Unfortunately it's not currently available for Ubuntu 24.04 LTS, Debian 12, or any downstream distro based on them; as some key packages in their repository are too old and Xlibre currently will not compile on them. I don't really know the details on if a way around that is possible. Shame as I like Mint.

I am still not sure about Xlibre. The HISTORY.md reads like a salty rant.

Based on his PR history over on Freedesktop it looks like Enrico kept breaking stuff, and he is just mad they kept declining his poorly made and untested PRs.

Freedesktop has declined four years of bug fixea and patches, including tearfree contributed from a multitude of contributions. The current people in charge of Xorg are also the ones developing Wayland, and have explicitly said in social media that they are actively trying to kill Xorg (because everyone who uses Windows knows that the best way to make people use your new product is to not make it better but to drop support for the old product so that users have no choice). Anyone who has tried to get the best performance while gaming on Wayland will have noticed that there has been politics that put the quality of the code on the backburner, hence why Xlibre has taken a stance against politics.

Even for years Metux has been the biggest contributer to Xorg, and when he broke things he always immediately fixed it. When you're rewriting a 20+ year old bit of spaghetti code built for a system that the Linux kernel dropped support for years ago, there's going to be some hiccups. He has fixed those hiccups every time. Since forking X11 it has gotten more stable and slick, and the people testing it have reported things like having better battery life and smoother experience on the Steam Deck and such.

Either way, the point of this is to provide the option for anyone to try out Xlibre and decide for themselves if they see any improvement or find it sufficiently stable.

not interested, thanks anyways 🙂...

Originally posted by Omega:

I am still not sure about Xlibre. The HISTORY.md reads like a salty rant.

Based on his PR history over on Freedesktop it looks like Enrico kept breaking stuff, and he is just mad they kept declining his poorly made and untested PRs.

This is a circumstance they want you to know, but the story was actually quite different.

The creator of xlibre had been working on xserver for a long time, adding commits that fixing bugs and introduced new features. This was the master/main branch in Git. After some time, he began pushing for a new release of xserver, which he had been working on for a long time and to which he had devoted a tremendous amount of work. To his surprise, freedesktop announced that the master/main branch would never be released, and only small point releases would be added to fix security bugs. So all his work was for naught, and no one informed him earlier that the master would never see a release. He then announced the creation of a fork called xlibre, which was a collection of all the previously unreleased commits from the master/main branch of xserver. However, when announcing their fork, he did it (in my opinion) unwisely. Instead of focusing on purely technical issues (they want to save xserver development because Freedesktop announced there would be no more major releases), they added ideological elements – which some may agree with, some may not – hence the controversy.
After announcing the fork, Freedesktop banned the author of xlibre from their own repository and then began reverting the code changes he made, explaining that they were bad and that the author was inexperienced (even though he personally worked at a company that used xserver and was responsible for its internal maintenance and modifications).

You already know the story; which side you trust is up to you. But if you're going to argue one side, it's worth understanding the broader context – including from the other person.

I'm writing this as a developer of one Linux distribution.

Originally posted by grahamf:

Freedesktop has declined four years of bug fixea and patches,

That is the thing, many of his bug fixes and patches just seem to break stuff. Which is a valid reason to decline them. I see examples where automated tests were failing, a PR should not have been opened in the first place if tests were failing, which indicates he did not test his changes.

I am fairly certain I encountered the dev once on Reddit, he was not very pleasant to interact with. He was zealously defending X, he seemed to utterly misunderstand it despite being one of the only people who has been spending any real time on it in the last couple of years. And he does not tolerate any criticisms against X.

I put to doubt his mental health.

I've been running Xlibre for a couple days now and I haven't seen anything broken. And it's not possible for one person to test every singlr system configuration, which is why most projects have a testing and stable branch as that's literally what the testing branch is for.

Xlibre is now available on ubuntu 24.04 LTS and Debian 12, meaning it's available on Linux Mint and other Ubuntu based derivatives

Lunatics complaining about wokeness and claiming any criticism is part of an organized smear campaign. I'm sure this will go far. /sar

The ownership problems with the AUR version have been resolved
https://aur.archlinux.org/packages/xlibre-xserver

https://aur.archlinux.org/packages
Xlibre is now the 8th most popular package on AUR

Honestly, I think it's too early to say how XLibre will turn out. I hope for the best because even though the world is moving to Wayland, legacy stuff still exists and I don't want to lose compatibility.

Of course, if XLibre is going to have a chance then people need to adopt it. I won't, I don't mind Wayland, but if you're curious then feel free to do so.

it just screams "reactionary fork" especially when people are claiming that it's already a de facto replacement for four-decade old X11 even though it's only been two months. only time will tell though. iirc a good deal of xorg developers gave up on maintaining X11 in favor of developing Wayland, and the latter is perfectly usable for me nowadays, more so than the former

I the strictest sense it's two months old, but it's purpose is to merge 4 years of patches into a 20+ year old stable project. It's already more stable then wayland, with everything being done having years of groundwork in place.

Its not like metux is deleting things willy nilly and rewriting it in rust, everything is something that had been carefully considered and worked on for a long time only to be rejected by freedesktop because they want to kill X11 in favor of Wayland.

Not unlike Microsoft ending support for Windows 10 in a couple months because they want everyone to move to Windows 11, even if users have significant concerns about Windows 11.

Originally posted by Trout:

it just screams "reactionary fork" especially when people are claiming that it's already a de facto replacement for four-decade old X11 even though it's only been two months. only time will tell though. iirc a good deal of xorg developers gave up on maintaining X11 in favor of developing Wayland, and the latter is perfectly usable for me nowadays, more so than the former

If you don't use it think again most games are using xwayland and that still uses a ton of patches from our project that we're merged to improve performance.

This sounds promising, would like to try it myself. Currently running Mint 21.3 but will have to backup one day soon and upgrade to Mint 22 line when also upgrade my Dual Boot of Win 10 to 11. Have personally noticed many glitches with the Wayland experimental option on Mint especially with VNC or RDP and even when learning how to use KiCad does not support Wayland, ontop of thatmy odd gaming performance is better on X11 session vs Wayland.