I totally agree. One of the more interesting of Phil's videos was the dynamic tail, but trying to recreate the scene has got me into all kinds of trouble: there are just too many nodes either not supplied or not seen in full. I wonder if Softimage/Autodesk are simply not aware of how much some of us rely on working scenes we can replicate to build our ICE skills.

Incidentally, if anyone posessing a superior ICE skillset wants to display their skill, why not create the scene for us? Or at least enough of it that we can follow. It seems Phil did all the hard work with the strand dynamics framework nodes, so the rest of it should be quite straightforward.