once. I know this is anathema to those who believe that the only alternative to white is black, and for whom anything less than perfect logical symmetry causes cranial asplosion. But hey, we got into weird territory right fro

      • That's all well and good but shouldn't it still be based on 100%?

        Lets take three users:
        User 1 uses Ubuntu and Red Hat
        User 2 uses Ubuntu
        User 3 uses Red Hat

        Given this you would assume that it was a 50/50 split but given the numbers what you are going to get is:

        Ubuntu = 2/3 = 67%
        Red Hat = 2/3 = 67%

        WTF kind of math is that? Shouldn't it really be:

        Ubuntu = 2/4 = 50%
        Red Hat = 2/4 = 50%

        • Re: (Score:3, Insightful)

          WTF kind of math is that? Shouldn't it really be:

          Ubuntu = 2/4 = 50%
          Red Hat = 2/4 = 50%


          No, it should be just as it was written. It's the percentage of *users* who answered the survey, not the percentage of all answers that were a particular answer.

          Given your sample data, ~67% of *users* use each of the operating systems.
          • Actually, I believe it was organizations but it isn't really telling us much. I'm just saying that perhaps they could have expanded on it more. For example, lets take the same three users/organizations based on installations:

            User 1: Ubuntu installed 97 times, RHEL installed 1 time
            User 2: Ubuntu installed 1 time
            User 3: RHEL installed 1 time

            Now, since there is only a checkbox and no room for a number, RHEL still comes out looking like a winner with 67% even though it is really only 3% installation
            • User 1: Ubuntu installed 97 times, RHEL installed 1 time
              User 2: Ubuntu installed 1 time
              User 3: RHEL installed 1 time

              s/User/Organization/g
              Organization 1's IT staff must possess competency to run both Ubuntu and RHEL; Org 2 only needs to know Ubuntu; Org 3 only RHEL. The number of machines with each distro isn't awfully important unless it's a really huge entity, and they're able to confine the expertise to smaller teams (e.g. the server admins know RHEL but the Helldesk techs only know Ubuntu). The s

        • Well, it's an online survey. The data probably gets collected in a db.
          The db has a record for every answer (1 record == 1 user).
          So it's more simple to do the math on the number of users rather than the number of selections: having done SQL, and being the lazy programmer I am, I'm pretty sure that's the main reason behind it.
        • No, because 2/3 of users do use Ubuntu, and 2/3 of users do use Red Hat.
        • More to the point (made up numbers):
          • User 1 uses Debian.
          • User 2 uses Debian and Ubuntu.
          • User 3 uses Fedora and Ubuntu.
          • User 4 uses Fedora.

          According to their system, each distro would get 50%, even though Fedora and Debian are (in this example) able to meet the needs of an organisation each on their own, while Ubuntu is only used in concert with another distro. More realistically, the results would be 37.5% Debian, 37.5% Fedora, 25% Ubuntu from these numbers.

          That said, if they don't include the number o

  • Server vs. Desktop (Score:5, Interesting)

    by Glowing Fish ( 155236 ) on Thursday November 22, 2007 @03:55PM (#21448499) Homepage
    Another interesting result from the LF survey is that in most company and organizations, the Linux desktop is more commonly used than Linux servers. From almost the beginning of Linux's business acceptance it has always been assumed that Linux was, is, and would continue to be more of a force on servers than on desktops. That appears to be changing.

    Is it just me, or is this possibly a misleading statement? Does "more commonly used" just mean more numbers? Or does it mean that organizations with Linux desktops aren't running Linux servers? Or just that they have more desktops than servers? Even if it is the first, I still don't think it means too much, because one organization running a gigantic Oracle database on big iron and Linux is going to probably be using Linux more than another organization running Linux and OpenOffice for word processing on 10 or even 50 desktops.
  • by cstdenis ( 1118589 ) on Thursday November 22, 2007 @03:57PM (#21448517)
    It's official, 2008 will be the year of the Linux desktop.
    • by turgid ( 580780 )
      Yes, well, every year since 1996 has been the Year of Desktop Linux for me, starting with olvwm on Slackware 2.x. Now I have Window Maker on Slackware 11.0.
      • Amen to that!

        I found IE4's integration just a bit too buggy and creepy and haven't used a Windows desktop since.
  • by chrb ( 1083577 ) on Thursday November 22, 2007 @03:58PM (#21448521)
    Fill in the survey [linux-foundation.org].

    Current results [linux-foundation.org]

    The results say the current number of respondents is 10941 (and counting). Where did the figure of 20,000 come from?

    • Re: (Score:2, Funny)

      by vtcodger ( 957785 )
      ***The results say the current number of respondents is 10941 (and counting). Where did the figure of 20,000 come from?***

      Rounding Error?

      Probably related to the logic that has 139.5% of the users reporting in already.

      In any case, we certainly are not going to blame these little arithmetical peculiarities on Linux. How about we blame Vista, Internet Explorer, the RIAA, George W Bush, and Intel? Don't worry. Ron Paul, Ubuntu, the second amendment and the free market will pull us all through this li

    • 10000 is only english speaking results. You have a few hundred for each other language too.
    • Thank you for providing a link to the survey. The survey was not on the main page; the link there merely led a page that blabbed all about how wonderful it was to have a survey, etc. but didn't point to THE ACTUAL SURVEY. Grrr! (Okay, now that I've said this, someone's going to point out some obvious link to the survey, but I had trouble getting to the survey.)
    • You have to add up all the language specific surveys. The french survey currently has 1500 participants, the brazilian 600 participants and so on.
    • We need a new moderation option: doing the editors' job for them.
    • Aggregate the languages to get the total number.
  • For those who would like to take the survey, here's a link. https://www.linux-foundation.org/en/2007ClientSurvey [linux-foundation.org] ---Alex
  • Novell downturn? (Score:5, Insightful)

    by brejc8 ( 223089 ) * on Thursday November 22, 2007 @04:29PM (#21448725) Homepage Journal
    I keep reading how this MS/Novell agreement is gaining customers but here I can see that:
    in 2005 Novell/SUSE got 28%
    in 2006 Novell/SUSE got 16%
    in 2007 Novell/SUSE got 11.7%
    • Re: (Score:1, Insightful)

      by Anonymous Coward
      They are gaining customers but the other distros (especially Ubuntu) are gaining more customers faster.
      • Yes, that's possible but the importance of SUSE is declining. Not some I would like to report to my boss.
    • by cp.tar ( 871488 )

      Well, that just means Linux is growing so fast that although Novell keeps gaining new customers, several other distros are growing even faster than that.

    • by Ed Avis ( 5917 )
      The survey covers how many people are running a particular distribution and replied to the survey, not how many are running a particular distribution, and certainly not how many paid money for it. Novell and other companies are mostly interested in the third number, a bit in the second, and not in the first.

      Customers are not people who downloaded a Linux distro gratis. They are people who have paid money for it.
      • by muszek ( 882567 )
        Yeah, and there's definitely no correlation between SuSe losing market share according to the same freaking survey as one year ago and sun losing real market share.
  • Problem (Score:5, Insightful)

    by Anonymous Coward on Thursday November 22, 2007 @04:53PM (#21448935)
    If you fill out the survey, it asks you about anti-virus, and specifically porting bigname AVs to linux.

    A few questions I pose:
    1) Why do we want the bloaty, slow, pieces of crap that are windows AVs ported to linux?

    2) Why do we want to port these, encouraging turning a blind eye to security and letting the AV do the work(such as it is on windows)?
    3) Why not just improve support on say, ClamAV?
    • by o'reor ( 581921 )
      What about those of us who have pointy-haired bosses or BOFHs that try to enforce security policies on all workstations -- both Windows and Linux -- and make the installation of commercial AVs mandatory on Linux workstations ?


      Well guess what : those guys are more likely to be heard when they complain about availability of AV programs under Linux, than your average fanboi has when demanding the port of Macromedia or Adobe tools.

  • by Anonymous Coward
    > Not surprisingly, the Ubuntu family of Linuxes is the most popular among organizations, at 54.1 percent.
    > This was followed by the Red Hat family -- RHEL (Red Hat Enterprise Linux/Fedora/CentOS) -- with 50.2 percent.
    > The Novell SUSE group -- SLED (SUSE Linux Enterprise Desktop) and openSUSE -- came in third, with 35.2 percent."

    Q: What did desktop linux users miss most?
    A: A reliable calculator!
  • It seems after the microsoft deal that it turned into the plague that infects everything.
  • by _Hellfire_ ( 170113 ) on Thursday November 22, 2007 @06:41PM (#21449649)
    linuxfoundation.org appears to be a domain squatter site.

    Whois shows:
    Last Updated On:26-Oct-2007 19:57:38 UTC

    Which is not the same day and month as the creation date, so I'm suspecting either someone has taken this domain over or it wasn't legit in the first place (I don't know as I don't think I've ever been there). Maybe check our links before we post them to the front page on ./? Hits on these types of sites just encourage domain squatting.
  • but gets only 5.4% in this survey. I don't understand what's going on.
    • Can't explain the Distrowatch popularity figures (never could). I think that you'll find that PCLinuxOS is a fairly small player by any other measure though. Try Google Trends [google.com] search on "Ubuntu,PCLinuxOS" as an example. The flatline at the bottom is PCLinuxOS (and it doesn't look much different if you compare Debian, Red Hat or others).
  • by dermoth666 ( 1019892 ) on Thursday November 22, 2007 @06:55PM (#21449749)
    I just took this survey earlier today, and after looking at the results it is obvious that it is totally biased.

    I'm writing from my phone so I won't go in-depth, but two things that bug me the most:

    1: It looks like many home users took the survey, but are being categorized as SOHO's

    2: At first it looks like the survey adress both desktop and server usage, but then the questions begin assuming repondent are using Linux on the desktop workstations. This isn't the case in my company, but he results to these questions are being used to show Linux desktop penetration.

    I also responded to some questions thinking "servers only" but it end up being both servers and workstation. In an organisation with more employees than servers, all running Windows, this obviously change the result!

    I'm not a Linux detractor, quite the opposite, but I'm being honest here. When you do surveys, please ask the right questions and make sure anyone responding to the survey won't bias it if the're not the targetted audience. To me this survey says almost nothing...
    • Re: (Score:3, Insightful)

      While the word I'd use is "skewed", I'd basically agree with you.

      There was an original announcement of the survey in the Linux-covering media, and I looked at it but didn't take the survey then, as it seemed only interested in business use. Later, there was additional coverage, asking where all the North American users were, as there had been relatively few such responses to the survey at that point. Most were and still are European, altho the North American response percentage increased from about 10% to
  • I used to care about stuff like this about 10 years ago when I was concerned whether or not this "linux thing" was going to last. I don't, anymore. Linux ain't goin' away. As a daily Linux user and abuser, this is the first time I heard about the poll, after 3 years of its existence and I still don't care.

    That's a good thing.

    --
    BMO
  • Shouldn't that be Debian family? Knoppix is also a somewhat popular desktop (at least it was at some point), and it's not exactly ubuntu, so it's not just a purist's argument (at least that's my excuse).
  • When I got to the proprietary apps question I realized I don't really want to see any of those on Linux.

    It was a realization that either Open Office and other Linux apps are already doing a good enough job for what I or my office would need, or I would rather those who do use those particular apps to convert their documents to support more open formats.

    I did write in one though, Print Shop. Maybe KreetingKard Card will improve.
  • How come in the "What is your job within your company or organization?" section, the only thing around software development is "software developer"? Where is software QA/test? And why is this such an overlooked part of software development? I've been doing it for 13 years, and it always seems to be the thing that is overlooked the most often before going out the door, yet is always the scapegoat when things go wrong. Why is it so hard to get out of the land of "programming" and into the world of "softwa

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...