I see Java's evolution more in terms of the overall platform, not just the language. The platform improvements that I look forward to most include simplification of the deployment of the Java platform, like what we're doing with the upcoming Update N release. I said I would also like to see further improvements in the platform libraries, to offer increased functionality and easier programming paradigms.
There was a response to this point that maybe we should be taking things out of the platform instead of adding them into it.
Some on the panel - I will claim this was a small minority, but I obviously have a bias here - would like to see deprecated or unused functionality stripped out of the platform. But the other side, on which I count myself, sees the platform as a clearly defined entity: Java is what it is and you cannot now start taking things out of it. If you do so, you will break existing applications.
So the issue boils down to whether we should create a new platform, which no longer has these dependencies. This decision then comes down to a question of resources: should we spend limited time trying to improve what we have or trying to create something new?
I personally feel Java still has a long life, so it is worth continuing to work on and improve what we have. And given limited time and resources, it seems like we'll do better by focusing the efforts on that problem for now, rather than giving up and starting afresh.
Having said that, if and when we (meaning Sun Microsystems and the larger community) have time, it would certainly be worth thinking about some future platform as well.
Chet is an architect in the Java Client Group of Sun Microsystems, and spends most of his time working on graphics and performance issues. He is also co-author of Filthy Rich Clients: Developing Animated and Graphical Effects for Desktop Java Applications.