The Linux Cast

Daily Linux Content on YouTube and Odysee

I Wasn't Wrong About Wayland

So in recent videos (as of writing), I've been talking about how I have been trying out Hyprland, which is a Wayland compositor. It might be a bit confusing, given that a few weeks ago I made a video about how I was going to protest Wayland until the end of time.

I still don't like Wayland. But I need to at least try it; otherwise, I'm one of those guys who doesn't like something for no reason. I really don't want to be the grumpy old man who hates change so much just because it's change.

So I have been forcing myself to use both Hyprland and the Wayland version of Plasma off and on for the last week (as of writing this). And I have to say two things.

First, it's better than it was. A year ago, I couldn't even think of making a video in a Wayland compositor and actually getting it to record properly. Now I can. I recorded my last NixOS video entirely in Wayland.

So it's better than it was.

But I was also right that it isn't really ready for me. There are still hoops that you have to jump through in order to use it, which just makes it not good/not ready. The biggest is the lack of Wayland support in OBS. Yes, you can get it working, but you have to be using the right version of OBS to get it working. And that version is not the official flatpak version that is maintained by the OBS project. Instead, you have to use the version from your distro's repositories. It's not a huge deal, but it also means you lose YouTube integration because the Flatpak is the only version that has YouTube integration built in.

Add on top of that, there is still the dumbass requirement of having to add in an environment variable in order to get screen capture to work properly. That's very user-friendly.

Sarcasm

Maybe the situation is different in KDE and Gnome. But I'm a Window Manager user, and that's the way it is right now. How can something like this be ready if it doesn't work out of the box for a lot of people? It makes no sense.

But I am trying it. I want to be able to say, “It's not ready,” and then say, “Of course, I've tried it,” when someone says, “You have to try it.”

The thing that I've always had a problem with is that if this change is happening, there has to be a user-facing benefit that makes it worth the pain that comes along with it. And I don't really see the benefits. It sounds great for developers. And yes, I guess a more secure system is a user-facing benefit, but it's not one many people spend that much time thinking about.

What do I get by using Wayland?

The answer, as far as I can tell, is no screen tearing. Which, with a little --vsync, I've had for years on Xorg. Maybe there's more there, but I just don't see it and even Josh couldn't tell me much beyond that.

I think my biggest hangup, right now, is the lack of good window managers. Hyprland is fine, but it's not stable enough for everyday use. And it's too frilly. I like the animations. For like 5 seconds. And then I have to turn them off because they just get in the way. And while it's not bad and is usable, I have been spoiled rotten by being able to choose from dozens of Xorg-based window managers for years. I can go to i3 and have manual tiling, I can do bspwm and have a bit of both, or I can go to my favorite and use Qtile, which remains awesome. I can do AwesomeWM or Xmonad (if I feel masochistic), or I can do DWM or Herbstluftwrm. You get the point.

And yes, there's Sway for the i3 lover in me and DWL for DWM (which is buggy AF, btw), and I think River is a bspwm-type compositor, but I'm unsure as I've never tried it. There are Wayland alternatives out there, but they're all either abandoned, in early development, or just bad. I get it, it will take time, but supposedly Wayland is here and XOrg is going to kill any sense of computer security I had, so I need to switch, right?

I'm sure with time, there will be a lot of compositors out there that will take the place of the plethora of options out there for XOrg. But until then, I'm not going to be happy with Wayland. The options right now are just not great. Qtile may be my favorite, and it does have a Wayland version, but it is about quarter baked at the moment. I've not met one person who has been able to use it daily. I hope that it gets there, but I have my doubts. AwesomeWM has come out and blatantly said they won't be doing Wayland. We know the DWM guys probably never will. Xmonad? I don't know, but I doubt that too. So we'll either be left with a bunch of forks, or we'll find ourselves with a bunch of new projects that we just have to hope we can trust to stick around for more than a few months.

As you can tell, I'm conflicted. The nerd in me likes the idea of something new to play with, but I'm also very cynical by nature. So I have no doubt that we Window Manager users will all be burned many times in the coming years by abandoned projects that we fall in love with.

It's another reason why I am so cautious about dedicating myself to Hyprland or something else. I have no idea if they will be around in a year. Yes, that's the way things go, but it also means a lot of disruption if I do get invested in it and it goes away.

Wayland is one of those things I'm going to have to keep coming back to, but I still think it's just not ready for me.

Or, to put it more accurately, I'm not ready for Wayland.

I hope everyone has a nice week. I know I've been crap at posting both blogs and patron pods lately. I will rectify that. You'll get this post today and a pod tomorrow.

Matt