Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

“This telegram must be closely paraphrased before being communicated to anyone”

history.stackexchange.com

Post reply on HN
Print view View original

61–70 of 144 posts

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#61

Post by Khoth »

BigJono wrote: post #52

> In this process, deletion rather than expansion of the wording of the message is preferable, because if an ordinary message is paraphrased simply by expanding it along its original lines, an expert can easily reduce the paraphrased message to its lowest terms, and the resultant wording will be practically the original message. This bit has me perplexed. If you had a single message that you wanted to send multiple t…

It's mostly talking about the case where someone receives an encrypted message which is intended to later be published openly. If it was padded by adding stuff, an attacker can try to reconstruct the original plaintext by removing the flowery adjectives, whereas if things were deleted the attacker doesn't know what to add.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#63

Post by bee_rider »

pyuser583 wrote: post #59

Known-plaintext attacks aside, if you're going to compress text, it must be done before encryption. I don't know if compression offers much protection against plaintext attacks. This also makes me wonder how helpful AI is in such situations. AI is essential an extremely effective, lossy, compression algorithm.

If nothing else it would make a great twist in a fiction setting.

These paraphrasing instructions could be followed. But the paraphrasing could be done using some LLM. A sufficiently advanced adversary manages to invert the model somehow, and as a result can get the original plain text out of the paraphrased message, which lets them do a known-plaintext attack, get the key, and use it on other messages.

Sort of technobabble (is the idea of inverting an LLM nonsense?) but fun.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#64

Post by bee_rider »

hiccuphippo wrote: post #25

So it would make sense for the first message in a chain to be very verbose and repetitive to make it easier to modify down the chain. Bureaucrats must've had fun writting those.

Repetitive and verbose but make sure you don’t use up all the synonyms for a concept, right? Everything you use is taken from your paraphraser.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#65

Post by Jweb_Guru »

conradludgate wrote: post #42
tlhunter wrote: post #30

Earlier quoted context omitted.

I'm not sure why drum55's answer is buried but they're correct that the Nonce concept in modern crypto addresses this issue.

It's not only the nonce. The nonce helps to ensure that the message re-encrypted doesn't have the same ciphertext, but the known plaintext can still be used to forge messages. What stops message forgery is the message tag that TLS has (using the AEADs like AES-GCM or ChaCha20Poly1305). That said, the nonce is still very important to avoid most key recovery attacks

Yeah the real answer here is that this is what AEADs are for.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#66

Post by manwe150 »

Zeebrommer wrote: post #15
haunter wrote: post #11

> Never repeat in the clear the identical text of a message once sent in cryptographic form, or repeat in cryptographic form the text of a message once sent in the clear And (more or less) that’s how the Enigma was cracked. Turns out starting weather report with ‘weather’ every single time is not a good idea.

Or ending it with the same salute involving the name of the leader, for that matter.

Seems like an interesting conundrum. If you encrypt all transmissions, you end up having a lot of boring repetition, like weather and sign offs to just fill space. But if you don't encrypt the boring stuff, then the transmission itself is a nice signal of something interesting about to happen. But if you try to just pad with completely random noise, the other end might worry they've decoded something wrong and ask for a new cipher pad increasing the chance of interception. So maybe they should have tried to find something almost random but with known structure instead of sending the weather? Seems similar to how we now know that choosing a random password from the dictionary adds encoding redundancy without reducing security. Or similar to the goal of getting ordinary people to use Tor for ordinary things?

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#67

Post by manwe150 »

Khoth wrote: post #61
BigJono wrote: post #52

> In this process, deletion rather than expansion of the wording of the message is preferable, because if an ordinary message is paraphrased simply by expanding it along its original lines, an expert can easily reduce the paraphrased message to its lowest terms, and the resultant wording will be practically the original message. This bit has me perplexed. If you had a single message that you wanted to send multiple t…

It's mostly talking about the case where someone receives an encrypted message which is intended to later be published openly. If it was padded by adding stuff, an attacker can try to reconstruct the original plaintext by removing the flowery adjectives, whereas if things were deleted the attacker doesn't know what to add.

In particular, the length of a message is not encrypted when encrypting the text. So if the encrypted message is shorter, you know exactly how much to remove to get back the original, and then just need to guess what to delete. If the message is longer, it is much harder to guess whether to add flowery adjectives, a new sentence, change a pronoun for a name, or some other change.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#68

Post by manwe150 »

beerws wrote: post #53

Ironically, stating this at the beginning of telegram would precisely cause what it seeks to prevent (vulnerability to known plaintext attacks). Which makes me wonder: how many permutations of this rule could be conceived (and needed) that on the one hand would keep the point clear to the receiver, but on the other hand prevent such attacks? In any case the best option is to not have (to repeat) this rule inside mess…

It could be sent in the clear, although since the point was to apply it to every encrypted message, that would likely already have been redundant with having originally been encrypted. Just consider it part of the decryption algorithm itself instead: step 1, attach warning text, step 2, initialize decryption state and decrypt.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#69

Post by vl »

manwe150 wrote: post #66
Zeebrommer wrote: post #15

Earlier quoted context omitted.

Or ending it with the same salute involving the name of the leader, for that matter.

Seems like an interesting conundrum. If you encrypt all transmissions, you end up having a lot of boring repetition, like weather and sign offs to just fill space. But if you don't encrypt the boring stuff, then the transmission itself is a nice signal of something interesting about to happen. But if you try to just pad with completely random noise, the other end might worry they've decoded something wrong and ask fo…

In modern crypto it’s solved by using random nonce to star with and by using (encrypted) hash of data at the end. Random nonce gives you different cypher text for same inputs, hash tells you if you actually decrypted what was intended.

Re: “This telegram must be closely paraphrased before being communicated to anyone”

#70

Post by nradov »

Zeebrommer wrote: post #15
haunter wrote: post #11

> Never repeat in the clear the identical text of a message once sent in cryptographic form, or repeat in cryptographic form the text of a message once sent in the clear And (more or less) that’s how the Enigma was cracked. Turns out starting weather report with ‘weather’ every single time is not a good idea.

Or ending it with the same salute involving the name of the leader, for that matter.

Standard US cryptographic protocol during the same time period was to begin and end every message with a few random words specifically to thwart such attacks.
Post reply on HN