Fwd: the 1130 (Was: Canadian history 
Author Message
 Fwd: the 1130 (Was: Canadian history

--part1_f7.f4f9277.28d53990_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

I asked Charles Brenner, who was instrumental, along with Larry Breed, for
developing the 1130 APL for some his recollections of his work. He replied
with the attached message. Charles nowadays is the "forensic mathematician"
highlighted in a New York Times story a year or so ago. He has discussed this
at APL conferences, I believe.

Eugene McDonnell

--part1_f7.f4f9277.28d53990_boundary
Content-Type: message/rfc822
Content-Disposition: inline


Received: from  rly-za02.mx.aol.com (rly-za02.mail.aol.com [172.31.36.98]) by air-za05.mail.aol.com (v80.17) with ESMTP id MAILINZA51-0915115220; Sat, 15 Sep 2001 11:52:20 -0400
Received: from  dfw-smtpout1.email.verio.net (dfw-smtpout1.email.verio.net [129.250.36.41]) by rly-za02.mx.aol.com (v80.21) with ESMTP id MAILRELAYINZA24-0915115207; Sat, 15 Sep 2001 11:52:07 -0400
Received: from [129.250.38.63] (helo=dfw-mmp3.email.verio.net)
        by dfw-smtpout1.email.verio.net with esmtp
        id 15iHjh-00074K-00; Sat, 15 Sep 2001 15:52:05 +0000
Received: from [168.191.203.164] (helo=roo.dna-view.com)
        by dfw-mmp3.email.verio.net with esmtp
        id 15iHjf-0000gP-00; Sat, 15 Sep 2001 15:52:04 +0000


X-Mailer: QUALCOMM Windows Eudora Version 4.3.1
Date: Sat, 15 Sep 2001 08:55:02 -0700


Subject: Re: Fwd: the 1130 (Was: Canadian history


Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed

As you recall, Eugene, I came to IBM as a summer employee following my
undergraduate graduation in 1967. You may wish to forward my comments.

The 1130 interpreter was a port of Larry Breed & Phil Abrahms' "LC
interpreter." LC -- "Little computer" I think -- was an elegant and simple
computer design, a prototype of which was actually built, and for which
Larry & Phil had written an APL interpreter as a research project. When IBM
marketing decided a APL interpreter might promote flagging 1130 sales,
Larry showed marketing's systems engineer Reve Carberry how to write an
1130 emulator for the LC. That was the initial port. The emulator of course
had to have one extra command beyond LC's command set -- an escape that
allowed introducing native 1130 code for speed and for IO. Larry says that
the first test was iota 5 (worked correctly) and the second was 2*3 (took
20 seconds to arrive at the answer 8, which is also correct).

I took up the project sometime during my extended (8 months) summer stay,
adding keyboard input, perhaps a few execution routines, and saw the
project through to fruition. Many years later I met Bob DeCloss at STSC.
Bob told me that he had cut his APL teeth on the 1130. I was pleased by his
positive recollection of the "triple shift" keyboard, which I had adopted
as an artifice to fit the large APL character set into the small 1130 keyboard.

Charles Brenner
consulting these days in forensic mathematics
http://www.*-*-*.com/


Quote:
>Perhaps you could provide your recollections of APL 1130 for these people.

>Eugene

>Received: from  rly-xa02.mx.aol.com (rly-xa02.mail.aol.com
>[172.20.105.71]) by air-xa03.mail.aol.com (v80.17) with ESMTP id
>MAILINXA39-0913161814; Thu, 13 Sep 2001 16:18:14 -0400
>Received: from  cherry.ease.lsoft.com (cherry.ease.lsoft.com
>[209.119.0.109]) by rly-xa02.mx.aol.com (v80.21) with ESMTP id
>MAILRELAYINXA28-0913161802; Thu, 13 Sep 2001 16:18:02 -0400
>Received: from PEAR.EASE.LSOFT.COM (209.119.0.19) by cherry.ease.lsoft.com
>(LSMTP for Digital Unix v1.1b) with SMTP id

>Received: from LISTSERV.UNB.CA by LISTSERV.UNB.CA (LISTSERV-TCP/IP release

>           2001 13:38:48 -0300


>Received: from malakbel.sun.csd.unb.ca (malakbel.sun.csd.unb.ca
>[131.202.3.10])
>           by hermes.csd.unb.ca (8.9.3/8.9.3) with ESMTP id NAA13176 for


>Received: by malakbel.sun.csd.unb.ca (8.8.5/971107-10:10) id NAA02551; Thu, 13
>           Sep 2001 13:38:44 -0300 (ADT)
>Precedence: gateway
>X-Gateway: UofNB vialist
>X-Date: Thu, 13 Sep 2001 18:38:01 +0200
>X-NNTP-Posting-Host: 212.198.57.108
>X-Accept-Language: fr,en

>X-Mailer: Mozilla 4.77 [en] (Win98; U)
>X-Trace: neon.noos.net 1000399099 10468 212.198.57.108 (13 Sep 2001 16:38:19
>          GMT)

>Date:         Thu, 13 Sep 2001 13:38:44 -0300



>Subject:      Re: the 1130 (Was: Canadian history

>Content-Type: text/plain; charset="US-ASCII"
>Content-Transfer-Encoding: 7bit



> > The 1130 was a disk based operating system, but the APL was card
> > based.

>Yes and no, in fact. True, APL did not work under the DOS of the 1130
>(as I constated sadly when the pack of cards I loaded erased the
>previous system), but it did need a disk, both for )LOAD and )SAVE
>operations, and also because the interpreter included three overlays
>(you could hear the disk noise when switching from editing mode to
>execution mode, for instance). The disk had to be dedicated to APL.

>Incredibly, the interpreter worked quite well on a 8K word (16K bytes)
>machine :-o

> > It might be that the card deck was supplied by NASA.

>I wonder if it was not written by Adin Falkoff and Kenneth Iverson, but
>I guess the latter of these gentlemen should know better than me :o)

>I was told, by the way, that APL was supposed to stay an APL internal
>use only product, but that the NASA exerted pressure on IBM to make it a
>commercial and supported product (this probably applies only to APL\360,
>the 1130 being too restricted in memory size, floating precision and
>speed for scientific production use with anything else than fortran or
>by the ALGOL written by Camille Bellisant and Olivier Lecarme in
>Grenoble). I never knew if the story was true or not.

--part1_f7.f4f9277.28d53990_boundary--


Thu, 04 Mar 2004 07:17:41 GMT  
 
 [ 1 post ] 

 Relevant Pages 

1. the 1130 (Was: Canadian history

2. the 1130 (Was: Canadian history

3. 1130

4. Canadian history available

5. (Fwd) (Fwd) Re: Band Matrices

6. (Fwd) MESSAGE ON PHONE SCAM (fwd)

7. [Fwd: envienla (fwd)]

8. Fwd: [LogoForum] Re: (Fwd) [apcs] Karel Goes Fishing!

9. Fwd: NetRexx language now available (fwd)

10. [Fwd: Re: RubyInline 1.0.4 Released! (fwd)]

11. Ruby Impressions (fwd) (fwd)

12. There's got to be an easy way to do this (fwd) (fwd)

 

 
Powered by phpBB® Forum Software