Announcement

Collapse
No announcement yet.

Upstream X/Wayland Developers Bash Canonical, Mir

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

  • #76
    Originally posted by Pawlerson View Post
    And gnome developers don't suffer from the NIH syndrome? Unity for example, was available before gnome shell. I wonder how did you get into such strange conclusions? Ubuntu is the most popular distribution and Ubuntu having X replacement that is supported by proprietary drivers will become even more popular. If something will end it's gnome shell and fedora/red hat leadership in dictating Linux future. And this is good, because they have no clue about desktops. Nobody cares about bsd crap here. There are no anti-bsd trolls here, but only bsd trolls. Linux users have no reasons to troll about bsd which is meaningless to us. There's no way it will go down. They have Valve behind them and it seems MIR will be supported by proprietary drivers which isn't so sure about Wayland. I didn't like Ubuntu to use SurfaceFlinger in cost of Wayland, but I MIR has my full support. And even with X Ubuntu is way faster than OS X. Furthermore, Ubuntu is using Linux kernel rather some crap and it seems you're an only one who doesn't see Apple's OS X will be dead in the near future.And even with X Ubuntu is way faster than OS X. Furthermore, Ubuntu is using Linux kernel rather some crap and it seems you're an only one who doesn't see Apple's OS X will be dead in the near future.
    Put these all together and it almost makes my point for me.

    Seriously, you should choose one target to attack and stick with it, rather than lashing out anything that is not brown. Still, your argument seems to be "I dislike GNOME Shell, therefore anything that Red Hat touches is evil" which is nigh on pathological (especially since saying GNOME Shell is a Red Hat product is a bit of a stretch).

    Originally posted by ryao View Post
    Would you elaborate on how this is "destructive"? As far as I can tell, the only one who loses here is RedHat and various proponents of a RedHat-based monoculture. RedHat took advantage of open source software to form the basis of their company, adopting various projects as their own. That enabled Redhat to sell support contracts as the company that lead development of the software, which has been very profitable for them. Now that organizations have decided to do the same, RedHat's supporters are crying foul.

    RedHat should never have published open source software if they expected to dictate how the source code is used. Various BSD people accepted that a long time ago, which is why they make no attempt to dictate how their software is used. On the other hand, Redhat and their supporters seem to have mistaken the GPL as a means to establish their company as the Microsoft of open source, but things do not work that way. RedHat has praised the benefits of open source software for years. Now that others are following suit, it is time to accept that no organization can has exert monopoly control over how open source software is developed and used.

    With that said, those that want a monoculture to exist in open source software should establish it through merit and not petty harassment.
    First off, Wayland is not even developed by Red Hat, so how is this going to break this "Red Hat monopoly"?

    Second, there is some serious skewing of the facts based on what you are implying. Red Hat does not own any of these projects, has no copyright assignment on them, and does not even lead or develop on some of them (see Wayland where they do not do any development). Red Hat does indeed make a lot of money off of free software (are you going to say this is wrong when you are defending Canonical's fumbling attempts to get a profit?) but to say they "took advantage" of the software is almost a laughable statement, considering how much developer resources they have given back, many of which do not even benefit them directly but instead feed the entire upstream ecosystem. They profit from the ecosystem, as do you and I, but if I were to compare their contributions to yours and mine (and yes I know you are a developer) they would be so small we could be said to be the leeches.

    Red Hat has never released proprietary software (Canconical has), never taken out a software patent (despite Alan Cox's jest back when he was a Red Hat employee), and has a good history of working with others in broader upstream projects instead of rolling their own and throwing a wrench in others (which Canonical certainly does). What does upstream mean? It means it is applicable for everyone, is peer-reviewed by everyone, and everyone can come on board assuming they have something to offer. This whole debate is not even about Red Hat, it is about the value of upstream contributions and mutual standards compared to competitive individualism and bringing Linux back to a state of tribalism.

    Red Hat has gotten to this position through merit. But the odd thing is that is not what is really relevant with regards to this discussion.

    Comment


    • #77
      Originally posted by grotgrot View Post
      I didn't, but browsing through doesn't seem to show that many packages. It looks like Debian (and hence Ubuntu) has about double the number of packages as standard. Ubuntu's PPAs fill in any missing gaps very well. The Fedora package list is definitely more complete than when I've looked in the past.
      Between the official repositories, RPM Fusion, and Google's YUM repo I have not had much trouble finding packages. I find Arch's repos to be in a rather similar state, not counting AUR which should not be counted the same way as it is fundamentally different.[/QUOTE]

      By the way, Ubuntu officially package Gambas3 yet (sorry, just took a little frustrated jab)

      Comment


      • #78
        Originally posted by ryao View Post
        Would you elaborate on how this is "destructive"? As far as I can tell, the only one who loses here is RedHat and various proponents of a RedHat-based monoculture. RedHat took advantage of open source software to form the basis of their company, adopting various projects as their own. That enabled Redhat to sell support contracts as the company that lead development of the software, which has been very profitable for them. Now that organizations have decided to do the same, RedHat's supporters are crying foul.

        RedHat should never have published open source software if they expected to dictate how the source code is used. Various BSD people accepted that a long time ago, which is why they make no attempt to dictate how their software is used. On the other hand, Redhat and their supporters seem to have mistaken the GPL as a means to establish their company as the Microsoft of open source, but things do not work that way. RedHat has praised the benefits of open source software for years. Now that others are following suit, it is time to accept that no organization can has exert monopoly control over how open source software is developed and used.

        With that said, those that want a monoculture to exist in open source software should establish it through merit and not petty harassment.
        Red Hat doesnt control Wayland so writing this here is kinda pointless

        Why do you prefer Canonical monoculture to Red Hat one?

        Which is better Wayland controlled by freedesktop.org or Mir controlled by single corporation with CLA
        Last edited by Ramiliez; 03-05-2013, 02:04 PM. Reason: typos

        Comment


        • #79
          Originally posted by Pawlerson View Post
          There's no way it will go down. They have Valve behind them and it seems MIR will be supported by proprietary drivers which isn't so sure about Wayland. I didn't like Ubuntu to use SurfaceFlinger in cost of Wayland, but I MIR has my full support.
          So surfaceFlinger = bad ... Wayland = good ... MIR = best in your mind? ~ you do realize Mir is way behind Wayland development at this point, right? You also realize canonical's move is VERY unhealthy for the actual FOSS projects from which they have built Ubuntu on, right?

          Originally posted by Pawlerson View Post
          Nobody cares about bsd crap here. There are no anti-bsd trolls here, but only bsd trolls. Linux users have no reasons to troll about bsd which is meaningless to us.
          Actually, there are a few anti-bsd trolls on this site (their user names even reflect that) and several threads (if you bothered to look), just as there are also BSD trolls. Some of those anti-bsd trolls, probably are linux users (one would think anyway) and please don't act as if you speak for ALL linux users ~ i have no problem with BSD (i used it for several years) and wouldn't characterize BSD as meaningless. Lots of good s/w has come out of the BSD camp over the years...

          Originally posted by Pawlerson View Post
          And even with X Ubuntu is way faster than OS X. Furthermore, Ubuntu is using Linux kernel rather some crap and it seems you're an only one who doesn't see Apple's OS X will be dead in the near future.
          While it is true that Ubuntu (or more broadly gnu/linux) can 'out-benchmark' MacOSX in some areas (particularly OpenGL), to say it is way faster, is simply false... in fact, there are many areas where MacOSX destroys gnu/linux for a variety of reasons. Obvious stuff like ease of use/configuration, much higher quality applications (in vastly higher numbers), etc ...but any graphical designer/photographer, sound engineer/musician, video editor, etc used to using MacOSX (with their high-demanding apps) will laugh in your face, when you go on about Ubuntu like this ~ Ubuntu (and linux in general, is YEARS away from being able to compete in these areas). Gaming may be improving for Ubuntu, but that's about it. That being said, I still prefer Linux - but you couldn't convince me to use Ubuntu, anyway. Since imho Ubuntu blows - always has, always will ~ Also, you need to back up your claims that XNU kernel sucks, with real data. (hell, you didn't even know what it was even called - and we're supposed to accept that you know what you're are talking about. HA!) XNU kernel is actually pretty decent. OpenGL tends to be the only sour-spot i see in MacOSX.

          I think you also under-estimate Apple user's 'loyalty'. Very few Apple customers are going to be buying an Ubuntu laptop from Dell (for example) - A). Ubuntu is buggy crap and B). Dell makes shitty h/w. C). The vast majority of Apple's users aren't interested in using commandline to get trivial things done, nor are many interested in having to hack around common problems Linux users tend to face. (h/w problems, upgrade problems, 6month release cycles, lack of needed apps, etc,etc,etc). Apple customers tend to not mind paying a little extra for their computers, since for the most part, they can avoid a lot of problems + have much better support from Apple (particularly in N.America) rather than spending hours in Ubuntu's crappy forums, googling how to fix issues/bugs, or what have you...

          Comment


          • #80
            Originally posted by strcat View Post
            That's not the full installation guide, it's a summary.
            Go to the front page - archlinux.org and look at the documentation section on the right. How is anyone supposed to deduce that the Official Installation Guide is the wrong thing to use in preference for the Unofficial Beginners' Guide?

            Nowhere on the page does it mention using the Beginner's Guide, and the latter is more detailed. Gentoo's guide still looks better and they solve the short summary/checklist versus long detailed instructions by only having one set of documentation but you can look at the structure to get your summary and the contents to get the detail. Also compare Arch and Gentoo's disk partitioning to see difference in completeness and writing style.

            Originally posted by strcat View Post
            There's no bug report for any issues with the install media in VirtualBox,
            If you use the net install stuff then the kernel hangs at the uncompressing message. It turns out that if you wait two minutes or so (getting an rcu message after a minute) then things do proceed just fine. Only once was it instant. This is repeatable for me.

            Originally posted by strcat View Post
            so it must work fine (for you too, otherwise you would have reported it - right? :P).
            You mean I would have gone to the trouble of creating an account (you can't report bugs without one), and then tried to divine what package I should report against, all for something I don't even use? I'm not a big fan of hoop jumping.

            In the end I gave up. I had two problems, both related to btrfs which I have been using on Ubuntu for over a year and is the filesystem I use on all my computers. The first is that the arch kernel refuses to mount a freshly created filesystem made with mkfs.btrfs. I worked around that by creating it as ext4 and then doing a btrfs-convert which worked fine. The second was that the final system was unbootable looking like some combination of the partitioning, grub, btrfs and mkinitcpio don't work together.

            Comment


            • #81
              Originally posted by Hamish Wilson View Post
              First off, Wayland is not even developed by Red Hat, so how is this going to break this "Red Hat monopoly"?

              Second, there is some serious skewing of the facts based on what you are implying. Red Hat does not own any of these projects, has no copyright assignment on them, and does not even lead or develop on some of them (see Wayland where they do not do any development). Red Hat does indeed make a lot of money off of free software (are you going to say this is wrong when you are defending Canonical's fumbling attempts to get a profit?) but to say they "took advantage" of the software is almost a laughable statement, considering how much developer resources they have given back, many of which do not even benefit them directly but instead feed the entire upstream ecosystem.
              I consider the use of software to be taking advantage of it and I consider that to be a good thing. Your claims to the contrary appear to be based on a completely different understanding of English vocabulary.

              With that said, the idea that Wayland is completely divorced from the RedHat monoculture is absurd when I consider recent comments that have been made by RedHat employees, including those that contribute to Xorg development. If these people had no involvement with Wayland, they would not care enough to make these comments.

              Originally posted by Ramiliez View Post
              Why do you prefer Canonical monoculture to Red Hat one?

              Which is better Wayland controlled by freedesktop.org or Mir controlled by single corporation with CLA
              I have yet to hear either party explain why I would want to use their display server over Xorg. Currently, Xorg does everything that I want.

              With that said, I would like a world where people can start open source projects in conflict with existing projects without any harassment for it. The response that the mere act of doing something different provokes in the open source community is childish. If you do not like what these guys are writing, either use something that you do like or write your own software.
              Last edited by ryao; 03-05-2013, 06:48 PM.

              Comment


              • #82
                Originally posted by ryao View Post
                I have yet to hear either party explain why I would want to use their display server over Xorg. Currently, Xorg does everything that I want.
                There was a talk on the whole topic at LCA recently - see the video at http://mirror.linux.org.au/linux.con...land_and_X.ogv

                Comment


                • #83
                  Originally posted by ryao View Post
                  With that said, I would like a world where people can start open source projects in conflict with existing projects without any harassment for it. The response that the mere act of doing something different provokes in the open source community is childish. If you do not like what these guys are writing, either use something that you do like or write your own software.
                  Ryao, what happened to eudev?

                  Comment


                  • #84
                    Originally posted by pingufunkybeat View Post
                    Ryao, what happened to eudev?
                    This has nothing to do with this thread.

                    Comment


                    • #85
                      Originally posted by ryao View Post
                      This has nothing to do with this thread.
                      ....very true - but i also thought the exact same thing reading your comment that was quoted. ~ it seemed like your comment was driven by your own recent experiences and to _some_ degree they do relate, in terms of fragmentation and competing (similar) implementations of critical linux infrastructure/components;

                      eudev = Mir
                      systemd = Wayland

                      (obviously, not literally but in terms of having some parallels(?) -> for sure).

                      Comment


                      • #86
                        Originally posted by ninez View Post
                        ....very true - but i also thought the exact same thing reading your comment that was quoted. ~ it seemed like your comment was driven by your own recent experiences and to _some_ degree they do relate, in terms of fragmentation and competing (similar) implementations of critical linux infrastructure/components;

                        eudev = Mir
                        systemd = Wayland

                        (obviously, not literally but in terms of having some parallels(?) -> for sure).
                        I disagree with how you equate eudev with mir.... First of all eudev is a requirement if you want to -not- use systemd..... There is no corrolation.

                        EDIT: I would even go further and say that there needs to be a complete udev replacement... It really wasnt that long ago when I was using a static dev that was far easier to use and maintain than what udev allows. Udev by its fundamental nature is broken POS that does -NOT- work and is is -NOT- adequate.
                        Last edited by duby229; 03-05-2013, 09:04 PM.

                        Comment


                        • #87
                          Originally posted by duby229 View Post
                          I disagree with how you equate eudev with mir.... First of all eudev is a requirement if you want to -not- use systemd..... There is no corrolation.
                          from that perspective sure, but that's not what i was saying at all. I was more talking about the fragmentation and drama involved and the direction i see Mir going (the way of the dodo bird).

                          EDIT: I would even go further and say that there needs to be a complete udev replacement... It really wasnt that long ago when I was using a static dev that was far easier to use and maintain than what udev allows. Udev by its fundamental nature is broken POS that does -NOT- work and is is -NOT- adequate.
                          again, your taking it from a different angle.

                          myself, i am quite happy using Systemd, it works great. it's easy to configure and better than anything i could use, imo. but to each their own.

                          Comment


                          • #88
                            Originally posted by ryao View Post
                            I consider the use of software to be taking advantage of it and I consider that to be a good thing. Your claims to the contrary appear to be based on a completely different understanding of English vocabulary.
                            While I do fully agree that English can be a bloody obtuse son of a bitch, I am pretty sure my understanding to the term is the more general one.

                            Originally posted by ryao View Post
                            With that said, the idea that Wayland is completely divorced from the RedHat monoculture is absurd when I consider recent comments that have been made by RedHat employees, including those that contribute to Xorg development. If these people had no involvement with Wayland, they would not care enough to make these comments.
                            Because they are invested in Linux by chance? Just like you and I are? Perhaps even more so because of Red Hat's stronger dependence on a strong upstream Linux community to continue to sell it's products. This is like saying that I as a Canadian had no reason to follow the last results of the US election because they do not directly involve me.

                            It was daniel's who commented that Red Hat had not direct involvement in Wayland though, so I am taking that from him. He should know though.

                            Comment


                            • #89
                              Originally posted by ryao View Post
                              With that said, the idea that Wayland is completely divorced from the RedHat monoculture is absurd when I consider recent comments that have been made by RedHat employees, including those that contribute to Xorg development. If these people had no involvement with Wayland, they would not care enough to make these comments.
                              Of course they care. Anyone who uses Linux has a reason to care. People aren't allowed to care about things they don't directly develop?

                              Comment


                              • #90
                                Originally posted by Hamish Wilson View Post
                                It was daniel's who commented that Red Hat had not direct involvement in Wayland though, so I am taking that from him. He should know though.
                                He would be wrong:

                                Kristian Høgsberg is working at Red Hat on the Linux graphics stack including drm, mesa, X, cairo and more. The recent couple of years Kristian has been focused on clearing out the roadblocks that prevent us from enabling a composited Linux desktop by default. Towards this goal he has been instrumental in implementing AIGLX, which has allowed compiz and other compositing managers to run on X, and DRI2, which integrates accelerated OpenGL with the COMPOSITE extension. Recent developments in the Linux graphics stack has led Kristian to wonder whether X is our ideal window system.
                                http://linuxplumbersconf.org/ocw/users/73

                                I know that Kristian is no longer at Red Hat, but his Wayland proposal for the plumber's conference was published when he was at Red Hat.

                                Comment

                                Working...
                                X