I know that this thread is old, but I would like to clarify things. I developed this method in 1993 when I was younger and much stronger. In reality, it was serendipity. I was not setting out to create a starter method. I was just trying to get my yeast mixed well with the starter medium. Having a been a body builder for most of my twenties, I still very strong, so I shook the starter like it owed me money. I started with a repurposed half-gallon glass Cran Raspberry bottle. When I moved up to a repurposed gallon glass apple juice jug, I noticed that the starter medium foamed like crazy (I was making quart starters at the time, so ratio was four to one). I also noticed that fermentations experienced shorter lag periods, attenuated better, and had few, if any off-favors. It was not until I started to think of the problem in engineering terms that I understood what was going on. Foam has a very high specific surface area compared to liquid; therefore, it is easier for oxygen to diffuse into it. In essence, I stumbled across an improved, low cost starter method just because I was strong. It took engineering and science to understand why the method worked well.
In reality, if one shakes and pitches, one has to shake a second time to disperse the yeast cells. I have debated whether or not the heavy initial shake damages cells, but empirical knowledge tells me that shaking to disperse the yeast cells is part of the magic. Inoculating and shaking often leads to a shorter starter lag time. The same thing happens when a carboy is shaken after the yeast culture has been pitched. In the early days, I used a repurposed 6.5 gallon acid bottle for a primary. Unlike the 6.5 gallon acid bottles that were sold after the homebrew trade became more established, the early bottles came with a screw-on cap, which allows the bottle to be laid on its side and rolled back-and-forth vigorously after the culture had been pitched. That step definitely shortened lag times by at least an hour.