I dunno. Google's already had bright minds working on it for some time. Trying to do it in a browser will probably always be inherently problematic. Why they don't launch a standalone "Google Voice and Video" client... if there's an opportunity there, and they really want to own it, that'd be the route they should go, not more browser-plugin stuff.

Heck, perhaps this is part of what facebook's recent spate of desktop-engineer-hiring is going after?

Part of what I see is that Google does not have a good model for making Google Voice profitable given their revenue model because it doesn't drive eyeballs to Google search and the advertising customer base, and because datamining the communications of Google Voice's users for the sort of keywords and profiles useful for search has a huge potential for social blowback - imagine how some people would react to web page ad placement based on last night's booty call - not to mention the possible legal problems given the diversity of wiretap laws, e.g. state by state in the U.S. never mind world wide variations.

[Edit] Given changes in Google's leadership and the apparent emphasis on core function which seems to be underway along with the limited rollout which Google has done for Voice, I would not be surprised if it is destined for the same fate as Google Health over the next few years.

Of course it drives advertising revenue -- in the same way GMail does. More people using Google accounts every day is a plus for Google.

I think "probably always" is not a safe phrase to use when talking about the future of browsers.

And whether or not you agree with that, I think it's clear that's where Google's money is.

Uh, how is a browser plugin different from a client-side application? They're both running native code. I don't see any technical problem.