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.
And it gives Linux something to compare against.
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.
And it gives Linux something to compare against.
Where did you get this idea? Companies shipping a Linux kernel never had to answer to the community, unless they want to. They can just ship a custom kernel only to their users and leave it at that.
Works really well for Google and Android where they offload handpicked versions of the Linux kernel to the phone vendors to maintain. In practice they don't bother with it so phones never get any updates after 1 year.
Maintaining custom Linux kernel fork is not possible if you want to keep it up with the upstream changes.
I don't understand what you mean here. It's my interpretation that keeping up with upstream is what was meant by answering to the community. It seems that the GP post was not concerned with this or with following the footsteps of Android. It certainly seems within Microsoft's means to deploy a custom kernel on Azure and then only periodically merge with upstream (I believe they are already doing this anyway).
It'd be much better if they upstreamed the changes instead of fragmenting the entire ecosystem with custom forks.
[1] https://www.zdnet.com/article/linus-torvalds-rejects-stupid-...
This reminds me of DirectX vs OpenGL in the late 1990s and through the 2000s. DirectX under Microsoft rule did not have to answer to any committee, thus the progress and groundbreaking technology that it enabled was really good. Meanwhile, The OpenGL consortium had to argue to get just one feature into the standard. It was very slow.
Another example was .NET vs Java. Even though Java was "driven" by Sun (and then Oracle) a lot of the new features were done via consortium and thus its development was slow. .NET instead got LINQ and other really nice improvements during that time. Now, I haven't programmed in .NET since around 2005 (I don´t do Windows), but I appreciate that they are pushing software technology.
You even get nice IDE features in VSCode.
C# is still a joy to develop with over Java, GoLang etc IMO.
The Linux kernel is great, but there is plenty of room for diverse opinions in user space.
Like CRLF newlines and "\" instead of "/" as path separator?
Jokes aside, MS is not the place I look for "something new" in OS design.
MS Research writes papers that are submitted to various journals (as does IBM, Intel, Google, all of whom have their own operating systems). An obvious example is the Midori project produced lots of novel research much of which has been used, even if the project itself was abandoned. https://en.wikipedia.org/wiki/Midori_(operating_system)
With Windows Phone, they did pioneer quite a bit. It was a phone first, and app running device second. Android and iOS feel like the phone part was a second-thought add-on in comparison. The Windows Phone did its best to make it so you could make a phone call or send a text, at the expense of email, running apps, etc. when the battery was low which took Android and iOS many years to implement. There were probably some other interesting things, but I don't remember them any more.
You'd be wrong, though...