You could actually install Windows on btrfs with that driver 🤨? Never tried it… but would be willing to try, since I’m sick and tired of Windows store corruption 😒.
@iHUNTcriminals but why comment on it? An elite user should know that different things work for different people not strut about acting superior because he doesn’t use a DE.
That’s actually a bit tricky for me especially via text, I’ve got a touch of the 'ole autism (like I have an actual diagnosis from a real medical doctor and not TikTok), but I get what you mean. I have a bad habit of assuming the worst about people
I wonder if they consulted Plasma devs about it. Sure they said that they aim to make Wayland ready for Plasma 6, but it didn't sound like it was an actual plan for 6.0. After all they got their hands full with Qt 6 porting, and there are still major roadblocks with completing Wayland support, while 6.0 is about to have its alpha release already.
Knowing Fedora devs however, I suspect they didn't. They switched to Plasma Wayland by default several Fedora releases ago, when it was in no way ready. I guess I will switch to a different distro when this time comes.
Yeah, as usual the opinionated crew are making something that one may even like feel like it’s forced down everyone’s throat (see: systemd, snap…) and making everything worse. I don’t see how any Linux desktop distro worth its salt can get by ignoring 90% of the PC GPU market share and essentially forcing them into an inferior desktop experience for pure ideology’s sake, and I LIKE Wayland. I even put up with all its quirks in a particularly quirky implementation (KWin). But this ain’t it if you want users to use your OS.
Linux has standards where interoperability is important. The more things needs to talk to each other the more they need a common standard to talk over. Things like X11/Wayland don’t have many alternatives as so many things need to talk over them. The only reason there are two standards here is because X11 has massive limitations that cannot easily be worked around.
For package managers applications don’t care about them. Interoperability only matters within a single distro. So people are more free to create what ever standards they want for their own distros. And when people can choose people have opinions and these opinions evolve over time. Which results in multiple competing products that effectively do the same thing.
And here is my hipotesis if the GNU project came up with a good and easy to work package manager in the early days of Linux
Probably, but creating a good, easy to work, fast and reliable package manager that meets everyones needs when you are discovering how you want it to work for the first time is extremely hard. And even if you created a perfect one at the start, requirements can change. This happened with X11, and even with package managers seeing the rise of things like flatpack, snap and appimage that all work fundamentally different from the traditional ones.
OK maybe what I meant was a packaging format and not a package manager, above there was a user that mentioned that all distros have their quirks and kinks, if GNU created a package manager that worked perfectly at first time maybe it’s adoption would go across the distros but as u said to make it perfect the first time is something hard and even harder on early days where nothing was set on stone and there would be always the odd one that would make their package manager.
But if we all agreed early on, one one packaging format (which of course would have to go through many iterations to reach a stable state ) all package format wars would be over and in well implemented ecosystem of package managers of each distro, it would be also an somewhat interoperable one.
The package format is almost irrelevant TBH. Most packages are not interoperable between distros due to the versions and names of dependencies. That is not something that gets fixed by a standard package format. Packages don’t even work well between different versions of the same distro. largely due to libc - anything that depends on that is built against a specific version and when you upgrade it you need to rebuild and install everything that depends on it. Similar problems exist for all compiled dependencies on a distro.
And while some packages of the same format can be installed on multiple distros (mostly those based of the same foundation) most cannot. This is what the newer package formats (like flatpack) are trying to solve - by including all dependencies inside the package.
So a standard format does not really solve those issues, so there is little advantage for one. At least not one of the old school formats. And the wars are not really over the format, they are over the tooling required for that format. At the end of the day RPMs, DEBs, and arch packages are just tarballs of files and some meta data (and there is even a tool that can convert between them - though anything with dependencies quickly becomes a complete mess). It is the build and install tooling that makes all the difference.
Oh … thx for the insight, it was a daydream anyways looks like the only solution is cloud native if one wants uniformity, still a bit hesitant to have a system so stable I can’t change it’s core filesystem.
I’ve never tried to run multiseat the way you do here.
I do however succesfully run multiple computers in one chassi using kvm/qemu with pci-e physical passthrough on gpu and usb controller to my virtual fedora gaming machine (using vfio drivers in the host). Definitely more overhead than multiseat but I do enjoy the easy backup and restore I have on my gaming machine.
Level1techs.com has a ton of good information if you’re interested in virtualizing instead, such as forum.level1techs.com/t/…/119639
linux
Top
This magazine is from a federated server and may be incomplete. Browse more on the original instance.