Post Your Methods Used For Fenn's Coded Words (SPOILERS)

[GeneticBlend]

🏅

[zaphod73491]

So, on to the next stage of the problem. Clearly some polyalphabetic substitution is required since the distribution is too flat for English.

[GeneticBlend]

Problem? We're having fun, aren't we? What "problem"?

Those pieces weren't made to look like puzzle pieces for no reason at all.

Merry Christmas.

[zaphod73491]

Dec 19, 2018 22:23:34 GMT -5 GeneticBlend said:

Problem? We're having fun, aren't we? What "problem"?

Those pieces weren't made to look like puzzle pieces for no reason at all.

Merry Christmas.

Didn't mean it in the sense of it being a burden, but as a generic synonym for puzzle or cryptogram. Jenny has done a good job designing her multi-stage puzzle. When the result of the first stage is still encrypted, then a good designer provides a hint or some inter-stage feedback that you've solved it correctly. In the case here, the evidence is in the extreme unlikelihood that all 80 codes would align with a word letter -- particularly for the 6-9 positions. The one typo correction leaves no doubt.

I have an idea of what she's done in the second stage, which I could theoretically brute force, but I'm sure that's well beyond what she expects from her audience. She has provided the key somewhere in the book (there is an obvious candidate of the proper length), so I will try some more elegant approaches first.

[jammy]

Holy shamolies Batman. This is kicking my rear end good.

So, I believe I have the first part correct, with the puzzle pieces. But the 2nd part is killing me. I have tried several different obvious choices, but none of them are working.

I tried using the tomorrow hint, but maybe I am not doing something right.

I have questions, but am afraid to post them here, as not to give away anything to anyone who is working on it.

[zaphod73491]

My initial assumption after solving stage 1 was that each puzzle piece might have its own Caesar shift, and those Caesar shifts might be encoded by a 16-letter word or passphrase (there is an obvious candidate in Jenny's book). Alas, no instant joy on that approach. You can certainly convert the stage 1 text into letters that have a distribution more like English with this approach, so it ~could~ be the right system. But without the key, it would be very hard to solve. (I'm thinking of trying a computer hill-climbing approach which I've used successfully on other ciphers, though typically not ones this short.)

[jammy]

That's not my take.

I wrote down the numbers on the puzzle pieces on a piece of paper, in the order they appear in the book.

The first number almost certainly is a letter on a page in the book. The second number, after the dash, refers to something 1-9. I tried pages, but that isn't it. I tried chapter numbers. Also not working. I tried pages with puzzle pieces at the top AND are related to TTOTC. Also nope...

I have been at this for 3 weeks now. I am stumped.

[ironwill]

I wish I could help you guys out but I'm stumped too. That tomorrow hint wasnt what I thought it was

[jammy]

I took it to mean page 1 is the first page of the book code - at least, I think that's what this is. But, that theory falls apart after a couple of pages.

I'll keep looking through the book to see if I notice anything.

[zaphod73491]

Jammy and ironwill: I thought from the discussion above it was clear how to solve the first stage. The number after the dash in each code is the letter position within a word. The number before the dash is the word number. The only real challenge is figuring out the right section of the book to start the word counting, and given the nature of the challenge the sensible candidate list is short. You'll find it.

[zaphod73491]

Addendum: Jammy, you are so close. You just had the wrong idea about what the nature of the numbers before and after the dash. Read my logic above about the frequency distribution of the digit that follow the dash on the 80 codes.

[zaphod73491]

So GeneticBlend: a question for you. How long after you decrypted the WXY-Z codes did it take you to figure out the next stage of the cipher (i.e. converting the resulting ciphertext letters to plaintext letters)? I wrote a hill-climbing program to decipher the 80 letters under the assumption that each of the 16 puzzle pieces had its own Caesar shift. It quickly converged to the "best" solution, but it was gibberish. I tried several rule variants on the final letter ordering, but no success. So I'm now inclined to believe that my going-in assumption may be incorrect: that there isn't a simple Caesar shift for each puzzle piece (or if there is, it isn't via the regular ABC alphabet). I feel I'm missing something important that you eluded to regarding Jenny's decision to use puzzle pieces for containing the WXY-Z codes...

[GeneticBlend]

Zaphod, it took me a VERY long time.

This may help:

These are puzzle pieces. So they must be put together in a certain order, just like any puzzle would be put together.

There is however, something in the book that will give you clues on how to order these puzzle pieces. This other "something" that is in the book has things in common with the puzzle you are trying to solve, and the puzzle pieces, so that you can recognize that it belongs as part of the Fenn Coded Words.

I will be honest. I did not like this part of the puzzle. It is VERY frustrating.

[zaphod73491]

Dec 27, 2018 19:58:22 GMT -5 GeneticBlend said:

Zaphod, it took me a VERY long time.

This may help:

These are puzzle pieces. So they must be put together in a certain order, just like any puzzle would be put together.

There is however, something in the book that will give you clues on how to order these puzzle pieces. This other "something" that is in the book has things in common with the puzzle you are trying to solve, and the puzzle pieces, so that you can recognize that it belongs as part of the Fenn Coded Words.

I will be honest. I did not like this part of the puzzle. It is VERY frustrating.

Hi GB: I'm pretty sure I know what you're referring to in the book (mostly because of its chase-relevant decryption, dimensions, and curious appearance without context) and for the last few weeks I've been trying to fit it into the scheme of things. But clearly my brain works differently from Jenny's, so I'm sure I've been trying things she never would have dreamed of, LOL. I'll keep hammering away...

[zaphod73491]

GB: I should add that I think of piece-ordering as an appropriate operation when you've got letters with an English frequency distribution. The stage-1 letters are NOT English, or any other language. So the letters themselves must change.