Raul Miller writes:

> Put differently, Iverson's approach (illustrated above) is to
> traditional math something like what arabic numerals were to roman
> numerals. It takes some time to get comfortable with reasons for
> representing "nothing" (and different flavors of nothing) but once you
> have grasped this kind of approach you are then faced with painful
> tedium when trying to do what you now feel is simple, when using the
> long winded arbitrarinesses of the older contrivances.

​
The introduction of the comparisons and further dyadic functions made
notions such as reduction and inner and outer product much easier to
assimilate, not by *simplifying* them, but by making them ubiquitous.

- Ken Iverson, *Revisiting Rough
Spots*<http://www.jsoftware.com/papers/rough.htm>,
1994

[Commenting on the empty vector:] I think we have the Arabs to thank for
inventing zero, but I know that we have Dr. Iverson to thank for inventing
nothing.

— Donald McIntyre, APL Users
Meeting<http://www.jsoftware.com/papers/APLQA.htm#nothing>,
1980-10


On Wed, Oct 2, 2013 at 7:33 AM, Raul Miller <[email protected]> wrote:

> On Wed, Oct 2, 2013 at 9:45 AM, Pascal Jasmin <[email protected]>
> wrote:
> > I guess that I should completely understand why $2 is not 0
> > or 1, but I can let my pretty little head just consider that a list
> > of 1 and a scalar must be different:
>
> Yes.
>
> In J,
>
> */$n must give the number of elements in ,n
>
> and
>
> */$$n must give the number of dimensions in n (which is also the
> number of elements in $n).
>
> If $2 were 0 then 2 could not contain the value 2.
>
> If $2 were 1 then 2 would have one dimension and be a list containing
> the value 2 instead of just being the value 2.
>
> Put differently, Iverson's approach (illustrated above) is to
> traditional math something like what arabic numerals were to roman
> numerals. It takes some time to get comfortable with reasons for
> representing "nothing" (and different flavors of nothing) but once you
> have grasped this kind of approach you are then faced with painful
> tedium when trying to do what you now feel is simple, when using the
> long winded arbitrarinesses of the older contrivances.
>
> Note also that J is not unique here.
>
> Thanks,
>
> --
> Raul
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to