Xlibre forks to the rescue – but Kubuntu gives X11 the boot
- Reference: 1750771925
- News link: https://www.theregister.co.uk/2025/06/24/mixed_news_for_x11/
- Source link:
[1]Xlibre 25.0.0.0 is out , the "Summer Solstice" release of this new FOSS X11 server forked from the X.org server. This is a preliminary release – as project lead Enrico Weigelt says:
Since this is the first major release of the Xserver since years (with about 3k commits in between), there might be some yet unnoticed bugs. So this .0.0.0 release is considered beta.
We [2]reported news of the fork a couple of weeks ago. There was a lot of response across various FOSS forums and message boards, some of it very positive. Weigelt's positions on matters such as [3]COVID vaccines and [4]diversity, equality, and inclusion are, to put it diplomatically, controversial, and these are not the only things, as has been [5]discussed in the Fediverse .
As we [6]covered last week , some other projects [7]such as Devuan have [8]come out in support , and that in turn is polarizing more people. Some of the responses that this vulture has received online have been very negative indeed, including several ad hominem attacks. Then again, vituperative flame wars are nearly as old as software itself. (The words "software" and "bit" were coined in 1952 by the [9]late John Tukey , recently [10]celebrated in XKCD .) Some flame wars are infamous enough to have their [11]own Wikipedia articles .
We've seen many criticisms and attacks on Weigelt's coding skills, but as well as fixes for the multiple security issues, the announcement of Xlibre 25.0 contains three new features:
Xnamespace extension: a novel approach for isolating clients from different security domains (e.g. containers) into separate X11 namespaces, where they can't hurt each other (for cases where Xsecurity from 1996 isn't sufficient)
Xnest ported to [12]xcb – no more dependency on old [13]Xlib anymore
per-ABI driver directories (allows distros installing multiple ABIs at the same time, e.g. for smoother upgrades)
Many people have invested far too much emotion in matters as arcane and ultimately trivial as Unix display protocols. Now that some of his work is out there, those interested can try it for themselves.
Some other news in the exciting world of Unix display protocols is less reassuring for X11 fans. KDE developer Nate Graham's blog is a good source of what's new in that community, and his latest post is [14]About Plasma's X11 session . He says more than 70 percent of KDE Plasma users who permit telemetry are already using Wayland, and the numbers have been pulled down slightly by last month's release of [15]SteamOS 3.7 , which defaults to X.org. Graham says the KDE developers intend to continue to support X11 for now:
… this is up to distros, not us. It wouldn't make sense for us to get rid of Plasma's X11 support while there are still major distros shipping it by default…
Well, on that front, we have more news. A [16]recent comment from the Kubuntu team's [17]Rik Mills says:
on latest Questing daily ISO we no longer install the Plasma X11 session by default.
He continues:
… it is highly improbable that we can support the X11 session in 26.04 LTS (or now even), so like Ubuntu desktop it is probably better to rip off this sticking plaster in 25.10 and concentrate on wayland with less distraction.
[18]KDE Plasma 6.4 ships with major usability and Wayland improvements
[19]Ubuntu 25.10 and Fedora 43 to drop X11 in GNOME editions
[20]GNOME 48 beta is another nail in X11's coffin
[21]WINE 10 is still not an emulator, but Windows apps won't know the difference
The comment is on the [22]issue of migrating X11 support individually to the various Ubuntu desktop flavors. That change results from the removal of X11 from default GNOME edition in the forthcoming 25.10 release, which we [23]covered earlier this month .
Software is human ideas and human thought turned into computer code. Commercial software exists as a result of capitalism, and FOSS is a response to the commercialization of computers. The [24]GNU Project began in response to the founding of the Lisp machine companies LMI and [25]Symbolics .
[26]
People have different ideas. To some, it's clear that Wayland is the future. Graham's blog post says "it's just obvious that X11 is in the process of outliving its usefulness" and states as fact that "Wayland is better for modern hardware."
[27]
This is not obvious or a matter of fact to everyone in this space – any more than the idea that [28]encrypted bootable images unlocked by a TPM chip are everyone's idea of a better Linux boot process. Despite mainstream acceptance, systemd remains widely controversial. There have been [29]problems with snap packaging and many favor Flatpak, but [30]that has had issues too , and in recent years [31]Flatpak development has slowed .
The widespread integration of controversial new tech in the big-name commercially backed Linux distributions, and changes such as its developers making GNOME [32]more dependent on systemd and thus Linux, are good news for the smaller, niche distros and desktop environments, or even entire OSes. The more corporate Linux gets, the more FreeBSD and the other BSDs will appeal. ®
Get our [33]Tech Resources
[1] https://lists.x.org/archives/xorg-devel/2025-June/059400.html
[2] https://www.theregister.com/2025/06/10/xlibre_new_xorg_fork/
[3] https://www.theregister.com/2021/06/11/linus_torvalds_vaccine_smackdown/
[4] https://github.com/X11Libre/xserver/pull/126#issuecomment-2972280899
[5] https://circumstances.run/@mawhrin/114682302951770021
[6] https://www.theregister.com/2025/06/10/xlibre_new_xorg_fork/
[7] https://x.com/DevuanOrg/status/1935307141511794726
[8] https://x.com/DevuanOrg/status/1935307143646769547
[9] https://www.latimes.com/archives/la-xpm-2000-jul-29-me-61253-story.html
[10] https://xkcd.com/3104/
[11] https://en.wikipedia.org/wiki/Editor_war
[12] https://xcb.freedesktop.org/
[13] https://www.x.org/releases/current/doc/libX11/libX11/libX11.html
[14] https://pointieststick.com/2025/06/21/about-plasmas-x11-session/
[15] https://steamdeckhq.com/news/steamos-3-7-has-been-released/
[16] https://bugs.launchpad.net/ubuntu/+source/ubuntu-meta/+bug/2114015/comments/14
[17] https://launchpad.net/~rikmills
[18] https://www.theregister.com/2025/06/18/kde_plasma_64_released/
[19] https://www.theregister.com/2025/06/12/ubuntu_2510_to_drop_x11/
[20] https://www.theregister.com/2025/02/18/gnome_48_beta/
[21] https://www.theregister.com/2025/01/24/wine_turns_10/
[22] https://bugs.launchpad.net/ubuntu/+source/ubuntu-meta/+bug/2114015
[23] https://www.theregister.com/2025/06/12/ubuntu_2510_to_drop_x11/
[24] https://www.theregister.com/2023/03/31/mit_cadr_software_recovered/
[25] http://www.symbolics-dks.com/
[26] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aFrLlpY295iF-EBYOOBHPAAAAZg&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[27] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aFrLlpY295iF-EBYOOBHPAAAAZg&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[28] https://www.theregister.com/2022/10/26/tightening_linux_boot_process_microsoft_poettering/
[29] https://www.theregister.com/2024/03/28/canonical_snap_store_scams/
[30] https://www.theregister.com/2025/02/25/fedora_obs_flatpak_dispute/
[31] https://lwn.net/Articles/1020571/
[32] https://blogs.gnome.org/adrianvovk/2025/06/10/gnome-systemd-dependencies/
[33] https://whitepapers.theregister.com/
I'm not sure, but it may be that there is some confusion. Do those people say that Wayland is fundamentally inferior as an idea/protocol or that Weston is not ready yet?
In terms of driver support for Wayland, that may very well be true but it doesn't necessarily relate to the architecture or purpose espoused of Wayland.
"...eddies around a sinking ship."
Is he?
He must have left the space time continuum...
double edged sword?
"The more corporate Linux gets, the more FreeBSD and the other BSDs will appeal."
Quite. As I've gotten older, and less dependent on big dumb companies for livelihood, I've found myself enjoying FreeBSD (and NetBSD) more. And I'm saying this having started with FreeBSD 2.2.8 and NetBSD 1.6 many years ago.
Red Hat back then was Red Hat Linux, not RHEL, nor even Fedora. It didn't "feel corporate", at least not in the most obvious ways at the time; e.g. you could just download the ISO, install Linux, and get on with it. No mucking about with licenses, or even registering if you didn't feel like it.
Then Red Hat did the first of many rug-pulls to come, Red Hat Linux went away and we got RHEL instead. Early RHEL didn't feel like a radical departure, at least not at first, but you could tell that things were changing. Eventually the RHEL admins at $WORK sometimes groused about the same sort of corporate shenanigans that the Windows admins did.
Fast forward to present, and Red Hat is part of IBM, about as Corporate[tm] as it gets. Though they were well down the road before then. Shenanigans abound.
Contrast with the BSD's. I'm running FreeBSD 14.3 and NetBSD 10.1 on several servers and other lab systems, and when I compare my installation and setup procedures for the modern releases to the versions I started with back in single digits, there is more alike than not. Also, doing sysadmin things on FreeBSD today isn't very different from yesteryear. I can't say the same for some of RHEL, or even my Debian systems these days after moving on from Red Hat, and the trend seems to be continuing.
Overall this situation is not always a good thing; e.g. it's true that the Red Hat's of the world have more resources and developers and such, so better hardware and driver support and so on than you typically find on a BSD. It's not been a problem for my own operations, but some previous $JOB's had issues around it.
Is the "corporateness" of today's Linux a good trade-off for modern hardware and drivers and related things? Depends on your mission. As a mostly retired sysadmin I haven't bothered with a corporate-released OS for a few years now and don't miss it at all. But if I still needed a for-pay job I'm sure I'd be keeping up with RHEL and/or Ubuntu.
Dear Nate,
Whereas:
> [Nate] says more than 70 percent of KDE Plasma users who permit telemetry are already using Wayland...
Whereas:
I don't permit telemetry,
Therefore:
This is to provide notice that I don't use Wayland on my machines (which run Debian KDE or Kubuntu).
Addendum:
Liam, Thanks for the head's up on Kubuntu's future change, which I may have to swap for another distro if Wayland still breaks my stuff when that future change comes.
It would be “arcane and trivial” if the touchpad worked
As someone who actually tries to daily-drive *nix of myriad flavors, I really just want some things to work. When it comes to touchpad support, an important feature since most everyone is on laptops these days, Wayland/libinput are just frankly inferior. It’s kind of incredible that the relatively shit xf86-synaptics driver is more usable than the modern replacement.
Ample arguments out there about whose responsibility it is to capture inputs, and the Wayland devs are very forceful in shrugging off that burden and pointing to individual applications as the issue.
Yet that’s philosophical. As an end user, I just don’t care; give me a rational argument why it’s *acceptable* for my mouse cursor to struggle, stutter, or lag, in 2025, on recent hardware, in everything from the DE to Chrome to QGIS.
Eddies around a sinking ship?
I find the polarization around software rather silly.
I understand you root for your team, but in the end, you "win the hearts and minds" of people by delivering what they desire, not by trying to silence or demonize the other teams.
X11 has a long history of not being able to deliver what people needed as its architecture was designed in different times, and the times have changed a lot. Xorg was unable or unwilling to adapt X11, so a new design was made. That is what engineering is all about, build better, faster, and cheaper products so more people can enjoy more services.
This is FLOSS, so everyone is free to start their own forks to develop the products they want. Fighting by word or fist with other teams is counter productive. In the end, developers and users will flock to those who are welcoming and deliver, not those who curse and cannot deliver. If Xlibre can deliver, great. If not, its a pity but such is life.
PS:
What most certainly will not help attracting general developers and users is to do explicit political grandstanding in your Readme. That Readme alone would prevent me from even considering to contribute if I was at all able to.
"Wayland is better for modern hardware."
Not according to Valve and CodeWeavers, creators of the Steam OS and Proton. So the most demanding use case, gaming, suggests Wayland is not better than X11 for modern hardware.