#X11 #Xorg is almost 40 years old. #wayland almost 15. And I still need to read thought-pieces that argue against adopting the latter with the same gate-keeping arguments.
The amount of effort put into whining together with the lack of effort put into maintaining Xorg (or any relevant software for that matter) is astounding and agonizingly agitating.
"If you are not part of the solution, you are part of the problem. Quit being a part of the problem!" - John McClane (slightly paraphrased)
I believe that thought process and quote doesn't really apply, because as an end user, I can't really do much other than complain when the major forces behind Wayland have adopted a philosophy that amounts to "I don't care if we don't support your obscure functionality that previously worked fine" and are extremely hostile to any kind of conversation about it. I've previously seen the "why would you want to do that?" argument being used, which is extremely frustrating. Just because a developer doesn't understand a use case doesn't mean other people don't want it, especially when it was previously supported. I can understand not wanting feature creep, but when you are replacing something, you need to be a superset of it, not a subset.
And if I don't whine, then I'll do what exactly? Fork it? GNOME and Plasma will still not support my extensions because they are "insecure" and the people behind it disagree with it. "Oh no there's no longer client isolation!"
I also noticed that advanced users who are fine with Wayland use sway or some other wlroots based compositor, because wlroots did the sensible thing and implemented a whole bunch of the "obscure" or "insecure" functionality.
But in the end, it doesn't matter because I don't like tiling WMs and people (for example, I don't know if it still applies, I haven't researched it) want ckb-next to detect the active window on GNOME Wayland because their distro shipped with GNOME and they got used to it.
@KitsuWhooa @fleischie28 I understand your frustration and even share some of it, but please remember that having a good answer to the question "why would you want to do that?" is the necessary first step to design a good protocol that solves actual use cases and ensures interoperability. The XY problem is real, and if "I want it to work the way X11 did" was the way to go there would be no need for Wayland at all.
@KitsuWhooa @fleischie28 TBH, I don't think this falls into Wayland's scope at all. This is something for a window management interface. X11 left it wide open for everyone as every client could do everything; now we're in a better world, but we kinda forgot about having an interoperable standard for that. There are extensions that do basic window management over Wayland, but their scope was so far mostly about things like taskbars.
Also, OMG this Thunderbird extension is such a bad hack 😂
Yeah, it makes sense for such functionality to not be in `Wayland` itself. But in my opinion there should have equally been a `Wayland-EWMH` type of thing that compositors implemented and was part of the standard one way or another. (I'm basically paraphrasing what you said, I know :p)
I believe most security concerns can be solved by sane defaults and toggles. Last I checked (Mojave) macOS let userspace applications capture keystrokes, but you had to explicitly give it permissions. I am fully aware of "consent fatigue" and its dangers, too.
> Also, OMG this Thunderbird extension is such a bad hack 😂
WebExtensions are quite limited because of similar rationales, and so extreme solutions are required. In the past all you had to do was install a TB extension and that was it.
But you have to do what you have to do when you don't want yet another icon taking up space in your taskbar/launcher.
Yeah, I definitely understand that as well.
> if "I want it to work the way X11 did" was the way to go there would be no need for Wayland at all.
It is not necessarily binary like that. Wayland can get rid of many assumptions of how graphics hardware works that X makes and are no longer relevant. It can have superior multiple monitor support and absolutely no tearing with no fussing about with different refresh rate monitors. But it should also let me make a program that hides Thunderbird because it's 2024 and it still doesn't support running in the background on Linux. [0]
There's no reason for client isolation if the user simply doesn't want it. I personally don't want it. I don't want control of my computer taken away from me because it can be abused by a theoretical threat. I think control being taken away is why people are upset.
[0] https://github.com/Ximi1970/systray-x