Speaking today at the Gartner Symposium ITxpo, Nadella stated:
“Windows 10 is a very important step for us. It’s the first step in a new generation of Windows as opposed to just another release after Windows 8. General purpose computing is going to run on 200 plus billion sensors. We’ve architected Windows where it can run on everything.”
In all seriousness, though, Windows wouldn’t be Windows when run on (most of) the hardware NETMF runs on.
Nadella is making a marketing statement, a product strategy statement. He’s saying that MS isn’t giving up things like phones and tablets, ARM devices, essentially, and Intel devices that play in the same space.
@ godefroi - Agreed, though I wouldn’t go so far as to say “silly.”
But there’s no reason whatsoever to try to shoehorn Windows into an mCu board.
I’m definitely happy that there seems to be more movement on the NETMF front, without worrying about whether or not it fits in the Windows shoebox.
While I haven’t jumped in yet, I am intrigued by the Galileo and Edison projects, and will be interested to see where they go. I’m also extending my knowledge beyond NETMF with mBuino and some Arduino-type stuff (in my copious free time ).
I think that having one Windows that runs on everything from servers to XBOX One to laptops to tablets to phablets to phones is quite ambitious enough, thank you very much. If Microsoft is able to pull that off, it will be quite the feather in Nadella’s cap, even if some of the work undoubtedly started under the much-maligned Steve Ballmer.
By that logic, Linux would be Windows since you can write .NET code via Mono. Really, I think we’re stretching things. NETMF is fine being just NETMF. It doesn’t need to be Windows.
I don’t know why… Linux runs on everything from giant server systems to ARM7 microcontrollers (ucLinux). It all comes down to what you define as “Windows”.
Well, the Windows kernel has been running on all those levels of devices for some time, now, so that’s not exactly new. What’s new is building a UX paradigm that can work across all those devices, along with an app model that makes it easy for devs to target them all with a single codebase (or at least with massive sharing of code, and minimal targeted UX work).
Yes, of course Linux can run on machines from small to large. But what’s the developer story there? Not something I’m interested in diving into, personally. But YMMV.
@ devhammer - Like I said, it all comes down to what you define as “Windows”. If “Windows” is a GUI, then we already have a non-Windows OS from Microsoft running on big systems (Windows Core).
Also, if there is one lesson to be learned from Windows 8, it’s that there isn’t a single unified UX paradigm that works for desktops and small systems (or at least, we haven’t found it yet).
Does it really matter if Windows, in whatever form runs from an ARM M0 all the way up to a 64 processor server? Are we not more interested in making it easier for the embedded processor to talk to the server? Isn’t the issue really about communication? Yes, it is nice to use the same development language from top to bottom. There is a danger however of pushing that too far. Languages usually evolve to solve application specific problems. HTML5 does is fine for web pages but do I want to try to use it to program my embedded processor? (Yes I know thats a dumb example but its all I could think of. :-[ )
There is no place, or possible way, any form of Windows will run on an M0 microcontroller or even an M4. Do you even need a OS for an M0? Maybe, but it would be tailored specifically for it and the application.
I think the focus should be on making the communication secure and seamless from top to bottom. If we have protocols, libraries and code example that make it easy for my C code in the embedded device to move data to the C# server and JavaScript/HTML5 web interface I would be happy.