18 ms·

Thus your choices for who will maintain Xorg: * The people who made it who also want to murder it * The crazy guy, who knows enough to be useful, but dangerou

by amiga386 11mo ago

Thus your choices for who will maintain Xorg: * The people who made it who also want to murder it * The crazy guy, who knows enough to be useful, but dangerous * Yourself None of these choices are good, and the Xorg developers know that. They're super-stoked about how much compelled Wayland adoption they're going to get.

mort96 11mo ago
What I see in freedesktop is a loose association of people trying to make the best libre display server ecosystem they know how to make, and they've collectively come to the conclusion that they're able to make better software by starting from scratch than by patching away at the X11 code base from the 80s. And otherwise it's just a boring standardization committee for protocols. It sounds like you're seeing something much more sinister for some reason, and I don't understand why.
- amiga386 11mo ago
  I think, rather, they've collectively come to the conclusion that brownfield development sucks and greenfield development rules. Which is true, if you're a voluntary developer. It's so much fun to play in new developments, and be involved in making history (that some other sucker will maintain in future) rather than maintaining someone else's past glories. Unfortunately, this means for third-party developers and users that the ecosystem they've built on top of is going to get thrown away, because the first-party developers have prioritised their own fun, over the utility and effectiveness of what they built. It did not have to be like this. The early freedesktop initiative managed to massively improve XFree86/Xorg without fundamentally throwing out X11. They threw out X11 so they do stuff like this: argue with developers that wanting to know the "primary" screen so they can open on it, is fundamentally, non-negotiably, not allowed, and fuck you for even asking for it (BTW it's standard on Windows, macOS and X11): https://gitlab.freedesktop.org/wayland/wayland-protocols/-/issues/179#note_2965661 https://gitlab.freedesktop.org/wayland/wayland-protocols/-/i... This is what they live for. This is why X11 has to die. GNOME has form for this. In describing the "Cascade of Attention-Deficit Teenagers" (CADT) development model in 2003, JWZ says: > I report bugs; they go unread for a year, sometimes two; and then (surprise!) that module is rewritten from scratch -- and the new maintainer can't be bothered to check whether his new version has actually solved any of the known problems that existed in the previous version. > It hardly seems worth even having a bug system if the frequency of from-scratch rewrites always outstrips the pace of bug fixing. Why not be honest and resign yourself to the fact that version 0.8 is followed by version 0.8, which is then followed by version 0.8? > But that's what happens when there is no incentive for people to do the parts of programming that aren't fun. Fixing bugs isn't fun; going through the bug list isn't fun; but rewriting everything from scratch is fun (because "this time it will be done right", ha ha) and so that's what happens, over and over again. He managed to predict Wayland, or at least the causes of Wayland, back in 2003.
  - mort96 11mo ago
    It seems like you're mixing up your dislike for some of the technical decisions made in Wayland (some of which I share, FWIW) with some weird conspiracy theory about how Freedesktop's main mission is to argue with developers.
    - amiga386 11mo ago
      It's more a lament. It used to be that no matter how crazy GNOME got, all GNOME-specific apps still worked in vague harmony with KDE apps, Motif apps, etc., because the common substrate was X11, and backwards-compatibility was a key concern. Coupled with FDO trying to bring more harmony and standards (XDG), times were good. But FDO seems to have morphed into the replace-X11-with-Wayland club, or perhaps more cynically I could say Red Hat/IBM and Canonical, who are paying the Xorg / Wayland developers, want returns on their investments sooner rather than later. I despair at how unfit Wayland is to replace X11, but the big guns are going to force its adoption anyway. People who wrote X11 applications will be compelled to rewrite all their work too, or their users will have to use them in the locked down X11 emulator that can't do anything more than Wayland lets it do, which will make some X11 applications become fundamentally useless, e.g. xwd will not capture your screen. As an example of why I think this is bad, I'd highlight we're still seeing intense ideological fights on basic capabilities, today. We're rushing into replacement while even the basics aren't settled. IMO, people like Sebastian Wick don't feel the need to yield an inch to backwards compatibility, and Wayland to them is all about these greenfields that they get to control - by fighting on bug reports, or refusing to implement protocols. This is what we're all about to be dropped into.
      - mort96 11mo ago
        > or perhaps more cynically I could say Red Hat/IBM and Canonical, who are paying the Xorg / Wayland developers, want returns on their investments sooner rather than later. Here's the weird vague conspiracy posting again. What exactly do you think Red Hat, IBM and Canonical's desired "return on investment" is with regard to Wayland, other than just a more polished desktop that works better for most people?

mort96 11mo ago

What I see in freedesktop is a loose association of people trying to make the best libre display server ecosystem they know how to make, and they've collectively come to the conclusion that they're able to make better software by starting from scratch than by patching away at the X11 code base from the 80s. And otherwise it's just a boring standardization committee for protocols. It sounds like you're seeing something much more sinister for some reason, and I don't understand why.

amiga386 11mo ago
I think, rather, they've collectively come to the conclusion that brownfield development sucks and greenfield development rules. Which is true, if you're a voluntary developer. It's so much fun to play in new developments, and be involved in making history (that some other sucker will maintain in future) rather than maintaining someone else's past glories. Unfortunately, this means for third-party developers and users that the ecosystem they've built on top of is going to get thrown away, because the first-party developers have prioritised their own fun, over the utility and effectiveness of what they built. It did not have to be like this. The early freedesktop initiative managed to massively improve XFree86/Xorg without fundamentally throwing out X11. They threw out X11 so they do stuff like this: argue with developers that wanting to know the "primary" screen so they can open on it, is fundamentally, non-negotiably, not allowed, and fuck you for even asking for it (BTW it's standard on Windows, macOS and X11): https://gitlab.freedesktop.org/wayland/wayland-protocols/-/issues/179#note_2965661 https://gitlab.freedesktop.org/wayland/wayland-protocols/-/i... This is what they live for. This is why X11 has to die. GNOME has form for this. In describing the "Cascade of Attention-Deficit Teenagers" (CADT) development model in 2003, JWZ says: > I report bugs; they go unread for a year, sometimes two; and then (surprise!) that module is rewritten from scratch -- and the new maintainer can't be bothered to check whether his new version has actually solved any of the known problems that existed in the previous version. > It hardly seems worth even having a bug system if the frequency of from-scratch rewrites always outstrips the pace of bug fixing. Why not be honest and resign yourself to the fact that version 0.8 is followed by version 0.8, which is then followed by version 0.8? > But that's what happens when there is no incentive for people to do the parts of programming that aren't fun. Fixing bugs isn't fun; going through the bug list isn't fun; but rewriting everything from scratch is fun (because "this time it will be done right", ha ha) and so that's what happens, over and over again. He managed to predict Wayland, or at least the causes of Wayland, back in 2003.
- mort96 11mo ago
  It seems like you're mixing up your dislike for some of the technical decisions made in Wayland (some of which I share, FWIW) with some weird conspiracy theory about how Freedesktop's main mission is to argue with developers.
  - amiga386 11mo ago
    It's more a lament. It used to be that no matter how crazy GNOME got, all GNOME-specific apps still worked in vague harmony with KDE apps, Motif apps, etc., because the common substrate was X11, and backwards-compatibility was a key concern. Coupled with FDO trying to bring more harmony and standards (XDG), times were good. But FDO seems to have morphed into the replace-X11-with-Wayland club, or perhaps more cynically I could say Red Hat/IBM and Canonical, who are paying the Xorg / Wayland developers, want returns on their investments sooner rather than later. I despair at how unfit Wayland is to replace X11, but the big guns are going to force its adoption anyway. People who wrote X11 applications will be compelled to rewrite all their work too, or their users will have to use them in the locked down X11 emulator that can't do anything more than Wayland lets it do, which will make some X11 applications become fundamentally useless, e.g. xwd will not capture your screen. As an example of why I think this is bad, I'd highlight we're still seeing intense ideological fights on basic capabilities, today. We're rushing into replacement while even the basics aren't settled. IMO, people like Sebastian Wick don't feel the need to yield an inch to backwards compatibility, and Wayland to them is all about these greenfields that they get to control - by fighting on bug reports, or refusing to implement protocols. This is what we're all about to be dropped into.
    - mort96 11mo ago
      > or perhaps more cynically I could say Red Hat/IBM and Canonical, who are paying the Xorg / Wayland developers, want returns on their investments sooner rather than later. Here's the weird vague conspiracy posting again. What exactly do you think Red Hat, IBM and Canonical's desired "return on investment" is with regard to Wayland, other than just a more polished desktop that works better for most people?