What has it gotten us?  All this thinking and trying and analysis?  The process of elimination but not much else.  Ad nauseum, it is entirely possible that I’m wrong and that K4 simply uses a ciphering system I’ve never heard of (or anyone else has for that matter) or that it uses a combination of known ciphers or that it does in fact utilize one that allows for some “masking” of plaintext values so as to rid us of our advantages of the English language.

Elimination then, of what?  Well, these are my feelings and opinions which I’ve tried to back up with decent arguments but evaluation of my success in that lies in other people’s opinions and feelings.  I think we cannot use standard cryptological methods to remove the “masking” block that Edward Scheidt put on K4.  Specifically, we can’t use transposition of any kind (flattened letter frequencies), monographic substitution (frequencies), digraphic substitution (odd # of letters, lack of repeats), monoalphabetic (frequency), or polyalphabetic (used in K1 and K2, just as vulnerable as they were).

What are we left with then?

It’s a riddle.

“Uh, yeah, we know that.”

No, literally a riddle.

“Eh?”

It’s a non-mappable system of masking K4’s text (PT or CT).  Another way of saying it is that it doesn’t use a derivable algorithm for encryption/decryption.  In other words, the previous plaintext is a riddle that we’re supposed to solve to know how to take the “mask” off of K4 to allow us to truly start deciphering what lies underneath.

Ed says in the WIRED interview that he’s not sure people know the method he used on K4.  This can either mean that no one knew what would be done or that it wasn’t a standard method.

I’ve spent the past couple months educating myself in as much cryptology and espionage history as I could absorb (it’s not limitless) and have basically tried everything I could on K4.

By trying out a lot of different cryptosystems and comparing what one can learn about them compared with the nature of the K4 text, I’ve reached the conclusion that the masking is not a property of type of ciphering but an actual effect.  I won’t pretend to understand right now all of the implications of what I’m suggesting but it means we need to solve the meaning of the plaintext before any forward progress can continue.

Is this a new assertion?  No.  I think I’ve said it previously on several occasions and I know that it’s been whispered and shouted for the past 20 years.

The difference is that I personally can feel pretty good that the reason I can’t solve K4 is not from a lack of experience with the cipher methods available anymore.  I’m no expert and I’m not perfect but it’s nice to eliminate the traditional analog methods from consideration.

To get beyond K4 we’ll need to find the keyword retrieval method/message, understand the plaintext messages better than we do now, discover some clue about the blocking, solve K4 and put it all together to know where to look for the final message.

Might as well hope for a pot of gold but it’s not like we’re the kind of people who give up easily.

It’s a riddle.

It can be solved.

Even if K4 can’t.