Also, d5 (may I call you "d"?), the big revelation from Williams' work--for me, at least--was how we, or at least some of us, need to adjust our basic thinking about stereo miking techniques.
From what I see in a lot of Web postings here and elsewhere, some people seem to hold firm opinions about which stereo miking techniques are their favorite--apart from the context of any particular recording situation. Like, in my previous message I said something positive about "Blumlein" stereo recording, and I know for a fact that there are people who would consider themselves charter members of "team Blumlein" if there were such a thing. That's kind of absurd in an almost dadaistic way--like holding up one end of a clothesline and expecting the other end to hold itself up so that you can dry your clothes on it. There's no such thing as a stereo recording method that's best in all circumstances; there can't ever be such a thing. Championing one method above all others only shows that a person doesn't get it (yet).
What Williams' work taught me was that every stereo microphone setup has a particular range (side to side) that it covers properly--he calls this the "stereophonic recording angle" of the setup. For setups with different microphone patterns, angles and distances between the mikes, this angle is different, and his charts and graphs let you figure out what that angle of any practical setup is.
The point is then that, if you're going to use a particular microphone arrangement, you need to match the "stereophonic recording angle" of that setup to the angular width of the direct sound sources in the recording venue. You can do that by choosing a miking distance that places the furthest left and right direct sound sources just within the coverage angle.
Now, in a way that's at least partly ass-backwards; in real life you mostly don't simply choose a miking distance just because there's a certain miking method that you want to use. But neither is it the other way around, exactly. If you're being realistic, you consider both things at the same time: You choose a miking method that will work with a miking distance that will work with that miking method (loop until ready), given the width of the sound sources and what the room sounds like if you mike at various distances. You have to multitask a little, in other words, and consider these factors at the same time--not "first A, then B."
A prior commitment to one miking setup just gets in the way of this process. Again using Blumlein as an example--in some respects it really is shockingly close to an ideal stereophonic miking system. Its Achilles' heel, however, is its narrow stereophonic recording angle (only +/- 45 degrees). For widely spread-out sound sources, you end up having to back your microphones pretty far away to fit that narrow angle. And that can be fatal with bidirectional microphones, since by definition they're exactly as sensitive in the back as they are in front; they tend to "want" relatively close placement so that they don't get swamped with room sound.
So if you consider this miking technique out of context, you will see only its wonderfulness while in real-life recording, there aren't many situations in which its great virtues can actually be used to maximum effect. When it works well, it's unbeatable as far as localization is concerned, and potentially very good in other respects (except for the relative lack of a sense of envelopment or spaciousness, which requires decorrelated left and right channels especially at low frequencies--never a strong point with any coincident recording method, unfortunately). But examples of its actually working well are fewer than you might expect.
Do you see what I mean? Williams' idea tends to debunk any one miking method or microphone type as an ideal unto itself; it puts microphone patterns and stereo miking setups into a practical context, and makes the engineer responsible for using them appropriately so that the eventual playback system can do the right things, more or less, relative to the original recording situation. And oddly, that's a bigger picture than a lot of people seem to want to take in.
--best regards