Of course, this is hardly surprising to anyone with more than two brain cells to rub together. First, go after the small fish who don't have the funds to fight Microsoft in court (Wistron, Velocity Micro, Itronix) or those already embroiled in a large patent lawsuit with Microsoft's partner in this endeavour, Apple (HTC). The key here is not to actually divulge to the public which patents these guys are supposedly infringing, because that way, the public and other companies can't help in invalidating them.
Now that you have pressured a slew of smaller companies into paying protection money, you have the basis to move on to the bigger fish. These bigger fish could, of course, take you to court - but then, Microsoft can claim the smaller companies validated the patent claims, even though they simply lacked the funds to take this to court.
The big fish is now in trouble, since their position is weak. This means they will most likely be forced to pay the protection money, which, coincidentally, is about the same per device as the cost of a Windows Phone 7 license, per device. To make matters worse - Samsung is not an American company, which means the chances of them scoring any victory in an American court against one of America's largest companies are even slimmer.
And thus, without ever having contributed a single dime to the development of Android or Samsung's devices, Microsoft will still get lots and lots of money off them, while at the same time raising the costs of Android to about that of Windows Phone 7. Which, of course, is a total coincidence.
It's funny how some argue in favour of software patents, claiming that software development is impossible without it, even though the fact of the matter is that software patents are a relatively new construct. Up until the mid-'90s, the US courts held the belief that software should not be patentable; it wasn't until 1998 that the courts ruled that a calculation which produced a "useful, concrete and tangible result" should be patentable (State Street Bank v. Signature Financial Group). Up until then, the USPTO refused to grant patents on software; after, they ceased resisting, and accepted software patents. Update: As pointed out in the comments, more accurately would be to say that the USPTO resisted software patents until the mid-'90s, with lower courts sometimes overturning USPTO decisions. Software patentability then developed further in the court system, until the 1998 decision, in which the patentability of software was established beyond any doubt. More here.
So, even without software patents, the computer and software industry flourished before 1998. Basically every computer and software technology we use today is older than 1998, so this means innovation and progress occurred just fine without software patents. Arguing that software patents are needed to foster innovation is akin to arguing that progress was hampered pre-1998.
As I've said before, ideas should not be patentable. A patent should cover an implementation, but since with software the implementation comes in the form of code, the implementation is already protected by copyright. Hence, software patents are not only idiotic, they are simply not needed.
It takes a lot of time and research to write a good science fiction novel, yet you'll see few people arguing that the idea of a space novel should be patentable. Yet, this is exactly what software patents are.