I can think of a number of ways do detect if the mouse if over (or
clicking on, etc) a arbitrary shaped bitmapped image (say, some sprite
where we want pixel-perfect click detection, not an enclosing
rectangle) , such as some kind of array with a mask in it, individual
lines of the sprite split into tiny rectangles, doing it all on the
server side via an XML request, but I wondered if anyone could offer
their experience or thoughts on the "best" way to do this that doesn't
eat too much client or server resources.