...the multi-touch (MT) interface (forget whether it's Apple's or Microsofts or Jeff Hann's, it doesn't matter from this perspective) is __huge__.
Why?
Because for the first time we have more than one human computer interaction device operating at the same time. I've been building software for over 10 years and it _always_ gets tricky when it gets to the UI.
Multi-thread all you want in the background, but when you want to paint something to the screen, you have to queue up a request for the __single__ UI thread. The display is sacred, only one thread can write to it at a time (For display read "window within an application."
Now look at MT...
Mutliple pointing devices (fingers, brushes, you name it) operating at the same time. At first I said to myself "Wait, but even though you 'pinch the image' to shrink it, it's really just one UI operation right?" Well, sort of, but then you see all the demos which show multiple things happening at the same time. I can be growing one image at the same time I am rotating and shrinking a video file. If I happen to have 4 hands I can do all that myself, or two people can do two different things, at the same time, on the same machine, using the _same_ display. I don't recall ever seeing that. Terminal Servers don't use the same display. Same machine, but different (virtual) displays. X-windows, same deal. Even on OS X or Windows if your mouse is dragging window A it can't be resizing window B at the same time as Window X is having text entered into it.
Speaking as a technical architect, I'm very excited at the potential to turn the world upside down with multi-touch technology. It throws away the "one input at a time" model we've lived with for so long.
And note, I'm not talking Apple versus Microsoft versus anyone else. I'm talking changing the way we interact with computers, at the technical level. The tools, APIs, everything which supports this is going to make us design software very differently. Who knows what we can do with it till we try.
Be well!
t