# Friend gave me a note I am not sure how to read. What does it say?

Tim Hansen gave me a note last week as we were leaving lunch and heading back to work. The paper was folded into quarters, and there were no directions to help understand what was written in the squares. He said that it should be easy to figure out, but the challenge might as well be shared with others.

The proper orientation for the message is unknown but probably is one of those shown below. Several images were created to archive the message as the original sheet of graph paper has now been discarded. The only useful peace of information Tim provided was that he called it a cipher of some sort.

• Does his name have anything to do with it or was it just plot/storyline? Commented Aug 6, 2015 at 2:48
• @warspyking There is no plot to this puzzle. This actually happened in real life. His name is probably irrelevant unless he embedded it within the message. If I am not mistaken, my friend created the cipher for a video game he is building. Commented Aug 6, 2015 at 11:08
• Ohhh sorry, a lot of puzzles here are based around this fake plot (for example look up Ernie in search). I didn't realize this was real haha. Commented Aug 6, 2015 at 11:33
• could the fact that it was folded into quarters correspond to anything? Commented Aug 7, 2015 at 16:40
• @SirParselot Probably not. It was just a convenient way of storing it in his pocket and passing it on to me. Commented Aug 8, 2015 at 13:29

I have a partial explanation, and an imperfect (but good enough) decryption method.

Building on the excellent work by GentlePurpleRain, I noticed a pattern, which is much more apparent if we apply a small modification to his setup.

First of all, arrange the 40 symbols by columns instead, like this:

AEIMQUY37@
BFJNRVZ48$CGKOSW159# DHLPTX260%  With this setup, the text looks like this:  6TM6+XSYX#+MW+2WV+5#4UT+%9 88UW+FI+PXQT+$U8#97+ZWN+6T M6+XSYX#+MW|RSMXYFMVXWMH+9 UXX+8U$U4+SMUZ+8U#56+18Q4+ BC@H37F5+NWZ+MP2+$Q2@U+JS+ A+D1E2#B%+MW+AFN@JK@DQ+23D Y03B+XS+12Q+EZEMPEFMPMX3|M VW+9ZZY$#+PX1O+OAPPHAIAJP+ LVI+Y$#D63B+1QP+9M4Y+6RUS|

As already established, + signs are spaces, and | are periods. Now, look at OAPPHAIAJP, which is the word that GentlePurpleRain already identified with SETTLEMENT.

If you look at the alphabet table, you will notice that the encryption is a simple shift to the right.

With this hindsight, I tried to decrypt the first sentence, and I managed to get this:

6TM6+XSYX#+MW+2WV+5#4UT+%988UW+FI+PXQT+$U8#97+ZWN+6TM6+XSYX#+MW| that+hcihw+is+ton+owned+cannot+be+dleh+ransom+rof+that+hcihw+is| Which, considering the reversed words, becomes obviously That which is not owned cannot be held ransom for that which is. Only problem is, I had to try different shift values for each word, and I didn't find any particular rule. In case anyone is wondering, for that line the shift where: -3, -4, -1, -2, -4, -5, -1, -3, -5, -2 (negative means to the left). I'll try and come up with a more precise rule. EDIT As pointed out by f'' in the comments, the shift values are defined by the word's length (minus one, depending how you count). I also have the full text: that+hcihw+is+ton+owned+ca nnot+be+dleh+ransom+rof+th at+hcihw+is|noitubirtsid+w ill+reven+gain+rewop+over+ noitaerc+for+eht+value+fo+ A+tcudorp+is+inversely+det aler+to+sti+availability|e no+offers+dlog+settlement+ dna+another+sih+warm+tfig| or better: That which is not owned cannot be held ransom for that which is. Distribution will never gain power over creation for the value of a product is inversely related to its availability. One offers cold settlement and another his warm gift. • Those shift values look like the length of the word, minus one. – f'' Commented Aug 7, 2015 at 15:56 • Wow. Now I really feel dumb, you are definitely right. I'll update the answer. Commented Aug 7, 2015 at 15:57 • If someone could put that full decrypted text in a multiline spoiler tag somehow, that would really be appreciated Commented Aug 7, 2015 at 16:09 • Huge congrats! I've spent hours staring at this, and had come to the conclusion that every word must be encrypted using a different alphabet, since there were no valid words that matched the patterns in the ciphertext. Kudos on figuring out how! (also, I spoilerized your solution for you) Commented Aug 7, 2015 at 20:22 • The shift depending on the word-length threw me off. Since I saw so many repeated words, I assumed the alphabet wasn't shifting. But when you always shift it by the same amount for any given word, the repeats still happen! Commented Aug 7, 2015 at 20:29 Since I find it very difficult to understand random symbols, I transcribed this cipher into text, using the following key: (The symbols I used are obviously not the intended decryption; I just used them to make it easier to peruse the text and find potential patterns.) As noted by @Mircea, the$+$symbols are likely word breaks, and the$|$symbols are likely sentence breaks. Also as noted by @Mircea, every second word (with one exception) is "upside-down". As indicated below, these words have been inverted as a whole (the entire word has been rotated, as opposed to each individual letter). I have indicated the "upside-down" words by enclosing them in parentheses. Here is the transcribed text (as it appears above): $9D$+ (0YG04) + DZ + (@ZP) + 24RF9 + (%3##FZ) + LC + (80E9) + TFS43I + (QZN) +$9D$+ (0YG04) + DZ | (OYD0GLDP0ZD6) + 3F00 + (SFTFR) + YDFQ + (SF42$) + 1SER + (KUJ6H1L2) + NZQ + (D8@)
+ TE@JF + (MY) + A + (51B@4K%) + DZ + ALNJMWJ5E + (@H5G#HK) + 0Y + (1@E) + BQBD8BLD8D0H |
(DPZ) + 3QQGT4 + (801X) + XA886ACAM8 + (7PC) + GT45$HK + (1E8) + 3DRG + ($OFY) |


With the inverted words inverted (so all words now read left-to-right):

$9D$ + 40GY0 + DZ + PZ@ + 24RF9 + ZF##3% + LC + 9E08 + TFS43I + NZQ
+ $9D$ + 40GY0 + DZ |
6DZ0PDLG0DYO + 3F00 + RFTFS + YDFQ + $24FS + 1SER + 2L1H6JUK + NZQ + @8D + TE@JF + YM + A + %K4@B15 + DZ + ALNJMWJ5E + KH#G5H@ + 0Y + E@1 + BQBD8BLD8D0H | ZPD + 3QQGT4 + X108 + XA886ACAM8 + CP7 + GT45$HK + 8E1 + 3DRG + YFO$|  My observations • There is one place (near the middle of the fourth line of the transcription) where there are two words that have the same orientation: DZ + ALNJMWJ5E. It is possible that DZ is actually inverted (because both of those symbols have rotational symmetry), but since DZ appears several other times in the text, it is unlikely. (Even if it were inverted, there would still be two words with the same orientation next to each other: (51B@4K%) + (DZ)). • There are several long words with very distinctive letter patterns in the text: OYD0GLDP0ZD6, BQBD8BLD8D0H, XA886ACAM8. Plugging these individually into a cryptogram solver, we get <nothing>, AVAILABILITY, and SETTLEMENT. If we reverse the order of the letters in OYD0GLDP0ZD6 to get 6DZ0PDLG0DYO, then it decrypts to DISTRIBUTI[VE|ON|NG]. This means it's likely that the "inverted" words are inverted as a whole, not as individual letters, and thus the 2nd transcription is the more accurate one in terms of helping with decryption. The existence of both (QZN) and NZQ reinforces this hypothesis. It is also possible that the "inverted" words are meant to be read from end to beginning, as a separate sentence of sorts, but I imagine that will be made clear once the text is decrypted. • There is a problem, though. AVAILABILITY uses the same symbol for L that SETTLEMENT uses for T. So, either (1) one of these is incorrect, or (2) the cipher somehow shifts the meaning of the symbols between each word/sentence. • There are several words that are repeated at least once in the ciphertext: DZ, $9D$, (0YG04), NZQ (and, in fact, the whole phrase $9D$+ (0YG04) + DZ). This would indicate that the meaning of the symbols likely does not shift between words, and the AVAILABILITY/SETTLEMENT problem above must have another cause or reason. • The only symbol that doesn't appear in the ciphertext is V. That means there are 39 distinct symbols in the text. They must represent more than just the 26 letters of the alphabet, but it is unclear what else they might represent. The fact that there are unique decryptions of the three long words mentioned above makes it likely that all the symbols they contain represent distinct letters of the alphabet, but then I don't know what the extra 13/14 symbols are for. Maybe some alphabet letters have two different symbols to represent them? Here is a frequency analysis of all the symbols: D 15 3.82% 0 12 3.05% 0 12 3.05% 1 6 1.53% F 10 2.54% 2 3 0.76% 8 9 2.29% 3 5 1.27% Z 9 2.29% 4 8 2.04% 4 8 2.04% 5 4 1.02%$  7  1.78%            6  3  0.76%
Y  7  1.78%            7  1  0.25%
G  7  1.78%            8  9  2.29%
@  6  1.53%            9  4  1.02%
Q  6  1.53%            #  3  0.76%
E  6  1.53%            \$  7  1.78%
1  6  1.53%            %  2  0.51%
T  5  1.27%            @  6  1.53%
A  5  1.27%            A  5  1.27%
H  5  1.27%            B  4  1.02%
3  5  1.27%            C  3  0.76%
L  5  1.27%            D 15  3.82%
J  4  1.02%            E  6  1.53%
K  4  1.02%            F 10  2.54%
R  4  1.02%            G  7  1.78%
P  4  1.02%            H  5  1.27%
B  4  1.02%            I  1  0.25%
S  4  1.02%            J  4  1.02%
5  4  1.02%            K  4  1.02%
9  4  1.02%            L  5  1.27%
#  3  0.76%            M  3  0.76%
C  3  0.76%            N  3  0.76%
2  3  0.76%            O  2  0.51%
M  3  0.76%            P  4  1.02%
N  3  0.76%            Q  6  1.53%
6  3  0.76%            R  4  1.02%
%  2  0.51%            S  4  1.02%
X  2  0.51%            T  5  1.27%
O  2  0.51%            U  1  0.25%
7  1  0.25%            V  0  0.00%
I  1  0.25%            W  1  0.25%
U  1  0.25%            X  2  0.51%
W  1  0.25%            Y  7  1.78%
V  0  0.00%            Z  9  2.29%

• It doesn't follow a English-like distribution (it's much more spread-out, rather than focused on the common letters like ETAIONS). This makes me suspect that maybe some of the most-common letters are represented by two or more distinct symbols.
• Every symbol but one is used. If in fact the 40 symbols at the top of the ciphertext are the entire alphabet, that means that every English letter but one is used in the ciphertext.
• Perhaps @Mircea's hypothesis about the Shavian alphabet makes sense. Since it contains many more "vowel sound" symbols, its frequencies would tend to be more spread out. If it is correct, then my decryptions of the longer words above don't apply, since there is not a one-to-one correspondence of each Shavian letter to an English letter.
• While thinking about this today, it did seem quite likely that some of the symbols represent the same characters. To help defeat frequency analysis, having more than one symbol for the most common letters would be rather helpful. Could guessing at some of the shorter words be of any help? Commented Aug 7, 2015 at 1:51
• Some great work here. Just an idea: multiple symbols for the same letter is something that happens all the time if you consider uppercase and lowercase, though I'm not sure that is what is happening here. Commented Aug 7, 2015 at 8:50

Some guesses up until this point:

the small block is the cipher - no 2 symbols are repeated.
+ is space (it does not show up the cipher)
also the interesting thing about the cipher is that it's 40 symbols - makes me think of Shaw/Shavian Alphabet
also if we accept that + is space, the letters in some words are taken directly from the cipher, in other words the letters are mirrored top to bottom cipher letters (for last image - first word direct, 2nd word mirrored, 3rd word direct, 4th mirrored...)

Close?

• That sounds good to me, but what about the | symbol? It does not appear in the 4x10 block of text either. Commented Aug 6, 2015 at 11:12
• My guess is that it would be a dot (end of sentence).
– Ivo
Commented Aug 6, 2015 at 11:25
• I fear that if this is using the Shavian Alphabet, it will be nigh-impossible to decrypt. Commented Aug 6, 2015 at 20:09
• @GentlePurpleRain It is seriously doubtful that my friend knows anything about the Shavian Alphabet. He thought I could solve the puzzle, and I had never heard about such an obscure alphabet before reading about it here. Commented Aug 7, 2015 at 1:53