4 ms·

> Wayland makes things much less customizable and much more brittle than X11 Another way to put it is Wayland with a few years of development by a few Linux pl

> Wayland makes things much less customizable and much more brittle than X11 Another way to put it is Wayland with a few years of development by a few Linux players (others were building alternatives, while yet others were seeing which one would win out, and yet others were hoping to stay on X11 forever) under its belt, is almost as solid as X11 with many decades of development by all the Linux players. The idea that X11 was a completely working option which had every capability it does today within years of its inception is wrong. X11 has been built and expanded for decades and even today is incapable of doing many modern functions that its competitors can. Wayland, OTOH, has almost reached parity with X11 rapidly, does things that X11 is not even capable of, and has a roadmap to reach complete parity.

yjftsjthsd-h 3y ago
That's not "another way to put it", that's a completely different claim. Now if you'd like to make the claim that Wayland is actually developing at a great pace because the first 10 years didn't really count, that's plausibly defensible in some ways, though I'm not sure how we could objectively tell[0]. > The idea that X11 was a completely working option which had every capability it does today within years of its inception is wrong. It's also not something anyone is arguing, and something that's not really meaningful because we have Xorg already. At best, you're arguing that Wayland actually has a chance to catch up even though it hasn't in the first 15 years. > Wayland, OTOH, has almost reached parity with X11 rapidly, Debatable but possible, > does things that X11 is not even capable of, Capable of without more extensions, but who's counting? > and has a roadmap to reach complete parity. Yeah, no. Wayland has roadmaps to someday hope to implement a lot of features, but explicitly is designed to not cover others and AFAIK has nobody working on others. There is not, and probably never will be, a wayland protocol to separate the window manager from the display server. There could be but AFAIK isn't any effort to let programs manipulate windows in a portable way (read: there will never be accessibility tools that work on all, or even most, compositors). Wayland is aiming for ~90% feature parity with X, and may someday reach it. --- [0] I can think of some metrics you could use, but they're all hopelessly entangled in confounding variables. For example, we could graph number of X11 window managers released per year against Wayland compositors released per year, but even if you could pin down good numbers they're happening in completely different ecosystems and contexts. Likewise, we could make up a list of features and when they were added, but it seems a bit unreasonable to ex. hold it against X11 that nobody cared about HiDPI for the first ~30 years it existed.
josephcsible 3y ago
> Another way to put it is Wayland with a few years of development by a few Linux players (others were building alternatives, while yet others were seeing which one would win out, and yet others were hoping to stay on X11 forever) under its belt, is almost as solid as X11 with many decades of development by all the Linux players. If Wayland is going to be caught up and have feature parity so soon, then why can't decisions to deprecate and remove X11 be 100% put on hold until then? > The idea that X11 was a completely working option which had every capability it does today within years of its inception is wrong. Nobody thinks that. But at the time X11 was missing all of those important features, there wasn't another full-featured option that everyone was rushing to deprecate and remove support for to force everyone onto the immature and incomplete X11, at the expense of things that already worked.

yjftsjthsd-h 3y ago

That's not "another way to put it", that's a completely different claim. Now if you'd like to make the claim that Wayland is actually developing at a great pace because the first 10 years didn't really count, that's plausibly defensible in some ways, though I'm not sure how we could objectively tell[0]. > The idea that X11 was a completely working option which had every capability it does today within years of its inception is wrong. It's also not something anyone is arguing, and something that's not really meaningful because we have Xorg already. At best, you're arguing that Wayland actually has a chance to catch up even though it hasn't in the first 15 years. > Wayland, OTOH, has almost reached parity with X11 rapidly, Debatable but possible, > does things that X11 is not even capable of, Capable of without more extensions, but who's counting? > and has a roadmap to reach complete parity. Yeah, no. Wayland has roadmaps to someday hope to implement a lot of features, but explicitly is designed to not cover others and AFAIK has nobody working on others. There is not, and probably never will be, a wayland protocol to separate the window manager from the display server. There could be but AFAIK isn't any effort to let programs manipulate windows in a portable way (read: there will never be accessibility tools that work on all, or even most, compositors). Wayland is aiming for ~90% feature parity with X, and may someday reach it. --- [0] I can think of some metrics you could use, but they're all hopelessly entangled in confounding variables. For example, we could graph number of X11 window managers released per year against Wayland compositors released per year, but even if you could pin down good numbers they're happening in completely different ecosystems and contexts. Likewise, we could make up a list of features and when they were added, but it seems a bit unreasonable to ex. hold it against X11 that nobody cared about HiDPI for the first ~30 years it existed.

josephcsible 3y ago

> Another way to put it is Wayland with a few years of development by a few Linux players (others were building alternatives, while yet others were seeing which one would win out, and yet others were hoping to stay on X11 forever) under its belt, is almost as solid as X11 with many decades of development by all the Linux players. If Wayland is going to be caught up and have feature parity so soon, then why can't decisions to deprecate and remove X11 be 100% put on hold until then? > The idea that X11 was a completely working option which had every capability it does today within years of its inception is wrong. Nobody thinks that. But at the time X11 was missing all of those important features, there wasn't another full-featured option that everyone was rushing to deprecate and remove support for to force everyone onto the immature and incomplete X11, at the expense of things that already worked.