Wave is computing 2.0. And I don't say that lightly.
It aims to do no less than gut and replace the core folder, file and application metaphors. It's a Newton-esque data object soup + internet, with the role of applications and workflow largely relegated to anthropomorphic services.
That's why the "wave = a + b + c" metaphors ring so hollow; it's a new conceptual model for computing that does not directly map to the old one.
And that is what makes its future so murky.
I think the browser and intervening decades have changed habits and expectations enough to give Wave a better chance than those who've presented such things before. But I also think the single smartest move Google did, was enable the building of seamless bridges in-to and out-of the Wave model.
Being Google will let them get their foot in the door. But allowing people to make a wave from a simple word document and save a wave off to a PDF or simple email will give them time to make their case.
But it's still an uphill battle.
I'll be very interested to see if Google releases some facsimile of Google Docs that are Wave-enabled and allow for full real-time editing/collaboration. Combine that with the email & SMS killer, and I believe Google has a viable alternative to Office + Exchange + Windows. The unifying element of the Wave acts to capture a company's workflow. In the end, business will want to use this instead of the MS solution because it provides more business value that the sum of the parts (IM + email + office suite + workflow), and it does so in a way that MS cannot. Companies can extend the Wave to add further business value to the whole approach with the Wave API. Finally, with the whole thing being open-sourced, companies will be able to run their own servers (something I think is critical to mass acceptance of the technology within corporations).
It is an ambitious project. I hope Google succeeds. I am watching it closely, and look forward to my own invitation. I already can see a number of areas where I would like to build my robots & gadgets.
The question is do we really need to replace email? It has obviously served us well for so long. If you think about the "what would email have looked like if invented today?" question that Google was trying to answer, you start realizing that there are so many limitations in the current model.
How? I think Wave will be a win when everyone with a Gmail inbox has a Wave inbox automatically. Then, if you're communicating with other Wave users, you get tons of added benefits (which slowly everyone else is going to want). Otherwise, they can simply interface it to send good old fashioned emails to "old" addresses.