< Another is to define
< TensorTrace(T, TrankProperty) ....
<
< and call it by
< TensorTrace(T, get(T,Trank))
<
< (I prefer that).
Richard, I think that this won't work for the
example Bruce supplied, because he put something
in the plist of X and then assigned something to
X. That's why I suggested using hashes: he can
think of X as being a metric, even if X is a
just a hash with a 'metric' slot, amongst others.
Of course, a structure is a more systematic way
to do the same thing.
Leo
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.