Hacker News new | past | comments | ask | show | jobs | submit login

I did one of these as a proof of concept for storing gpg keys long term, securely and offline.

My version didn't use a motor and required a human to pull the tape through, had a set of guide holes down the middle to indicate when a bit was to be read and used LEDs as both the light sources and the photo receptors.

All in all a rather fun project which worked much better than it had any right to.

Never did find a material which had the longevity of aluminium foil with the durability of 35mm film stock.





A friend had a commercial paper tape reader (don't recall who made it) from the S-100/8080/CP-M homebrew era built exactly like that. It was the 'inexpensive' version (they had motorized versions), and was finicky but worked OK for small amounts of code. CP/M actually has a built-in papertape reader device in bios called RDR: (tho it was often repurposed for other kinds of devices).


For something as short as a gpg key, you can easily use stamped brass “tape” 0.1-3mm in thickness. You can roll it like tape if it’s thin enough or just store it as a bar if your pattern is dense enough.

(Also note that OP’s version doesn’t require a motor either and can be actuated by hand, per TFA.)


Here's from 2014 with some capacity calculations [0]. "ssh private key (900 bytes): 15 feet of tape."

[0] https://heepy.net/index.php/Data_storage_capacity_of_teletyp...


If we prefer paper but relax the teletype constraint, Oleh Yuschuk's PaperBack [0] allows encoding 500 KB on a sheet of printer paper.

[0] https://ollydbg.de/Paperbak (posted various times to HN)


CNC machines which use paper tape, use a form of Mylar.


How did you write data onto the tape?


For the paper version, which is the only one I tested, I used a co2 laser to burn the holes, a jig to center tape and a pin to index the holes.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: