The Linux Thread - The Autist's OS of Choice

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
maybe it's time to make a version X12? If there was one, what would be the biggest changes?
  • Support for passing HDR metadata with an API instead of relegating that stuff within DRI3
  • Per-monitor fractional scaling
  • Good support for mixed refresh rates across different monitors
  • Ability to switch bitcolor at runtime without restarting the X server
These are essentially what come into my mind right now.
 
  • Support for passing HDR metadata with an API instead of relegating that stuff within DRI3
  • Per-monitor fractional scaling
  • Good support for mixed refresh rates across different monitors
  • Ability to switch bitcolor at runtime without restarting the X server
These are essentially what come into my mind right now.
Would it work to rework Xorg so that each monitor has its own X server and X servers can be added or rebooted per-monitor? Or somehow split the server so it can still handle applications spanning multiple monitors but still have components that are monitor-specific?

All those wish list items seem like something where coming out with X12 would be easier then a whole new windowing system.
 
The problem is that Wayland is a deeply flawed technology to be used on a desktop system while it could be useful on a kiosk or mobile platform.
sed -e 's/Wayland/systemd/g' -e 's/kiosk or mobile platform/a fleet of corporate servers/g'

One thing that I always asked myself if why couldn't the X11 developers use these 15 years (yes Wayland is that old) to actually improve an existing technology instead of starting from scratch?
They weren't paid to do that. Companies wanted a display server for embedded Linux systems, polishing X11 up for task was too much work, there was no need for a complex system in a kiosk and now here we are (source: my ass). Paying customers are satisfied, desktop Linux users will assume the square peg shape to be forcibly shoved into the round hole of The Future™️ according to these retarded dipshits.
 
What does this offer over base debain?

Desktop-focused presets, and a smaller base of starting applications. Really, it's just another desktop distro built on top of Debian, lighter than Mint, but also sits between "beginner" and "intermediary user" as far as desktop usage is concerned. I've tried this on a Raspberry Pi a couple years ago and was quite pleased with its starting defaults.
 
Desktop-focused presets, and a smaller base of starting applications. Really, it's just another desktop distro built on top of Debian, lighter than Mint, but also sits between "beginner" and "intermediary user" as far as desktop usage is concerned. I've tried this on a Raspberry Pi a couple years ago and was quite pleased with its starting defaults.
Oh nice so it has a fleshed out arm based repo then?
 
Would it work to rework Xorg so that each monitor has its own X server and X servers can be added or rebooted per-monitor? Or somehow split the server so it can still handle applications spanning multiple monitors but still have components that are monitor-specific?

All those wish list items seem like something where coming out with X12 would be easier then a whole new windowing system.
The best way would to make X11 monitor aware, that is, make sure that each monitor for example has it's own data structure containing information about it.
Hell, Windows 95 could handle multiple screens even better that whatever this is (if I'm not mistaken).
Speaking of HDR, I do not understand why the metadata is hidden from the userspace tho
 
  • Thunk-Provoking
Reactions: Doctor Neo Cortex
KDE has decent multiscreen handling as of 5.27 or so when kscreen was rewritten. If KDE 6 will break X11 on launch I'll probably switch to LXDE plus autoxrandr or something similar. I'm kind of getting tired of new=better mentality in software world, maybe it's just what getting older feels like. Why rewrite whole desktop environment every decade, muh security is not a valid reason, why didn't you think of security implications ten years ago?
 
I don't get what the security problems would be if everything is running on the same computer. A remote X11 connection could be encrypted yes, but locally it's fine as is right?
"B-but then other programs can sniff the screen content!!!" YEAH you absolute niggerfaggot! What if I NEED to share the contents of the screen or a window? What if I NEED to record the screen, a window, etc. Wayfaggots actually believe that screensharing or screen recording must be done with pipewire because it's secure! Oh fuck off with that bullshit.
Absolutely niggerlicious.
Edit: For clarity
 
Last edited:
Companies wanted a display server for embedded Linux systems, polishing X11 up for task was too much work,
X11 in embedded applications has already been a thing for decades.
I was on a plane once where on startup one of the engine monitoring displays showed the standard X cursor and background on startup for a second or two. At the time it almost certainly wasn't Linux, but it absolutely was X11.

Considering X once ran on systems with 4MB of RAM, I can't imagine it couldn't be slimmed down to beat Wayland.
 
Not like Win32 apps are sandboxed.
IIRC Win32 doesn't have much more sandboxing than Linux anyway. I'm thinking more of something like what Android has, but not retarded like what Android has. Linux can also already do this with namespaces and shit, it's just rather clunky and the OS environment doesn't lend itself to doing it easily.

Overall it doesn't fucking matter that much because you shouldn't really be running random shit on your computer anyway.
 
Wayland reminds me of GNU Hurd.
It's good to have lofty ideals, but for pity's sake, shit or get off the pot - I've turned grey waiting for you cunts to stop fiddling with shit and release something that just works.

Shit like this is why Microsoft are still in business.

Manjaro+KDE+Wine+Deus Ex iso=fuck twatting around with settings I'm off to nuke Area 51.
 
This stuff does need to be sandboxed, but there needs to be a new security paradigm first. Users and groups don't really work the best.
Hard disagree, given the time I've wasted fighting with Windows' ACLs and SIDs. There's a virtue in simplicity, when it comes to security.
 
Back