RB, you wrote above, "So here is a practicing engineer who goes on to challenge the conventional view of audio as frequency dependent. To him (and me) it should be time dependent." Can you amplify on that statement? Seems to me that frequency and time are not separable.
Yes, music comprises sound pressure levels which vary with time. In nature, the only thing I can think of that interprets changes in these levels as frequencies is the ear canal.
Yet, historically audiophiles think in frequencies - the bass, the midrange, the treble etc. And engineers use Fourier analysis to derive the frequency components in the original time signal. Fourier analysis depicts sound as superimposed sine waves.which lend themselves to measurement. Given the ability to measure, we get distortion readouts and introduce feedback to reduce the distortion in the frequency domain. But feedback smears the time domain! It is what SoulNote calls the Curse of the Fourier.
Take speakers for example. If you just look at the frequency domain, you probably value a flat frequency response. But that is much less than half the story. Look at the speaker's response in time to an impulse, for example in a waterfall chart. They really tell more of the story.
But look at how time stuffs up the sound from physically separate drivers, especially as it is characterised in the comb filter effect. Even if you carefully time align the drivers to get coherent direct sound, the reflections are smeared in time and you are locked in to a small sweet spot.
That's why I like pseudo point-source speakers - your pseudo-line source speakers may perform the same coherence trick but I have not done the maths to fully convince myself.
The PRATs might have been right all along!

