Lonelygirl15

Posted: **Sat Jul 14, 2007 5:56 pm**

<sitting in front of computer with jaw dropped into lap>

You guys are so incredible. I may have to just sit here and chew on this pretty purple crayon now...unless you're using it, VB? <batting eyelashes>

Hmmm... Maybe the library paste is more my speed today anyway.

Don't forget, btw, with "stones" - when we're in a cemetery, regardless of whether we're londonfiling (LFing?) or WoWing...them's tombSTONES...so we could be going to THOSE stones, too? Did any of them seem noteworthy?

Posted: **Sat Jul 14, 2007 9:42 pm**

Since we're not GOing to them stones, bot instead GOT them, I'm inclined to believe they're small stones that were hidden in that stone bucket, which Kris found and is bringing with him for Mr. Gone. The thing is, if he's going to SF and Gone is in Georgia, how will he get them to him?

Posted: **Sun Jul 15, 2007 9:27 am**

Darn! Back to my Crayolas... <chuckle>

My first thought when you asked SF vs. Georgia was the fact that there's an international airport there - but why go to the west, just to go east again? Misdirection?

Posted: **Mon Jul 16, 2007 8:14 am**

Wow a lot has happened while I was gone.
Good work guys!

Now I have some catching up to do.

Posted: **Mon Jul 16, 2007 3:06 pm**

QtheC wrote:For the music code, my method may seem a bit complex, but I think it may be close. .

it is complex. The solution is simpler than that. Deagol got the notes right. From there, just ask yourself what would be the easiest way to encode a text using this technique. However, don't kill yourself trying to decode it. I should tell you that the information it reveals is outdated now. And I'm sure this code will be used again in the series.

Posted: **Mon Jul 16, 2007 3:27 pm**

impulse wrote: it is complex. The solution is simpler than that. Deagol got the notes right. From there, just ask yourself what would be the easiest way to encode a text using this technique. However, don't kill yourself trying to decode it. I should tell you that the information it reveals is outdated now. And I'm sure this code will be used again in the series.

So, for the musical folks out there.
How would you easily, encode a message with music?
In other words, try working this problem backwards. How would you do the encoding?

- - - - -
On another topic - here is Deagol' s notes again. First without sharps, then repeated with the sharps

Emaj(chord)
EA EE FA Emaj(chord)
AB GB FC GB Emaj(chord)
FFBEEEFAFFCGGGBA Emaj(chord)
FE EE GB BB BC FC Emaj(chord)

Emaj
EA EE F#A Emaj
AB G#B F#C# G#B Emaj
F#F#BEEEF#AF#F#C#G#G#G#BA Emaj
F#E EE G#B BB BC# F#C# Emaj

Posted: **Mon Jul 16, 2007 3:41 pm**

By the way, while I'm around and talking, great work on the baconian code deagol! And congratulations to silverblue for noticing it right away. You guys never miss an opportunity to impress me.

Posted: **Mon Jul 16, 2007 5:54 pm**

impulse wrote:
QtheC wrote:For the music code, my method may seem a bit complex, but I think it may be close. .
it is complex. The solution is simpler than that. Deagol got the notes right. From there, just ask yourself what would be the easiest way to encode a text using this technique. However, don't kill yourself trying to decode it. I should tell you that the information it reveals is outdated now. And I'm sure this code will be used again in the series.

That word "easiest" bugs me. The solution always appears easy once you find it, but like a needle in a haystack, there are a multitude of possible "easy" solutions one can conjure up. We haven't hit on the right one after working a long time on the problem, so it has not been easy to find.

It could be a code that assigns single notes to numbers that then translate into letters.

It could be intervals of notes that encode to numbers that then translate into letters.

The intervals could be determined in different ways, depending on which combinations of notes you choose.

Maybe everything is in reverse order for some reason.

The numbering base (if there is one) could be decimal, hex, octal, or something more unusual like base-6. Numbering can begin with zero or with one.

Pairs of notes could be decoded directly to an alphabet ... ignoring the interval values, just taking them as pairs of things and doing a substitution code, for example:

Code: Select all

          second
first      E     F#    G#    A     B     C#    D# 
 E         a     b     c     d     e     f     g
 F#        h     i     j     k     l     m     n
 G#        o     p     q     r     s     t     u
 A         v     w     x     y     z
 B
 C#
 D#

The faster line of notes could be taken as containing twice as many intervals, (encoding 8 letters), or as containing only 4 letters, but requiring a different range of numbers... or they could be a way of encoding numbers, or capital letters, or they could be taken as single notes relative to the Emaj key the way Deagol did like a Tabulature code, etc.

FFBE, for instance, in the fast line could easily be
2251 (single note intervals from Emaj) or
14 (intervals for pairs of notes FF and BE) or
44 (intervals for pairs of notes FB and BE) or
47 (intervals for pairs of notes FB and FE) or
FF could decode directly to a letter and BE could also
etc.

The code may be simple once it is known, but that does not mean it is simple to find.

Posted: **Tue Jul 17, 2007 4:09 am**

Impulse sent me a PM saying the subtitution code example I gave was close to the solution, but missing an important detail.

I'm beginning to feel like this is a waste of time and a guessing game from a semi-infinite set of possibilities with no actual path to a solution. To illustrate...

The code we have is:
Emaj
EA EE F#A Emaj
AB G#B F#C# G#B Emaj
F#F# BE EE F#A F#F# C#G# G#G# BA Emaj
F#E EE G#B BB BC# F#C# Emaj

or dropping the sharps,
Emaj
^ EA EE FA Emaj
^ AB GB FC GB Emaj
* FF BE EE FA FF CG GG BA Emaj
^ FE EE GB BB BC FC Emaj

Taking the notes pairwise (which may or may not be correct, especially for the third line - so that's one degree of uncertainty already), and just counting the pairs that appear we have something like this:

Code: Select all

          second 
first      E     F#    G#    A     B     C#    D# 
 E         ^^*   .     .     ^     .     .     . 
 F#        ^     **    .     ^*    .     ^^    . 
 G#        .     .     *     .     ^^^   .     . 
 A         .     .     .     .     ^ 
 B         *     .     .     *     ^     ^     .  
 C#        .     .     *     .     .     .     .  
 D#        .     .     .     .     .     .     .

where 
^ = a pair from lines 1, 2, or 4 of the code  
* = a pair from line 3 of the code (since it may be not be correct)
. = no pair of this type found in the code

so, there the message is not long enough to be subject to frequency analysis, but we can get an idea of the set of pairs used, and notice EE and G#B are the most common codes in the small sample that we have. No D# appear at all in the code, either as the first letter or the second of a pair.

The second line of the code has two of the common GB's in it: AB GB FC GB. If this is a direct translation to readable text, this might be something like GOTO or HERE.

If the text is further encoded after the subtitution code (such as by a vigenere cypher or something like that), then this puzzle is completely bogus without hints as to the solution. It would have far too many degrees of uncertainty making the solution space unreasonably large.

So, assuming that is not the case, then the problem may be to discover the subtitution code. Again, there seem to be a huge number of simple possibilities. None I have tried seem to work.

For example:

Is BE the same as EB, or are they different codes?
Is the matrix 5x5 (with some codes representing two uncommon letters such as j/k) or 6x6 or 7x7 or 8x8 or ???
What order are the letters filled into the matrix?
Should it be based on first note + interval value or some variant?

Code: Select all

7x7 matrix, filled in with first note + interval value order
          second 
first      E     F#    G#    A     B     C#    D# 
 E         a     b     c     d     e     f     g 
 F#        n     h     i     j     k     l     m 
 G#        t     u     o     p     q     r     s  
 A         x     y     z     v     w     a     b 
 B         c     d     e     f     g     h     i
 C#        j     k     l     m     n     o     p
 D#        q     r     s     t     u     v     w

GB gets q which seems unlikely if this is readable text.

Code: Select all

6x6 matrix, filled in with first note + second note order
          second 
first      E     F#    G#    A     B     C#    D# 
 E         a     b     c     d     e     f      
 F#        g     h     i     j     k     l 
 G#        m     n     o     p     q     r  
 A         s     t     u     v     w     x
 B         y     z     a     b     c     d
 C#        e     f     g     h     i     j
 D#        

GB again gets q which seems unlikely if this is readable text.

Code: Select all

8x8 matrix, filled in with first note + secondn note order
          second 
first      E     F#    G#    A     B     C#    D#    E' 
 E         a     b     c     d     e     f     g     h
 F#        i     j     k     l     m     n     o     p
 G#        q     r     s     t     u     v     w     x  
 A         y     z     a     b     c     d     e     f
 B         g     h     i     j     k     l     m     n
 C#        o     p     q     r     s     t     u     v
 D#        w     x     y     z     a     b     c     d

GB gets u... at least it is not q.
/ EA EE FA / AB GB FC GB / FF BE EE FA FF CG GG BA / FE EE GB BB BC FC /
= / d a l / c u n u / j g a l j q s j / i a u k l n /

Code: Select all

7x7 matrix, filled in with first note + interval order
and assuming that half the matrix is used (so AB = BA, etc.)

          second 
first      E     F#    G#    A     B     C#    D# 
 E         a     b     c     d     e     f     g 
 F#        .     h     i     j     k     l     m 
 G#        .     .     n     o     p     q     r  
 A         .     .     .     s     t     u     v 
 B         .     .     .     .     w     x     y
 C#        .     .     .     .     .     z     .
 D#        .     .     .     .     .     .     .

GB gets p which seems unlikely if this is readable text.

Posted: **Tue Jul 17, 2007 4:13 am**

impulse wrote:
QtheC wrote:For the music code, my method may seem a bit complex, but I think it may be close. .
it is complex. The solution is simpler than that. Deagol got the notes right. From there, just ask yourself what would be the easiest way to encode a text using this technique. However, don't kill yourself trying to decode it. I should tell you that the information it reveals is outdated now. And I'm sure this code will be used again in the series.

What is the solution?

Why is the information 'outdated' five days after the video appeared? Was there a 'tracking code' for a package or something that has now been delivered?

Posted: **Tue Jul 17, 2007 8:54 am**

QtheC wrote: Why is the information 'outdated' five days after the video appeared? Was there a 'tracking code' for a package or something that has now been delivered?

Impulse explained why the puzzle is outdated in the OoG discussion. http://lonelygirl15.com/forum/viewtopic ... 2&start=15

impulse wrote:
chershaytoute wrote: Will we ever get the opportunity, through the next pieces in the puzzle, to find the past pieces?
You will. Obviously some information is important for the future and these elements will come back into the story. However, some of the clues have a limited "relevance time". For instance, it's too late to try to decode the musical clue in the last video. It was a tough one but the information it reveals is sort of outdated already. (Found in time, that information might have led you to actually meet Kris and Mary)

I am guessing, the main characters had a limited window of opportunity to meet fans.
OoG - I feel two people responsible for these videos have been in Europe, summer backpacking?, and they may have traveled back home, San Francisco?, last weekend. It would fit the story line neatly, if Kris and Mary would meet with fans at an airport, since in the story they are traveling to the US.

QtheC wrote: What is the solution?

I prefer hints myself.
My wife and I worked on this a lot last night. We had a simple approach to it. We tried to assign numbers 0-9 to keys. That way we ended up with something to decode DEC to CHAR, or maybe hex to something. No luck with our attempts.
Since 24 letters are not easily represented with the notes played, numbers seem like the easier path.

It would be nice to know if the notes should convert to numbers or a readable phrase or a tinyURL or a YouTube video ID?

Posted: **Tue Jul 17, 2007 9:11 am**

I sense your frustration. So I'll try to help but I won't give you the answer. As you suggested, let's back engineer this code. And I hope you will see the methodology is quite logical and simple, even if the resulting code is difficult to break.

hint#1

Imagine, you want to send a message. In our western civilization we use an alphabet made of 26 letters and eventually 10 numerical characters (0-9) so you have an easy way to write down numbers. So we need 36 characters to be able to write efficiently a simple message.

Posted: **Tue Jul 17, 2007 9:17 am**

hint#2

now, you're looking for an easy way to encode and send this message so you want to use a limited set of characters or symbols or, why not, musical notes, to represent these 36 characters you need. What do you do?

Posted: **Tue Jul 17, 2007 9:52 am**

QtheC wrote: Why is the information 'outdated' five days after the video appeared? Was there a 'tracking code' for a package or something that has now been delivered?

Yes. That would really help if you had a time line, including all the known players and the different known videos posted on the different known accounts.

Posted: **Tue Jul 17, 2007 2:11 pm**

impulse wrote:hint#2

now, you're looking for an easy way to encode and send this message so you want to use a limited set of characters or symbols or, why not, musical notes, to represent these 36 characters you need. What do you do?

My wife tells me, this tune stays in one octave. One octave is 12 notes. It doesn't make since to try 3 octaves to get the 36 notes. Because we wouldn't know where to start and our code would come from one area of the range...
we would end up with junk like (hahhia) and never get letters like r, s, or t.

What makes more sense is to just modify 12 notes of an octave 3 different ways. Maybe whole notes, half notes, and quarter notes.

Mostly, I just want to share there are 12 notes in an octave and we need 36 different elements to decode. 36 can be divided by 12 so lets try something there.

Any feedback from the puppetmaster?

Lonelygirl15

impulse- "londonfiles_012" [07/11/07]