I think that mitre3 (the measurement direction sumti slot) should be
dropped. First, no independent unit word has it. This includes other SI
unit words, non-SI unit words for distance (such as "minli" and
"gutci"), and arc angle measurement words (which have the same problem).
Second, from the standpoint of careful speech, mitre1 should really be the
length of a path between two endpoints with enough detail provided for
determination of the intermediate points, possibly with the
metric/distance function defined as well. Likewise, grake1 should be the
mass of an object, and not the object itself. We can sumti raise for
simplicity, but there will always be questions in such cases. For example:
What is the distance between the top of Mount Everest and the center of
Beijing? Well, the answer will depend on whether we are flying, following
the contours of the land, or cutting a tunnel through Earth. Moreover,
whether we take the shortest route by distance for each option, or its
great-circle complement, or the path which is cheapest in terms of fuel
prices or duration, or whether we stop over in Paris on the way, or follow
a highway, vel sim. will determine the length as well. To me, all of this
should be wrapped up in mitre1, not mitre3, which cannot handle it anyway.