PBS Discussions

NOVA scienceNOW4 · Joined: 27 Dec 2006 Posts: 50 Location: Boston

As long as there have codes, there have been code breakers. Some of the best code breakers alive today have tried to crack a dauntingly elusive code on the Kryptos sculpture, which was created by the artist and amateur cryptographer James Sanborn and installed at the Central Intelligence Agency nearly two decades ago. Want to take a stab at solving it yourself or comment on its stubborness? Here it is—good luck:

Dcase · Joined: 24 Jul 2007 Posts: 1

I do not know the Kryptos code of this thread.

However, code toward the end of the TV segment on this topic may read: 'NOVAscienceNOW'

ripley256 · Joined: 24 Jul 2007 Posts: 1 Location: Home

I agree. Just going by the pattern of capital and lowercase letters "NOVA science NOW" seems to be a good fit to the code at the end of the segment. For anyone who missed it the code was: WNLWfeyyjpbPHG

lbsnake7 · Joined: 24 Jul 2007 Posts: 1

yeah.

dparrent · Joined: 25 Jul 2007 Posts: 1 Location: Las Vegas, NV

I agree as well. That is what it would appear to be (NOVAscienceNOW), but what is more difficult is to determine the method used to code it. Based on the segment and in review of the provided code WNLWfeyyjpbPHG it looks like substitution was used. But it must not have been the Caesar Cipher unless I did something wrong. Even if Substitution and Transposition was used jointly, there are 3 instances of the letter N in the assumed answer and there are not three instances of any letter within the entire code. I'm stumped... has anyone figured it out Question

_________________
DcP

fephisto · Joined: 25 Jul 2007 Posts: 1 Location: IOWAS

First of all, let me take a crack (bad pun) at the Kryptos final fourth part cipher. I'm pretty sure it's a reversal, given the question mark at the beginning (or, it's a clue that the cipher includes translating to Spanish first and then doing something else).

I'd say cryptology is as much psychology as it is actually techniques and mathematics. So, I definitely agree with the consensus that it's NOVAscienceNOW. As for the technique they used, going along the object of psychology, it's more than likely going to be one discussed in the show, otherwise we're left with a whole range of functions, thus:

transposition
transcription
Some various mean of obfuscation

By frequency count, as dparrent noted, it can't be a simple transcription or translation or even both, as obviously the number of characters is preserved. So, it's probably some other obfuscation method, of which the two I remember mentioned were phonetic sounding or dropping vowels.

Of course, seeing as how the cipher has 11 letters of the alphabet with 3 letters reapeated more than once, and the NOVAscienceNOW has 9 with 4 letters repeated more than once, it's possible that the letter n just had two substitutions, but then again, probably not, as ciphers have to be reversable; but then again, it might be reversable by having the rule of the coding be "every every numbered n transpose the letter n to x instead of y" r something like that. But since they showed Kryptos in 3 parts, I think that NOVAscienceNOW was encrypted in 3 parts.....Oh well, enough theorizing for now.

norak · Joined: 25 Jul 2007 Posts: 10

"NOVA science NOW" is the first thing that came to my mind also without any attempt to break the code. It's just too obvious. When breaking a code, it helps to get into the mind of the originator.
_________________
Enn Norak

Khaiya · Joined: 25 Jul 2007 Posts: 1

Well, it says now that it is not NOVAscienceNOW as I had first thought as well, and it also says that it is not an easy solve, so it can't be a straight substitution, otherwise I'd be guessing it was something like this...

THAT????ingNEW
or
THAT????ingNOW

but that doesn't really work out anyways... hmm...

my reasoning was the double letter quite possibly could have an "ing" following it... or the couldve also have been "message" but then the first word couldn't be "THAT"... But I also couldn't think of any other word that starts and ends with the same letter but has two different letters within it.

But yah, its probably neither of them considering they've said it's not easy to solve and simple substitution can be pretty easy.

Sonny · Joined: 25 Jul 2007 Posts: 1

It would be so cool if we could solve this puzzle here. Using the name of the show would be such a lame puzzle and I trust that NOVA Science NOW would not play a joke on us. I might go so far as to think the upper case letters are a red herring.

There are 14 letters. If you consider both lower/upper case letters; there are only double 'y' lower case, double 'W' upper case and the rest occur only once. In ASCII order, G, H, L, N, P, W, W, b, e, f, j, p, y and y. ASCII codes respectively are 71, 72, 76, 78, 80. 87, 87, 98, 101, 102, 106, 112, 121 and 121). This doesn't look like anything. If you thought 98 character were hard, then 14 is really tough. I think it must be much simpler.

Maybe the upper and lower case is as someone suggested, a reference to the 3 previous puzzles. Is there 3 words the sum up the other kryptos?

Felgar · Joined: 25 Jul 2007 Posts: 6

Theognostic · Joined: 24 Jul 2007 Posts: 2

After racking my brain for awhile this afternoon, the closest I can come is THATbaffledEGO referring of course to my inablity to figure this thing out. I also came up with THATpuzzlingNUT. I'm missing something somewhere.

norak · Joined: 25 Jul 2007 Posts: 10

You may be on the right track, in which case the last 3 letters PHG may decrypt into a 3-digit area code. I guessed that the first W should initially be moved to the end of the first group of 4 capital letters so that it reads NLWW which could translate into CALL by simply counting back 11 letters for each of the letters NLWW. If the 7 lower-case letters are a 7-digit telephone number, then some numbers can be represented by two different letters. I'm running out of evening hours now but I think that we should try to figure out the key to the area code if that is what PHG refers to.
_________________
Enn Norak

Mist · Joined: 26 Jul 2007 Posts: 4 Location: St. Louis

heh, i also registered just to try and help figure this thing out.
I'm convinced that the first word is "CALL" (substitute 15 letters to the right then unjumble) because where u turn in the answer is secret and has to be in the message.

As for the number you have to call, i guess you can't do a simple substituion with A being 1 and B being 2 and such because you've got no way of knowing if your answer is right. The only way I can think of off the top of my head is each letter already having some sort of numerical value.
Maybe you have to use substitution so that all the letters are those found in roman numerals... i'll look into that.
_________________
Peace
~Mist

Mist · Joined: 26 Jul 2007 Posts: 4 Location: St. Louis

Well, I've pretty much just scratched the idea about Roman Numerals. I did some research and found that PBS is stationed in Arlington, Virginia. And according to "http://arlingtonva.areaconnect.comzip2.htmcity=Arlington&qs=VA&searchtype
=bycity", the area code of Arlington Virginia is 703. But there is no Roman Numeral for zero, so that throws out that idea (if it is 703 we're trying to call).

I'll keep looking, only after i get some sleep though.
_________________
Peace
~Mist

obiwan121 · Joined: 26 Jul 2007 Posts: 121

"WNLWfeyyjpbPHG"

A puzzle contained in simple phonetic substitution and concatenation.

The key is deciphering that there are missing "d's."

When'll we finally jump PBS, see "d's," and be paid by Dr. deGrasse Tyson?

Lanka has my direct reply.

"It's the second mouse that gets the cheese."
Laughing

Felgar · Joined: 25 Jul 2007 Posts: 6

That reply Obiwan, is almost as cryptic as the code itself. Smile

Are you saying you think you've got it and there are missing D's? Are they missing in the code or in the solution?

Edit (continuing on the "Call __ Now" theory):
With the help of excel here are the 26 variants on the substitutions available from the middle portion. Even mentally unscrambling the results or reading phoenetically I don't see anything interesting. Which would mean if it's a substitution at all, it must be transposed prior to substitution...

New Edit: Oops, I missed the J. Here's an updated list. Also note that re-ordering before or after makes no difference if it's just a letter-shifting routine.

feyyjpb
gfzzkqc
hgaalrd
ihbbmse
jiccntf
kjddoug
lkeepvh
mlffqwi
nmggrxj
onhhsyk
poiitzl
qpjjuam
rqkkvbn
srllwco
tsmmxdp
utnnyeq
vuoozfr
wvppags
xwqqbht
yxrrciu
zyssdjv
azttekw
bauuflx
cbvvgmy
dcwwhnz
edxxioa

Mist · Joined: 26 Jul 2007 Posts: 4 Location: St. Louis

And using the alphabet on a phone doesnt work either.

feyyjpb = 339.9572
PHG = 744

i tried calling different variations of the number, including with the 703 area code, but nothing came up. Sad

_________________
Peace
~Mist

Felgar · Joined: 25 Jul 2007 Posts: 6

Yeah, my idea about the number being a word that you then enter into the keypad is falling apart. Since there are no letters assigned to 0 and 1, neither of those digits could be used in the number or area code. Would rule out 703.

The original idea would be if the code translated to: CallPBSNovaNow, which would be Call 727-668-2669 ... Doesn't seem to work out.

Spareaker · Joined: 26 Jul 2007 Posts: 15

Here's my thinkin'

The upper/lower case mapping of NOVAscienceNOW to WNLWfeyyjpbPHG is a hint that NOVAscienceNOW is a 1:1 key for a substitution.

Be forwarned - this is a hunch. It has not yielded results. (If NOVAscienceNOW is a cypher key, it is not the whole decryption.)

But it works like this:
A simple Caesar Cypher shifts each letter by a set number of places. As a result, each letter has a static counterpart. And, because of this, frequency counts and other patterns in the language pose a weakness -"th"-es together (especially followed by an "e"), "ing" word endings, and double letters.

To throw off the counts & break up these patterns, the encypherment is shifted by a key. It could be an agreement: "We'll start with A=W, and shift the Caesar Cypher by one space each time.

Or you could use a "key phrase," as I suspect NOVAscienceNOW gives us (in both sense of the word).

In this, A=N for the first letter encypherment, A=O for the second, A=V for the third, etc.

And I think that this is what the coincidence of the upper/lowercase sequences of 14 letters each is telling us. But, again, this is just a hunch. When I applied this directly (using Excel too, Felgar!), it was not the immediate answer; nor were "+1 shifted" variants (1st character: A=N, B=N, C=N, etc.).

Some itterations yielded tantalizing clusters of maleable letters (nice mix of vowels & consonants, no q's without u's, not too many "rare" letters, etc.). But I'm not very skilled at anagraming.

Enjoy

Felgar · Joined: 25 Jul 2007 Posts: 6

How about just reading pheonetically like it were a license plate. Does anything stick out then Spareaker?

Edit: Try entering some of you best looking variants here. Smile

http://wordsmith.org/anagram/

Bluewater · Joined: 26 Jul 2007 Posts: 3

I think it's possible that the dollar bill shown in the video might have some information contained in it that is central to solving the puzzle.

Mist · Joined: 26 Jul 2007 Posts: 4 Location: St. Louis

The dollar looks normal, but here are a couple notes about it.

Printed in Cleveland, Ohio (D)

and its code is: D 42941506 B
(now looking at the picture... the fingers of the woman holding the dollar appear to be pointing to this code...)
_________________
Peace
~Mist

naveed · Joined: 26 Jul 2007 Posts: 1

dbain · Joined: 27 Jul 2007 Posts: 1

Seems too few characters to do a meaningful frequency analysis... Does seem that this code will be solvable with pencil and paper. (Being lazy, I also played on Excel.)

Tried using NOVAscienceNOW as a key but nothing interesting after a quick perusal.

Did find something interesting if you do all combinations using a simple Caesar Cipher:

cnlwfeyyjpbphg
domxgfzzkqcqih
epnyhgaalrdrji
fqozihbbmseskj
grpajiccntftlk
hsqbkjddouguml
itrclkeepvhvnm
JUSDMLFFQWIWON
kvtenmggrxjxpo
lwufonhhsykyqp
mxvgpoiitzlzrq
nywhqpjjuamasr
ozxirqkkvbnbts
payjsrllwcocut
qbzktsmmxdpdvu
rcalutnnyeqewv
sdbmvuoozfrfxw
tecnwvppagsgyx
ufdoxwqqbhthzy
vgepyxrrciuiaz
whfqzyssdjvjba
xigrazttekwkcb
yjhsbauuflxldc
zkitcbvvgmymed
aljudcwwhnznfe
bmkvedxxioaogf
cnlwfeyyjpbphg

The one I put in ALL CAPS--"JUSDMLFFQWIWON"--could be: "Just email FFQ with I won." No clue who FFQ is. Just a coincidence? Seems like clue on this board that once solved the winner will know who to contact makes this more plausible. Or not?

Anyway, back to the drawing board...

Spareaker · Joined: 26 Jul 2007 Posts: 15

Exhausting, isn't it? It's easy to get Excel to generate strings of 14 characters, but then you have to go down the list and try to pronounce them. (And some of my "pronounciations" would have gotten my mouth washed out with soap, when I was a kid!)

Then there are the possibilities with clear English words embedded in a bunch of gibberish, and I keep trying to force meaningful context around them.

Exhausting!