Roger Hui <[EMAIL PROTECTED]> wrote:
> x!y in J has the same argument order as every
> APL that I know of.   The description in "The
> Design of APL" has a typo -- it should have said
> n!m .

> > One thing about J that I was always curious about was that
> > when I first encounteed APL around 1970, in X!Y, X was typically
> > the larger argument, but in J, the parameter order has been
> > reversed.
> > 
> > At first I questioned my memory, but the above article does
> > mention that ordering. I am curious why the order was changed?

My memory of this argument order is very strong, but every
piece of evidence I can find (including just digging up a
copy of APL\1130 and running it under an emulator) confirms
the current J ordering, so I have to conclude that I learned it
wrong in the first place, and have been been labouring under
a false assumption for almost 4 decades :)

(By the way, I do think that the "standard" ordering makes
more sense than the backwards order that I remember, since it
follows the elegant symmetry that (!y) -: (1!y) similar to
many other J verbs).

-- Mark D. Niemiec <[EMAIL PROTECTED]>

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to