Interesting indeed. Nothing that was completely new to me, but I don't think I've seen it explained so coherently before.
The one thing I'm still wondering about is LuLa's explanation of ETTR, which says it works due to the way digital sensors are able to record more unique values the higher in the bit range you go. To me DPR's "law of great numbers" explanation makes intuitive sense, but I'm not sure if that does anything to diminish LuLa's explanation.
I haven't yet read the article, but I have practiced "exposing to the right" on the advice of experts and found it simply doesn't work for me. I find blown highlights in digital photography much uglier than any failure in shadow capture. A personal bias. Whatever the numbers say, it does't work for me with the cameras I own.
Lawrence: the point of ETTR is to expose as highly as possible without blowing the highlights.
If your camera can't accurately predict when the highlights will clip, then it's probably not the best technique to try; but if the camera does allow you to predict it, maximizing exposure while staying below the blowout limit is pretty much ETTR.
If your camera does allow you to predict when highlights will blow but you're blowing them anyway, you're not ETTR'ing, but just throwing away information.