

It shouldn’t break as long as you download the extensions through the distro’s repos. Sure, it’s a relative small collection, but it happens to hold everything I personally need.


It shouldn’t break as long as you download the extensions through the distro’s repos. Sure, it’s a relative small collection, but it happens to hold everything I personally need.
Interesting. If you’ve used this on Linux, could you share your experience with and thoughts on it?
You might be absolutely correct on that.
Though, I do wonder what would prevent a stateless system accompanied by a healthy dose of integrity tests from pulling this off.
FWIW, some (perhaps even most/all) uBlue derivatives actually do automatic major version updates. Though, thanks to the bootc-model, they’re dealing with a whole lot less state(/moving parts); hence smooth updates are somewhat expected. The built-in rollback functionality doesn’t hurt either.


Yup, Bazzite is one of uBlue’s images.
Thanks for the answer!
I don’t know to what degree it’s sensible to regard the following as a metric. However, it does help in putting things into perspective. My belief is that Vanilla OS is directly competing with the whole uBlue ecosystem, even if one is Debian-based and the other Fedora-based. As such, I am curious to know how the custom image ecosystem of Vanilla OS is. Like, does something like this exist for Vanilla OS? (FWIW, there was another website that made browsing through these a better experience. But, sadly, I couldn’t find it.)


My two cents: do yourself a favor and switch to one of uBlue’s images. See: https://ublue.it/
As for the why: while Fedora has put a tremendous efforts into making Fedora Atomic possible and continues to have ambitious goals regarding it, the fact that on the latest major release, a serious regression for at least one of the atomic images was not considered a blocker, showed to all of us that they aren’t as serious about it, at least at this point, as one would like or love to see.
The project has been relatively stale in the last couple of years. And I would argue that were it not for bootc and the whole ecosystem of derivatives that it has made possible, there would have been very little to be positive about (momentum-wise).
By contrast, the uBlue images are actively trying to solve issues and problems that the model has. This has led to it being a very polished experience with a great community around it that’s very actively engaged and that tries to be helpful.
All in all, take it from a Fedora Atomic user of almost four years that the uBlue side of things is definitely a whole lot greener.
To me, it seemed as if Vanilla OS lost a decent amount of momentum in their transition to Vanilla OS 2. And I’d say they’ve yet to regain what has been lost. I don’t want to sound all gloom and doom, but is it sensible to be enthusiastic about this project’s future at this point? Like, are the developers still as invested, enthusiastic and motivated as they were in the beginning?
LOL, I probs shouldn’t feed the trolls, but here goes nothing.
We just bought a new PC. And because my family members are kind of adamant on using Windows, we will probably install the latest iteration that M$ has offered.
Funny enough, I am also the techie within the family. As such, I am expected to set it up so that it works as conveniently as possible.
Oh boy… From my reading on the subject, it seems to have become a gargantuan effort to rid Windows 11 of all of its anti-consumer mishaps. Only after doing so, can I actually start setting it up for our use case.
By contrast, if I’d install any beginner-friendly Linux distro, i don’t have to undo anything; setting up starts right after the installation.
I can’t fathom why anyone would prefer the former over the latter, but please feel free to give your take on the matter.
Stuff like Bazzite has literally become plug-and-play. And breaking it is harder than your average Windows install.
As for software: WinBoat is a very easy (almost) one-click installable piece of software that bridges whatever gap you’re imagining.