๐ [Rant/Discussion] What's with all the excitement around Wayland?
๐ก Newskategorie: Linux Tipps
๐ Quelle: reddit.com
The main arguments I've heard concerning how Wayland is better than Xorg are:
- It has a cleaner/smaller codebase since Xorg is so old
- The X11 specification is too large, too many features that aren't being used often
- It's more secure and prevents keylogging
I want to address each one separately.
To start off, the first 2 arguments are basically the same thing. Xorg has more features and as a result it has more code, whereas Wayland gives applications a pixmap to draw whatever they want however they want, and thus saves the code that would be needed to draw lines and arcs and rectangles and such that Xorg offers. This isn't a feature. The point of making a smaller codebase is so that it's easier to add more features, but Wayland has no intention of doing that and so it becomes almost useless. The argument that X11's specification is too large is also not a good reason, since we've been using the same implementation of it (Xorg) for nearly 20 years.
The third argument is more reasonable, but Wayland's approach of blocking apps from accessing inputs completely isn't a good approach either. In my opinion, it's actually worse. The way I think of it is: If the only two options were to let everyone board a plane whenever they wanted or to ban everyone from using planes, which one would be better? (I know this isn't an obvious question. What I'm saying is, if you would answer "allow everyone" then you would agree that Xorg's approach is better, but if you would answer "full ban" then you would argue that Waylands approach is better). Even if Wayland allowed users to choose which apps could/couldn't access the inputs of other apps, it wouldn't be worth completely replacing all of Xorg. There are some security addons available for Xorg that distributions/desktop environments can just install by default.
โ
I know that all I've said so far doesn't really give a reason for not using Wayland, so I'll cut to the chase and share why I think X is better. If you read the note beside this wikipedia link, you'll see that before X was created, there was a windowing system called "W". X was supposed to be a faster alternative to W, and it got it's name because "X" was the letter that came after W. The way W worked was similar to wayland: applications would write to a pixmap, send it to the windowing system with a specified size and position, and have that data passed on to a window manager which could draw it onto the screen. X worked by writing this data directly to the screen and allowing window managers to draw around it (or over it if they liked). For window managers that actually change how windows look (like compiz), X wasn't any faster than W: X would write to the screen buffer and the WM could get that data and write that data to the screen, where W would write data to a random memory location and the WM could get that data and draw it on the screen. It was pretty much the same process. On the other hand, window managers that only draw a titlebar and borders would gain a huge speed boost by not having the computer draw a window contents twice.
The way I see it, Wayland is just a remake of W, but with "ayland" added to it, and the word "modern" stamped all over the place. People on the internet are saying that X is too old and that Wayland is the new "modern" replacement, but they never give a reason other than the smaller codebase and less features (which is somehow supposed to be a good thing). They rarely even bring up the security benifits.
Am I missing something or was everyone else missing something?
EDIT: The reason I tagged this as "discussion" and concluded with "am I missing something" is because I really don't know if there are any other arguments to using Wayland. Rather than disliking this and leaving you could explain your point of view to me because I genuinely want to know what it is.
[link] [comments] ...