Interestingly, while Toole and others (Zilch, for example, but also some other more notable individuals) make some solid claims about what does and does not matter with loudspeaker sound in real-world listening rooms, only Allison and Berkovitz managed to do some real research is real-world rooms. Sure, maybe Toole and others are correct about some things (I doubt some of this, but let's just speculate), but the fact is that Toole's work at the NRC and Harman, at least from what I gather, NEVER did do any real-world evaluations in real-world home listening rooms like what we had from Allison and Berkovitz. A lot of special measuring and listening has been done in the specialized rooms that Toole and Olive set up, but nowhere I know of did those guys go out into the world of consumers and actually measure, let alone do listening evaluations, in rooms that most people (including people like those who are members of our little CSP group, here) live with. Ideal situations were embraced by the Toole faction, and conculsions were reached by means of idealized listening environments that few people experience at home.
Yes, AR company data as it related to direct-field performance did not involve measuring in situations where cabinet diffraction and driver interference would have an impact. On the other hand, in real-world listening rooms other factors (boundary reflections, reflections from furniture, and placement anomalies) would also skew the results, so competing brands and brands in production right now that stress direct-field exactness, would have problems with those artifacts. In other words, there is really no way to predict the impact of the placement locations, or even listener positions, and so the best way to get at least a decent idea of what a system can do is provide good driver curves (on and off axis; the wider angled the better), and then provide power curves that illustrate what the speakers can do to input flat power in the bulk of home-listening environments. Anything else will vary widely from room to room. Heck, most speaker and driver manufacturers who do provide system or driver curves rarely show response characteristics much beyond 45 degrees off axis, and when we do see driver or system curves that show wide-angular response the results are often not all that good. Indeed, the driver curves that AR provided four decades ago (and Allison provided three decades ago) are still equal to or superior to any produced by contemporary manufacturers or speaker builders.
While it is true that the concert slope typical of pre-ferrofluid AR speakers of the era were sloped downward due to thermal issues, the fact is that many, if not most, classical recordings of that era were, as Allison has noted in some of his writings, much "hotter" in terms of brightness than what one would encounter at live performances in good halls, and when sitting in the best seats. (Jeeze, I do hope that some people here make it a habit to attend such concerts.) It may have been serendipity that things worked out as they did, but the fact is that the situation at that time did require some upper midrange and treble attenuation, and the AR line at the time delivered the required goods. Later on, Allison, who was using both ferrofluid and silicone grease as midrange and tweeter cooling agents from the word go with his Allison Acoustic systems, made it possible to select either a flat output or a "concert curve," with a middle option that kind of split the difference. Still later, once the digital age took hold and pop music began to eclipse classical even more than previously, Allison began to provide his speakers with no slope options at all, with the dialed-in curve both smooth and flat.
All I can offer up as a counterpoint to your comment "Why would anybody argue that engineering advances since then haven't been true progress?" is that if Villchur could make a pair of cabinet-diffracted, driver-interferenced speakers like the AR-3 work so well during those LvR concerts (proving to him at least, and to Allison, too, that maybe diffractions and driver interactions were not a big deal at all), just what kind of "meaningful" advances have we had in the last 40 years? I think that in terms of performance in real-world rooms Allison, with his woofer-placement approach, went beyond what AR and Villchur did prior to his retiring from AR (and Villchur, I should note, was impressed by those advances), and I think that liquid cooling for drivers certainly did make it possible for systems to take better advantage of advances in digital recording and playback technologies and the advent of home theater sound. And thanks to certain subwoofer manufacturers, low-bass performance certainly improved.
As for Villchur maybe doing things differently if he, as you noted, "could have achieved a speaker with better direct field response at the time and could have sold a very different looking grille-less system," well, no. I think that he proved to himself, and also proved to a lot of astute audio journalists and consumers (with the latter opting to buy a huge number of AR speakers), that the issue of diffraction effects, driver interference, and the so-called need for perfected performance in the direct field in opposition to perfected performance in the reverberant field, was not much of a serious issue at all. If his LvR concerts had been flops perhapse he would have changed direction, but they were not flops, as many contemporaries were ready to admit.
These days, with musical tastes being somewhat different, and with recordings in most pop-music cases NOT being made to simulate as best they can a real-world, acoustic-instrument performance space, Villchur might take a different approach. However, given HIS musical tastes I rather think that these days he would opt to get into a different business altogether.
What makes you think that Villchur doesn't like contemporary music or contemporary recordings? I have never heard of anything to suggest that. In fact, AR was born in an era dominated by incredibly poor recordings and very cheesy musical tastes.
I haven't the slightest doubt that if EV were designing commercial products now, he would consider the nature of current music, the increased interest and understanding related to localization and imaging, and offer appropriate products.