Reading ancient paper digital media (was Re: Hamurabi Focal source)

Ethan Dicks ethan.dicks at
Wed Mar 31 14:46:25 CDT 2010

On 3/31/10, Tony Duell <ard at> wrote:
> OK... I was worried for a moment that somebdy had tried to archive paper
> tapes by scanning them a foot at a time on a flatbed scanner or something
> equally daft

I did that exact thing with punched cards as an experiment.  I did not
(and still don't) have a punched card reader, but I have a
correspondence course in "Data Processing" I picked up at a thrift
store that came complete with blank coding forms and punch cards as
teaching aids for the homework.

I put the punchcard in a flatbed scanner with the unprinted side down
(the "back" side) with a dark backing sheet to contrast the holes.  I
scanned it to a TIFF then converted it to a GIF and used Tom Boutell's
GD library to import it into a C application that did some simple
image transformations (edge detection, etc).  I didn't complete the
code to the point of converting spots to bits, but I did scale and
locate the card in the scan area and was on the verge of writing the
code to check for holes when I got distracted and set it aside.

Making an 8 or 9 channel papertape reader from scratch is not an
impossible exercise (witness the ancient Byte article referenced here
every so often).  Making a punchcard reader from scratch is a very
different level of effort, so back then I figured that it'd be easier
to use a flatbed scanner than try to make a 12-level reader and
mechanical feed system on my own.  These days, though, with
inexpensive fabrication tools (access to laser cutters, home CNCs,
Makerbots, etc) it might _not_ be as hard to make a punchcard reader
from scratch as it was 10 years ago.


More information about the cctech mailing list