Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!linus!philabs!seismo!rochester!ritcv!jrc
From: jrc@ritcv.UUCP (James R Carbin)
Newsgroups: net.social
Subject: My first time (working with a computer)
Message-ID: <521@ritcv.UUCP>
Date: Sun, 11-Sep-83 22:35:27 EDT
Article-I.D.: ritcv.521
Posted: Sun Sep 11 22:35:27 1983
Date-Received: Tue, 13-Sep-83 10:15:36 EDT
Lines: 45


I'm mot sure why this discussion is in net.social, but since it is, I'll
add my two cents.

My first computer was an IBM-702.  The "Preliminary" Programmers Manual
(there never was a "final" manual) was about 80 pages long with the first
chapter a primer on data processing complete with an explanation of
punch cards and how they could be stored on magnetic tape (7 track) at
200 bpi.  Peripherals consisted of a 200 c.p.m. card reader, 100 c.p.m.
card punch, 10 tape drives, and a 200+ l.p.m. printer.  Memory consisted
originally of 10,000 characters (yes - characters not bytes) of memory
which was later expanded to 20,000.  It had a drum to store routines
and tables with a capacity of (I think) 10,000 characters).  General
Electric acquired it in 1953 to process the Corporate payroll and it
ended up doing all of the d.p. for G.E. Corporate Accounting Operations
from 1953 to 1963 when a conversion to third generation hardware was
started.  It was finally retired in 1967 (maybe 1968) and supposedly
was given to the Smithsonian.  Don't laugh, it ran just about 365 days
a year, 24 hours a day, throughout its lifetime.  In fact, an IBM C.E.
had to delay his retirement until it was retired because IBM didn't
want to have to train someone new on such an old machine.  No O.S.
or anything like that, a crude assembler that wasn't very often used,
but a RPG that could produce 7 different reports with one pass of
the file as it took approx. 20 minutes to read a 2400 foot reel of
mag. tape.  Some of the files on the machine were very large for
its vintage - 550,000 shareowners records (GE stockholders) with 
each about 2,000 characters long.  Sorting that d*mn file could take
in excess of 24 hours.  They even had a bell installed on the console
to wake up the operators on the third shift if the processor stopped.

Do I wish for those days - NO!  But it sure was a hell of a way to
start a career in C.S. (They didn't even call it that in those days)
and I don't regret in the least the opportunity that I had.
By the way, they build 17 IBM 702's and you could not be sure that
any two were totally compatible.  Also debugging consisted of the
programmer obtaining time and then sitting at the keyboard executing
his program until it stopped.  Then you would attempt to patch it
via the console and continue (making notes of what you had done)
and finally obtaining 1) printed output and 2) a memory dump which
was the only documentation that was maintained for each program.

Boy have we come a long way!!!!!!!!!

Jim Carbin
RIT