I also think it's fine for projects to target a specific more hobby-esque audience. Not every project has to be for everyone.
It took me a long time to learn not to expect too much, and my choice of projects has been very different since learning this.
Some projects seem like they're held back by trying to be kinda-mainstream-ish, rather than being fully able to try whatever they want.
It's an experimental hobby distro. If you want something grandma can use, you want to use use something everyone uses.
Chimera is fairly niche and goes against all standards. You're always going to run into Chimera specific issues. Chimera also has a nice transparant build system that one can tinker with.
As for Manjaro, I don't think it's hard to make a distro based on Arch that just works. Manjaro developers just suck.
This. They cant even be trusted to renew their certificate 4 times in a row. The level of incompetency reaches new highs.
You are spot on about this project I think in your analysis and I think it simply isn't ever going to be something everybody will want to use. That's okay, like you said, some don't want to participate in operating system development, you are mostly looking to be a user. There are plenty of free and open source operating systems that will mostly provide that kind of experience for you. The intent with something like Chimera is probably to dogfood these sorts of improvements so that other more user friendly distributions can use it as an upstream source in the future. So maybe one day we will see a more user friendly version of this.
I tend to focus a lot on having a clear purpose, just because I meet so many devs that don't quite understand how an non-tinkerer types see computing, and why the rest of us like Android, Web, Ubuntu, and Snap so much.
I have a hard time imagining someone making even a super polished derivative end users would want, but that's perfectly fine if the audience is hobbyists with a deep interest in OS stuff, and I wouldn't be surprised if the actual mainstream distros adopt some of the ideas if you discover something relevant to everyday use.
i do like freebsd, i used it for around a decade so it probably left some mark (just like other systems i have touched over the years, like void and debian), but chimera does not try to explicitly be like anything else - i simply do what i feel are fitting design choices
Not sure if Arch should be grouped with those. On the Steam statistics it's more popular than ubuntu, and that's not counting SteamOS.
In that sense, both of them are products, not projects, to the vast majority of users, and that influences what the core teams choose to focus on.
excellent observation
I wish Linux distributions could advertise to existing Linux users without inadvertently causing this sort of confusion among the non-Linux tech media. Many Linux users don't want the OS to take off, because things will get materially worse in many regards if it gets popular. You often only find the new users and tech media theorizing about the 'year of the Linux desktop', or about Linux's product weaknesses in the narrow lens of comparing it to Windows.
Fedora is able to balance this public image issue fairly well. In their landing page, they emphasize that it is for "developers and makers of all kinds". But because Fedora Workstation actually is fairly close to being an easy-for-all-users distro, in their "Is Fedora For Me" page, they have this paragraph that essentially explains that it might also be for non-developers as long as they are curious/enthusiastic about computers in general:
> The Fedora distribution is made for enthusiastic and curious computer users that like to learn and experience newer versions of software and therefore might not suit everyone. Although the Project works hard to make Fedora as usable as possible for the widest possible audience there are inherent limitations on how fast a new software can be translated, gain accessibility features, provide the most comprehensive documentation and guidance. If you on the other hand need a stable Linux distribution that supports even the latest hardware, are willing to put some effort to help the free and open source software (FOSS) community and file some bug reports this might be a perfect system for you!
to a degree your system package graph also defines your system configuration for things that are coarse enough to allow it; it's also easily auditable
you can run steam in flatpak (it's there), chromium will probably come natively in not that long, it does not seem like that big of a nightmare besides the ways chromium is a nightmare in in general (you can run it in flatpak too right now)
there is no apparmor right now, but it'll come
If you do embedded, those systems often run Linux, and it appeals to anyone who likes the idea of one standard platform for everything, so you might as well use it on the desktop. Especially with RasPi work, which is similar enough to Ubuntu that it really does seem more or less unified.
It's also free, and can run on old computers.
It doesn't have any commerical incentives to leave out features, so a large linux distro can support every protocol, file format, and filesystem with no hacking and fussing and buying expensive compatibility apps.
It supports lots of old windows stuff via WINE.
Is this an experimental or hobby distro or does it intend to be something grandma can use?
If I had to choose this or BSD, it looks like I'd very strongly considering this or something like it, but it definitely doesn't seem like an Ubuntu or Windows replacement. Simplicity is just not a factor end users consider.
As a dev who likes modern software and isn't really into lightweight hacker friendly stuff... the odds of me using anything that's not at least 1% or total desktop market share is pretty low. Anyone who's not an Arch/Kali/BSD/suckless enthusiast type has a finely tuned sense of "Oh no, that makes me feel the same way that thinking about a project car does, I'm outta here".
It could drill be a wonderful worthwhile project... but it's going to be an uphill battle to interest anyone but tinkerers and I always kind of worry about people, because this stuff takes a lot of time and without managed expectations it can really be a disappointment when it doesn't take off. I've tried several times to start new projects and it's been pretty hard realizing they're unlikely to go anywhere anytime soon.
I think getting any mainstream adoption would require changing the entire software ecosystem in general. I would imagine there would be some kind of issues trying to port current mega apps like Steam or Chrome to this.
Plus, dynamic linking seems like it's not that well suited to user friendly distros meant for end users.
Unless you have snap or Flatpak or something like that, I don't see how you can have cutting edge large complex third party packages without at least occasional compatibility issues.
My experience with dynamic distros generally hasn't been terrible, but it always seems like there's some issue or other.
Does this have any kind of AppArmor type sandboxing?