How come no one's making a new OS from scratch anymore? Everything is based on ancient 70s concept

How come no one's making a new OS from scratch anymore? Everything is based on ancient 70s concept.

Other urls found in this thread:

en.wikipedia.org/wiki/ReactOS
en.wikipedia.org/wiki/Haiku_(operating_system)
redox-os.org/
twitter.com/NSFWRedditImage

Those who do not understand UNIX are condemned to reinvent it, poorly.

>It looks like Plan 9 failed simply because it fell short of being a compelling enough improvement on Unix to displace its ancestor. Compared to Plan 9, Unix creaks and clanks and has obvious rust spots, but it gets the job done well enough to hold its position. There is a lesson here for ambitious system architects: the most dangerous enemy of a better solution is an existing codebase that is just good enough.
> —Eric S. Raymond

Sucks, I know.

Why did Microsoft abandon DOS/W32 archiecture when they could have improved on it, removing the 640KB barrier and continue to improve upon it?

What about Fuchsia

The only ones I know of are TempleOS and KolibriOS, both of them being a hobbist / niche thing.

Why reinvent the wheel?

Google FucksYa, nigger!

Read the question again.

Because Windows has controlled the market, and there aren't a whole lot of programmers talented enough to make a decent, modern-feeling OS from scratch. They're all modified Linux. Except for TempleOS, but that doesn't count because it's straight out of 1985.

because reinventing the wheel is stupid user

Cred Forums Pass user since August 2013.

Because there are plenty of features an operating system could have that you're ignorant for. Plan 9's name spaces and networking are examples of this.

None of these could ever be implemented cleanly on the primary operating systems.

>pass user since aug 2013
Enjoying wasting money just so you don't have to squint at blurry numbers, user?

What was the incentive to create MacOS and Windws at the time they were created that no longer exists today?

Is the answer simply because everything is done online and not locally now?

Seems like the incentive would still exist to create something new to respond to the problems of managing large networks of computers.

>Cred Forums Pass user since August 2013.
The fuck?

It's not wasting money when I'm supporting a service that I don't want to see go under.

Cred Forums Pass user since August 2013.

what

offtopic

seems like this creates exactly the same problem as tripfags.

>Cred Forums Pass user since August 2013.
wtf is this

Because it's Cred Forums's birthday, they feel like autobragging about whoever threw money in Hiroshimoot's direction.

vaporware

>Why did Microsoft abandon DOS/W32 archiecture when they could have improved on it, removing the 640KB barrier and continue to improve upon it?

MS-DOS was based on UNIX.

>Those who do not understand UNIX are condemned to reinvent it, poorly.

Imagine that.

>Everything is based on ancient 70s concept.

Do you know why UNIX was created? The guys working on Multics wrote a game program on the Multics system. When Multics was cancelled, they wanted to continue playing the game they created. Unix was created to play a game.

Just like what happened with MINIX until:

>Those who do not understand MINIX are condemned to reinvent it, poorly.

And thus Linux was born.

Because the existing OS have a massive head start. They've had years of work, thousands of developers, and millions of dollars put into them. It would take far too much work just achieving parity with the existing systems, let alone implementing any useful new features.

>he doesn't know about Urbit

Jesus, Cred Forums, get with the fucking times.

/thread

urbit's not an os

legacy
x86 is crap but guess what cpus might be a lot faster than intel's, but that doesn't matter when there's 10000 software for intels cpu and 's cpu can't run that software because it's a different architecture.
simliarly, nobody is going to use your great OS with 0 software written for it.

see

Actually, if you went to university you'd realize all the cutting edge shit coming out today was already invented in the late 70s.

Stuff like in memory dbms using non volatile memory which didn't exist, and was already optimized for maximum efficiency across memory that didn't even exist at the time, is just now being implemented in 2016.

tl;dr that "ancient" 70s stuff is actually cutting edge, much like that "ancient" lambda calculus that was invented in the early 1900s then wasn't implemented until 40 years later.

As for new OS, there's a few. Google is making it's own proprietary OS right now. CoreOS and many other parallel OS exist like NixOS for mass scripting and deployment. Somebody is writing a rust based timesharing OS but the fundamentals are all the same.

>What is ChromeOS
>What is iOS
>What is Android
>What is Web2.0
>What is rolling release

used by nobody
modified Linux
alright, that one is valid
not an OS
not an OS

1 word
>drivers
we have a lot of hardware support built into linux and starting from scratch seems like a painful idea

Also, counter point to what you say is that there are projects that are written from scratch
en.wikipedia.org/wiki/ReactOS
en.wikipedia.org/wiki/Haiku_(operating_system)

Microsoft also spent a decade formally verifying all their drivers.

>modified Linux
fucking triggered

The self-defeating name didn't help either.

Names are important. If you name your OS "ShittyOS", people won't use it.

>android
>modified linux
Just because it's build on linux doesn't mean it's a proper linux distro. It doesn't even come with GNU.

>verifying
kek no, they sign them, which means that only companies allowed by MS can write drivers.

Isn't Haiku based on the Linux kernel?
Also both React and Haiku are trying to imitate ancient existing OSs.

>Isn't Haiku based on the Linux kernel?
no

>Implying

redox-os.org/

...

Its about time. Chances are he saves a great deal of time

>CIA Nigger detected