Curious about origin of J source

2 views
Skip to first unread message

Michael Norris

unread,
Sep 24, 1993, 2:41:26 AM9/24/93
to

Suspicious.

I just downloaded the J source code. It looks suspiciously
similar to the source for Sharp APL/HP that Rob Hodgkinson
and Laurie Gellatly wrote about this time last decade.
I can't find any actual credits. Anyone know if it's really
a hack (in the kindest sense of the word) of SAPL/HP?

Just Curious.

-m.

Roland H. Pesch

unread,
Sep 24, 1993, 6:50:51 PM9/24/93
to
As a former APLer rather than an active one, I usually refrain from
posting in this group; but I couldn't let this one go by without a
reply---especially since the last time I talked to Arthur Whitney, he
said he's given up on netnews for now.

mich...@cs.uq.oz.au (Michael Norris) writes:

>Suspicious.

>Just Curious.

>-m.

First, a direct answer to your query: no, there is no genetic
connection to SAPL/HP. Except for about a page of code which inspired
it in the first place, J is the work of Roger Hui.

That one page of inspiration was an afternoon's hack by Arthur
Whitney, who dashed off a very simple APL-like interpreter in an
afternoon as an example (Ken Iverson had asked him what it was like to
program in C). That original tiny interpreter is reproduced in its
entirety, for the curious, in an appendix to Roger's book on the J
implementation.

Arthur has a very distinctive C programming style, anathema to most C
programmers but often appreciated by APLers. Roger adapted [sic] that
style.

It is probably that style which strikes you as similar to SAPL/HP; as
I recall from talking with both Arthur and Rob, Arthur was actually a
major (perhaps _the_ major) author of the SAPL/HP interpreter you
describe as written by Rob Hodgkinson and Laurie Gellatly.

To further disabuse you of any suspicions about SAPL/HP direct
influence on J: in the interim, Arthur had---

+ worked for a number of non-APL companies here in California,
+ written, as a contractor, two "package" modules for SAX (the SHARP APL
for Unix interpreter I was involved with, at one time)
+ written another major interpreter, A, at Morgan Stanley

Say hi to Rob for me, if you happen to be in touch!

/Roland Pesch

--

----------------------------------------
Roland H. Pesch pe...@cygnus.com +1 415 903 1410
:Cygnus Support::1937 Landings Drive::Mountain View CA::94043:

Reply all
Reply to author
Forward
0 new messages