Newsgroups: comp.lang.apl
Path: watmath!watserv2.uwaterloo.ca!mach1!torn!utnut!cs.utexas.edu!swrinde!gatech!howland.reston.ans.net!usc!elroy.jpl.nasa.gov!jato!csi!csi.jpl.nasa.gov!sam
From: sam@csi.jpl.nasa.gov (Sam Sirlin)
Subject: Re: A modest proposal
Message-ID: <1993Apr19.174640.9008@csi.jpl.nasa.gov>
Sender: usenet@csi.jpl.nasa.gov (Network Noise Transfer Service)
Nntp-Posting-Host: kalessin
Organization: Jet Propulsion Laboratory, Pasadena, CA
References:  <735171646.AA03641@blkcat.UUCP>
Date: Mon, 19 Apr 1993 17:46:40 GMT
Lines: 39

In article <735171646.AA03641@blkcat.UUCP>, David.Michelson@f70.n109.z1.fidonet.org (David Michelson) writes:
|> I have a modest proposal for ASCII transmission of APL code:
|> It looks as ugly as J, but soon becomes easier to read through 

Ah yes, the bad old days of key or bit pairing!

|> Does anyone around still remember the
|> old o_u_t ?

Of course! How about the days of APLSV on a 360 interupting you if you
take too much cpu time at one shot?

|> An example:   an algorithm for converting field widths to column ranges
|>  @1 0   Y  0 1 O__    O_?  1 0  J.-        -?           0, L
|> neg1  drop     flip tranpose  outer add  plus reduction   quad
|> 
|> that is: (negative-1 0) drop of a 0 1 rotate in Lio-th dimension of the
|>             transpose of 1 0 outer-added to the sum scan of 0,quad input
|> Ugly as sin, but complete, and each of us can write a tiny translate
|> function for our own Lav (that is quad-AV, of course).

Ugly as sin is definitely an understatement! J shines by comparison.
If we're going to have an ASCII standard, why not make it a READABLE
standard. For example:

 .ng .drop  .crotate .tranpose  .jot .+ / .quad

(I have a complete list, and alternates, other ASCII etc) While the
above proposal works (as long as no cases were mixed in the original)
for machines, it would make reading postings difficult, especially for
those who don't have an APL symbol set or (non-standard!) keyboard at
hand. Then there's the problem of giving up the upper case letters!
All modern APLs don't anymore, giving your scheme no way to
distinguish Variable and variable.

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

