Im not sure how many people thought this would _acctually_ happen. This is one of these things that seems cool from an Linux lover's point of view (because it somehow proves them right that Linux is the absolute best) but makes absolutely no practical or business sense. Two major groups buying windows are business clients (who would rather you not change things so that their 20 year old custom software still runs in Internet Explorer) and the average user (who couldn't care if the Kernel was NT, Linux, or a literal peice of Corn as long as things work).<p>Everyone who thinks that using the Linux Kernel is better than NT is already using Linux. Everyone who needs some sort of intermediary support uses Dualboot/WSL/etc. Windows would never sacrifice it's backwards compatibility. This is the company that won't change the bug that makes excel think 1900 is a leap year so that it doesn't break people's spreadsheets. [0] Not to mention porting windows would take a LOT of work.<p>[0] <a href="https://docs.microsoft.com/en-us/office/troubleshoot/excel/wrongly-assumes-1900-is-leap-year" rel="nofollow">https://docs.microsoft.com/en-us/office/troubleshoot/excel/w...</a>
The two canaries to watch for are a systematic switch in the default line ending character from CRLF to LF, which will be driven by Microsoft developers fed up of git hassles, and a switch in the default path character from "\" to "/". In most places those are already <i>supported</i>, but not default.
What I am really waiting on is a Microsoft buys out Canonical headline cause it seems plausible. It would definitely give them an interesting edge as the primary maintainer of Ubuntu in Azure so there is definitely incentive for it. What I would love to see is Microsoft create something Wine like and charge about as much as a basic Windows license for it.
I don't get all the hate for win10 -- I love *nix from a the pov of a clear, orthogonal internal design (at least from 10k feet) and it doesn't make since to deploy at scale on Win (unless you really need to) -- but why would I ever want to spend time dealing with all the configuration idiosyncrasies and driver issues on a day-to-day basis, if what I'm using a OS for is to host browsers, editors, IDEs, terminal windows, etc., and productivity apps? Okay, win8 was DoA, but win10 seems pretty unoffensive FWICT. I just want to pick up my machine (and just as importantly, someone else's random machine) and have it just work as I expect.
I would not want Microsoft to "rebase to Linux". Why? Ideas, and innovation.<p>If Microsoft come up with something new in operating system design, they don't have to answer to the community when adding it into the kernel.<p>And it gives Linux something to compare against.
I wonder if it’s the other way around. With WSL2 you can give Microsoft credit for actually getting Linux on the Desktop.<p>Probably not what most expect from Linux on the Desktop. But hey, if you still can’t properly support 4K scaling in 2020, it’s never going to happen. It will probably always be a fringe group.<p>Unless something really changes within the mentality of the open-source / Linux community.
Quite right, Microsoft just realised that they should have cared more about POSIX NT subsystem, and since 20 years later Linux ABI compatibility is more relevant than POSIX, they just got into it, following the BSDs, Solaris, AIX, IBM i, z/OS, ClearPath and a couple of other platforms that also offer support for virtualized Linux kernels.
Microsoft already gave up on re-implementing Linux on windows, and went with running a genuine Linux kernel in WSL2. The only way to guarantee a fully backwards compatible Windows layer on Linux would be to run it on top of a genuine Windows kernel, at which point what are you actually gaining? Even if they did do it, I don't see how it would meaningfully change anything, or what advantages it might offer them.<p>WSL is simply about better supporting Azure developers and customers. That's all there is to it.
Reading the article the first time was confusing to me, as an unqualified "Raymond" in the context of Microsoft means Raymond Chen to me, it was only when mousing over one of the links that it finally clicked that it was esr.
I love linux, but I would actually rather see MS open source the Windows kernel than replace it with the Linux kernel.<p>I'd also love to see a "Linux Subsystem for Windows", to make it better to run windows applications on Linux. Wine sort of fills this role, but I'd really love to see a MS-backed fully compatible system. And maybe it could work similar to WSL2, where it is an actual windows kernel running inside KVM.
It almost is, but not exactly a rebase. They are bringing in official SCP/SSH support, proper terminals with unix shells. X11 gui support (with WSL).<p>I feel like Microsoft has read the writing on the wall. Developers love nix-based systems (Mac, Linux). They are trying their best to catch developers at the door with a "but wait! there's more!"
That never was a goal of Microsoft. Their goal is to get Microsoft apps, technology, and development toolsets embedded everywhere, and to a large extent, they have already succeeded in that. Projects like .NET Core, SQL Server on Linux, Visual Studio Code, Windows IoT, all have cross-pollinated to the Linux world and have gotten Linux, macOS, and ARM hardware users to adopt Microsoft technologies. All those investments drive usage of other Microsoft services which results in more revenue for Microsoft, which is all they ever wanted.
There's another angle to this.<p>What if Microsoft created a new linux-based OS precisely to start from scratch and let go of all the Windows cruft?<p>Windows 10 is mature enough to not need any significant work and still maintain backwards-compatibility, but maybe Microsoft is preparing the terrain with a new OS for the next 20 years.<p>Apple is moving fast towards universal apps for mobile and desktop. Maybe Google is doing that with Fuchsia too. Microsoft failed with Windows Phone but it probably learned from its mistakes. I'd be surprised if Windows 10 + Android is their long term goal.
Why would they take a beautiful, modern microkernel OS like Windows 10 and graft it on a clone of an Apollo-era monolithic-kernel OS? You’d have to be nuts to even think of that!
It would be an unbelievable technical headache to rebase to Linux. Think of the bugs that exist in Windows 10 <i>now</i> with their current testing regime.
Between Apple Silicon and Chromebooks, it's looking like ARM-based computing is becoming a thing, or at least enough of a thing that if I was Microsoft, I'd worry about how much Windows is tied to x86.<p>And while Windows does have an ARM port, all the drivers and board support packages seem to be tied to Linux, and Linux based operating systems like ChromeOS and Android.<p>What I suspect to see is a product that combines the Linux kernel, ELF versions of Edgium and Office, a port of the Windows shell that runs on top of this, and maybe support for running Win32 binaries in emulation, with as-it-is backwards compatibility guarantees.<p>If something like this comes to exist, I'm guessing it will be sold as Windows A, or something other name like that. Will it be Windows? That's probably a philosophical question, more than anyththing.
<i>> Raymond is correct in one key part of his blog. I do think the era of the desktop OS wars is ending.</i><p>Mhm. Let's repeat this when percentage of Windows usage will be much lower and there will be more games released for Linux on day one and MS will stop swallowing Linux friendly studios like inXile and Obsidian. Right now I don't see it being the case by far.<p>MS is still very hostile to Linux gaming with DirectX lock-in and platform politics. The above claim just ignores reality.
Well, they did a similar move with Edge! (which is now a Chromium derivative [1])...<p>There's many business reasons against rebasing the Windows Kernel to Linux, but there's also many positives too<p>never say never<p>[1] <a href="https://www.theverge.com/2019/5/6/18527550/microsoft-chromium-edge-google-history-collaboration" rel="nofollow">https://www.theverge.com/2019/5/6/18527550/microsoft-chromiu...</a>
I don't know much about the direction Windows is heading (I have not used Windows since early 2000s), but if there ever was a good time to do it, it would be now that Apple is slowly killing macOS and turning macs into giant phones.<p>If Windows had UNIX kernel and POSIX with custom UI, it would be a viable platform to switch to.
Why not?<p>It wouldn't be the first time that Microsoft ditched it's technology. (DOS - OS/2, Sliverlight - JavaScript, EdgeHTML - Blink, etc.)<p>I'd guess that the OS market is done, they want to get rid of most of their Windows legacy stuff, and focus on making real money with the cloud.
what people keep forgetting is maybe the next desktop will be run as a service. games which are compute intensive are already run on the clould. aws has a desktops product. so yeah microsoft wouldn't switch to linux kernel but would have other microsoft apps stream off a linux server in the cloud.
> It is unclear if the Windows user space could even be rebased from NT to the Linux kernel and maintain the compatibility that Windows is known for, specifically what enterprise clients with mission-critical applications are paying to get.<p>Given how buggy, and unreliable Windows is, wouldn't be this a straight improvement for their "mission-critical" stuff?
Windows is a legacy operating system with an elderly demographic. They just need to keep Windows as Windows for its business and consumer customers without trying to make it something that it isn't.
> Microsoft has plenty of paying customers to continue supporting Windows as-is, some for decades<p>I have to respond to this thought because -- I don't think its true.<p>Aside from government and military, which of those paying customers truly have revenue streams so guaranteed that those entities cannot be replaced or outcompeted over the course of decades by somebody approaching the problem with fresh eyes -- or even just some internal division at the same company approaching the problem with fresh eyes and building something _better_ because it was allowed to change as opposed to something very stable because it is _not_ allowed to change?<p>For government -- do we really want/need our government throwing money away because they are more or less not allowed to development new solutions to old problems?<p>For military -- ... yeah well ... I don't know what to think about military ... Military is not a sector that has any measurable goals aside from self-perpetuating its budget so its hard to argue against boondoggle maintenance projects as having some capacity for objective success with regard to that goal ... But -- perhaps there is a future where the system could reasonably decide that the wisest course is to _not_ spend billions of dollars keeping 1980s era visicalc programs running unmodified so that they can properly manage the floppy disk supply chain for nuclear missile silos ...