I think if this discussion would have happened in 1990, I
would have argued that i. and e. should be reversed from
their APL predecessors. Because they were not, they have
never been a stumbling block for me having memorized the
order of iota and epsilon in 1968 APL.
My stumbling block was the later introduction of E. (which
I think does have the order right) - it is too bad that
backwards compatibility is such a strong/emotional
argument.
I argued with Ken for years about the axis order of base
and antibase ( #. and #: ) after many years, Ken returned
from teaching one summer saying that he agreed and we
should change it - of course, it was impossible to really
consider. Fortunately, j provided that opportunity, too
bad others slipped through....
- joey
PS - I do find myself using i.~ and e.~ quite a bit, and
that does make me appreciate ~ . Probably I should
just adopt the attitude that the correct spellings
are i.~ and e.~ ...
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
