Hacker News Reader

Comment by marwis | Hacker News Reader

X11 is also just a protocol and you can implement it securely and expose access control through extensions. In fact it already exists (XACE) just nobody is using it.

View on HN

XACE is basically an extension of SELinux into userspace for X11. Given that today Fedora/RHEL are the only distros that enable SELinux out of the box, such an approach would have been doomed to failure (or doomed to provide a product differentiator for RHEL in the best case) - not to mention the sheer joy and excitement that debugging SELinux AVC denials would produce for end users and desktop application developers.

https://www.freedesktop.org/software/XDevConf/x-security-wal...

Exactly! It astonishes me that Wayland folks think that replacing the whole graphics stack is somehow easier than using an X extension that already exists. Really speaks to the profound immaturity of the project.

My god, those "wayland folks" were the same people developing X. They were not some unrelated people pushing an agenda. It's always fascinating to see all the arm chair experts saying things how things ought to be done without ever having written a X extension or anything of that sort.

So the fact that someone was a developer of X, means their resilient to following some agenda or non objective opinion? There's a really famous and often quoted talk about X vs Wayland from one of those developers and it was clearly made in a way to present Wayland in a better light by purposely portraying X11 in a bad way, for example by leaving out crucial details. Like in one example they picked on the slow startup time of applications and attributed it to the blocking calls made to X.org, without even mentioning that asynchronous alternatives are not only available, but toolkits like Qt were already using them exclusively at that time. The verdict of course was that X11 is broken and fixing that is almost impossible.

X11 was designed to everything like OS, it's display server, gui library, font manager, printing server and more, X11 was ment to be like qt or gtk, you tell it that you want to display input and it will draw it for you, the issue now it that basically every app now display every window as one bitmap, because we do not want to use 98% of the x11. Also why do you think there wasn't much development activity in X for years now? There is so much wrong with it's state that nobody wants to work with it. It's like talking: why someone would like to use rust or go or c# when we have cobol? why not just write cobol extension?

I'm not really sure what you're getting at. Like I never said there's no point in using or developing Wayland. I was addressing the statement, that Wayland was build by former X11 developers, which supposedly makes them purely objective authorities. Of course, that's not the case as, can be seen by the example I gave.

No, you're arguing a strawman. What I'm saying is, if you have never even tried writing an X extension, likely never properly looked at the X protocol (let alone it's implementation) you have no leg to stand on to make technical criticism of the people who did the work and then decided it was too much work and build something else.

And, have you written a Wayland extension and a X11 extension and learned the difference? No? Then why should we listen to you?

There isn't really much difference. It's about the same amount of work either way to ship an extension.

In Wayland, you make the extension, then you patch every compositor, every toolkit and every application.

In X11, you make the extension, then you patch the X server, Xlib, XCB, every window manager, every toolkit, and every application.

The question was related to the relative difficulty of replacing the whole graphics stack vs. patching in support for an X extension that has existed for longer than Wayland. One of these things is far, far less work than the other.

As someone who has done both, I respectfully disagree.

I call bullshit. If you had done so, you would have shared links to them to prove your point.

There's no code I can show you that would be any more convincing than just looking at the code to the X server and then looking at the code to any wayland implementation. There's not anything particularly special about these projects.

Reply on news.ycombinator.com