One of the most attractive aspects of this to me is that I would not be developing just for a single device tablet OS. The applications will run across both tablets and desktops, and even a Microsoft "failed" Windows release is sure to see a massive uptake, especially as it will be the default OS on the vast majority of new computers.
It's not as if Microsoft haven't tried to do this before and failed and they're taking a pretty different approach to what's working in the market at the moment.
It's possible that Win 8 can be a success on the desktop / laptop but fail as a tablet OS.
On the flipside, you're more likely to be developing for a lot of different versions of that OS, and – over time – you'll be developing for the lowest common denominator among them.
One of the most attractive aspects of this to me is that I would not be developing just for a single device tablet OS. The applications will run across both tablets and desktops, and even a Microsoft "failed" Windows release is sure to see a massive uptake, especially as it will be the default OS on the vast majority of new computers.