I think the “die” part is a bit of an overexaggeration, it’s true that they need to implement protocols if they want the features those protocols exist to provide, but the X way seems to have caused more problems in the long run. Niche projects will attract people who want that niche, hopefully despite missing features, and some people who can help implement those features.
It’s worth noting that DEs can do things like use XDG portal implementations from another DE, I think it might’ve been Niri that does that.
if x11 had HDR I’d switch back to my i3 setup.
Then I can wish you that you get a worthy compositor/DE/whatever to replace i3 on Wayland. Either that or i3 somehow getting a Wayland implementation. I’m lucky in that I was comfortable on plasma with X, so switching to Wayland was mostly painless, though there were significantly more issues back then.
It is an exaggeration in that Wayland isn’t outright killing x, but a lot of the talk surrounding both is that it’s supposed to.
I’m not going to defend the state of X and the problems it’s caused because it’s a fugly old project, just when you look at something like pulse -> pipe wire, there wasn’t a big paradigm shift with it. It did the same job and better (to my knowledge!).
I didn’t know that DEs could use others portal implementations, that seems neat?
Yeah it’s the specifics of that that’s the issue for me though. I tried hyprland and many things are really cool about it but lots of things just break, and maintaining separate configs for what feels like core features was annoying. Then you get things like HDR performing differently between it and plasma, something that I’d hoped Wayland would just own. I’ll probably try niri and sway at some point, they both look cool
It is an exaggeration in that Wayland isn’t outright killing x, but a lot of the talk surrounding both is that it’s supposed to.
I interpreted that killing to refer to niche DEs rather than X, and while that isn’t happening yet, yeah, I imagine a future where software stops supporting X entirely.
maintaining separate configs for what feels like core features was annoying
That part specifically sounds like the fault of the specific DE? Wayland might be split into protocols, and some functionality might be handled by the portal, but ultimately it’s the compositor handling how it’s all implemented, and the configuration could well all be in one place.
Yeah, I guess I see the two as linked. I don’t think i3 is going to reimplement itself to stay available, nor would I expect them to
Also yes I phrased myself very badly there: it’s a decision hyprland has made, where there are separate first party packages for core functionality that need their own configs. I’m sure it’s to allow for your own drop in replacements it just felt clunky to me.
I don’t really have an issue with Wayland’s actual composition beyond I get lower input latency on X (which seems to put me in the minority), but they are both substantially lower and nicer to use than windows so I don’t care
I think the “die” part is a bit of an overexaggeration, it’s true that they need to implement protocols if they want the features those protocols exist to provide, but the X way seems to have caused more problems in the long run. Niche projects will attract people who want that niche, hopefully despite missing features, and some people who can help implement those features.
It’s worth noting that DEs can do things like use XDG portal implementations from another DE, I think it might’ve been Niri that does that.
Then I can wish you that you get a worthy compositor/DE/whatever to replace i3 on Wayland. Either that or i3 somehow getting a Wayland implementation. I’m lucky in that I was comfortable on plasma with X, so switching to Wayland was mostly painless, though there were significantly more issues back then.
It is an exaggeration in that Wayland isn’t outright killing x, but a lot of the talk surrounding both is that it’s supposed to. I’m not going to defend the state of X and the problems it’s caused because it’s a fugly old project, just when you look at something like pulse -> pipe wire, there wasn’t a big paradigm shift with it. It did the same job and better (to my knowledge!).
I didn’t know that DEs could use others portal implementations, that seems neat?
Yeah it’s the specifics of that that’s the issue for me though. I tried hyprland and many things are really cool about it but lots of things just break, and maintaining separate configs for what feels like core features was annoying. Then you get things like HDR performing differently between it and plasma, something that I’d hoped Wayland would just own. I’ll probably try niri and sway at some point, they both look cool
I interpreted that killing to refer to niche DEs rather than X, and while that isn’t happening yet, yeah, I imagine a future where software stops supporting X entirely.
That part specifically sounds like the fault of the specific DE? Wayland might be split into protocols, and some functionality might be handled by the portal, but ultimately it’s the compositor handling how it’s all implemented, and the configuration could well all be in one place.
Yeah, I guess I see the two as linked. I don’t think i3 is going to reimplement itself to stay available, nor would I expect them to
Also yes I phrased myself very badly there: it’s a decision hyprland has made, where there are separate first party packages for core functionality that need their own configs. I’m sure it’s to allow for your own drop in replacements it just felt clunky to me. I don’t really have an issue with Wayland’s actual composition beyond I get lower input latency on X (which seems to put me in the minority), but they are both substantially lower and nicer to use than windows so I don’t care