Announcement

Collapse
No announcement yet.

Gnome kills KDE

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #1
    Is this incredibly bad trolling or do you seriously believe that quoting statistics out-of-context, and indeed, ones which are pretty irrelevant to your claim, is proof of something 'dying'?

    Comment


    • #2
      Ok, one thing... let's forget the "taking statistics out of context" thing for a while. Let's assume, hypothetically, that these graphs represent accurate statistics that accurately reflect the amount of manpower working on each desktop environment.

      So... wouldn't this mean that in 2010, KDE was 2.5 times "healthier" than GNOME, and their "health" level has since dropped and only recently, in the beginning of this year 2013 dipped below Gnome's "health", and is now only 30% lower than Gnome? According to this statistic, KDE has had lots of extra "health" earlier, and has only recently dropped to the same "health level" as Gnome...


      Anyway, I think this is stupid, there's no point in being against either DE, they both have a userbase, they both have their own vision, and there's plenty of room for both to coexist in peace and harmony on the Linux desktop ecosystem. Both have good and bad points, but at least they're both aiming towards Wayland support, and both contributing towards an open, collaborative future of a next-generation graphics system on Linux.

      Comment


      • #3
        You just came here to troll, didn't you? Anyway I have a nice diagram for you which proves that GNOME is dying:

        Comment


        • #4
          Originally posted by Honton View Post
          KDE has so much code, that more people are needed. Anyway as it is today, KDE is lacking man power.


          KDE has about 1/3 less contributors, but those change over twice the number of files, changing over three times as many lines of code.

          Comment


          • #5
            Originally posted by Honton View Post
            I can't think of a better way to kill quality. Can you?
            Developers experienced with a code base are not only able to work faster than new developers, but they are able to produce higher quality code while they do. Anyone who knows this would never make a remark such as yours. The information provided does not permit us to draw conclusions on code quality.

            Comment


            • #6
              Desktop is irrelevant when all the distributions are dying.. Ohloh

              Comment


              • #7
                lol @ gnome

                Comment


                • #8
                  Originally posted by AnonymousCoward View Post


                  KDE has about 1/3 less contributors, but those change over twice the number of files, changing over three times as many lines of code.
                  I would assume that's because they're making a huge shift to KDE5 and Qt5 so they're refactoring everything... right?

                  Comment


                  • #9
                    Originally posted by ryao View Post
                    Developers experienced with a code base are not only able to work faster than new developers, but they are able to produce higher quality code while they do. Anyone who knows this would never make a remark such as yours. The information provided does not permit us to draw conclusions on code quality.
                    More lines of code means more defects.


                    This whole Gnome vs KDE debate is kind of stupid, btw.

                    Comment


                    • #10
                      Originally posted by Honton View Post
                      Nice graphs. Im happy to inform you that both KDE and Gnome have mangers overseeing their Ohloh stats. So don't worry, KDE is really dying
                      KDE doesn't have managers it have had.

                      Comment


                      • #11
                        Originally posted by Honton View Post
                        Every code change is a potential regression. I did not state anything about the developers's code expericence. But Im glad you made me do that. KDE have lost many experience hackers lately, where as the Gnome camp have lost less experienced hackers. So thanks for bringing this up KDE is dying because less people care about KDE.
                        Not really. Things like comments and white space fixes are not regressions. Bugs in new applications are not regressions. Adding code to test suites tends to prevent regressions.

                        Anyway, you have no statistics on the actual distribution of developers and not all developers are equal. You could have a situation where all significant changes are made by one person and all others are one-liners. Without actual statistics on the distribution, the numbers posted are meaningless. You should take a basic course on statistics. A proper education should bring you to realize the meaninglessness of what you are posting.

                        Originally posted by kigurai View Post
                        More lines of code means more defects.


                        This whole Gnome vs KDE debate is kind of stupid, btw.
                        When all things are equal, yes. However, these are just statistics on modifications. Neither the composition of the code is provided nor the composition of the changes are included. If the changes involve new applications, then any defects in them would not affect existing applications and therefore would be fairly innoculous. Also, things like comments and white space also do not contribute to defects. Adding code to test suites has the effect of reducing defects. There are plenty of factors, including some that I did not mention, but my point is that trying to reason based on these numbers is meaningless. The information people want is not there and trying to reason about it anyway lets you fill-in whatever result you want. It is illogical.
                        Last edited by ryao; 09-16-2013, 03:01 PM.

                        Comment


                        • #12
                          Originally posted by Honton View Post
                          Every code change is a potential regression. I did not state anything about the developers's code expericence. But Im glad you made me do that. KDE have lost many experience hackers lately, where as the Gnome camp have lost less experienced hackers. So thanks for bringing this up KDE is dying because less people care about KDE.

                          Comment


                          • #13
                            *sigh* - there was already the same trolling in another thread. There I was still questioning whether it's just ignorance, but given that thread title I'm now quite convinced that it's trolling. @Michael: you should do something about it. I think the trolling in your forum is harming your platform more than the generated clicks.

                            Now I'm going to cite myself from the other thread (which is worth reading giving that more obvious issues with the comparison are shown) where this smart graph has been posted:

                            Originally posted by TheBlackCat View Post
                            Still spreading lies I see. For everyone's information, this only tracks the master branch of git repositories. A lot of KDE development right now is going on in frameworks 5-related branches, which will not show up in those numbers until they are merged later this year or early next year.
                            Not entirely true. I just checked on ohloh.net and for kdelibs the frameworks branch is also tracked. For other repositories it isn't. E.g. kde-workspace is only tracking master and thus for example the 160 commits to kwin last month missing.

                            In general the numbers from ohloh.net have to be taken with a salt of grain. New repositories are missing. E.g. kscreen is missing, bodega is missing, the gsoc project I'm mentoring is missing, etc. etc. It's not auto-synced with KDE's project files.

                            Also the drop down in the stats a few years ago is rather pointless. It's when KDE started to switch from svn to git. As a matter of fact people used the svn repository also as a "scratch" area. With git that completely goes away. Creating a local repo is rather trivial and even if you want to backup it on KDE infrastructure you just create a scratch repository which will never make it to ohloh.net.

                            Number of commits is also pointless as in SVN times the number of commits was about twice the number of what we have in git as we did backports. Nowadays we do merges. So the number goes down obviously.

                            So in summary: awesome graph! Unfortunately not telling what it looks like.

                            Comment


                            • #14
                              Originally posted by mgraesslin View Post
                              *sigh* - there was already the same trolling in another thread. There I was still questioning whether it's just ignorance, but given that thread title I'm now quite convinced that it's trolling. @Michael: you should do something about it. I think the trolling in your forum is harming your platform more than the generated clicks.

                              Now I'm going to cite myself from the other thread (which is worth reading giving that more obvious issues with the comparison are shown) where this smart graph has been posted:


                              Not entirely true. I just checked on ohloh.net and for kdelibs the frameworks branch is also tracked. For other repositories it isn't. E.g. kde-workspace is only tracking master and thus for example the 160 commits to kwin last month missing.

                              In general the numbers from ohloh.net have to be taken with a salt of grain. New repositories are missing. E.g. kscreen is missing, bodega is missing, the gsoc project I'm mentoring is missing, etc. etc. It's not auto-synced with KDE's project files.

                              Also the drop down in the stats a few years ago is rather pointless. It's when KDE started to switch from svn to git. As a matter of fact people used the svn repository also as a "scratch" area. With git that completely goes away. Creating a local repo is rather trivial and even if you want to backup it on KDE infrastructure you just create a scratch repository which will never make it to ohloh.net.

                              Number of commits is also pointless as in SVN times the number of commits was about twice the number of what we have in git as we did backports. Nowadays we do merges. So the number goes down obviously.

                              So in summary: awesome graph! Unfortunately not telling what it looks like.
                              Thanks for sharing. This is all information that I did not have.

                              Comment


                              • #15
                                Originally posted by Honton View Post
                                Another month, another Ohloh stat showing KDE is dying and Gnome stays fit and healthy.

                                And for those in total denial I can only say KDE commit digests show same trend and those covers the pie-in-the-sky called KF5.

                                Byebye KDE!
                                Honton troll is back! Too bad for you it's KDE and Unity that kills gnome. It's unusable piece of shit.

                                Comment

                                Working...
                                X