Date: Wed, 3 Nov 2010 09:10:13 -0400
Reply-To: Nat Wooding <nathani@VERIZON.NET>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Nat Wooding <nathani@VERIZON.NET>
Subject: Re: SAS on mainframes
In-Reply-To: <051F377763004C428751BF293E22DCD7@Aragorn>
Content-Type: text/plain; charset="US-ASCII"
Daniel
From my experience, if you had any paper tape, it would be so dried out and
brittle by now that it would resemble the dead sea scrolls. Shortly after I
started work in 1973, the company needed to install some water temperature
monitors and purchased some that recorded hourly PHOTOGRAPHS, yes,
photographs, of a thermometer. We were paying the vendor some $200 to read
each film (16 mm movie film) and since they were changed monthly and we had
about 20 devices, the cost was pretty steep. I had been putting together a
means of using a HP digitizer and HP programmable calculator to read strip
charts and we realized that we could literally set a microfilm reader on the
digitizer, mount a track and simple sight in front of the reader's screen,
and mechanically couple the sight to the digitizer's cursor. Output from the
calculator went to an ASR 33 and we snail mailed the paper tape to the
mainframe. For grins, I once fed the output into one of those
typewriter-sized terminals that had rubber cups into which one stuck a
telephone receiver and sent the output directly to the mainframe.
The vendor was not happy when we told them that we were reading our own
films.
The rest of this is a short history of the use of punch cards for the newer
generation. They should light a bunch of candles or incense or whatever
every night in gratitude for never having had to deal with cards and tape.
As for punch cards, I never heard the "5081" used but rather learned to
simply call them "IBM" cards. For those who have never seen one, they had 80
columns with 12 rows where rectangular holes could be punched. There were
versions that were perforated so that the holes could be punched out
manually (in the US, think of the infamous hanging chads in the voting
forms) and other versions with small bubbles that could be filled in and
read with a mark sense reader. The use of punch cards started in the late
1800's when the US Census needed to move from a manual counting system to
something automated. At some time in 1900's one of IBM's competitors
introduced a rival card with round holes and a different number of columns.
I don't think that these lasted very long and I never saw such cards. One
neat feature with cards was that institutions could have a logo printed on
them and I saved a couple of these. NCSU had the university's bell tower on
its cards, for example.
To use cards in running a job, you placed them in a reader which was the
size of a chord organ or small piano. The cards were moved from a hopper
through a reader to a stacker by small rollers that, over time, left their
marks on the leading edge of the cards. And, after a while the cards would
start to jam. The immediate fix was to smooth the edge of the damaged card
with your fingers but eventually, you had to repunch it. I seem to recall
that there were keypunch machines that would duplicate cards. By the way,
there was such a need for people to enter data onto cards that a significant
percentage of jobs were for keypunch operators.
Short jobs might fit on a stack of cards a couple inches thick and these
could be carried in your hand and secured with a rubber band (or two, if you
were smart). Longer jobs usually went in a box that had once held new cards.
Often, the tops of the decks were marked with felt-tipped pens to help in
sorting them when the deck was dropped. There was a provision for putting a
number on each card but when you were writing a program, this was not very
useful since the numbers could change several times a day.
There were sometimes situations with smaller computers in which the
programmer would load his/her own cards but often you would have to go to a
data center and leave your card deck. Someone there would load it and when
the output finally appeared on a printer, it would eventually appear on a
shelf for you to find. I once heard that at the University of Iowa, someone
wanting special treatment would put a piece of hard candy under the rubber
band.
If all of this sounds woefully inefficient in terms of programmer time,
believe me it was. At NCSU, we had to walk about a hundred yards to reach
the data center (after waiting for the one elevator in our six story
building - I climbed a lot of stairs). When I first started work, we had to
cross a street to reach the company's main building. You were often lucky to
get in more than 4 or so runs a day if that many.
I think that I have droned on long enough but I hope that it has been
interesting. For anyone who is interested in the history of machinery that
used such cards, the following link tells how the US army kept track of
personnel during WW2
http://pattonhq.com/ibm.html
and
http://www.alteich.com/tidbits/t042202.htm
gives lots of links to sites with more history than most would ever want to
read.
Nat Wooding
-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Daniel
Nordlund
Sent: Wednesday, November 03, 2010 2:16 AM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Re: SAS on mainframes
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of John
> Burton
> Sent: Tuesday, November 02, 2010 6:56 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Re: SAS on mainframes
>
> Does anyone remember the 5081 card and the green plastic flow-charting
> template?
>
>
>
> --
> "be seeing you",
> Ray Burton
> Richmond VA
Ray,
I have my green plastic flow-charting template (probably bought in 1968) in
a desk drawer immediately to my left as I write this email. That probably
tells you more about my home office than I should admit to. :-) However,
unlike Alan CHurchill I do not still have any 5081 cards. Nor do I have any
of the paper tape that I used for capturing experimental data, which was
then loaded via a KSR 33 teletype into a Data General 1200 minicomputer for
analysis. Ahh, those were the days.
Dan
Daniel Nordlund
Bothell, WA USA
