RR, while your analysis is keen, it seems to stem from not having 'got' WinFS. Think about how you use your computer and categorize your data. In order to 'find' information, you need to use the right app, name it the right way, store it on the right partition
and in the right folder structure (often 3-5 levels deep in order to sort the volume of data we all work with on a daily basis).
Now, while you may rarely use the Search feature, you are constantly searchING. You are trying to remember what app you used, what you called the file and where it's located. It happens as part of your internal process.
WinFS fundamentally changes that. I'm not going to get into the whole "is changing users' processes good" debate (I have that one with Buzz B all the time, it's always fun). You asked why it's useful. WinFS is fundamentally useful because you never need to
do an internal search again. You could use the wrong app, name the filename all wonky and store it somewhere useless and still get a snapshot of whatever it is you're working on.
Want all data for your "NextGen2.0 Website"? Here you go. Instantly. Oh, look, there's a file with the schematics and process deifnition. I'd forgotten what I named that. Cool, thanks Longhorn!
Beyond WinFS, the largest benefit of Longhorn will be the ability to tap into the more advanced features offered by Windows 2003 Server (and, eventually, Longhorn server). Windows 2003 makes a hell of a lot of sense, and Longhorn will give you the ability to
interface with that seamlessly.
Beyond that? Security, greater integration with DSI, etc are all important for your business.
The whole innovation thing has always bothered me, personally. If WinFS isn't an innovation, how can the iPod be called an innovation? There were metatadata schemas before (WinFS) and there were MP3 players before (iPod). They were both evolutionary and, in
my mind, both innovative because they both take it to the next level.
Personally I see things like DSI, .NET, WinFS, Avalon, Indigo, VS Team System, XBox, etc as innovative. That's not even counting softer innovations in product lines. I see those as just as innovative as anything else out there from companies like Apple, Sun,
Novell, IBM, Oracle, Siebel, PeopleSoft, etc.
I don't really see a lot of "wow, that's completely new" innovations from any of the big companies. From the new companies, sure, but not from the big ones.