Cryptologist [Lambros Callimahos] was a victim of his own success. He wrote a trilogy of books called Military Cryptanalytics covering code breaking in 1977. The first two volumes were eventually published, but the NSA blocked the public release of the third volume back in 1992. But last December, it finally saw the light of day.
Of course, some parts of the book are redacted, including parts of the table of contents. That’s pretty bad when even your chapter headings can be classified. [Richard Bean] over on Phys.org has some notes about the book along with some examples of hard-to-solve crypto puzzles.
One key part of the book, apparently, is cryptodiagnosis which is the approach to solving a message encrypted using an unknown method. In other words, while it is hard to break, say, an Enigma message, just knowing it is an Enigma message is a pretty big hint. If you get some random encoded message, where do you start? The answer is in the book.
We always think it is interesting that looking at what’s in a book that has been released can give you clues about what is still secret, just like smoke might show an invisible being by highlighting where they are not. However, we imagine the intelligence community has the same thought and throws some red herrings in just to confuse us.
Of course, one easy way to break codes is to break the machines people use to create them. Then again, sometimes the best encryption is to simply hide in plain sight.
Generally the easiest way to break encryption is not to break the machines people use, but to break the people. [obligatory XKCD reference]
Of course, it’s even easier if you build the machine yourself in the first place: https://www.bbc.co.uk/news/world-57394831
Yep break the people who use the machine. It’s cheap and fast.
This of course presumes you know who the people are AND that you have access to the people.
When I was in the U.S. Coast Guard, a Morse code radioman, inroute from Galveston Texas to Chareston South Cariolina in the Gulf of Mexico, I copied a broadcast message that was composed of about a hundred five letter coded groups e.g. ‘ XNAPO BZIDL CZYRA’ etc. The message was repeated several times. It was on a frequency near 500 kHz possibly a harmonic of , say 1.4 kHz The communications office and I gave it to the Navy sub base radio station in Charleston. We were then told to destroy all copies and not to speak of it. The best part was that afterwards, he took me to the officers club and we ‘spliced the main brace.’ Years later I found out the Navy had an ELF antenna site the the UP of Michigan that ran miles to sent messages to our ‘Boomers’ while submerged anywhere.
Sounds like OTLP.. back in the day I spent many a long night reciting five letter groups over HF from the field wondering If some random ham was thinking “wtf?”
I am not familiar with the term ‘spliced the main brace’. Is that military code for anal sex?
Sailors talk for having a drink, which being sailors almost always means getting drunk. Comes from the old days where if the main brace (which holds up the heavy main sail) was damaged, it’d have to be lowered to the deck and spliced back together, as it was labour intensive and quite a skilled task, those sailors would be awarded with an extra allotment of rum/grog/whatever they’ve got at the end of the task
Too many pages are empty. If it is not classified anymore, why even censor? Not being a buzzkill, it is highly interesting from an academical standpoint. But you could just use pyCryptoDome and have “unbreakable” ciphers like XChaCha-20 with AEAD, that no known computer to this day could break.
Just because you trust an algorithm, doesn’t mean it’s free of back doors and anomalies that others know about.
@ Bruce Perens:
Sadly there is no remedy against seeding FUD without proof.
But I take this as scots free card for using a fallacy as well: Daniel J. Bernstein would disagree with you.
He is a very capable mathematician and cryptographer.
Remember in the SF flick ‘Independence Day’ the invading Alien could not copy Morse Code and it saved our bacon.
“Sources and methods”
They redact so as not to give away anything they know that isn’t general public knowledge. If they have an edge on you, they don’t want to give it away.
e.g.: Differential cryptanalysis (what does the output do if one character in the message changes and the key is the same?) was classified for a long time until some one published a paper.
e.g.: the DES algorithm was published, including the S-boxes. The NSA told IBM that the method behind the design of the S-boxes was classified.
e.g.: The fact that we (Poles, French, Brits and US) broke the Enigma was classified until the 70s. Why 25 years after the end of the war? Because Enigmas and their descendants were being sold to second world countries and the NSA was having a field day reading their communications.
Mom spent WWII in OP-20-G at Nebraska Ave during the war.
Where the redaction is of the form:
31. Solution of plaintext autokey systems involving known cipher alphabets:•
when the introductory kev consists of several letters.
32. Analvsis of a case involvin :•unknown components.
____________________________ .,.
34. Analysis of digital plaintext autokey systems.
35. Concluding remarks on autokey:systems.
You can pretty much guess what section 33 was….
Care to give us a hint?
I am just scratching the surface on cryptography and most of this is greek to me at the moment.
My guess is,
“Analysis of analog plaintext autokey systems.”
From :
“The first two parts of the series have been declassified. The third part has so far been released only in such a severely redacted form that it hasn’t been considered worth reprinting. (The unredacted portions repeat material from the first two parts.)”
> He wrote a trilogy of books called Military Cryptanalytics covering code breaking in 1977
Three books ?
So “Military Cryptanalysis, Part IV, Transposition and Fractionating Systems, ISBN 0-89412-198-7” was not not part of the quadrilogy ?
(ref: https://en.wikipedia.org/wiki/Military_Cryptanalytics )
My bad, looks like the last three books were never completed “However, parts IV–VI were never completed”
It’s possible to have a good guess at the software library used to generate RSA keys, by looking at the key alone.
I wrote system for steganography text into text
Like this? https://people.apache.org/~jim/NewArchitect/webtech/2001/08/java/index.html
I don’t see a link to the first 2 volumes, are they available?
Looks like the text of all 4 books is available for DL here :
https://www.nsa.gov/news-features/declassified-documents/military-cryptanalysis/
@YGDES said: “I don’t see a link to the first 2 volumes, are they available?”
Good question. Excerpting [1] by B. Schneier: “Parts I and II, by Lambros D. Callimahos and William F. Friedman, were released decades ago — I believe repeatedly, in increasingly unredacted form — and published by the late Wayne Griswold Barker’s Agean Park Press.”
You can buy the books by Callimahos in physical form, there a list with searchable ISBN numbers in [2]. In paperback they cost roughly USD $200 each. Other than Part-III [3] I have not been able to find Part-I & II by Callimahos online for download, but that doesn’t mean they don’t exist somewhere (put your DuckDuckGo-Fu to the test). The books by Lambros D. Callimahos [4] are actually a derivative corpus based on the series of four books published by William F. Friedman [5] in 1938 – 1941 under the title Military Cryptanalysis, all of which are free to download as .pdf’s at [6].
* References:
1. Military Cryptanalytics Part III, January 4, 2021 by Bruce Schneier
https://www.schneier.com/blog/archives/2021/01/military-cryptanalytics-part-iii.html
2. Military Cryptanalytics
https://en.wikipedia.org/wiki/Military_Cryptanalytics
3. MILITARY CRYPTANALYTICS Part III By LAMBROS D. CALLIMAHOS (.pdf)
https://www.governmentattic.org/39docs/NSAmilitaryCryptalyticsPt3_1977.pdf
4. Lambros D. Callimahos
https://en.wikipedia.org/wiki/Lambros_D._Callimahos
5. William F. Friedman
https://en.wikipedia.org/wiki/William_F._Friedman
6. Military Cryptanalysis Part I – IV by William F. Friedman (.pdf)
https://www.nsa.gov/news-features/declassified-documents/military-cryptanalysis/
Putting a little perspective in here…FWIW, the FBI kept track of former WWII Army Signal Corps crypto operators etc for quite some time after the war.