New X.Org Server Release While Maintaining Separate XWayland Being Discussed

[kpedersen]

Please remember valve supports dos titles.
https://store.steampowered.com/curat...red-by-DOSBox/
Yep 221 dos based games on sale in the steam store. Some of the games on sale in the steam store were on sale before the first copy of Windows sold.

Well kind of, it distributed emulators running DOS. I would be very much more impressed if it still maintained Windows 95-98 era games (which even today are a real weak spot to emulate. Too resource intensive for full emulation, too hardware specific for virtualization). Neverwinter Nights and some of the older Loki games are going to prove challenging for stores like Steam to maintain. Though I think I am dragging this off topic here.

I have tried to be clear that its X11 bare metal that has been unmaintained and is going away unless something changes.

Yeah, I do get that. And I believe you have been clear. Much of it I actually agree with you. Though I think some of the time durations may be much longer than some people may believe and be more specific to the Linux platform. That said, only time can tell I suppose.

[Azrael5]

@kdepoerson you tell about many other pieces of software are unmaintained, but you forget to tell that Xorg is not a common software. Indeed, it is the major graphic stack of Linux Oses since the last years. So, to unmaintain it has a huge meaning: Xorg is dead. The proof is the transition to XWayland and Wayland at end. Why this transition? Because Wayland is more efficient and secure than Xorg is. Ok, transition phase has its problem. It's inevitable because it faces with change. Now the question is: why do you don't understand that an innovation can improve Linux? Because you are not objective. It's not a match between Xorg and Wayland. It's simply a progress.

[ssokolow]

and I understand that next they're adding their own X11 to the mix.

If you mean WSLg, that's a Wayland implementation combined with the same XWayland we use from what I remember. (It's not in the text, but I know I heard that, so it's probably in the associated announce video.)

[kpedersen]

Now the question is: why do you don't understand that an innovation can improve Linux? Because you are not objective. It's not a match between Xorg and Wayland. It's simply a progress.

It is entirely because I don't just think of Linux. But instead of all UNIX-like platforms. Xorg did a great job normalizing between them. I am not yet convinced that Wayland can achieve the same. I don't actually think the open-source community is stable enough for that these days. It is pulled in too many directions.

[ssokolow]

Uhm....not sure what you are talking about. I was referencing a comment from @Azrael5

My intent was to provide another example in support of your side of that exchange.

[tomas]

It is entirely because I don't just think of Linux. But instead of all UNIX-like platforms. Xorg did a great job normalizing between them

Yes, I understand that. But the thing is, while desktop Linux may be considered to be niche, it's nothing if you consider other UNIX-like operating systems, there you can really talk about being niche in a niche, no matter if we consider old proprietary Unix like Solaris or if we include FreeBSD. If we just consider UNIX-like operating systems, then Linux completely dominates everything else when it comes to desktop use case. Almost all development targeting the desktop happens on Linux and is aimed for Linux. So when the Linux community decides that it is time to start retiring the old X server in favor of Wayland, then that will affect the other UNIX-like operating systems, whether they like it or not. Now, I'm quite certain that FreeBSD will in fact adapt to wayland one way or the other, so the problem (if any) will be solved. For example, look here

https://forums.freebsd.org/threads/h...wayland.77128/

and here

https://wiki.freebsd.org/Graphics/Wayland

[kpedersen]

then Linux completely dominates everything else when it comes to desktop use case. Almost all development targeting the desktop happens on Linux and is aimed for Linux. So when the Linux community decides that it is time to start retiring the

I don't disagree but back in the 90's you could almost swap out word for word in your statement "Linux" with "Windows". We were told all the time "Windows is the future, everything else is niche". Yes, at that time they were absolutely correct. But thankfully we never listened to them (imagine what sh*t we would be in now if we did!). Same with Linux now (luckily a more pleasant OS) there are more operating systems to deal with than just one. There always will be and that is a good thing.

Linux can do what it wants but that doesn't really have much impact on what all other operating systems do. Open-source is united in some ways but very separate in others. Calling things "modern" or "dead" always really sounds dumb. However even more so if only one platform is making the change.

[tomas]

I don't disagree but back in the 90's you could almost swap out word for word in your statement "Linux" with "Windows". We were told all the time "Windows is the future, everything else is niche". Yes, at that time they were absolutely correct. But thankfully we never listened to them (imagine what sh*t we would be in now if we did!). Same with Linux now

No, not at all like Linux. Windows is and was a proprietary operating system that tried hard to destroy all competition and lock-in users and developers to their platform. All technology that Linux is based on, from the kernel to graphics system and desktop environments, is 100% open source. If other UNIX-like operating systems want to go the Wayland route they are free to do so. The specification and the source code is there. But of course they have to do the work, the Linux community will not do it for them. If they rather stay with X or develop something else entirely then they are also free to do that. But again, they themselves have to put in the work, the Linux community will not do it for them.

Regarding calling something "dead" or not, I don't think it's a productive discussion. X is far from "dead", but of course it matters when Linux as a platform decides that it's moving to another display system. Linux is not just "one platform making the change", it is THE platform which completely dominates when it comes to desktop usage among UNIX-like operating systems. So while it might be true to consider Linux as "one platform making the change", in reality it means like 95% of all UNIX-like desktops is going Wayland in the next couple of years.

[Charlie68]

Whenever I read about these discussions, they are always the same words ... but in reality the problem is always the same, not everyone is ready to switch to Wayland yet, not because they don't want to, but simply because a lot of work needs to be done on the composers and it's not something you do overnight.
So yes, Wayland is key today, but not everyone is ready yet. Let's take XFCE, I think he recently finished the transition to GTK 3, he certainly couldn't think of wayland when they were still on GTK 2.
The open source world of Gnu / Linux is made up of many projects, some have companies or large communities behind them, others do not have this strength, so it explains why wayland is still underused, only Gnome is using it and KDE is now starting to see an acceptable wayland session. Until today Nvidia did not support wayland, some software is still not on wayland, not even the latest Ubuntu LTS uses wayland and Lts versions must be supported for at least 5 years.
So talking about Xorg dead or undead doesn't really make any sense.

[mdedetrich]

I find it hilarious that people are saying BSD's are irrelevant and hence are a non factor when it comes to Wayland migration when the whole point of Wayland is that its just a pure protocol that was deliberately designed to know nothing about the OS it runs on.

The only reason right that Wayland is "tied" (very loosely speaking) to Linux now is that since Wayland is only a protocol, all of the DE were forced to re-implement everything from scratch and so unsurprisingly they decided to use Linux specific technologies (such as GBM) since it was easier which kinda forced the hand for the various BSD's.

It turns out that releasing a protocol without any proper implementations creates these crappy situations, and ontop of this X11/Xorg abstracting over the hardware/OS did turn out to have more merit that people give it credit for considering that now there is so much code duplication out there (so much for Linux developers striving for "technical excellence".