Announcement

Collapse
No announcement yet.

Qt LTS Releases To Be Restricted To Commercial Customers, Other Commercial Changes

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

  • #21
    Ah, 144Hz's job of infiltrating the Qt company board has succeeded.

    Comment


    • #22
      We already have GTK+3, Qt, FLTK, Tk, wxWidgets, several mobile-focused options, that GNU project that copied Mac OS and fell behind over a decade ago, the Enlightenment toolkit, and more.

      Nobody uses anything but GTK+3 and Qt for serious projects, partly because the alternatives don't visually blend in with the main desktops or because using them is even *more* painful. Or, in the case of formerly-popular codebase like GTK1 and GTK+2, are too limited or unable to handle the high-resolution screens of today.

      You want to stop this nonsense? Buy out the Qt Company, or at least a controlling share. Or convince IBM to force RedHat to either cease development of GTK+, or redesign and rewrite the entirety in Rust, at the expense of other development.

      Personally, I'm in favor of the latter, as I have hazy memories of there already being some interest in ditching both C and their special Vala language in Rust's favor. Hopefully they solve the First World Internet Availability Syndrome of the Rust ecosystem, first.

      Enough of the forks! Stable, secure, scalable-resolution multiplatform GUI libraries are not an easy project to begin with, and we have limited programmer resources. Most who are experienced with GUI library dev are already employed by the aforementioned companies, or prefer web-based junk-tech that requires an entire browser running per app.

      Yes, Electron, I still have problems with your very existence.

      I could start an entire rant about how the Linux world needs to abandon the bazaar method for certain layers, but that would be a blog-post type thing and I abandoned that many years ago. I'll leave it at saying that hard forks should only be done when project leadership puts personal powermongering over the good of the programmers and project... and that in such circumstances, the first thing the new project must do is not to write new code but to set up a system so that admins can be excised from either the position of power or from the project as a whole (depending on how toxic or whiney said punished admin is).

      Comment


      • #23
        Originally posted by onicsis View Post
        Too bad . Can't be sure today about what's will happen tomorrow. Thinking on long term the answer right answer to Qt [or GTK] it's something else. From scratch
        There are a few GUI toolkit projects in the rust context, e.g. Druid, OrbTK and Iced. Just to name a few. Maybe something good comes from those initiatives.

        Comment


        • #24
          Originally posted by frank007 View Post
          Linux need a new, completely free (as in freedom) gui libs. Both the Qt libs and the Gtk+/- libs aren't. Maybe Linux (not the kernel) needs a full reboot. The gtk1 gui libs are the only good for a fork.
          There are others, like FOX and EFL for example (which needs some theming to look good though) and a few others I can't think of right now. But none of them have a company or large team of developers (except for EFL, but EFL is very controversial among FOSS devs).
          Last edited by Vistaus; 27 January 2020, 12:47 PM.

          Comment


          • #25
            Originally posted by NateHubbard View Post

            Aren't systemd and wayland already kind of a complete reboot, in the way you're describing?
            They aren't GUI libs though.

            Comment


            • #26
              I can't blame a framework developer whose sole revenue is from commercial licences for wanting to keep the lights on.

              I can't blame them either for wanting to get a better idea of the usage their framework enjoys (the Qt account stuff). I find it somewhat curious that people would revolt over the requirement of having a Qt account though; on the face of it, it doesn't seem like an entirely unreasonable requirement? What am I missing?

              From my perspective, this entire thing is somewhat similar to the model RH uses where you pay for stability if you need it as -- let's face it -- there's simply no such thing as Free Beer in the long run if quality and stability are your main concerns. Maintaining a standard of quality requires quality engineering resources. Employing consistently high quality engineering resources is not free.

              For RH, the only difference is that they/IBM can afford to offer CentOS gratis and that they in turn do their best to make it harder for Oracle to (ab?)use RH/CentOS as an upstream for their Unbreakable Linux.

              Will be interesting to see how this plays out during the next couple of years.

              Comment


              • #27
                Originally posted by Vistaus View Post

                There are others, like FOX and EFL for example (which needs some theming to look good though) and a few others I can't think of right now. But none of them have a company or large team of developers (except for EFL, but EFL is very controversial among FOSS devs).
                I don't like the EFL libs, too much useless thing in it. +1 for the FOX libs.

                Comment


                • #28
                  Originally posted by treba View Post
                  Urgh, I hope this will not impact security on Linux too hard. Does the license allow e.g. the kde project to distribute their own LTS version with backported fixes, so all distributions can pull from it?
                  These are moments when I'm very happy that we have GTK.
                  I don't think that this is a reasonable reaction to the topic of the article.
                  The Company behind KDE wants to get money - e.g. to pay developers - may be a good thing.
                  It is a question of ballance. But as long as the distributions have access to bug fixes and that the code stays free SW, all is OK.

                  I am currently still using XFCE and I will try to switch to KDE or LXQt to get a reasonable configuration and theming back, which got totally broken for my use cases.
                  GNOME has the guideline that users should not be overwhelmed by too many option - which might be OK if experts do have all options.
                  But deliberately developers dropped options of former GNOME releases and not even have all of the ancient CDE desktop options.
                  So they don't support common workflows. From my point of view this is a contradiction with the free software reasoning.

                  Same for Mozilla, who broke on purpose AddOns of both Firefox - no longer getting a button if viewed code is correct and with one click a list of possible problems (extremely helpful for web developers) - and Thunderbird - as FireTray being the sole option for getting the number of unread messages shown in system tree and being able to keep it running - even when closing window with x to just put it in the tray. And Mozilla is not responsive - they should know by now what they did. And all for security ... which is just laughable. Mozilla lost users not by inferior speed, but due to getting all the customizations destroyed which kept that loyal users in first place (I am using both programs - with frozen Thunderbird 60.9.1 as only option to keep working with FireTray, though - and I will look for options when switching HW and DE - for me Chrome is still inferior in many ways - but it's good to have it around as alternative and to check the website also with that renderer [and its limited support for Unicodes - cough ]).

                  To light one of many things which IMHO may be more crucial to the Linux desktop:
                  If I can not keep my system from using HDMI (broken sound of screen; several games switch on HDMI - prefered when starting in fullscreen ... why those options are related which should be absolutely unrelated is really puzzling) and just force SPDIF/TOSLINK (i.e. IEC958) to my amplifier and not finding any clue from internet research as Unix expert, there seems to be something severely broken (maybe in HW, but probably in SW stack).

                  So for me GPL is VERY important and supporting the needs and workflows of as many users as is possible on the developers side (keeping backward compatibility at nearly any cost) - as is the rule of the Linux kernel (with don't break user space). A shame that this rule is not propagating to basic libs and from there to the programs - wouldn't that be nice?
                  And still the desktop is the most important place - where more real time would really be needed (especially when performance is trashed by the mitigation avalanche of HW vendors) - and I am not sure PulseAudio, systemd, GTK/GNOME, Wayland in current form or flatpak/snap even in concerning its basic scheme to handle things are the way forward to help the desktop (and not only the involved companies which is my current impression).
                  Time will tell.

                  Comment


                  • #29
                    Look on the bright side, it could be worse and Larry Ellison bought out the Qt Company.

                    Edit: More info here, perhaps Michael could update the article for a fuller explanation....
                    Last edited by Slartifartblast; 27 January 2020, 01:25 PM.

                    Comment


                    • #30
                      Originally posted by frank007 View Post
                      Linux need a new, completely free (as in freedom) gui libs. Both the Qt libs and the Gtk+/- libs aren't. Maybe Linux (not the kernel) needs a full reboot. The gtk1 gui libs are the only good for a fork.
                      GTK is free already

                      Comment

                      Working...
                      X