An operating system I’ve been writing since ~June 2017. Although it’s a long shot (and very optimistic), I ultimately intend it to replace Linux and Windows as a desktop operating system.
Very optimistic, but there’s quite a few things here already. The code is on gitlab, where you can find more information, too.
“Optional POSIX subsystem.” These words tell me it will adopt the same kind of package manager system present on all Unix clone operating systems, and always be as irrelevant as those on the desktop.
Gitlab gives me an error code. Is this from scratch? Is it yet another unix-like OS?
I am always interested in alternative approaches, but ultimately it is disappointing from an end-user perspective. Computing isn’t simple anymore. Current hardware needs so many sub-systems to be implemented to be able to function bare metal, that new operating systems either need between 20 to 30 years to get up to snuff or they are condemned to live in virtual machines. Just look at Haiku and ReactOS.
Even if an alternative OS manages to overcome the hardware hurdles in a adequate way, it leaves the question of applications. An OS without applications isn’t very usable. What happens too often is that applications from other OSes are ported to it. It quickly gives you applications to use, but most of the time the ported code doesn’t use the unique capabilities of the new OS. Since most app categories are already filled by ports, barely any native programs are developed. So you just switch out the underlying OS to run the same programs you ran on your previous OS. What does swapping the OS tangibly bring at that point?
It seems like the 80’s was the heyday of alternatives and experimentation. Lots of different computers, lots of different OSes and (mostly) unique software to go with it. Beautifully clunky peripherals and software on as little as one floppy and sometimes as much as 35. New hardware brought wild new capabilities. Now it is all fairly homogenized, whether x86 or Arm, the peripheral connectors are mostly the same. Software is either a *Nix or WinNT. The web is a code heavy monstrosity.
Alright, I’ll take my anti-nostalgia pills now.
r_a_trip,
I agree with the entirety of your post, this summarizes the state of alt-os in a nutshell.
Most alt-os projects simply will not get the attention they need to be supported. It doesn’t matter how good they are if they don’t have the software and the only way they’re going to get the software is by becoming compatible on some level with a preexisting popular platform like linux. Committing to compatibility is an ongoing effort that takes away from limited resources and projects end up always having to play catch up.
I sometimes think plan9’s had superior engineering than all the unix clones, but it never became relevant because they couldn’t achieve market penetration. I’ve pretty much given up the alt-os scene, not because they don’t have merit, but because they aren’t relevant and aren’t supported.
Heck even running an alternative android stack in the form of lineageOS/AOSP has got me pounding my head on the wall at times. The latest has been a fight to get banking apps to work. The last thing I want is to give in to monopolies/duopolies, but It can get pretty futile and some days I ask myself what’s the point in trying when the whole world is pressuring you to give in 🙁
Yeah… 🙁
First of all, I think the OS looks really good, the tabbed design is cool, I tried running it and it all works! It can run Unix tools like GCC, but is about GUI. Very impressive. It’s a neat project.
But, yes, what would it take for me to use it, or Haiku? Probably Firefox. That is sad, I guess.
I wonder if one needs to be thinking about new uses for computers, and maybe some of them will need to be more than just an app on an existing platform.
There’s lots of experimentation while there are open ended problems, once solutions are found you see less and less research activity on problems which, frankly, have been solved.
That’s why today most of the experimentation is on the wild west is on the internet: the cloud, apps, and software as service. Because there’s not fully established clear solutions (although we’re getting there). Then things will move over to IoT and stuff like robotics and augmented interactions with reality. You also have lots of wild stuff going with the 3D printing, etc.
Honestly, how many times do you need to reinvent the wheel when it comes to ISA, Desktop metaphors, or kernels.
Oh, how we laughed at the time…
Good luck to Nakst, but I can’t help but feel disappointed that it looks like yet another Windows clone.