edit: nevermind, didn't realize they kept the 'plate spinners' on during the experiment. The rings in the appalam are artifacts from rotating an anti-node through the food. would have to do something different to get equal heating using ~two points vs. a line
Like a slow-speed random orbit sander? :) [1] Some of the dead areas looked too big for even that to help, and it would mean the maximum plate size would be smaller. Hmm...
OK, I'm going to say that the right answer is to move the microwave source linearly (or change how it's aimed? Or otherwise change its focus) at a frequency that would maximize even coverage of the heating area based on the rotation speed of the turntable.
Another (more limited) possibility is a way to mount a fixed spoon that could stir soup or other appropriate liquids while the bowl turned.
edit: nevermind, didn't realize they kept the 'plate spinners' on during the experiment. The rings in the appalam are artifacts from rotating an anti-node through the food. would have to do something different to get equal heating using ~two points vs. a line