Although Wayland has largely replaced Xorg, and most major Linux distributions and desktop environments have either already dropped support for the aging display protocol or are in the process of doing so, efforts to extend Xorg’s life or replace it with similar alternatives continue. Recent examples include projects such as XLibre Xserver and Wayback. And now, a new name is joining this group: Phoenix.
It is a new X server project that takes a fundamentally different approach to X11. Written entirely from scratch in the Zig programming language, it is not yet another fork of the Xorg codebase and does not reuse its legacy code. Instead, according to devs, Phoenix aims to show that the X11 protocol itself is not inherently obsolete and can be implemented in a simpler, safer, and more modern way.
Phoenix Emerges as a Modern X Server Written From Scratch in Zig
https://linuxiac.com/phoenix-emerges-as-a-modern-x-server-written-from-scratch-in-zig/
17 Comments
TheTechnician27@lemmy.world · 50 pts · 238d
This is 100% true, but the efforts are negligible and not even worth consideration.
^is an exponentiation operator in C and who got kicked off Xorg for being a moron who did functionally nothing of any importance while carelessly breaking things like the ABI. Enormous quantity but zero quality to speak of. It will go nowhere and only has any crumb of relevance because of the maintainer's virtue signaling.As GNOME and KDE drop X11 and DEs like Cinnamon adopt Wayland, more and more actively maintained applications will stop giving a shit about X11. Even if they don't explicitly not support it, none of the developers will be using it, and most of the userbase won't either; thus, applications' support for X11 will just rot away if it isn't outright deprecated. Obviously X11 will always have a base of legacy applications, but you're going to be seriously hard-pressed even two years from now to find someone who would use X11 over Wayland – except for specific and severely outdated hardware, conspiracy nutjobs, and the rare case where XWayland doesn't properly support a legacy application.
Resplendent606@piefed.social · 19 pts · 238d
Whenever I see a new project trying to modernize Xorg, my first question is whether it's actually about the code or just another protest/spite project. We’ve already seen how XLibre is less about display servers and more about a weird crossover of technical incompetence and fringe politics (championed by people like Lunduke). Is Phoenix actually a serious technical effort, or is it just the latest attempt to build a sanctuary for people who were kicked out of the Wayland and Xorg dev circles?
poinck@lemmy.world · 5 pts · 238d
corbin@infosec.pub · 2 pts · 236d
Yeah, this is just building a new engine for a car already at the junkyard. This will have no practical utility when, or if, it reaches parity with X11.
victorz@lemmy.world · 2 pts · 238d
I dunno, I thought the description of the efforts and ambitions here sounded really promising. It's not even meant to be fully X11 compatible.
I'm excited to hear more down the line. If it turns out to be nothing, fine, but I'm excited for the possibilities this could bring.
anders@theres.life · 2 pts · 206d
Just switched to Xlibre 25.1.1 from using Wayland since 2022.
The performance is a lot better on Xlibre than on Wayland and especially much better than Xorg (which is plagued by mouse lag and tearing)
In my experience it doesn't seem like metux is a bad dev. He has done what the Xorg devs said they couldn't do.
loppy@fedia.io · 29 pts · 239d
I think this is important to highlight:
5714@lemmy.dbzer0.com · 20 pts · 239d
In 2038, there will be a 32-bit computer with this, sysvinit etc. dying a well-deserved time death.
dgriffith@aussie.zone · 25 pts · 239d
32 bit computers can handle 64 bit timestamps, it's just a matter of defining time_t to be 32 or 64 bits at compile time. The compiler will deal with all the mess of splitting the 64 bit value up to calculate on the smaller registers in 32 bit architectures, just like any other variable defined as int_64.
Linux kernels have had support for 64 bit time on 32 bit systems since version 5.something, so generally speaking there'll still be retro 32 bit hardware running past 2038 just fine.
fonix232@fedia.io · -5 pts · 239d
*2058
4am@lemmy.zip · 7 pts · 238d
https://en.wikipedia.org/wiki/Year_2038_problem
fonix232@fedia.io · 7 pts · 238d
I'm very well aware. But you're clearly not aware of the fact that this X server simply wouldn't be running on an old enough kernel for the 2038 problem to be relevant.
jeena@piefed.jeena.net · 8 pts · 239d
But why the name? Even Firefox had switch from Phoenix.
qaz@lemmy.world · 3 pts · 238d
Didn't they switch from Firebird because of the database?
ptmb@piefed.zip · 7 pts · 238d
It was first named Phoenix and they ran into issues with the phoenix BIOS because some of their BIOSes had a built in browser. They renamed to Firebird and then found out about the DB with the same name, and that's when they finally renamed to Firefox.
qaz@lemmy.world · 8 pts · 238d
morto@piefed.social · 3 pts · 239d
The plot thickens...
msokiovt@lemmy.today · 1 pts · 239d
Another X Server in Pheonix in Zig? I'll keep an eye out for this one.