What if Microsoft open sources NT? Just NT and not the rest of the things that come with Windows. Apple open sourced XNU and a bunch of other shit>https://opensource.apple.com/
>>110275163>What if Microsoft open sources NTit would be very useless on it's own
>font rendering in the kernel
>>110275216linux is pretty useless on its own. you usually download a customized GNU with a linux kernel. what if someone made GNU / NT or whachamacallit / NT or whatever
>>110275244priorities and having standards
>>110275163They don't have to. Just let Claude reverse engineer it from docs. We can call it Windows NT-Reverse (NTR for short)
The IO subsystem is dogslow, linux runs circles around it.
>>110275163>gamesrun on linux>officemigrated to cloud, runs in the browser now>software devliterally a linux VMWindows will be theseus'd into another linux distro.
so microsoft kernel """"developers"""" are beta cucks letting their work freely ruined by hindus and hr foids?LMAO
>>110275163Extremely based and factual statement
>>110275329..... or just use Windows after purging all the trash. I've using a debloated 11 since 2022>Windows will be theseus'd into another linux distro.I don't think NT is going anywhere anytime soon
>>110275256>We can call it Windows NT-Reverse (NTR for short)
>>110275163They are right about legacy Windows being interesting. XP apps were lighting fast because they used a shared toolkit and under the hood options to solve problems.Low level windows code I assume means a light fast program that works on Windows 2000 to 11.
Remember that guy who made a ultra light notepad for Windows 11. He did that just by calling existing libraries and building them into that. Where as most normie devs use chromium electron apps for hello word. Mental Yakub Outlaw weeps at this.
>>110275428>Low level windows code I assume means a light fast program that works on Windows 2000 to 11.if you write WIN32 and remain feature locked to Win2k APIs and use old MSVC then yeah. I doubt binaries outputted by recent MSVC would work on Win2k even if you don't use any recent WIN32 APIs
>>110275163>muh magical NT kernelWindows sucks at I/O, networking, scheduling, sandboxing and these are functions that reside in kernel. This was confirmed by countless Phoronix benchmarks. Apart from that it's difficult to compare Linux and NT kernel directly since Linux is a monolithic kernel that houses basically all the drivers and subsystems while NT kernel does bare minimum and all relevant functionality moved to another systems including third party drivers.>>110275246>linux is pretty useless on its ownWrong. You can run minimal system consisting of kernel + some userspace applications (this is basically average embedded Linux setup). It's NT kernel that is useless on its own: it doesn't have drivers, it doesn't have anything, it can only operate as a part of bloated Windows OS. Microsoft even tried to create new debloated version called Windows S but failed miserably, presumably because at this point entire Windows codebase is a spaghetti code that even seasoned MS programmers don't know how to deal with.
>>110275451you have 0 idea of what you're talking about. sad.
>>110275438>I doubt binaries outputted by recent MSVC would work on Win2k even if you don't use any recent WIN32 APIsVisual Studio 2026 only support Windows 11 and later and binaries it produces can't run on Windows past 10.
>>110275461>saar, you don't know what you talking about, windows de best!!!
>>110275494the fact you had to turn this into a windows vs linux fight in your head is also sad.I would joke that you're 12 but in reality you are 100% a grown man with an underdeveloped, childish brain that needs to pick sides and fight over the thing you like being objectively superior to the other thing
>>110275163>Apple open sourced XNU and a bunch of other shitYeah and it's pretty much useless.I remember PureDarwin being basically dead. It seems to be "less" dead now but no one cares about it still.
>>110275526unlike NT, there's nothing interesting about darwin. it can't even run macos program
>>110275539nobody did anything with the leaked windows xp source
>>110275539it's pretty interesting. it can't run a macos application because the userland is closed source, which would be the exact same case if only NT was open sourced.there is very little incentive to recreating the entire userland from scratch on top of either kernel, which is why it will never work
>>110275521>0% arguments>100% buzzwordsyep, certified MSTF shill
>>110275591kys at the earliest convenience
>>110275582the threat of a trillion dollar company gaping your anus kind of discourages that from happening
>>110275582Legally no one can do anything with it. Windows' source code is already semi-public but being publicly accessible does not mean you can use it for anything you want. The only people that would touch it are Windows devs trying to develop software for Windows.ReactOS or WINE devs cannot touch it without contaminating their own projects. The only people that care about leaked source code are hackers and Russians/Chinese/Indians who will make illegal bootleg versions of Windows
>>110275163Dave Cutler was right about UnixIf only we could go back to Windows XP
>>110275616To your points, it’s not even a case of just using it, you can’t even suggest or intimate that you have so much as looked at it.
>>110275616>Russians/Chinese/Indians who will make illegal bootleg versions of Windowsif they dont care why would anyone else? Even without using proprietary code nobody bothered to do cool kernel extensions using non public internal apis
>>110275656I found some rando who claims to have Windows NT 4 on ARMv7 using the power of AI sloppa. NT 3.5 and NT 4 were among the leaked stuff from 6 years ago
>>110275163Threads like this only exists to create Linux vs Windows flamewars. In reality, you can't even compare Linux to NT since these two kernels have completely different set of features and different way of deployment.
>>110275670That's believable considering Windows NT 4 was available for many system architectures at the time.
>>110275689You say that as if it's an impossible feat. We got ARM 64 bit today and had ARMv7 with Windows 8 back then, there has been Itanium Windows as well.picrel is a very early Windows 8 build running on a galaxy s4 (the power of ai sloppa)https://x.com/xandreithefrog/status/2117023193261080680
>>110275685>Linux vs Windows flamewarsImagine being this autistic to get triggered by that crap
>>110275717>Itanium Windowspity the poor souls who tried to support and develop for that
>>110275256
>>110275717Uh, no, I said it was believable for Windows NT 4.
>>110275656>Even without using proprietary code nobody bothered to do cool kernel extensions using non public internal apisNo one bothers to make anything cool because Windows is such a great system where if you want to deploy something kernel related you need a permission from a big daddy Microsoft. Otherwise, you have to edit bootloader config to load system in a dev mode for loading unsigned kernel drivers. Last time someone tried that (Denuvo hypervisor bypass) normies were scared shittles, even though source code for driver was shipped alongside driver itself and everyone could read and build kernel modules by themselves (well, by everyone I mean competent people, it's not like average wintoddler knows anything but how to double-click exe).
>>110275163>What if Microsoft open sources NT?It would help the open source NT project: https://github.com/LiberNT/LiberNT
>>110275163The slop running on top is what actually makes it Windows. If you rewrote everything else, you'd end up with a different OS that would have similar compatibility to WINE
>>110275771How? NT 4 never released on ARM not even the ARM they had at the time
The NT kernel is not what makes Windows good, it's the other parts of the OS built around the kernel, and the way they're architectured. The parts that make modern Windows shit tends to be superfluous userspace garbage.
>>110275163>it's the best except for all the things that suckYou can say this about literally anything.
>>110275801>to compile NT kernel you need Unix environment
>>110275879>The NT kernel is not what makes Windows goodI'm sure if anything makes Windows good at all it will only be the kernel lol.
>>110275670perhaps armv7 can run wince mockup
>>110275872If the OS was available for many architectures then it means it was most likely designed with a certain ease for multi-architecture development, making the realization of an ARM port easier than, say, later versions of Windows that focused on x86 and Itanium.
>>110275163isn't it already open-source? IIRC it is (for some clients/orgs), just not FOSS.I'm pretty sure you can find it in the internet... hell, people were building early versions of Vista in this very same board some years ago from some leaked source code.
>>110275616Have one LLM write up a spec based on it. Then have another LLM write up an OS that fulfills that spec.
>110276675Modern operating systems, including NT, got that thing called a hardware abstraction layer where you put platform dependent code and the rest of the operating system works with it. you probably could port NT 3.5 to run on RISC. it's the same operating system, but slightly older>making the realization of an ARM port easier than, say, later versions of Windows that focused on x86 and Itanium.they are the same operating system bruh.
>>110275163>2026>still can't handle >5 buttons on mouse and need to do vendor hid parsing shittery or retarded keyboard macro binding shit
>>110275163>What if Microsoft open sources NT?High IQ individuals posters from xitter and /g/ will unite over this topic and discuss it until there will be some topic limitation. thats what will happen. that will even happen without open sourcing anything.
>>110276743Yes, it's so simple that ReactOS is successfully working in all of its ports.
>>110276743to run on a RISC-V*>>110276777microsoft simply didn't care about other architectures at the time. back you had>under powered shit like arm on nokia phones>proprietary server shit like power pc>obsolete shit like dec alpha>intels itanium
