Well I think as folks realize that they're not dealing just with a browser plug-in and as they get over the 'Skip Intro' syndrome and realize that it's not just for entertainment or media, but it's a full client, that there are client capabilities there, and that they can pass on more sophisticated knowledge of the user back into the deeper tiers of their enterprise, I think we'll start to see thicker clients, or thicker uses of clients, even beyond the browser. I think we'll see a lot of cases where traditional Web applications will escape the browser and we'll see little 'mini-apps' in different scenarios, better integrated with operating systems, with task bars or tool bars or even residing in their own container, providing a forum that's just a little bit better fit for the type of data that will be aggregated. So I think we'll see architects and developers deal with issues like client-side data repositories, replicated data from client-to-client, client-to-server and a lot of issues that are almost akin to putting a mini-server on the client. So I think we're going to see a lot of practices and patterns emerge around those sorts of approaches and I think we'll see a growth in client-side development patterns which is an area that is notably absent from today's J2EE pattern catalogues.