I wonder if Android and the iPad keeps Steve Ballmer awake at night.
The iPad is successful, not only as a consumer device but also as an enterprise device. If the results of the iPad are anything to go by, the Tablet is poised to be a very popular form factor.
But the iPad is effectively just one device (albeit the first one) in this new emerging Tablet space. It seems the majority of these devices will be running Android, and none seem to be running Windows.
Now consider this. Windows has risen to be the dominant enterprise operating system, despite of the fact that for most of it's existence the Windows platform has been inferior to the Unix/Linux/Mainframe platforms.
Consider how first DOS then Windows started out purely on - what was then - the low end of the computing market; the Desktop, and gradually started moving into more and more sophisticated applications, with each version getting incrementally better until Windows Server 2008, which can be used quite safely for high end computing tasks.
It was able to do this, partly thanks to Moore's law but more importantly by capitalizing on the familiarity of Windows by the people making the decisions, not to mention the shallow learning curve offered to people who had practically grown up with the OS.
Windows is an example of a disruptive technology, clearly illustrating the fact that's it's often easier to drive a low end product into the high end, than a high end product into the low end thanks to mass appeal.
Unfortunately Windows is no longer a low end product.
Windows 7 is optimized for desktops, it's not really designed for Tablets and touch screens, and so far it hasn't really featured much on any of the upcoming Tablets.
Now iPad aside basically every new Tablet is running Android, a low end operating system and based on the iPad's success; the Tablet is the new disruptive technology.
So lets suppose for a minute that Android becomes the dominant Tablet OS, and that the Tablet becomes popular enough that at least most knowledge workers and decision makers have one.
It's not inconceivable that those same knowledge workers and decision makers may easily decide to start using Android for other tasks, initially maybe for things like cheap Android powered terminals (even terminals calling Windows Apps via Citrix or Terminal services) for call center agents and then later for full desktop replacements.
Now you may be thinking "Huh! Android is a mobile OS" but consider that if it made the jump from a phone to a tablet, why couldn't it jump to a full desktop? I don't think many people envisaged Windows getting onto the server either.
Of course this is just speculation, the only successful tablet so far is the iPad, and it's too early to say if any of the multitude of Android Tablets (most of which haven't even been released yet) will be as successful. Furthermore it's still early enough for Microsoft to develop a compelling product for the Tablet space.
Microsoft - despite it's wide market range - Is dependent on it's desktop dominance. Steve Ballmer must be a very concerned man.