I am nowhere near as ambitious as Henry Rich. However, it seems like
his "J for C Programmers" might deserve a companion volume:
"J for Lisp Programmers"
Some issues are "obvious":
car {.
cdr {:
cons ,
Some are "messier":
Dyadic verb definitions play a fundamental role in the language rather
like cons cells play a role in lisp -- but at very distinct levels of
abstraction.
You can make a lambda work-alike in J, though the native J syntax for
this kind of thing is bulkier than in Lisp:
http://www.jsoftware.com/pipermail/programming/2012-June/028338.html
(and mind the error at the bottom of that page -- that hanged session
was a symptom of a bug, and not a symptom of a feature).
Or the use of induction instead of tail recursion.
Some issues are "misleading":
Efficiency, for example. (I much prefer to focus on simplicity most of
the time -- efficiency is a mix of choice of requirements, along with
hardware used along with measurements about resources used by
implementations... and efficiency is difficult to talk about
rationally, without a very narrow focus).
Anyways, it seems like there's at least some material here that might
be worth developing.
--
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
