XLibre to the rescue?

Coffee break
1

I keep waiting for someone else to start talking about the XLibre fork around here, but nothing yet. The Xlibre project seems like good fit for the misfits around here. Case in point, the CoC:

image
image479×161 3.17 KB

I’ve never been keen that in a “free” (as in freedom) software environment seemingly everyone decided to halt work on X and push Wayland on us no matter its readiness or feature support. It is interesting to me to hear claims that X was being intentionally held back by the gatekeepers.

I’m excited about this new fork and I noticed that I’m not the only one sniffing around:

13 Likes
2

I’m excited about it too. We’ll have to see what it is when the first release comes out later this month.

8 Likes
3

I won’t pretend to be an expert on this kinda stuff, but wouldn’t this explicitly require different DEs and window managers to target Xlibre instead of Xorg?

A lot of the active projects in these fields are already moving to Wayland for various reasons, and those staying on X are either doing it out of preference, or just because they don’t really care enough to go ahead with the switch.

While I’m glad someone is keeping X alive and well, I can’t really see many devs going ahead with the switch unless it’s effectively a drop in replacement, unfortunately (Again, no export).

2 Likes
4

The OMlx team is working on it. I believe they are trailing it in cooker or at least berolinux is at this point.

1 Like
5

I’m no expert either. There is X11 which is the protocol, this was not forked. Then there is the X.Org server, a reference implementation of the X11 protocol. So now we have two implementations of the X11 protocol for Linux. XQuartz is an implementation for MacOS and Cygwin/X is an implementation for people still plugging into the matrix.

Rather than having target Xlibre instead of Xorg, I think it is a matter of having the window managers and GTK and Qt keep their support for X11 and distributions like OMLx could ship one or both X11 servers and people having an issue with the way X.org is killing X can use and support XLibre.

I think if Weston was started today they would have used Rust instead of C. That is the kind of stereotype I have in my head of the Wayland crowd.

5 Likes
6

Side note, I have wondered how things would have looked differently if Zig was a little further along in it’s development than rust, being it plays so much nicer with C then rust does. But part of life is luck and timing and Zig seems to be late so who knows.

I can’t speak for Enrico Weigelt of Xlibre, but he’s a smart guy and I assume it will be a drop in replacement with backwards compatibility once he’s worked the initial kinks out of it. Then he’ll get to implement all the improvements Red Crap was gatekeeping out of Xorg’s X11 for so many years.

5 Likes
7

That’s awesome news. I was going to go with Apache OpenOffice as an alternative but I’ll give this a try now.

1 Like
8

Best CoC ever.

4 Likes
9

An interview with Enrico Weigelt that is worthy of a read:

2 Likes
10

This made me laugh more than it should have.

11

if it follows the OSI model…technically they might be able to drop it in like MySQL>MariaDB. My guess is this will keep some legacy&embedded systems still going until they finally give up the ghost.

2 Likes
12

I am someone that was happy about Wayland. I found the technical advantages of Wayland to be useful. But I do understand that having options are a good thing, and so being able to have X forked and continued is a win for all.

I will be interested to see how things go on it. Is it possible for one person to maintain, fix, and improve such a massive code base? I don’t think so. I hope that support pours in and even though this is supposed to keep away from corporate I hope that this is not just a grassroot volunteer based effort. Software needs paid development to truly succeed.

2 Likes
13

Considering there is already 47 forks I think that support is pouring in rather quickly.

3 Likes
14

it wont be the first time xorg / xserver has been forked

xenocara for bsd is a fork of xorg

however i would like to see artix, devuan, void and open mandriva attempt to take up the mantle of a new xserver

sam bent has an interesting video on it if your interested

15

No, it isn’t a fork.

The goal of Xenocara is to provide a framework to host local modifications and to automate the build of the modular X.Org components, including 3rd party packages and some software maintained by OpenBSD developers. It is not a fork. We are tracking X.Org modifications and try to push back our changes whenever they are good for upstreams too.

16

I’m mainly on wayland, because I was having horrible colour issues on Xorg. I actual just got it working, I had to create /etc/X11/xorg.conf.d/30-screensetup.conf

Section "Screen"
  Identifier    "Default Screen"
  Monitor        "Configured Monitor"
  Device         "Configured Video Device"
  #               24 for 8-bit or 30 for 10-bit
  DefaultDepth    24
EndSection

My monitor is 10-bit capable, but the colours appear washed out with that. HDR & VRR also do not work under X11. I don’t actually care much about those, but VRR would be nice.

There are definitely pluses to xorg, most don’t affect me, but I would happily switch to XLibre if development is ongoing

17

I’m cool with the idea of Wayland and all, but don’t want to be stuck with it while there are still show stopper bugs affecting my gaming.

1 Like
18

So I am happy that Xorg lives on and that forking saved it.

However, what Mr. Bent here advocates for is a contradiction. He says that basing things on technical merit is better, but then ignores how Wayland is a technically superior professional way to handle the display stack.

Saying Xorg worked, therefore it is good, is devoid of any technical criteria.

The world has changed. Computing has changed. Security matters and now ironically so does privacy.

Xorg is a MASSIVE project. No one person can manage it. Fixing the security issues it has is incredibly difficult. So difficult that trying is kind of insane.

Rewrites happen in the software engineering world for a reason.

Wayland doesn’t do things yet that Xorg shouldn’t have been doing in the first place.

When you really understand this it deflates a lot of this drama.

Yes Redhat is crazy, manipulative, and controlling. Or is it possible that they are simply passionate about all the work they have invested in this?

Their is a modicrum of truth to the crazy for sure.

Going against OSS ideals is also crazy.

But the need to bad mouth Wayland as the evil incumbent instead of the technically superior software seems a little hypocritical.

I will say however, that their actions around this deserve the scrutiny that it is getting. How Redhat, Gnome, and even Ubuntu are handling this is also just crazy.

Also Mr. Bent’s comments on how these companies think they own Linux is apt. They do appear to think that.

4 Likes
19

I didn’t realize that Xlibre was already in Cooker. :grinning:

Count us (OpenMandriva) in – our plan is to support both X11 (XLibre – we’ve already switched Cooker (our development branch) over to it, and aside from some issues with the mouse cursor, it works great - writing this in Plasma 6 running on XLibre) and Wayland as long as it remains viable – or, of course, until something better comes along - I think everyone looking at things objectively will agree that neither X11 nor Wayland are perfect. Wayland fixes some real X11 problems [most of which could be solved with new X extensions too], but at the same time introduces a load of problems of its own, some by design and some by oversight. Obviously there’s no chance of the breakages-by-design getting fixed ever. An alternative that is better than either one might still appear.

ttyl
bero

I’ve been testing out OMLx for just over 4 months now. It’s starting to look like one of those “It’s temporary unless it works” situations.

20

Yeah I think you just need to stick around @SteveM OMLx has obviously chosen you and it’s destined to be this way :wink:

1 Like