[Puzzle] JayNineteen Youtube Profile Update [5/7/07]

Luminous · Post by **Luminous** » Wed May 09, 2007 7:23 pm

ignatzmouse wrote:A random thought. We have 24 letters of ciphertext and six shifts. Perhaps we're meant to be grouping the ciphertext into blocks of four letters, and using those to generate codons?
Code: Select all
ijoo pyaa ttzb qkip yerg gjqa
Not that I have a great idea how to turn "ijoo" into a 3-letter {A,C,G,T} sequence.

Why not group them into threes ?

ijo opy aat tzb qki pye rgg jqa

I understand the number sequence would fit if you grouped them into fours.

but I don't know if there is a way to generate a codon from fours - TOSG?

ignatzmouse · Post by **ignatzmouse** » Wed May 09, 2007 7:35 pm

OOG...

Luminous wrote:Is it possible that in order to maintain cohesion with LG15, the PM has changed direction, and is now tying the storyline into Hymn of One rather than OpAphid?

Could be, although to me this looks like a gamejack by a PM with different priorities. So far there hadn't been a huge amount of OpAPHID-specific stuff in Facility J: there was the OPA name and logo but that's about it. Of course, who knew what was up the PM's sleeves...

Luminous · Post by **Luminous** » Wed May 09, 2007 7:51 pm

I really like what you did here Mouse, breaking the gibberish phrase down into groups that sync with the numbers.

ijoo pyaa ttzb qkip yerg gjqa
4(+13) 1(+4) 8(+17) 4(-11) 1(-14) 8(-16)

So when the sequence is broken into groups of four, the numbers fit exactly.

ijo opy aat tzb qki pye rgg jqa
4(+13) 1(+4) 8(+17) 4(-11) 1(-14) 8(-16)

When it's broken into groups of three, as it would be for codons, we have two groups of three left over. Not sure what to do with this, but it's an interesting observation.

ignatzmouse · Post by **ignatzmouse** » Wed May 09, 2007 8:03 pm

Luminous wrote:I really like what you did here Mouse, breaking the gibberish phrase down into groups that sync with the numbers.

ijoo pyaa ttzb qkip yerg gjqa
4(+13) 1(+4) 8(+17) 4(-11) 1(-14) 8(-16)

So when the sequence is broken into groups of four, the numbers fit exactly.

ijo opy aat tzb qki pye rgg jqa
4(+13) 1(+4) 8(+17) 4(-11) 1(-14) 8(-16)

When it's broken into groups of three, as it would be for codons, we have two groups of three left over. Not sure what to do with this, but it's an interesting observation.

Yes, if there was some wacky 4-character-to-3-character decryption scheme, then we could get our old friends the codons back. Can't think of anything though. It could just be that it's six letters of text followed by the codons, or er something else.

Luminous · Post by **Luminous** » Wed May 09, 2007 8:16 pm

So, based on this breakdown, it looks pretty likely that ijoopy . . . is the message text.

ignatzmouse · Post by **ignatzmouse** » Thu May 10, 2007 2:27 pm

One thing that's funny is the frequency distribution of the message:

Code: Select all

01234567890123456789012345
abcdefghijklmnopqrstuvwxyz
**  * * ***   **** *    **
*     * **    ***  *    *
*

It's very "bursty": lots of doubletons and empty slots, I'd expect a random distribution to be flatter.

In particular, the message only occupies 14 of the possible 26 slots. This led me to wonder "is there a 4-character one-time pad which would encrypt an {A,C,G,T} sequence this way". The answer is no, oh well. The (rather dull) reason why follows...

Let S be the set of numbers in the message "ijoopyaattzbqkipyerggjq" (treating the letters as numbers 0...25):

Code: Select all

S = {0,1,4,6,8,9,10,14,15,16,17,19,24,25}

The only possible values for x s.t. all of x+a,x+c,x+g,x+t are in S are:

Code: Select all

i: +a=8=i, +c=10=k, +g=14=o, +t=1=b
y: +a=24=y, +c=0=a, +g=4=e, +t=17=r

This leaves:

Code: Select all

S' = {6,9,15,16,19,25}

But there's only one x s.t. three of x+a,x+c,x+g,x+t are in S', which is:

Code: Select all

n: +c=15=p, +g=19=t, +t=6=g

There would have to be two such x for the one-time pad.

This means there's no 4-character one-time pad which encrypts an {A,C,G,T} sequence to "ijoopyaattzbqkipyerggjq".

I dare say you were on the edge of your seats about that one.

Rearrange the words "at", "straws" and "clutching" to form an appropriate famous word or saying.

deagol · Post by **deagol** » Thu May 10, 2007 4:26 pm

I'm not getting what you mean by a 4-character one-time pad. A one-time pad key is at least as long as the message to be encoded.

ignatzmouse wrote:Rearrange the words "at", "straws" and "clutching" to form an appropriate famous word or saying.

CHAR TWISTS CAN GLUT...

Oh wait, you said rearrange the words... sorry.

deagol · Post by **deagol** » Thu May 10, 2007 7:40 pm

I tried something similar to what you did here, but instead of having ijoopyaattzbqkipyerggjqa as the message to decrypt, I tried it as a key to turn some english text into a DNA sequence, which I think is how those telegram post stamps were used before. There's two conversion tables, depending if the english text was encrypted or decrypted with the ijoopy key:

English encrypted into DNA:

Code: Select all

a:srmmlcaahhbzkqslcwjuurka
c:utooneccjjdbmsuneylwwtmc
g:yxssriggnnhfqwyricpaaxqg
t:lkffevttaausdjlevpcnnkdt

English decrypted into DNA:

Code: Select all

a:ijoopyaattzbqkipyerggjqa
c:klqqraccvvbdsmkragtiilsc
g:opuuveggzzfhwqovekxmmpwg
t:bchhirttmmsujdbirxkzzcjt

The idea is to look for english words from the table, choosing amongst the four possible characters in each position. For example, using the first table I can make the following words for the beginning of the message:

stool (accca)
stoor (acccg) ha, though I doubt he'd be referencing me like that.
UK mole (ctacac)
UK Monica (ctaccgca) hey maybe Bill is involved

UK more (ctacgc)
UK Moria (ctacgga) another Tolkien reference

UK moric (ctacggc) as in moric acid
UK MSN (ctagc)
UK of reach (ctctgcaca)
UK of leg (ctctacg)
UK of let (ctctact)
UK of neat (ctctccat)
UK of net (ctctcct)
UK of Reagan (ctctgcagtg)
UK sole (ctgcac)
UK son (ctgcc)
UK sore (ctgcgc)
UK SS reach (ctgggcaca) not sure if MI6 (Security Service) is or was ever known as the UK SS. I know MI5 is known as SIS (Secret Intelligence Service).

UK FM reach (cttagcaca)
UK FM net (cttacct)
UK folic (cttcagc) as in folic acid
UK for (cttcg) I didn't look deeper on this one as this branch seemed quite foliaged
UK foe (cttct) idem
YT (gc) as in a YouTube account, with any of the words shown above for UK, or even more since it can be a proper name. Didn't check further.

Anyone care to check the other table? I'm tired.

Luminous · Post by **Luminous** » Thu May 10, 2007 7:47 pm

Whoa Deagol! This is amazing. If I could understand what you just did here I would be willing to give it a shot

deagol · Post by **deagol** » Thu May 10, 2007 8:06 pm

It's pretty simple, though tedious: pick one of the four letters for each column in the table, making up something meaningful.

For example: here's the second table...

English decrypted into DNA:

Code: Select all

a:ijoopyaattzbqkipyerggjqa
c:klqqraccvvbdsmkragtiilsc
g:opuuveggzzfhwqovekxmmpwg
t:bchhirttmmsujdbirxkzzcjt

So I'll pick the 'I' for the first column (a: or 1st row), then the 'p' for the second column (g: or 3rd row), then the 'o' for the 3rd column (a: or 1st row again), etc. Looks like the only thing remotely in english is ipoopacat and then I run into an impossible combination of consonants. Looks like the first table was more prolific.

Luminous · Post by **Luminous** » Thu May 10, 2007 8:09 pm

deagol wrote:It's pretty simple, though tedious: pick one of the four letters for each column in the table, making up something meaningful.

Geez! That's worse than an anagram! Do you really think that will get us anywhere?

deagol · Post by **deagol** » Thu May 10, 2007 8:20 pm

Luminous wrote:
deagol wrote:It's pretty simple, though tedious: pick one of the four letters for each column in the table, making up something meaningful.
Geez! That's worse than an anagram! Do you really think that will get us anywhere?

Right. I don't know. This is kind of what imouse has been trying the last couple of days. Check my edit just in case.

ignatzmouse · Post by **ignatzmouse** » Thu May 10, 2007 8:32 pm

deagol wrote:I'm not getting what you mean by a 4-character one-time pad. A one-time pad key is at least as long as the message to be encoded.

You're right, that was a bit vague wasn't it... I meant a one-time pad drawn from a 4-character alphabet. (This is just on the basis that it's a bit weird to only have 14 characters used out of 26, but not so weird to only have 14 characters used out of 16, which is the number you'd get from a 4-character alphabet for both the plaintext and key. Spot the desperation.)

ignatzmouse · Post by **ignatzmouse** » Thu May 10, 2007 8:36 pm

deagol wrote:
Luminous wrote:
deagol wrote:It's pretty simple, though tedious: pick one of the four letters for each column in the table, making up something meaningful.
Geez! That's worse than an anagram! Do you really think that will get us anywhere?
Right. I don't know. This is kind of what imouse has been trying the last couple of days. Check my edit just in case.

And just to make life more interesting, there's the 6 extra characters, so you're probably best off starting in the middle of the table to avoid the possibility that the plaintext starts or ends with some pading text. Or throwing your hands up in horror, which is what I've been reduced to.

deagol · Post by **deagol** » Thu May 10, 2007 8:37 pm

Yea this is probably not going to work. I'll highlight the problematic section from each table:

deagol wrote:English encrypted into DNA:

Code: Select all

a:srmmlcaahhbzkqslcwjuurka
c:utooneccjjdbmsuneylwwtmc
g:yxssriggnnhfqwyricpaaxqg
t:lkffevttaausdjlevpcnnkdt
            ^^^^^

English decrypted into DNA:

Code: Select all

a:ijoopyaattzbqkipyerggjqa
c:klqqraccvvbdsmkragtiilsc
g:opuuveggzzfhwqovekxmmpwg
t:bchhirttmmsujdbirxkzzcjt
          ^^^^^^

I'm pretty sure no word or phrase can come from those sections, in english or any other language. I can't be 100% sure though, the other day I was amused about this common, 7-letter english word that has no vowels (well, that's probably inaccurate since it has a 'y' acting as a vowel, but still).