If Live actually works and if it actually works in Firefox as well as IE and it works well, then Live will have been worth it. Otherwise it's just more roadkill on the journey to a world of cross platform web apps. It's amazing that a company with the resources of Microsoft can't be bothered to hire a few people to create Live as a set of cross platform web apps right from the start. Old monopolist habits die hard I guess! It's not rocket science and there's plenty of prior art (i.e. flickr, Gmail, et al!). Oh well, with "cool, smart people who get things done" like Joshua and Dare working there, I am sure that this lack of cross platform readiness is just a very temporary interregnum.
First, I think Bill sets the stage nicely by explaining the age of software on your PC is over. There is an explosion of devices, and the only way to tie them together and give the next wave of features that people want is with software as a service. I don't feel the discussion of this as a "sea change" is very convincing -- as he mentions, this vision is a lot like the 1999 company meeting, and the decks look a lot like many of the .NET decks. Maybe it comes from having worked in Netdocs and MSN and seeing lots of "service-izing" of Windows and Office firsthand, but it doesn't seem all that new to me.