One mentioned 0123456789ABCDEF, and the fact that one has software
already
which does sums in this hexadecimal notation.
That works for software, but not so much for human communication.
The main obstacle to use hexadecimal in human communication are not
the symbols used in its notation, but that every single human language
uses 10-based numbers in *speech*, with at most minor deviations (e.g.
eleven, twelve). The overall system is always 10-based. (10, 100,
1000, (10000 in indian and chinese), 1000000) etc. have their own
names, numbers composed are to be understood in the decimal system
(nine million fivehundred thousand seventy two translates exactly and
directly to the _decimal_ notation 9500072), whereas it would be
pretty hard, even with Nyström's numbers (give the poor guy his dots,
we have Unicode after all) to immediately convert the above number
into writing.
BTW, this is also the main reason why decimal and SI _is_ more usable
than any pre-SI system, including the Imperial/English you claim to be
superior in several respects.
Szabolcs
