Apple just released an HTML5 Showcase showing what's possible with the current generation of browsers (read: why the need for Flash is overstated).
I think what's clear is that Flash is still the way to go if you want to do vector-based interactive animations (i.e., games), but Flash is no longer necessary (or, at least, should be the fallback) for the things it's been used for in its most popular contexts, like video playback. That's a good thing, since video is a media type that browsers should understand, just like images, and it's in the spec. The sooner people are coding to the spec, the better.
But even the interactive animation context is under assault. With the canvas element, you can already create interactive animations using bitmap graphics and crude vector graphics. It won't be long before HTML5 games start showing up. If Adobe wants to make the case for Flash's long-term relevance, they need to stop putting all their chips in fighting against the video tag - they lost that one when YouTube jumped - and start focusing more on staying ahead of the canvas element. Adding "real" 3D support (ala Shockwave3D) to Flash would go a long way toward that.
Or, better yet, make a Flash-like IDE for HTML5/Canvas, and migrate Flash users to that, and become the de-facto IDE for developing next-generation interactive web content just like they dominated the last generation. Unfortunately, I don't have a lot of confidence that Adobe is that forward-thinking from their track record with products like Director. But maybe.