Ok, hopefully you will admit, technology to produce abstract images better would be valuable and have a big impact. But what kind of research should we do? In the case of realism, the research agenda is pretty easy to state ? develop techniques that model and simulate the everyday world. Unfortunately, we are caught in the trap of solving the problems we know how to solve, rather than the more important ones.
I believe that the intellectual problems of creating and manipulating abstract imagery are deep and intellectually interesting. I see two major general directions. One direction is towards psychology, particularly cognition and perception. We need to understand better how people use images in their work. The other is towards information processing, things like databases, logic, text processing and machine learning. I call this second one the mathematics of information visualization, and it is different than the mathematics underlying physical simulation. Both directions involve forming bridges between computer graphics and other disciplines ? a nimble trait that has always served us well.
The goal of this talk was to convince you of the importance of abstract imagery. I decided to give this talk because it will be challenging to make progress, and I will need your help. Thank you for listening.