On Feb 26, 1:34=A0am, neit...@[EMAIL PROTECTED]
(Martin Neitzel)
wrote:
> > One way of looking at is that J links the items together but in APL2
> > the link is missing.
>
> Hi Gosi, isn't this reasoning about as crappy as stating "Pascal
> has FOR loops but in APL they are missing"?
>
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0
=
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Martin
This is ALL wrong :-)
1) APL has :For loops
2) If you think a function to link things together "helps", write one
- you don't HAVE to use "stranding"
3) Modern APL also has functions for indexing (squad and pick), you
are not limited to square bracket indexing
I was a SHARP APL user for the first 10 years before moving to APL2, I
also used to think that the lack of rigor in APL2 was going to rot my
brain but I have to admit that after writing applications for 15 years
I have yet to discover a case where I thought "oh boy, if only I'd
been forced to box and unbox all the time, this would have worked
better".
I do still feel a bit disturbed about the "space is (sometimes) an
invisible function" , which makes expressions potentially ambiguous,
and I guess I can see how this abiguity MIGHT limit the user (it is
certainly a nightmare for the guy writing the parser). But if "missing
links" are bad, why does J use space as an even more powerwul ivisible
thing to construct hooks & forks, "verb trains", whatever they are
called?
Morten
|
