Announcement

Collapse
No announcement yet.

Ubuntu's Unity Written In Qt/QML For "Unity Next"

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

  • #21
    Originally posted by pingufunkybeat View Post
    Do these people have absolutely no direction? What the hell is going on there?

    First they adapt GNOME 2 with a patchset from hell. Then they adopt GNOME3 elements, but write their own interface called Unity. Then they design a 2d look-alike called Unity 2D written using Qt. Then they drop it. Now they are rewriting Unity in QML.

    It must be hell for their developers. Tomorrow someone might come in and say "Forget it boys, today we're rewriting everything in C# and FLTK". Who makes these decisions?
    Personally I found canonical's gnome 2 FAR better than vanilla gnome 2, I couldn't stand using gnome without ubuntu's patches (especially the indicators which were a huge improvement over gnome 2's shitty system tray)

    I'm not saying canonical is perfect or that they always make good decisions, but Here's a breakdown of why their decisions were made:

    When ubuntu was first started gnome was the simplest and easiest desktop around, so they based ubuntu one that since it fit with their goals. As time went on ubuntu and upstream gnome began disagreeing about how to better the gnome desktop, ubuntu wasn't able to get indicator's or notify-osd accepted upstream because they didn't fit with gnome's vision of the future gnome 3, so they added the ayatana patched to their gnome 2. When gnome 3 came around canonical and gnome further and further disagreed on how to do things, but canonical still heavily depended on gnome for the desktop, so they wrote their own unity shell and continued to use gnome underneath.

    Now its clear that canonical and upstream have very different ideas on how to do things, so it is in canonical's best interest to completely break from gnome and write unity in QT/QML (QT/QML is also straight up technically superior too IMO, and I am a gnome user). I think this is a smart decision given what they are trying to do. No one is forcing you to use ubuntu anyway.

    I do think they could have handled this more smoothly of course, instead of doing all their own stuff with unity, compiz, and NUX they should have stuck with using a modified gnome-shell (gnome-shell is written in javascript and is very extensible/flexible, they could have forked it like mint did. unity could have easily been implemented using gnome-shell technologies, and it probably would have been better than using compiz) until they were ready to switch over to mir/QT/QML, would have been a lot less wasted effort.
    Last edited by bwat47; 04 March 2013, 04:45 PM.

    Comment


    • #22
      Next step, base Unity on KDE Plasma

      Interesting development. Qt is so much more advanced and nice to use I wonder why they didn't choose it from the very start. Well now the only thing they have to realise is that they don't need to reinvent the wheel and just use KDE Plasma as the basis for their Unity implementation (in similar way that now Plasma Netbook and Plasma Active offer completely different interfaces yet still use a lot of common Plasma code in the background).

      Comment


      • #23
        Originally posted by coder543 View Post
        "millions."

        Really? I'd like to some sources... I really doubt this number.
        It's an estimate. Let's say that you can get a decent programmer for $40000 a year. That's 10 guys working for 2.5 years.

        Comment


        • #24
          This is the first good decision that Canonical has made in a while. Qt is so much more advanced than GTK and pretty much always has been. Honestly, only the tin foil hat folks and people who have an aversion to object oriented programming would dislike this decision.

          Comment


          • #25
            Originally posted by Tsiolkovsky View Post
            Interesting development. Qt is so much more advanced and nice to use I wonder why they didn't choose it from the very start. Well now the only thing they have to realise is that they don't need to reinvent the wheel and just use KDE Plasma as the basis for their Unity implementation (in similar way that now Plasma Netbook and Plasma Active offer completely different interfaces yet still use a lot of common Plasma code in the background).
            Or even simpler, just drop Ubuntu entirely and go contribute to Kubuntu instead

            Comment


            • #26
              Originally posted by coder543 View Post
              Ah, you've put your finger on the problem precisely! that explains exactly why Ubuntu is getting the first serious media coverage it has ever gotten...

              Not really, no. QML is very flexible, and combined with Mir, if Mir works out, then they'll be able to pull this off. If Mir doesn't work out, they might be up a creek without a paddle, as the saying goes... but otherwise, this idea is just crazy enough to work.
              "Crazy enough to work" only works in movies, not real life.

              Comment


              • #27
                Originally posted by pingufunkybeat View Post
                It's an estimate. Let's say that you can get a decent programmer for $40000 a year. That's 10 guys working for 2.5 years.
                Their developers are working on many different projects; your math is a very, very rough estimation technique. If this were a company the size of Microsoft, that team of developers would only be working on that project, and the math would make sense. Here, it is all very subjective, and I still say it wasn't a waste. They accomplished what they were hoping to, and that makes it worthwhile. They obviously could've gone about things more efficiently to some small extent, but every company has room for improvement.

                Comment


                • #28
                  Originally posted by dee. View Post
                  "Crazy enough to work" only works in movies, not real life.
                  Stupidity is defined as trying the same thing over and over again and hoping for different results. If they kept doing what they had always been doing, then they have no hope of changing things and becoming a big player. It works in real life too.

                  Comment


                  • #29
                    Originally posted by coder543 View Post
                    Stupidity is defined as trying the same thing over and over again and hoping for different results. If they kept doing what they had always been doing, then they have no hope of changing things and becoming a big player. It works in real life too.
                    Nope that's insanity not stupidity. Stupidity can manifest in lots of different ways. For example putting a cow in a centrifuge and expecting to get milkshake.

                    Comment


                    • #30
                      Originally posted by BO$$ View Post
                      I don't get why do people have something against OOP? Is it something in the linux culture to hate modern programming practices? If it's not assembly you're not man enough or what?
                      While I do prefer Qt, I actually prefer plain old C. OOP is not necessarily always better despite what they teach in school these days. Some people are just a little more rigid in their style and can't stand OOP. I see uses for both styles. C++ is not the most elegant language and that seems to turn off a lot of people.

                      Comment

                      Working...
                      X