Newsgroups: comp.lang.apl
Path: watmath!watserv2.uwaterloo.ca!torn!utcsri!rpi!usenet.coe.montana.edu!saimiri.primate.wisc.edu!doug.cae.wisc.edu!umn.edu!csus.edu!sfsuvax1.sfsu.edu!emclean
From: emclean@sfsuvax1.sfsu.edu (Emmett McLean)
Subject: Re: J is NOT APL (Summary)
Message-ID: <1993Jan30.012741.9349@csus.edu>
Sender: news@csus.edu
Organization: San Francisco State University
Distribution: NA
Date: Sat, 30 Jan 1993 01:27:41 GMT
Lines: 100


  Here is a summary of recent J is NOT APL e-mail with a
  few comments of my own.  Thanks to those who mailed.

> From: "Curtis A. Jones" <jonesca@vnet.ibm.com>
> 
> Yes, Manugistics APL*PLUS II is pretty expensive.  Their ordinary
> APL*PLUS costs what?  $600?  And APL2/PC costs $600 and includes
> a 32-bit interpreter and the application package maker...
> And if you're a student, $600 is a big chunk of your computing
> budget, but you can't complain about the price of TryAPL2 which
> should be pretty complete for most student activities.  So
> you may be exaggerating price as an obstacle to using APL.

  For the record, my point wasn't that I thought $1700 was
  unreasonably expensive.
  
> 
> And any comparison ought to include utilities included in
> the packages such as graphics, interfaces to other systems,
> file facilities, editing and debugging facilities...
> 

  1. J's modularity makes it difficult to directly compare it
  to APL. If I've defined the verb foo, which is used in an explicit
  definition verb I can test it interactively, but in APL I'd have to  put
  a stop before the line number in question and then step through
  explicitly defined function. Also, Erik Iverson is optimistic that
  in October the new release of J will included a debugger so things
  may change.

  2. I hope this may change. Now that IBM is laying off many
  of it's employees only one of them would have to pick up the  J
  source and make a new version with graphics which will knock
  our socks off.  But then this may be wishful thinking.

  3. Since J is written in C any advances in graphics theoretically
  can be transfered to J quickly whereas in APL you'd be stuck with
  the same old technology.

  4. Manuguistics has been introducing alot of OOP constructs
  to their product, so the similarities are getting increasingly
  smaller.  In fact, a Manuguistics representative in a recent APLBUG 
  meeting mentioned that they may re-name their product INCA
  for "It's Not Called APL".   I was suprised at how many APL'ers thought
  this was a good idea.
> 
> From: tomh@wes.on.ca (Tom Haapanen)
> 
> 
> I bought Iverson's APL/386 for Windows for $30.  Not much more than J!

  0. I've heard a lot of positives about ISI's APL.  But...

  1. J is a more deeply thought out language than Iverson's-Sharp`s APL.

  2. It is faster, more portable and more amenable toward electronic
     communication. It's functional capabilities if one is familar with them
     provide the language far more power.

> 
> >     c.) If you are working in an environment where you are the *only*
> >         person using a non-imperative language, you won't be at a loss
> >         if the printer you must use does not print APL symbols.
> 
> With Windows 3.1, it's not much of a problem anymore.  With Iverson's
> included TrueType APL font, APL prints quite happily on our LaserJet --
> or on any other Windows-supported printer.


  What follows below is *really* well written!!
> 
> From: SILER@WILMA.WHARTON.UPENN.EDU
> 
> Dear EM
> My first computing experience as an adult was with APL in Philadelphia
> 20 years ago when K. Iverson was also in town.  Although I haven't
> made a living wiht APL it's certainly influenced how I problem solve
> so when I saw that Iverson was to speak at a daylong NYSIGAPL I
> decided to attend, partly to hear APLtalk again, partly to see a hero
> and partly to learn about this new J thing.  My strongest impression
> of the day was of some (not all) old APLers badmouthing J and
> nitpicking during K. I.'s  J presentation.  It was exactly like 20
> tears ago when I'd hear him do his APL exposition in front of COBOLers
> and FORTRANers- same closed minds, similar silly questions.

  Yes, not all.


  Sad but true: The *only* reason I learned J was because I quit my
  job and J  was the only thing I could find which ran on a UNIX box.
  Most firms I've interviewed with do not have APL. It's turned out
  that except for having demonstrated a liking for programming,
  most employeers chuckle with sceptism when I sell myself as 
  a programmer.  Today things aren't much changed, I know C and Modula,
  but I'm not an expert UNIX programmer( Modula's *real* useful -:)).
  But knowing J, and knowing J will run on most any machine I use,
  provides me with confidence  APL can not provide.
  
  Emmett
