Newsgroups: comp.lang.apl
Path: watmath!watserv2.uwaterloo.ca!mach1!torn!cs.utexas.edu!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!jato!csi!sam
From: sam@csi.jpl.nasa.gov (Sam Sirlin)
Subject: Re: J is NOT APL (was Re: Interpreter advice sought.)
Message-ID: <1993Jan26.184600.24394@csi.jpl.nasa.gov>
Originator: sam@kalessin
Sender: usenet@csi.jpl.nasa.gov (Network Noise Transfer Service)
Nntp-Posting-Host: kalessin
Organization: Jet Propulsion Laboratory, Pasadena, CA
References: <1993Jan23.113019.23895@fid.morgan.com> <C1EIJn.I3@quadsys.com>
Date: Tue, 26 Jan 1993 18:46:00 GMT
Lines: 67


In article <C1EIJn.I3@quadsys.com>, roland@quadsys.com (Roland Besserer) writes:
|> hui@fid.morgan.com (Roger Hui) writes:
|> : There are some simple tests of this assertion.  For example,
|> : try posting the text of an APL function to this news group, 
|> : using the system editor on your machine to construct the message.
|> 
|> This is certainly not a valid point. 

Why? Perhaps Roger has been too terse. For better or worse, a
printable set of ASCII (a bit less if you worry about IBM machines) is
the standard of communications over internet and phones. All computers
can display this set, while probably almost all can't handle something
like unicode, and fonts in general are a pretty hopeless mess to try
to port from one machine to another. There are some exceptions for
some special cases, like TeX and postscript, but these are far from
universal. Yes, if you live in your own world you can get something to
work and even work well for any machine, but try to talk to some other
machine. The WSIS is still not universally available (or even quite
standard). APL, while ahead of other languages in so many areas, has
needed a readable, transportable representation for a long time (it
still isn't in the standard! and WSIS is only informational). If you
haven't tried to exchange ws, or talk about APL over this kind of
medium, let me tell you it's very frustrating! If APL is to come out
of the closet we will need to share information much more than in the
past, in line with the large archives of Fortran and C that lurk out
there in net-land. 

There are a number of solutions. If you don't like Iverson's rather
terse approach, or the similar APL slash bang, there are the many
keyword systems (for example, ahem, mine) that have been invented over
the years, and J. Mitloehn's system for putting togther a ws.
Hopefully there will eventually be a grass roots standard that will
eventually be adopted...

|> To hopelessly confuse the syntax 

Huh? This isn't a requirement for an ASCII system, and isn't the case
with J either I think. Though I do think ammend is much more confusing
than the old standard, reliable, selective assignment, this descision
had nothing to do with the character set.

|> and limit
|> it to agonizingly obscure sequences of punctuation characters for the sake
|> of being able to post verbose sources is nonsense.

Well one person's "agonizingly obscure sequences of punctuation
characters" is another person's wonderfully terse and expressive
system. I agree that I love the original APL characters (though I like
having lower case as well), and much prefer a real "times" sign to
.times or *, but I'll use the others to talk over the net or exchange
ws.

Yes, I'd love an APL on my sparc with the APL character set. But I
can't really afford one, and ISI certainly offer J at the right price,
for which much thanks. So I use J. You certainly don't have to.

At first J really looks strange, but I must say if you use it you do
get used to it and it has the same spirit as APL, very different from
the Fortran family (C, Pascal, Algol, Basic, etc.). There's even lots
of new ideas and functionality. The only thing that I still find
missing is selective assignment...

-- 
Sam Sirlin
Jet Propulsion Laboratory         sam@kalessin.jpl.nasa.gov

