Among many enjoyable moments of my collaboration with Mike Bennett
was the time when he made a beautiful observation that the group
action mechanism in vector spaces over finite fields yields an
immediate proof of the (d+1)/2 exponent in vector spaces over finite
fields. He walked into my office and told me about it in a very calm
manner. I did not believe Mike and politely diverted the
conversation to something else. He came back a couple of hours later
and politely insisted, so I asked him to write the argument on the
board which convinced me very quickly. This was typical of my
interactions with Mike Bennett. He was a hard-working and insightful
student who frequently came up with interesting and unusual ideas.
My joint papers with Mike and others on group actions in the context
of Erdos-Falconer problems in vector spaces over finite fields
continue to influence my research to this day, and his dissertation
serves as a very fine reference source for many ideas and
calculations.