He formed a mathematical representation for acoustic quanta by relating a time-domain signal s(t) to a frequency-domain spectrum S(f). He then mapped an energy function from s(t) over an “effective duration” dt into an energy function from S(f) over an “effective spectral width” df to obtain a characteristic cell or acoustic quantum.
From Curtis Roads: Microsound (Chapter about Gabor matrix)
And the beauty is, this has been done and is done for more than 50 years now. But in contrast to math or to a lesser degree physics, it doesn’t get uninteresting just because somebody already thought of the basic formula for this. Because in music finding a specific set of parameters is equally interesting and valid as finding the general description. Because in music a every set of parameters is connected to specific emotions and meaning and these are what listener react to. There are still unknown emotional landscapes to discover.