Hacker News new | past | comments | ask | show | jobs | submit login
Has World War II carrier pigeon message been cracked? (bbc.co.uk)
16 points by drucken on Dec 16, 2012 | hide | past | favorite | 6 comments



No.

This is just some random person with no knowledge of cryptography making up a set of nonsensical abbreviations to might fit the ciphertext. It's no more likely to the the answer than any other random guess.

It's embarrassing that anyone is taking this seriously.


As an armchair expert, I doubt this is what the code means.

Rationale: "Short-and-Sweet", and the use three letters for at "K" sector, Normandy? Firstly, why waste a letter on 'at'? Secondly, how many "K" sectors were there where this could have been from?

Moreover, this decoding is not exactly filled with actionable info. All I get from it is "there's somebody somewhere who knows a bit about the enemy's positions". What does one expect as response? A return pigeon "please tell us more"?


Betteridge's law of headlines.

This article has basically zero content. Where's the full decryption? And the 'suggestions' on the right-hand-side are basically pointless. "Determined Jerry's Headquarters" - fantastic, but communicates nothing of importance to the recipient.

To me it looks like a message encrypted with a one-time pad. "PABLIZ" also looks more like PABUZ to me, which makes sense since I think they were typically broken up into five letter groups.


This is a decent image of the ciphertext: http://i.huffpost.com/gen/874813/thumbs/o-PIGEON-CODE-570.jp...

I see two different handwritings and implements on there.

The main code is blocky. The U's have square bottoms. They're even written like we might write "LI", as if the writer preferred a fountain point nib and had a habit of avoiding 'pushing' up vertical strokes. By the third line, it looks like his ball point is running dry and he doubles over the lines. He was taught to write the code squarish and blocky -- he writes "27 1525/6" much more naturally.

It looks like he wrote the note at 15:22 and handed it off to someone else who "lib.[erated]" them at 16:25. Probably the Sjt. wrote the plaintext, encrypted it, and handed two copies of the ciphertext to a (lucky) Pigeon Officer to crack open a crate of unused pigeons and drive them well away from the front lines before releasing them. This took about an hour.

Sjt Stott probably wrote "1522" so small to leave room for the Pigeon Officer to write the release time into the same box. (British and their paperwork). But the Royal Pigeon Service Officer didn't want to cram in the letters, so he just underlined "origin" as if to say "this is the correct time of origin of the pigeon I filled out my part of the form right" and wrote above it by hand.

The form was designed for a single user, but on D-Day you had one guy filling out the form and handing it off. The Sjt. may not have had much practice sending messages via pigeon or this particular form much before.

"27 1525/6" sure looks like "signing off at 15:25 on the 6th" (D-Day was June 6, 1944.) What happened in the three minutes between 15:22 and 15:25? Is that the time it took to make a second copy? If so, then why is the time in the message later than the time at the bottom?

Note that the Pigeon Officer knew the IDs of both pigeons before releasing either one of them.

So if the BBC is representing Mr. Young accurately when he "says Sgt Stott would have sent both these birds - with identical messages - at the same time, to make sure the information got through." then I think Mr. Young may be off the mark.

Young and BBC also make the suggestion that "PABLIZ" correpsonds to "Panzer Attack - Blitz". But clearly these are 5-letter groupings so the correct ciphertext is "PABUZ" which doesn't fit nearly so well.

This is a great book to describe the practical use of crypto by agents and troops in the field in WWII: http://www.amazon.com/Between-Silk-Cyanide-Codemakers-1941-1... Leo Marks - Between Silk and Cyanide: A Codemaker's War, 1941-1945


So which bit indicates which one-time pad should be used to decode the message? Or did they just use the date-and-time for that?


I think that's indicated by 'AOAKN' which is repeated as the first and last group of the message. Putting it first allows the receiver to begin decryption immediately (or get the part that was sent successfully before the sender was 'silenced'), duplicating it at the end allows you to decrypt the part you got if you tuned in to the transmission late.




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

Search: