Re: APL, J or K?

Marshall Spight wrote:
> Anyone here have any experience with the APL, J or K programming
> languages? (Yes, I recognize the redundancy in "APL programming
> language.") What about with Kdb?
>
> They look fairly interesting, if somewhat thrown together. It
> doesn't appear that any of them are examples of great design,
> but the underlying math seems quite interesting.
>

APL goes a long way back. The seeds for APL were planted in the 50's by Kenneth Iverson, and then implemented in the 60's by IBM. I would not call APL "thrown together". Au contrair, it is probably one of the most thoroghly designed languages existing. (Iverson was awarded the Turing Award in 1979 for his work on APL)

It still has an ultra-loyal following and some evolving implementatios, see e.g. http://www.dyalog.com/.

It has a fairly steep learning curve, not least because of it's use of lots of extra symbols, see e.g.
http://www.users.cloud9.net/~bradmcc/cgi-bin/vuImag3.pl?i=149

(I am under the impression that the IBM selectric "golf ball" typewriter was designed in part to accomodate for the APL-symbols)

But the power is immense in the hands of the knowing and I personally know people that have built extensive systems single handedly using APL. See e.g. http://www.users.cloud9.net/~bradmcc/APL.html

The way to think of APL is as a vector processing machine, and it is hence a not so distant cousine to a "true" relational language (a relational language processes relations instead of vectors) and could be, in a sence used to get a feeling of the power obtainable from a TRDMBS.

The J and K-languages are similar to APL but use the normal keyboard.

Regards,
Lauri Pietarinen Received on Sun Aug 14 2005 - 09:06:48 CEST