In the red book it says you need to jitter the object temporally instead of spatially what exactly does that mean?
its too slow, dont bother. It involves using the accumulation buffer == bad.
On Voodoo5 cards you can use the T-buffer, they have an extension for it (GL_3dfx_tbuffer, i think?).
Basically what they’re saying is that you need to know the positions of all objects in the scene at specific points in the past relative to the current moment. If you bounced a ball then you would need to know the position of the ball at, say, 0.1 sec intervals back in time from the current frame for, say, 2 seconds which would create 20 temporal objects. These temporal objects are what need to be jittered and sampled in order to create motion blur this way.
search for motion blur on the boards, there have been numerous solutions.