Filip Norrgård’s blog @ filips.net
Filip’s blog about mobile technology, tech itself and other stuff

Solving the Kryptos Code (Ok, Not Really!)
24 January 2005 in Root | Comments enabled

Jim Sanford is the creator of a statue called Kryptos, placed in the front of CIA HQ that has got an encrypted text that has been puzzeling crypthographers for some time. I got a few ideas for the cryptographers out there to try out, when Sanford said recently in a transcribed interview to Wired:

It’s also well-known that I did use some matrix codes Ed gave me, and I have also designed visual systems for encoding, which are much harder for cryptographers to crack because they’re individualistic.

To add to my theory, Elonka mentions on her webpage dedicated to the Kryptos:

In the “NDY” and “AHR” rubbings, you can see that the letters “Y”, “A”, and “R” are a couple centimeters out of alignment. All the other rows on the sculpture seemed to be properly aligned though. So was it an error in the sculpting process, or a deliberate misalignment related to some part of the code?

Might be deliberate. I’ve seen something like YAR before and that was in vector transposing and transformations using matrices in high school. YAR could signify that one should be using some sort of matrix calculations to solve the last section of Kryptos — or not. What I can recall, I used Y for the vector, A as a vector lenght increaser/decreaser and the R was the rotation matrix.

John Wilson’s Kryptos page says that the third section of Kryptos

is solved by using a modular transposition decoding: [ decrypted=(192*encrypted+192-1) mod 337 ].

If you consider the fact that Jim Sanford upped the ante from section 1 to section 3 by requiring more mathematical calulations to decrypt the code, then it seems quite logical that he would have chosen something as complicated as matrix calculations (or, at least to me ).

Can it be so that the tableau side is the matrix that should be used to transform the vector that can be made out of section 4? Or, should it be considered a pattern for the identity matrix where “krypto” indicates the location of the 1’s? Or, should one only take the inverse of the vector in section 4? The possibilities are infinite.

These are just the sudden ideas that I got from reading the Wired interview and looking at the two Kryptos pages mentioned here, so, I can be completely off-course. Fact is, matrix calculations are relatively new to the mathematical world and it might even be that Sanford did not use matrices in encoding the forth section. These are just my two cents and I wish you all cryptographers out there the best of luck in decoding this statue.