My question about totally new operating systems is: does any "end-user" software work on it? Can I run Python? A Bash-like shell and some approximation to GNU coreutils? Will audio output actually work? Will my monitor work at a sensible resolution?
It seems like such a tremendous effort just to get to the point where anyone other than a serious OS developer can use it. I wonder if there's any way to reduce the effort required, or if that's just how things have to be.
I think the question is "how is this better or different than Linux or Windows", which it aims to replace? What are its design choices? Does it introduce any new concepts? For an example, see https://fuchsia.dev/fuchsia-src/concepts
There's a difference between making a Unix-compatible/Unix-like (a la Redox, Minix, Linux, the BSDs, macOS, etc) and having POSIX-compatibility + interfaces (a la Haiku, Serenity, Windows + Cygwin or WSL, etc).
One is just reinventing the wheel, even if it means "having useful software"; the other is attempting something new while offering tools to make porting useful OSS simpler.
As far as I see, Windows has always been different for the sake of it; it is a prime example of reinventing the wheel. What makes it good is its market share means things are almost always guaranteed to run on it, otherwise it is terrible and always has been since MS-DOS. If it didn't have "useful software", what else would it have going for it?
You could sum up most hobby OSes in this manner. My main metric for when it's moved from simply hobby to a potential functional OS is a) are there additional contributors and b) does it have a USB stack.
Quite a few alternative OSes check A, very-VERY few also check B. And only one, that I know of, checks B but not A.
From the link someone else posted:
Although it's a long shot (and very optimistic), I ultimately intend it to replace Linux and Windows as a desktop operating system.
Unless you also have the means to manufacture your own end-user devices that have feature parity with Apple devices, there is no practical way to replace MacOS. Getting non-Apple OEMs to ship your OS with their PCs is probably not happening but at least in the realm of being conceivable.
I do not see any impressive innovation in its design and the OS doesn't introduce any new concepts or paradigms, which makes it, in essence, no different from regular monolithic operating systems, implementing POSIX API with regular system calls. Just a monolithic kernel crudely implementing some well known device drivers as well as already existing filesystems with a block cache, and a POSIX-based userspace with GCC and binutils.
That is said, it doesn't mean this isn't an incredible piece of work, especially if written by a single author!! That is definitely a huge amount of engineering that needs some good skills and persistence. I believe it is a very rewarding experience. The product also seems to be really working fine with very impressive stuff.
Always interesting to see a non-linx os around, I guess next goals could be getting it self-hosted. Also the source code of a modest enough size that it's not impossible to actually get your head around it.
When you run the .start.sh file, You're greeted with this message:
This project also include the work of other projects.
All trademarks are registered trademarks of their respective owners.
They licenses may be found in the following files:
- util/nanosvg.h
- shared/stb_image.h, shared/stb_sprintf.h, shared/stb_ds.h and util/stb_truetype.h
- res/Fonts/Hack License.txt, res/Fonts/Inter License.txt
- res/Icons/elementary Icons License.txt
- res/Media/Licenses.txt
- Ported applications have their licenses in their respective folders.
It's not clear where the allegation "distributing a ported/patched GCC[...] under the MIT license" even comes from. That same notice appears in the file LICENSE.md (not that it would matter much even if it didn't).
So a toy OS like Android. You need linux to compile it. An it says nothing about the goals, what is different compared to other OSs, why would i want to use it.
My question about totally new operating systems is: does any "end-user" software work on it? Can I run Python? A Bash-like shell and some approximation to GNU coreutils? Will audio output actually work? Will my monitor work at a sensible resolution?
It seems like such a tremendous effort just to get to the point where anyone other than a serious OS developer can use it. I wonder if there's any way to reduce the effort required, or if that's just how things have to be.