One other reason why people are using Windows instead of Linux is the consistency. While there are some changes (xp->Vista->7) most of the stuff is still intuitive and most things are where they used to be.

On Linux, and particularly Ubuntu, the whole thing changes arbitrarily. This might be OK for some people, but some it sucks. I am pretty good with computers, been using them for 17 years or so. I am also a developer, and I am used to multiple OSes, and I still don't like changes for the sake of change.

So microsoft will release windows on all form factors (pocketable, tablet, netbook, notebook, desktop, servers, etc) and they will have a single appstore that handles all their software distribution.

One can purchase a license for, say, ms office on their desktop and when they use a tablet, she can login to MS' appstore from the tablet and downloads her copy of MS Office for ARM on the tablet.

It expands the playing field for them, their customers and their partners/developers.

Windows will have a lot (*LOT*) more difficulties in doing a centralized appstore a la MacOS. For MS products, yes, maybe, but for generic 3rd party apps ?

I see the Windows franchise beginning to sink at this point, as well as MS. This move seems good on paper, but the paradigm is changing and the windows brand that has been established for literally decades is going to have to undergo a phenomenal shift in order to be accepted on anything but a desktop OS. Not to mention that there are HUGE technical hurdles to overcome with this transition, as well as entry into a competitive space with major players already firmly entrenched and gobbling up new users at a record pace.

I don't see it being a total fail, this move simply stinks of let's-lean-on-our-established-brand-to-remain-relevant. I can't see MS taking a leading position in this new world of computing -- not even second place. Remember, with an architecture change, legacy application support will vanish, and the new Windows will have to start from scratch with no apps with competition that has tons. AFAICT, legacy support and brand recognition have kept customers locked into the Windows bubble.

Of course, if they pull a NaCl with some funky x86 to ARM translation, they may be able to surprise the critics (me). Though I'd be surprised if anyone could use a Win32 app on a low-res phone...

If they are just capitalizing on the trend of ARM slowly entering the High Performance Computing market, and still targeting desktops, then this is something different entirely. Interestingly, there is still relatively very little established desktop OS competition other than OSX and now Chrome OS. I have my doubts that this market will remain both the same and the same size.

But I'd bet that MS would still be beat (about the ears) in the server market by an army of Linux variants.

Look whether you like it or not a feature on desktop pc's doesn't become relevant unless Microsoft introduces it to the masses. While you can run Ubuntu on Arm but wheres the proprietary drivers and software the only run on x86?, where the Adobe flash? The importance of Windows on Arm is that it will make arm based desktop/Laptops/tablets popular and bring key Applications like drivers and Flash to Arm based desktops.

OH SNAP ... NOW I GET IT!

Microsoft's strategy for Windows 8 is to bring forth two different operating systems.
The first Windows 8 will be built for the ARM platform and it will support new programs made for Windows 8.
The second Windows 8 will be built for the x86 platform and it will support new programs made for Windows 8 and it will crudely support Windows 7 and Vista programs (and perhaps Windows XP, exclusive to the Ultimate version).

(this is not like Windows and Windows CE, since Windows CE software were unsupported on Windows)

If someone developed a Windows 8 software using the tools Microsoft will make available, then the produced software will work on both ARM and x86 version.
If someone developed a Windows 7 software using the Windows 7 tools, it will work on Windows 7 and Windows 8 (x86 version).

Do you see it, Microsoft can create a cross-platform architecture and keep the legacy codes, it just needs a sneaky tactic like the one mentioned above.
I would be very surprised if this is not the tactic they will be employing.

I highly doubt Win 8 will have a different architecture than Win 7. And no sane developer would write programs that only work on Win 8.

If what I deduced earlier is Microsoft's goal, then I highly doubt your comment.
Why?

If I were a developer, and lets say I am targeting laptops I would choose to develop for Windows 8 instead of Windows 7.
+Because that way I know my software can be in service for longer.
+My product will have less competition (Windows 8 is new, there's less software made specifically for it).
+My software can run on more machines (the ARM built devices, the x86 devices, and x86 devices that are upgrading).

But you are not a developer. I am a developer.
No (read; very few) windows developers developed specifically for Vista until a few years after it was launched, because why give up 75%-50% of the market share?
Developers don't have unlimited resources, and they'd much rather develop for something that has a bigger market.

Of course, some will develop for Android or IOS, but those are different devices. If you develop for desktop, and you must invest money in the development, then you go for the mainstream. This is why there are so few Linux and OSX applications compared to windows.

One problem is that x86 is x86 while ARM is a whole jungle of different configurations.

So, Windows on ARM, is simply a question of what configuration it will run on, what will be the standard. If it happens, it will probably be on Intel chips and this will set the standard for both Intel and MS on ARM.