unforum

[miamo_tutti]

arnezami, am I correct in saying that even if you discovered part of the keystream from guessing the correct placement of a word or phrase, it would still be pretty much be a word-guessing game to decrypt the remainder of the messages? So the ideal thing would be for the portion of the message you decrypted to make obvious the use of a certain a word, which you could then use on the other message, etc. etc. until both are decrypted?

Also, how do you figure the whispered hint works into this?

[arnezami]

[themandotcom]

arnezami, I am just wondering, is that the only way that you can do the (A+K)-(B+K) thing? What my idea was is that the entropy quote would be encrypted with the same key stream as the code on the card? Is it possible to do without trying to guess a word, as that would be hard for us.

[miamo_tutti]

If this has already been mentioned, it bears repeating:

Using EODMFXRUTH as the keystream beginning at the 27th letter of the top row of ciphertext (FHWY...) yields: ASSLYVSBED

Using FHURGIFULP as the keystream beginning at the 27th letter of the bottom row of ciphertext (MMOS...) yields: GETAFMVFOU

...assly v's bed...?
...as sly v's bed...?
...get a fm v fou....?

Compared to the consonant-stricken outputs I'm so used to, these actually look like you could make sentences out of them. They use two different keystreams, however, which conflicts with the two-messages-one-keystream method.

[fruppet]

Off Tangent

[Kuroiken]

I've been doing a little bit of testing using the Solitaire Cipher system.

I made a few assumptions before i started:

The 10 'thermal sensitive' letters are removed from the code
There are two texts coded using the same key, each is 54 letters long
The two coded texts are used the same direction (neither string needs to eb reversed)

in a simple spreadsheet i made a few functions that would allow me to move a piece of text character by character and generate the two possible outputs from the coded texts. In many cases a lot of garbage was spewed out.

There were, however, some points where certain words yielded something that might be part of a word from the other text.

Most notably, 'THE':

at position 5 in second string, gives 'PUT' in first string.
at poistion 8 in first string gives 'OF A' in second string.
at position 14 in second string gives 'COI' in first string.
at position 15 in second string gives 'ALL' in first string.
at position 31 in first string, gives 'hap' in second string.
at position 31 in second string, gives 'fot' in first string.
at position 34 in second string, gives 'OF P' in first string.
at position 46 in first string, gives 'EXA' in second string.
at position 46 in second string, gives 'IRI' in first string

'AND':

at position 42 in first string gives 'JET' in second string.

'SENTE':
at position 12 in first string, gives 'ISEMA' in second string.
at position 26 in second string, gives 'RRIDO' in first string.
(i thought maybe corridor, but this didn't work)

Neither Cube nor Receda gave anything legible.

'EARTH':
at position 30 in the second string, gives 'omyic' in first string.

can't think of what other words to try out, any ideas?[/list]

[miamo_tutti]

Re: Off Tangent

[fruppet]

Re: Off Tangent

[UKver2.0]

I'm betting he's already figured it out. I (as well as many others I'm sure) sent him the complete details of the puzzle a while back. You know, he works in the same city as I do. Maybe I should run by his place and convince him to share. Heheh... I'll show him my South Park video as a warning. But seriously, I'd take this as a definite hint.

[Kuroiken]

If anyone else wants to try putting some words in, i have uploaded the xls spreadsheet.

Just put the text in the commented row in various places to check for results.

[poozle]

[miamo_tutti]

[Stratman]

At last we return to what I (and others) have been thinking all along (but still got nowhere) - the passkey is retrieved from the phrase somehow.
To be pedantic, and especially for non card holders as the accurate wording may be important - I keep seeing the phrase written slightly wrong, as in Fruppet's email to Bruce Schneier.

It is...
If entropy wins, outward looks should leave you cold

ie not...
will leave you cold

Still struggling on...

[miamo_tutti]

[arnezami]

Hi guys

I've found some time to make a little program so eveybody who wants to can now start trying to find words using the A-B exploit.

For those who don't know what I'm talking about read my former post here:

The A-B exploit and why this could be our core problem.

I've given a pretty thourough analysis there about the idea there could be two messages on the card encrypted with same key(stream) and why MC might have chosen this approach when designing the card.

Anyway. Here is my program you can use to actually try to decode the encrypted message using the A-B exploit. You need to choose what is ciphertext A and B and then choose a guessing word or phrase. It will then try this guessword on every position of both ciphertexts and will decode that part of the other ciphertext (using A-B).

Here are some screenshots:



As you can see I've put it some common ways of dividing the characters on the card into two ciphertexts. If you know more I'll add them. I've also put in some basic guess words but you can enter them yourself.

If you have any suggestions or questions don't hesitate to ask .

Regards,

arnezami

PS. @themandotcom: I don't see how the entropy hint being the output (ciphertext) of the solitaircipher could be helpful in any way. The problem is the plaintext would then be random characters and we would never be able to recognize it as "the plaintext".
PPS. @Kuroiken: seems you've already tried some words (using the xls file). I can confirm your findings (tried several). Would be nice if were able to find some longer words (or phrases). Hopefully we will find them with this proggy. If you have any ideas/suggestions please tell me.