The no future of X.org, FreeBSD becomes headless?

RedHat announced two days ago they are abandoning X.org starting 2025. This is a done deal, there are no ifs.

Not a single BSD distribution including FreeBSD, NetBSD or OpenBSD, not a single remaining UNIX including AIX and Solaris has released any statements in regard to that.

What are you guys going to do? Your graphical display server is now declared abandonware.

Wayland for FreeBSD is considered highly experimental. Other BSDs and Unixes don't even incorporate it in any shape or form.

Have FreeBSD and UNIXes essentially become RedHat's hostages in terms of what software you're offering? That's a depressing thought.
 
Wayland for FreeBSD is considered highly experimental. Other BSDs and Unixes don't even incorporate it in any shape or form.

Wayland is considered highly experimental. It's function right now is just "good enough" and does not offer what is needed for a fully robust graphical system. Quite frankly, I believe the headlong rush that the Linux world is doing into Wayland is suicidal. I'll take the mature Windows over the currently half-baked Wayland implementations any day. I like things to "just work."
 
RedHat announced two days ago they are abandoning X.org starting 2025. This is a done deal, there are no ifs.

Not a single BSD distribution including FreeBSD, NetBSD or OpenBSD, not a single remaining UNIX including AIX and Solaris has released any statements in regard to that.

What are you guys going to do? Your graphical display server is now declared abandonware.

Wayland for FreeBSD is considered highly experimental. Other BSDs and Unixes don't even incorporate it in any shape or form.

Have FreeBSD and UNIXes essentially become RedHat's hostages in terms of what software you're offering? That's a depressing thought.
The article says that the upstream project, fedora, has no such plans. There are too many over reactions to red hat announcements and it gets old.

"What are you guys going to do? Your graphical display server is now declared abandonware."

Please read the article you posted.
 
I tried Wayland multiple times. It is just not ready. My system runs way too smoothly with X11. I understand the benefits of Wayland but I could not imagine using it for my daily driver right now...
 
Unless X.Org Foundation is fully backed by RedHat, I don't see a simple reason why anyone should be concerned about what RedHat does. I understand that their PR message is "We own Linux and everything around it", but something tells me that it's a typical big tech lie we see everywhere these days. I wouldn't be surprised if they would announce next year that they are replacing Wayland with "gen AI" (now I'm exaggerating, I know).
 
Maintaining Xorg ourselves is actually not so bad. The only code that needs updating these days as hardware progresses is in the modesetting driver. This is almost identical to the code in the modesetting Wayland compositor "backends". The complex stuff is in the lower libdrm layer which Wayland is still tied to.

Now that Linux can finally stand on its own two feet, gain its own identity and for better or for worse move over to Wayland, as cracauer@ mentioned, we can finally start to clean the smell of wee off Xorg that Linux and freedesktop.org has left.

Not a single BSD distribution including FreeBSD, NetBSD or OpenBSD, not a single remaining UNIX including AIX and Solaris has released any statements in regard to that.
If BSD isn't a UNIX, then you must be referring to UNIX as the certification. In that case, Inspur and Huawei's EulerOS are both UNIX'03 certified and being based on Linux, will support Wayland no problem.
 
The article says that the upstream project, fedora, has no such plans.
Somebody misunderstood something here... For starters, Fedora is downstream from RedHat...

Second, yes, Fedora does have plans to go Wayland-only:

Third, I am looking forward to the day that FreeBSD irons out Wayland to the point that it installs and runs as seamlessly as Xorg does. Xorg requires very minimal config to compile and run successfully on FreeBSD - as long as you have the correct GPU driver loaded, you're good to go. Wayland does require a bit more work (especially monitoring the environment variables that need to be set reliably and correctly, and the very presence of /var/run/user/NNNN/). Once that house of cards is stable, and common config work is something that can be included in a default .conf file, that would be nice.
 
What does that mean if Red Hat abandons something? are they the standard usually for all Linux/Unix OS to look at or something ?
 
my experience with Wayland is excellent on FreeBSD. it is much better protocol and X.org development is just stopped for some years ( except security fixes and some bugs fixes ). problem is within legacy software which is not migrated to new protocol. it is not Wayland protocol problem per se. here is an comparison example about the difference:

X11 screensaver application ( client ):
- client requests all keyboard/mouse input
- client makes window fullscreen
- client places the window on top of all other windows
- server ( the xorg ) has no knowledge what client is trying to do.
Now, consider screensaver crashes due to various reasons ...

Wayland screensaver:
- client informs the server ( wayland compositor ) that it is a screenlocker
- server informs client of window size
- server restricts input to the screenlocker client
- server handles terminations, crashes and other.
I understand sentiments and nostalgic veil about the software of old days but is not going to be developed any longer. I am aware about some corner cases such as nvidia, VR devices and others but again it is not wayland protocol problem.
 
Correct me if I'm wrong, but REL is unlikely to be used on the desktop in the first place, no? Isn't it mostly a server and enterprise used OS?
I work for a Fortune 200 company that uses hundreds of Red Hat (RHEL) workstations for mission critical work.
 
I'm using Xfce I believe it's not yet compatible with Wayland so I've no other choice unless I change to other DE.
This is an important point. Is my preference able to work with Wayland? That is pretty much a yes/no question. If the answer is no, then why should I want wayland?
 
Somebody misunderstood something here... For starters, Fedora is downstream from RedHat...

Second, yes, Fedora does have plans to go Wayland-only:

Third, I am looking forward to the day that FreeBSD irons out Wayland to the point that it installs and runs as seamlessly as Xorg does. Xorg requires very minimal config to compile and run successfully on FreeBSD - as long as you have the correct GPU driver loaded, you're good to go. Wayland does require a bit more work (especially monitoring the environment variables that need to be set reliably and correctly, and the very presence of /var/run/user/NNNN/). Once that house of cards is stable, and common config work is something that can be included in a default .conf file, that would be nice.
If you'll read the article that was posted it says Fedora does not have plans to drop X11. That's what we were working off of here. :D And if you read from Red Hat's website you will see that Fedora is upstream. It's basically the testing for Red Hat. It's been that way since Fedora Core was launched.

Think about it like this. The Fedora project is the upstream, community distro of Red Hat® Enterprise Linux. Red Hat is the project’s primary sponsor, but thousands of independent developers also contribute to the Fedora project. Each of these contributors, including Red Hat, bring their own new ideas to be tested and debated for inclusion by the larger community into Fedora Linux. This also makes Fedora an ideal place for Red Hat to put features through its own distinct set of tests and quality assurance processes, and those features eventually get incorporated into a version of Red Hat Enterprise Linux.

https://www.redhat.com/en/topics/linux/fedora-vs-red-hat-enterprise-linux

Wayland has been our default session for a long time and has matured considerably. At the same time, the X11 session isn't getting as much testing, and is an additional resource burden.
It would therefore be good to have an exploratory conversation about whether we want to make any changes with regards to the X11 session. Options that we could consider include:
  • Hide the X11 session by default, so that fewer people use it.
  • Remove the X11 session file. Users would be able to add it back and use it as an unsupported configuration.
  • Build GNOME components (e.g. mutter/gnome-shell/gnome-session) without X11 support. Technically, users could recompile those packages themselves.
These changes would only affect GNOME and the session. Apps would still be able to use X11 through Xwayland.
Note that there is currently no anticipated time frame or schedule around this discussion or its outcomes. Additionally, no decisions will be made without getting advice from the relevant maintainers and domain experts.

https://pagure.io/fedora-workstation/issue/395

Edit: This whole discussion is rather nonsensical in my opinion. An upstream to Red Hat having discussions on Wayland vs X11 has zero impact on FreeBSD. Linux based systems can't agree on anything therefore there will be a continued need for old and new technologies to continue to receive support on their platforms. What that has some users concerned about is a mystery to me regarding FreeBSD.
 
The article says that the upstream project, fedora, has no such plans. There are too many over reactions to red hat announcements and it gets old.

"What are you guys going to do? Your graphical display server is now declared abandonware."

Please read the article you posted.



RHEL10 is dropping X.org.

Fedora 40 (to be released in 5 months) is already deprecating X.org for Gnome and KDE sessions. Fedora is an upstream project for RedHat. If RH drops something, Fedora is going to do that much earlier.

What should I read outside a ton of ill-placed sarcasm in your message?

What does that mean if Red Hat abandons something? are they the standard usually for all Linux/Unix OS to look at or something ?

RedHat leads Linux and it seems like Unix as well. They are the driving force behind and provide almost all the money, resources and backing for: X.org (which they are abandoning in less than 2 years), systemd, pulseaudio/pipewire, Gnome (and a multitude of related libraries), Wayland and a ton of other projects. They are investing in the Linux kernel a lot.

If RedHat decides something other Linux projects follow. There have been no precedents to the contrary outside of Devuan, a Debian fork with no systemd.

Basically Linux is RedHat. Without the company Linux would have been nowhere near where it's now (that includes Steam Deck which wouldn't be possible otherwise).

Maintaining Xorg ourselves is actually not so bad. The only code that needs updating these days as hardware progresses is in the modesetting driver. This is almost identical to the code in the modesetting Wayland compositor "backends". The complex stuff is in the lower libdrm layer which Wayland is still tied to.

It's weird "it's not so bad" yet *BSDs and UNIXes have made very few contributions to Xorg over the past decade:


It's basically been a RH show.
 
This is an important point. Is my preference able to work with Wayland? That is pretty much a yes/no question. If the answer is no, then why should I want wayland?
The FVWM chief dev said he has no interest in supporting wayland. So the question is answered for me: Wayland? No, thx! :)
 
Back
Top