Humans would probably, if observing a signer move his hand up and down, be interested in the general location of the movement, the speed of the movement, the duration of the movement and how it temporally relates to other events. In this particular case, both of the original representations are viable. Recall that the concepts learnt will also be explained in terms of these parameters, hence if they are not in a form that a human comprehends, then the definitions generated by this method won't be correct either.