I know some people have already solved the LOST numbers puzzle and congrats to them!
First off, great work on the blog guys! There's a lot of things left in the magazine I haven't figured out yet, glad to see I'm not the only one who's still puzzling through it!
Anyway, NICK!!! You've definitely found something!
I've been running "The Numbers" through a cryptogram solver without much luck. I knew it was more complicated than just a simple substitution. With the Kryptos article, I started looking at Vigenere ciphers. That's what we're working with. With a Vigenere, the letters change depending on where they fall in the code... a 7 might be A the first time and B the second time.
Anyway, I finally stumbled across an online Vigenere decoder here:
Anyway, I plugged in the numbers as Nick has transcribed them, then started taking random guesses at code keys. Tried KRYPTOS, PALIMPSEST, ABRAMS, LOST... then, WIRED. That got me somewhere! It decoded something useful, but, it wasn't enough, the key had to be longer. So then I tried WIREDMAGAZINE and here's what I got:
U S E L E T
T E R S B A
C K W A R D S F
R O M E N D
H Z A Y K C
B Y F L B Z
Z P X B X S S
B U M C I N M
The first page translated to USE LETTERS BACKWARDS FROM END.
And the second page translated to gibberish. And it's late, so I'm going to bed. But it's a start! I guess the next step is to reverse some letters and try again, and maybe with the same key or maybe a different one? I tried a few things but nothing's worked so far. I know you guys will be eager to pick this up and run with it in the morning.
Great work everyone! Oh, and if you're reading this Chris Baker... hope I'm not giving away too much!
Oh, one more note... the original key for that page they just GAVE to us, actually...
at the bottom of the page, it gives the page number as "104 WIRED MAGAZINE"
Every other page in the issue gives the page number as "### * WIRED * MAY 2009"
My guess is we don't need anything NOT on this page to solve this... just another thought.
I'm now playing with page 2's letters both in their original form and post-vigenere decrytped form. I've even tried "brute-forcing" the key via this ingenious little app: http://math.ucsd.edu/~crypto/java/EARLYCIPHERS/Vigenere.html
Still nothing. The problem with my brute force approach is that the ciphertext is only twenty-something characters in length -- which messes with the letter frequencies.
@JM: how did we miss the PAGE NUMBER?! Nice find.
I've also tried a few of the "probable" keys in a Playfair Cipher with the page 2 letters, but it doesn't look promising. I just needed a break from wrestling with Vigeneres.
Has anyone tried any other ciphers?
A good resource, if you feel like playing with other schemes: http://rumkin.com/tools/cipher/
Good work, all! Keep the brains churning...
Alright, im doing a bit of work trying to solve the second half of the Vigenere Cipher (All the stuff I remember from Digital Fortress keeps coming up).
Whatever the key is for the second half, its got to be something to do with the weird image/text on the last page. That section is called "THE END" and previous issues of Wired haven't been called that. If the key for the first half is any indication, it is something that should stand out from what is normally present in the magazine and isn't too complex.
As you guys have found, the math of the vigenere means that one slight variation in the ciphertext could mess up the plaintext translation for the key, so we have to be careful.