NEW PARADIGM

What we need is a Hitler license, one which is basically two clause BSD but has a third clause requiring Hitler be praised, and commie jews gassed, visibly on startup. If it's a "web app" it must be on the landing or login page.
"Anti-Semitism" is kryptonite to the corporation.

Also Zig Forums masterrace you commie lain faggots!

I want that license.

Why isn't there a single hardware abstraction layer that all hardware producers would be forced to use? So you would have many audio or lan cards, but all of them would have to use same interface, for OS every card would be managed in same way.
there would be no (((drivers))) that have direct memory access and other shit, only a default system driver.
and what's even the point of so many different lan cards? they all are the same. the only difference is speed they support and backdoors they implement.
instead of having 1000 different capitalistic lan cards, we could nationalize that and have one communist 100mbit card, one 1000mbit card etc.

but not millions like stupid linux

open source design choices are bad

You have been watching too much Hollywood.

There is. It's called a kernel. You can read more at kernel.org.
So regulations are terrible until the free market doesn't give you something you want? Imagine if every processor was forced to implement some government mandated interface. We'd still be using 8086.
Then go back to windows, faggot, and see how much better it is. Or better yet, try redoing it yourself and see if you do better

>twitter.com/zerodium/status/1039127214602641409
>mozilla.org/en-US/security/advisories/mfsa2018-24/

>(((javascript)))
why am I not surprised?

was javascript made in israel?

Stallman will set us free, anons.

Chuck Moore's colorForth was well under 100K lines of source (Probably under 100K bytes), and had all those features, plus USB. There's even source code for Intel 10/100 Ethernet cards and a TCP/IP stack floating around. Terry is a fucking fraud

Nice visionary post OP. But count on Zig Forums fanboys to miss the point as usual. What's your answer though, OP? How do we get from the current shitty situation to at least something decent?

openbsd solves most of your issues. they're obsessed with code quality. the whole os, not just the kernel, but the whole os, has the be approved of by dictator theo. there is nobody dropping in to shit on the system, instead the team does everything they can to unshit various pieces of software. like x, and openssl

Cool stuff tbh.
IDE hard disk driverbsy 1f7 p@ 80 and if bsy ; then ;rdy 1f7 p@ 8 and if 1f0 a! 256 ; then rdy ;sector 1f3 a! swap p!+ /8 p!+ /8 p!+ /8 e0 or p!+ drop p!+ drop 4 * ;read 20 sector 256 for rdy insw next drop ;write bsy 30 sector 256 for rdy outsw next drop ;

If I recall correctly the tiniest and yet most featured OSes are all written in assembler and/or pure machine code.

shill

By understanding how we got here.

Here's a non-troll posts I made in

Reading material:
kalzumeus.com/2011/10/28/dont-call-yourself-a-programmer/
This is why Haskell isn't used much and Javascript grows like a cancer.
AKA, the pajeets or SJWs.
Remember what I said about exploiting white 20something newgrads?

If you want to understand Silicon Valley, read Peter Thiel's Zero To One, attached.

Yeah I love my nice free and open computer made and designed in venezuela/cuba/north korea, oh wait there is none. The ones made are made by capitalist corporations in fake communist China.

Does is have a flight simulator? Can it talk to god?

This overly complex bs vs 8-bit unsigned encoding with support for spinning 3d tanks embedded in source code? TempleOs wins.

HolyC compiled into x86 machine code vs ColorForth compiled into bytecode that has to be executed by a virtual stack machine? TempleOs wins again. (I might be wrong here, won't bother waste more time on this)

No interrupts vs Have you ever written an interrupt routine faggot?

It was interesting, but comparing it to TempleOs is quite frankly an insult to what Terry has done.

Glad I scanned your post and saw this so I could skip reading the rest of it.

8 bit ascii is why colorForth does in 1% of the code what takes TempleOS 100K lines. ColorForth is not overly complex. "support for 3d spinning tanks embedded in source code" is not a templeos feature. It's a feature of the hyperdoc (whatever it was called) format. stupid shit that can be implemented by anyone.
HolyC is interpreted, colorForth is pre-parsed and compiled with practically no overhead. The "virtual stack machine" runs more efficiently than TempleOS's bloaated compiled output, since it's not running an optimal single-pass optimiser on every line of input. You can implement nterrupts in regular Forth ( wilsonminesco.com/0-overhead_Forth_interrupts/) porting it to colorForth wouldn't be hard, just a waste of time as they literally don't matter.

Chuck Moore uses colorForth to draw the doped substrates for the VLSI designs of his own chips (ShBoom, F18A (GA11)). ColorForth has TCP/IP and the simplest possible codebase. ColorForth uses pre-parsed input to implement one of the simplest interpreter/compilers (and single-pass optimisers) in programming. The system is so simple a pig could be taught to write clean colorForth. What did Terry do? Write another C compiler at the age of 50? Wow. Read Chuck's book faggot. colorforth.github.io/POL.htm

pasting the code here doesnt make sense since it should be in color. But Chuck Moore is a genius.

also youre fucking stupid. what is 4 + 28? wonder what that number correlates to

I got really drunk last night.

Terry was probably more around 30 years old when he started his OS. And it's not a C compiler, give he doesn't care about the C standard.

Attached: LoseThos510.jpg (640x480, 74.8K)

Attached: out.mp4 (640x360, 11.69M)

It took him 18 years to write an OS? You worship this guy?

I don't worship anyone, but I like that he tries to recapture the 80's hobbyist era. It's more authentic than the raspberry pi shit and other commerical so-called "retro" products. And it didn't take him this long to write it. He had other versions long ago with different names, but pretty much the same deal.

It didn't take long to write the OS. It took 18 years to refine it.

Don't shit on the rpi I turned mine into a thinkpad flasher and boot-lock disabler.

Truly Lord Theo is Baste!

The Thirty Million Line Problem - Casey Muratori (youtube.com/watch?v=kZRE7HIO3vk)
polite sage

I feel bad for getting involved in this.

The way to manage millions of lines of code is to:

a) refactor
b) create libraries that can be individually audited
c) accept limitations and develop automated tools for fuzzing and auditing that exceed human limitations

linux is millions of lines of code, but you don't need every fucking module or every fucking branch to build it, for your platform your only looking at a much more managable amount of code and drivers, which you can put on your own tree and audit to your heart's content

3. you clearly have schizophrenia. I don't know which lolcow you are but I'm sure someone will figure it out over the course of this thread.

if you’re the colorforth guy, can you supply me with any OS made with this stuff? i know moore learned under mccarthy, so maybe a forth + lisp machine is only natural

it works great on openbsd

>kalzumeus.com/2011/10/28/dont-call-yourself-a-programmer/
Interesting read, thank you.

I wonder whether there's an easy way to measure how many lines of code are in a kernel compiled with specific options. I suspect in my case it's more along the lines of tens or hundreds of thousands of lines of code, rather than millions. I compile a custom kernel, and I remove support for tons of stuff I don't use, like most of the filesystems, network drivers I don't use, etc. On my desktop, I don't use a wireless card of any kind, so I got rid of mac80211 support and all drivers for PCI and USB wireless cards. That's probably tens of thousands of lines of code right there.
One could theoretically do that, yeah. But most people on Zig Forums can't audit shit. I can't.

Here's the problem though - HTML/CSS is cruft upon cruft, with shit being added to it with every standard. Are there any alternatives at declaring simple web layouts that you describe? Right now, it's just flexible boxes inside flexible boxes. If something like that exists, it's just a matter of bolting a good API on it for realizing AJAX and whatever other things one might need. I think modularity is the key to browsers. Prototyping a minimal program that is responsible for gluing everything together, where you could swap out a JS/layout engine would be a huge effort already.

U R A BASTARD. in case u didn't get it, I called u a BASTARD. }}??? : it means i understand u have trust issues - u were born to a whore unaware of who impregnated her.{{
ur rant doesn't belong in /tech.