Here's a take on the future of HTML5 from a leader in the gaming community:
"It's very early days for HTML5. If you think about the order of it maturing, it's first going to mature for periodicals, magazines, newspaper-type applications. The Financial Times already moved there and the NYT already moved there. A year or two beyond that you'll start to see other simple apps like the Bloomberg financial apps, weather apps, that kind of stuff all makes sense for HTML5. Beyond that, another year or two down the line, you're going to have simple casual games that are going make sense to be programmed in HTML5. Glu's approach is to play at the exact other side of the spectrum here. We build apps that require 100-200-300 megabyte downloadable clients. Things like 3D FPS. For our style of games, you're five years away if not 10 years away from being able to deliver a comparable bandwidth and hardware experience using an app model vs. an HTML5 model. We choose to partner and build a style of games that take advantage of the latest hardware."
I'd argue that line of business applications require UI capabilities somewhere between casual games and the kinds of games that Glu does. So we may be looking at anywhere from 2 to 10 years.