Bull punch card deciphering

[creaky]

Hello,

Anyone know how i can decipher a Bull punch card?[

Long story, I purchased an antique desk top penholder with a punch card mounted within it. I would be interested to find out what information is on the card. I was told the pen holder was French and made for Fredrik Rosing Bull of Compagnie des Machines?

 

[pbirkel]

Try studying https://www.chilton-computing.org.uk/acl/literature/chapman/p013.htm and see how that applies to your card.

 

[g4ugm]

Try studying https://www.chilton-computing.org.uk/acl/literature/chapman/p013.htm and see how that applies to your card.

I think those only cover modern square hole cards. That card has round holes and these cards could have arbitrary codes.

 

[pbirkel]

I think those only cover modern square hole cards. That card has round holes and these cards could have arbitrary codes.

Sounds like the pre-IBM era where hole positions were assigned to fields with arbitrary geometry so as to pack the maximum amount of data into a card (e.g., the card designs used in the US decadal census). In that case there's no possibility of decoding an unmarked card without knowing the card-specific field layout and the encoding of data for each field. Possibly was hand-punched using a pantograph-type device that incorporated a layout-card to guide the placement of punches for each field.

 

[daver2]

There is an excellent Bull museum in France. They may be able to help...

Try here first: http://feb-patrimoine.com/PROJET/index.htm

Dave

 

[daver2]

It does just look like a 10 per column row card for a tabulator.

Is there only one hole (at most) in every column?

Storebrand OKA Ormestad Egli

Storebrand OKA Ormestad Egli

Dave

 

[g4ugm]

Sounds like the pre-IBM era where hole positions were assigned to fields with arbitrary geometry so as to pack the maximum amount of data into a card (e.g., the card designs used in the US decadal census). In that case there's no possibility of decoding an unmarked card without knowing the card-specific field layout and the encoding of data for each field. Possibly was hand-punched using a pantograph-type device that incorporated a layout-card to guide the placement of punches for each field.

I don't think there is a pre-IBM era! A pre 80-column card era yes, but IBM not so sure? The IBM names dates from 1924 (according to WikiPedia) but the 80-column card from 1928.

It does just look like a 10 per column row card for a tabulator.

Is there only one hole (at most) in every column?

Storebrand OKA Ormestad Egli
Storebrand OKA Ormestad Egli
Dave

Those are interesting links! But some columns do have multiple punches.

 

[daver2]

Worth a first guess...

Dave

 

[creaky]

Thank you all for your replies, Daver2 some of the columns have 2 holes..

 

[RobS]

I changed the perspective on the card in the picture to make it easier to see where the holes are located.



From the spacing it does look as though there may only have been ten hole positions per column or maybe eleven but probably not twelve. There are one or two holes in each column which makes it unlikely that there is any free format text encoded on it whatever the coding system used is. The suggestion already made that a template is needed to decode the meaning of each hole seems the most appropriate. In a document that I wrote many decades ago I mentioned that information equals data plus structure but here we only have the data without the underlying structure needed to give it meaning. Just a last minute thought - could it be that the card is face down and the layout is printed on the other side? That seems unlikely given the position of the clipped off corner though.

 

[RobS]

I think those only cover modern square hole cards. That card has round holes and these cards could have arbitrary codes.

Before I even started working with computers, so around 1962 maybe, I recall being taken around a site where small round holes were punched into standard eighty column Hollerith type cards. The card punches used had two settings so that one could be used by the first punch operator to punch holes at the top of the spaces for rectangular holes and then a second operator could repunch the same data into each card as holes at the bottom of the spaces. A separate automatic verifying machine then checked that all the resultant holes were ovals signifying that both operators had punched exactly the same data. Entirely correct cards could then be fed into normal readers. At least that is what I recall of the process.