The density of storage on these is so super low that to store this sentence would overflow onto a second card.

You can read them manually, you've maybe got 2 kilobytes worth, so not a whole lot, but it's probably worth a go to see if you're reading opcodes or names and addresses, program vs somebody's customer list or student directory or something.

And one more thing. First few cards is instructions to do specific calculations is run through to set up the computer, remember these early computer did not have storage at all. Then they run the data cards which computer calculates card by card to create new punch cards or onto tape or printer for result output.

Cheers,

Also, some cards (the yellow one) have the contents printed along the edge.

Possible coding information here: https://homepage.divms.uiowa.edu/~jones/cards/codes.html

Given the printing on the cards, there's some connection with SOAP II assembler:
https://en.wikipedia.org/wiki/Symbolic_Optima … ssembly_Program
https://en.wikipedia.org/wiki/Symbolic_Optima … _CodingForm.png

Hi,

Thank you for your replies.
I had a better look at the cards yesterday. Unpacked and read everything. I also tried to sort them.
First of all, the previous owner underestimated the count. It's probably around 2000 cards, and not 200.
Some are totally blank (~200 maybe).
About half of them have text contents printed on the side. (as snufkin noticed).
This allowed me to have more information on them.

1. They are standard IBM 80 columns cards. With EBCDIC encoding
   • One card = one line
   • You read them sideways. The "cut" corner is top left
   • Reading :
     • One punch on the column
       • Punch in [0-9] line = a digit 0 to 9
     • Two punches on the column
       • Punch on 1st row (no label) + Another punch in [1-9] line = Character A to I (A=1, B=2... I=9)
       • Punch on 2nd row (no label) + Another punch in [1-9] line = Character J to R (J=1, K=2... R=9)
       • Punch on 3rd row (label 0) + Another punch in [2-9] line (note that line 1 is not used) = Character S to Z (S=2, T=3... z=9)
2. It's mostly FORTRAN (even if SOAP II is mentioned (PASO 2 in french))
3. There are some comments and output strings (in French)
   • Seems physics/electrical engineering related programs. (talks about frequencies, Bode plots, ...)
   • Looks like the IBM computer was connected to a plotter to draw graphical results/curves (There are output string saying something like "Place the plotter in the centre of the sheet")
4. A few have annotations. Like corrections and formulas, hand drawn in the margins

But something bothers me. The cards "decks" are a bit mixed up. And there is no way to tell which ones are from the same program.
Luckily, some have the same colour. And (i don't know how to put it in English) some have markings across the "sides" of the deck.
But most are just shuffled in the box. So it will be very difficult to sort the cards to have a complete program. And the annotations/corrections make me think that the cards may be from a discarded batch.

I will try to build a reader for the cards that are not labelled. Just for curiosity. Some may contain data/strings/names.