Apple comes out with an interesting digital music service, and they become a media darling. Customers like it too, to the tune of 2 million tracks purchased in the first two weeks.
Microsoft was working on their own, subscription-based alternative for a while now. It was supposed to ship ahead of Apple's music store. They're also playing catch up, and trying to fight the meme that Apple has put out there: we've purchased music for decades and want to continue to do so; "renting" music is just a bad idea.
Among the things Microsoft has done are: improve their DRM technology, work out how to make subscription based services work with DRM (i.e. timeout features) and work with multiple vendors (i.e. provide a "standard" supported by 15 portable players).
From a project management/product management perspective, Microsoft did everything right: talk to the stakeholders (customers, consumer electronics vendors, music labels) and come up with a solution acceptable for all parties, support diversity in the market place (multiple music rental sites, multiple MP3 player vendors), and generally try to please everyone.
Yet no one seems to want what Microsoft is working on. Furthermore, Apple's stovepipe model (iTunes store only works with iTunes 4, which only works on Mac OS X 10.2, which only works on newer Macs; the only supported portable player is the iPod) is a big win in the market. Superficially, Apple looks like it is doing all of the wrong things, especially when it comes to expanding vendor lock-in.
So, what is Microsoft doing wrong? What is Apple doing right? Or did Apple just figure out that the "stakeholders" in a digital music service are irrelevant, except for the end-user who just wants a compelling "total product"?