I'm most familiar with the example of the Kin, and it was not an example of "
experimentation and innovation", but a case study in how Microsoft has largely lost its ability to execute on a proven concept.
For that matter, the Zune was a (late) iPod clone; sure, it had some experimental features (that didn't pan out, but, hey, they tried), but it was hardly seriously innovative.
It also was yet another example in how you'd best not trust Microsoft, as part of it they threw their PlaysForSure ecosystem under the bus.
Hmmm, getting whole sectors of the world to trust Microsoft in the future is going to be a tall order for the next CEO, assuming he doesn't just punt on many of them, which might be the wisest thing to do.