> It is estimated that the one-half of the round-trip delay happens in each direction. Anyone know how true this is? Particularly for mobile/asymmetric bandwid

> It is estimated that the one-half of the round-trip delay happens in each direction. Anyone know how true this is? Particularly for mobile/asymmetric bandwidth connections - what might the errors bars be?