Saturday, May 17, 2008

Breaking the Fermilab Code

A story appeared on Slashdot about a mysterious fax received at Fermilab written in an unknown code. The full story is here. I looked at it and immediately noticed a few things:

1. The first part looked like ternary (base 3) with digits 1 (|), 2(||) and 3(|||).

2. The last part looked like binary with digits 1(|) and 2(||)

3. The middle bit looked like either a weird substitution code, or I wondered if it might be machine code.

4. In the last part the digit 2 (||) never occurs more than once, perhaps it was actually a separator and the last part is not binary.

The first step was to convert the bars into numbers. Here's a copy of my marked up print out:



The first part has the numbers (or at least I thought):

323233331112132
333231322123312
111331132312233
333212123213113
311333313331111
211333323232211
232313331121231
33231312

Noticing this had 113 digits (which is a prime number) I went off on a wild goose chase around primes, and then around the interpretation of this number in hexadecimal as a string in ASCII, Unicode or binary... waste of time.

Then I started thinking about ternary again and wrote down the largest ternary numbers that can be expressed with 1, 2, 3, ... digits:

23 = 210
223 = 810
2223 = 2610
22223 = 8010

One of those stood out: with three digits the maximum number is 26 and there are 26 letters in the alphabet! Then the only question was was how to map the three digits used in the code (1, 2, 3) to the three ternary digits (0, 1, 2).

To simplify things I wrote a small Perl program that tries out all the possible mappings and outputs the ternary interpreted as a string (with 001 = A, etc.):

use strict;
use warnings;

my $top = $ARGV[0];

$top =~ tr/321/abc/;

my @chunks;

while ( $top =~ s/^([abc]{3})// ) {
push @chunks, $1;
}

my @digits = ( '0', '1', '2' );

foreach my $d0 (@digits) {
foreach my $d1 (grep {!/$d0/} @digits) {
foreach my $d2 (grep {!/[$d0$d1]/} @digits) {
print "($d0$d1$d2) ";
foreach my $c (@chunks) {
my $v = 0;
my $m = 1;
foreach my $d (reverse split( //, $c )) {
$d =~ s/a/$d0/;
$d =~ s/b/$d1/;
$d =~ s/c/$d2/;
$v += $d * $m;
$m *= 3;
}
print chr( 64 + $v );
}
print "\n";
}
}
}

With my initial interpretation of the top part of the coded message I got the following output:

(012) CIBYS@KDUGZBSGI@PUOXH@FBZQ@CJQJFBWKAF
(021) FRANK@SHOEMAKER@WOULD@CAMV@FTVTCAPSBC
(102) JDNXUMEISOZNUODMFSGYQMPNZHMJCHCPNTELP
(120) PVLBEMUQGK@LEKVMTGSAIMJL@RMPWRWJLFUNJ
(201) THYLOZGRKUMYOUHZCKENVZWYMDZTFDFWYJGXW
(210) WQXAGZOVES@XGSQZJEKBRZTX@IZWPIPTXCOYT

A ha! The 021 block (which corresponds to the mapping 3 -> 0, 2 -> 2, 1 -> 1) seems to have a partial message: FRANK@SHOEMAKER@WOULD and then it's garbage. Going back to the original message I realized that 113 is not divisible by three and that I'd either missed a symbol, or had two too many.

After much fiddling around I discovered that the correct interpretation of the top block is that two of the threes are wrapped from one line to another (there appears to me some indentation in the message that indicates this, take a look at the original, but this could be just random).

323 233 331 112 132
333 231 322 123 312
111 331 132 312 233
333 212 123 213 113
311 333 313 331 113
113 333 232 322 133
231 333 112 123 133
231 312

Rerunning my Perl program output the full message:

(012) CIBYS@KDUGZBSGI@PUOXH@FBXX@JDRK@YURKG
(021) FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE
(102) JDNXUMEISOZNUODMFSGYQMPNYYMCIVEMXSVEO
(120) PVLBEMUQGK@LEKVMTGSAIMJLAAMWQDUMBGDUK
(201) THYLOZGRKUMYOUHZCKENVZWYNNZFRQGZLKQGU
(210) WQXAGZOVES@XGSQZJEKBRZTXBBZPVHOZAEHOS

So much for the first part. The second part took me off into Z-80, 6502 and 6809 machine code wondering if it was a program and then nowhere. I still don't understand what this part is trying to say.

The third part looked initially like binary but on closer examination I decided that the 2s (||) were actually separators and the message should be interpreted as number separated by 2s by counting the 1s (|). That yields:

31211112111312
32213123123331
12213111332312
23333333233123
12313123332311
33223232312312
112

(Once again there was a wrapping 'problem' in the message where a run of 8 |s was actually 3 |s then 1 || and 3 more |s.) Using the little Perl program reveals:

(012) GZWXUNGG@YOZAGI@ABKKG@KRLJGGY
(021) EMPLOYEE@NUMBER@BASSE@SIXTEEN
(102) OZTYSBOOMXGZLODMLNEEOMEVACOOX
(120) K@FAGXKKMBS@NKVMNLUUKMUDYWKKB
(201) UMJNKAUUZLEMXUHZXYGGUZGQBFUUL
(210) S@CBELSSZAK@YSQZYXOOSZOHNPSSA

So, the same mapping between digits is used.

That leaves some final questions:

1. Who is Frank Shoemaker?
2. Why is base spelt incorrectly?
3. Is the extra S in BASSE a reference to the middle section where three symbols start with S.
4. If #3 is correct, then those three symbols could be intepreted as FC16 which is 252. Could this be the employee number of the author?
5. Why is the letter A missing from the middle section when all the other hexadecimal digits are there?

Labels:

91 Comments:

Blogger ignoti said...

http://www-boone.fnal.gov/cgi-bin/boone_search?FND=All&KEY=shoemaker

It would appear that Frank Shoemaker is a physicist who worked on the BooNE neutrino experiment.

4:07 PM  
Blogger Douglas said...

http://www-tele.fnal.gov/cgi-bin/telephone.script

Gives an email address for a FRANK C SHOEMAKER.

Random thought: the S could be the missing symbol for A, so the SDt => AFC_16 => 2812.

4:09 PM  
Blogger John Graham-Cumming said...

Employee 2812 appears to be EVERETT, ADAM ALEXANDER at Purdue University.

4:22 PM  
Blogger John Graham-Cumming said...

Adam Everett replied to my email saying that he is not the author, that his ID is not 2812 and that he believes that 2812 is not even a valid ID number. Looks like we'll have to look harder at 252.

5:33 PM  
Blogger Dag �gren said...

http://www.gmilburn.ca/2008/05/17/fermilabs-strange-letter-progress/

7:00 PM  
Blogger Alex said...

"basse" might refer to the region in france?

7:22 PM  
Blogger John Graham-Cumming said...

I wondered about whether 'basse' might need to be interpreted in French. There is the region 'Basse Normandie', but more generally 'basse' in French means low, deep, bottom... If the writer is a French speaker then this could be a clue where they are saying that the employee ID is written in hex and at the same time making an illusion to the fact that it's the hex at the bottom of the block in the middle.

Interestingly, there is a Fermilab person with ID 252 (0xFC) called Pierre Piroué (he's a professor at Princeton).

I sent him a mail.

8:01 PM  
Blogger robf said...

I'm still at a loss for the symbols... Why, the symbol "key" itself must have a message and not simply for the "basse" sixteen employee number of whomever. An oddity in the symbol table is the lone Phi - 7... it's the only symbol I readily recognize. Flipping through about 30 some odd symbol sets from ancient times (cuniform, goth, runics, etc..) as phi pops up a few times in these sets, but I am wondering what made the choice for those specific symbols, or was it arbitrary from the author. this bothers me.

9:42 PM  
Blogger Craig said...

Could be that phi is somehow alluding to the golden ratio

6:23 AM  
Blogger Craig said...

Could phi be referring to the golden ratio (for some odd reason)

6:23 AM  
Blogger Craig said...

Also I am wondering why "basse sixteen" is even mentioned when it's obvious to just guess this (for the 2nd paragraph) without even decoding the others (maybe this is noise to). I too am worried about "basse"... and "s.."...possibly the hex code has to be substitututed back in somewhere. For example may it modifies "basse" to reveal another number (only word would by forty or fifty, giving forty or fixty sixteen as the employee number). Just random thoughts :-)

6:28 AM  
Blogger John Graham-Cumming said...

BTW I sent an email to Pierre Piroué and he was kind enough to answer quickly, but, unfortunately, my theory about him was wrong. He denies being the author.

7:02 AM  
Blogger John Graham-Cumming said...

The capital Phi is interesting, and in physics it does have meaning (electrical or magnetic flux).

7:10 AM  
Blogger marduc said...

you ought to read up on slashdots article for an interesting take on the BASSE word (in the comments)..

8:51 AM  
Blogger Craig said...

I suppose you've seen this:

http://www.fermilabtoday.com/pub/about/whatis/wilsonhall.html

note the references to "basse"

9:01 AM  
Blogger John Graham-Cumming said...

Marduc... thanks for pointing me to the new Slashdot story. I hadn't realized that they had linked here.

Marduc/Craig... the basse speculation is interesting. Perhaps this person simply worked in that building.

9:43 AM  
Blogger Mike said...

Look at the first picture. It's a table of ANSI character graphics. I believe these are the characters in the second stanza.

http://technet.microsoft.com/en-us/library/ms947792.aspx

9:46 AM  
Blogger John Graham-Cumming said...

Mike... I don't see all the symbols there, but perhaps a search of the Unicode tables would be interesting.

9:57 AM  
Blogger echaozh said...

maybe the ss in the word basse should be replaced by fc, making the word bafce (765902)?

10:19 AM  
Blogger pauljs75 said...

Caught this on slashdot, this seems like an interesting puzzle. Dunno if I can figure much out, but its fun to watch people solve this.

Googling "phi value encryption" gave something on RSA Alogrithm in the results that might be interesting. Dunno if there's other methods that use the phi value, but that's the first one that popped up.

Also note that the first two lines match up in spacing and pattern. So the legible text and line above it are possibly a key to the remaining encrypted text. I'd suspect that if you figure the offset of the readable line and the line above and use some encryption formula (just might be RSA) to further shift that, it might say something legible.

What I'm wondering is if there's something more interesting to it, or just some exchange between bored physicists with ciphers/cryptography as a side hobby. Another guess is that it's some thought out promotional stunt for using RSA encryption.

10:31 AM  
Blogger inertnet said...

How about this: http://en.wikipedia.org/wiki/Basse_chiffree

11:15 AM  
Blogger Johan said...

maybe the 'ss' is refering to the German sign which looks a bit like the Greek beta symbol and is often replaced by 'ss' in typed text.

any Germans working at Fermilab?
Or perhaps it is coming from someone from a German research institute that worked with Frank Shoemaker?
This link has lists research centers:
http://www.rarf.riken.go.jp/rarf/np/nplab.html
for some reason, GSI drew my attention...
doing a search on 'element' brought me to the link for element 111:
http://www.gsi.de/forschung/kp/kp2/ship/el111_e.html

you were talking about '2812'. Notice the pages that are mentioned in the link 281-282... probably a coincidence

11:38 AM  
Blogger joachim said...

The indentation in the first section is not random, nor is it significant: it's simply a 46-wide grid with linebreaks occurring sometimes before a space, sometimes after, and sometimes in the middle of a symbol.

Why 46? Might be just the number of cells per line on a sheet of ruled paper. My A4 pad with a 0.5cm grid inside significant margins has 47 cells to the long side.

11:45 AM  
Blogger derbes said...

I was an undergrad physics major at Princeton 1970-74. Frank Shoemaker taught the big, tough Halliday and Resnick Phys 103 course, and later my group of majors advanced E&M. We knew he was a master at magnet design and construction.

Pierre Piroue was on the faculty then, so these guys know each other.

Shoemaker must be quite elderly now, in his eighties (I'm 56, and he had to have been at least 40 when I took his course in 1970.)

My guess is that the coder has a Princeton connection, but who knows?

David Derbes
U of Chicago Lab Schools
loki@uchicago.edu

11:48 AM  
Blogger John Graham-Cumming said...

I don't have much time left to work on this now, so I'm going to mostly sign off with the thought that this author has used ternary twice (top and bottom), so perhaps the middle is also ternary (although I've spent hours trying to decode it like that).

1:02 PM  
Blogger Bry said...

I believe the extra S in BASSE is designed to point out the significance of the S in the 3-char code. It's the only character not present in the 2 lines above. So what does sFC mean?

Subtract FC?
Swap F and C?

1:37 PM  
Blogger Joe said...

The characters appear one, two or three times in the second section (excluding the three non-key glyphs).

1:37 PM  
Blogger Chris said...

IF Basse means low in french, maybe it denotes that the Hex is LSB, MSB.. i.e. CF(16) = 207 decimal...

2:08 PM  
Blogger Derek said...

There must be a significance to the repetition of symbols in the middle section.

2:09 PM  
Blogger ghemeon said...

Frank Shoemaker is working on the miniBooNE project which is looking at quantum noise ...

2:13 PM  
Blogger Greg said...

I'm thinking the symbols in the middle are more of a pictorial puzzle. Perhaps if the employee number is converted to base 16 the corresponding symbols would reveal another message. For example E8C5 would form the word OIL if the 8 and C are combined to form the letter I (and you have to look in a mirror).

2:49 PM  
Blogger Stephen said...

BASSE looks like a "calculator" word to me. If you spelled this in a 7-segment LED and turned it upside down it would read:
35549 or 35569

3:13 PM  
Blogger tedalan said...

what if the actual word "BASSE" should be interpreted as hex? Referencing S in the middle to correspond to "FC" yields BAFCFCE. The sentence could then read "EMPLOYEE NUMBER BAFCFCE 16". just a thought.

3:15 PM  
Blogger pauljs75 said...

Took another look at something in relation to how the placeholders line up in the message. So I'm gonna guess at something different.

Cipher: CIBYS@KDUGZBSGI@PUOXH@FBXX@JDRK@YURKG
Key:????(Don't know yet)?????
Message: FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE

Now on the next four lines, notice how the placeholder @'s on the top two also line up with the @'s on the bottom two.

So let's try making the top two lines the code, and the bottom two lines the key. (or vice versa)

Code: JDNXUMEISOZNUODMFSGYQMPNYYMCIVEMXSVEO
Key: THYLOZGRKUMYOUHZCKENVZWYNNZFRQGZLKQGU
Message:
?????

I'm not sure if a simple alpha grid is the right way... (ABC across, then from A, ABC down. Fill out the grid and rollover.) That seems like it'd be too easy to solve, but it looks like it might say something. So go down vertically on the left to J, then across to where T is. Now go back vertically to the top from where T is, and you get K. Etc, etc. Looks like it spells out some actual words, or perhaps a person's name. (At least from the first dozen letters or so I looked at. I don't have a straight edge and a printout to make following the grid easier at the moment.) But if it's that simple, then what's the middle part about?

On another tangent, somebody at SlashDot was saying "basse 16" could be actually referring to a building in a pecular way. (Something about the Wilson Hall design relating to some cathedral in France. Crazy stuff (almost like DaVinci code or something), but interesting none the less.) What if "employee number" wasn't an ID thing, but rather how many people were working there at the time the code was made?

3:30 PM  
Blogger Patrick Wall said...

Just a guess, but if we assume that the code on the top is talking about the top of the substitution code in the middle, then it's just noise. The bottom part of the substitution code should have the employee number in hexadecimal if you interpret it this way. If you look at the bottom part, it's all in hex, except for the S. So I'm just ignoring the S as an extra for now.

Converting Hex to ascii takes a pair of hex numbers, and without the S there are exactly 13 pairs of numbers or 13 letters. So translating:

F0 BE 58 F2 FD 63
6C 79 D2 E3 93 E6
FC

gives us:

ð¾Xòýc
lyÒã“æ
ü

This is just a load of gibberish, so maybe there's a trick to it. The only hint I can think of is the extra S, which you can convert into the hex number 53. Maybe applying it in some way to the hex numbers will decipher it, but I'm too tired to think about it more.

3:31 PM  
Blogger thoma said...

has anybody noticed that the geographic position of basse-terre(http://de.wikipedia.org/wiki/Basse-Terre_(Stadt) the english wiki doesn't seem to know this...) a city in france is exactly

16° 0′ N, 61° 43′ E !

5:07 PM  
Blogger mantar! said...

And now ...

Something else to look at :)

Notice that in your final Perl decode of section 1, reading *down*, column-wise, the same string of "@@MMZZ" occurs after each(?) word break.

Have at it!

5:24 PM  
Blogger Eric said...

Wilson Hall at fermilab has 16 stories, on the 7th floor it the 2 side columns are joined by cross overs.

http://www.fnal.gov/pub/about/campus/wilsonhall.html

Basse Ouevre has 7 columns on the back of it. in the 1st and 3rd stanza there are 7 sets and and an 8th partial. the Basse Ouevre's columns are in a semicircle, the center one could count as a 2 if you viewed the building from one side the other allowing the center column to double itself visually.

5:30 PM  
Blogger Ben said...

the noise in the background of the image is totally a map of the sky! Dippers are clearly visible in between the first and second section...

Anyone know how to do an analysis to figure out the date when the northern sky looked like that?

6:02 PM  
Blogger Eric said...

start in march 05, and get a sky map that spins around showing the different stars at different times of the year. The original was sent from chicago.

6:23 PM  
Blogger Eric said...

march 2007.... for some reason March 5th was in my mind?

6:29 PM  
Blogger Ben said...

actually if I'm not mistaken, the image is mirror imaged left-right ... though not being from up north I'm not 100% sure... Anyone with an eye for the north sky handy?

6:36 PM  
Blogger Rupert said...

Maybe the middle part contains a telephone number?

"FRANK SHOEMAKER WOULD CALL THIS NOISE" might be interpreted as "Frank S. would make a phone call to the number embedded in this (what seems to be) noise". Once connected, you'd have to ask for employee #252 -- or dial the extension.

The difficult part is finding a phone number in there...

Here are some ideas, but I haven't come up with any solution yet.

Replacing the 24 symbols with their respective frequency yields
311 311 323 232
311 222 312 233
or, with the usual modulo-3 mapping
011 011 020 202
011 222 012 200

that makes decimal
4 4 6 20
4 26 5 18
-- which doesn't look anything like a phone number.

Maybe each column gives one digit:
044255072806 or 044677056802
-- still not good

Maybe interpreting the numbers as letter as before and then dial that string using the letters on phone buttons?
DDFTDZER -> 33383937
-- still no luck.

(at least not if we assume that it is a US number)

All that assumes that the hexadecimal mappings are there only to decypher the employee number 0xfc -- and to give a hint to look at the frequencies, as it would otherwise have redundant entries.

Still, phone number or not, I think we are missing something obvious. The top and bottom part have been rather straight forward, almost simple. I don't expect the middle part to have some super-complex multi-layered encryption.

6:40 PM  
Blogger Daniel said...

Random thought:

What if BASSE SIXTEEN is a phone number? It works as a 10 digit phone number with a 2 digit extension... if you map the letters to the keypad of a phone.

Just a thought.

BTW: comes out to (227) 737 4983 #36 a Maryland Number... Lots of employees from University of Maryland there...

6:41 PM  
Blogger Aigarius said...

Has anyone tried to really decode the:

F0 BE 58 F2 FD 63
6C 79 D2 E3 93 E6

???

If we assume that it is a sequence of bytes and then decode it using the tertiary code of the other two parts, we will get ...

012 212 000 010 210 102 010 120 200 201 010 211
011 121 202 210 202 022 110 201 002 220 012 220
EW@CUKCORSCVDPTUTHLSBXEX

:(

6:51 PM  
Blogger Charles said...

Is it possible that it's not a phone number but an integer sequence?
http://www.research.att.com/~njas/sequences/Seis.html

this link shows at least the second and third sets of numbers related to eachother
http://www.research.att.com/~njas/sequences/b006218.txt

6:57 PM  
Blogger Aigarius said...

About the phone number idea: 0 is often the extension to get to an outside line, +44 is the country code of UK, but in UK no phone numbers start with 6 or with 25. Dead end there, but don't assume that any phone number is from US ...

7:02 PM  
Blogger Ben said...

Bingo, confirmed that my suspicion of the stars being flipped looks like it's spot on... I've got a good mapping of stars to 'noise' when it's flipped... Perhaps the code is actually written on a transparent film from a telescope?

If you're having trouble spotting it, polaris is to the right of the last line on the first paragraph.

Also, it might have some kind of filter, or perhaps not be optical, as stellarium doesn't have the small triangle of stars above the 4th character of the 2nd paragraph.

(oh, there seems to be virtually zero chance to match this to a time/date when it was taken as there seem to be many many dates when it appears to match up... either that or there's some trick to date-by-stars I'm unaware of...)

...what's bugging me is what do the stars have to do with frank shoemaker, or is it perhaps just coincidence that it was the nearest piece of scrap paper to the author at the time?

7:36 PM  
Blogger Ben said...

Oh, agarius, how did you get that sequence of trinary numbers from the hex code?

When I did it I got:

120000102112012022120002120111020010020110021100111002112011100010112020

7:38 PM  
Blogger Aigarius said...

Could someone redo the decoding of the middle part assuming little-endian 16bit encoding (basse is supposed to mean 'low' or 'little' in French)?

8:24 PM  
Blogger inertnet said...

>F0 BE 58 F2 FD 63
>6C 79 D2 E3 93 E6

I think that should be read as:

F6 0C B7 E9 5D 82 FE 23 F9 D3 6E 36

or:

6F C0 7B 9E D5 28 EF 32 9F 3D E6 63

8:35 PM  
Blogger 6smail said...

Hi,
I just think loundly:

what if we consider that to be simpler than what it seems.
I think "basse sixteen" means the hex 16 in little endian.
starting from this we can have :
'employee number 620' (to octal)
'employee number 1021' (in hex)

please try to find if these ID refer to the author

My second thought is that the middle part is some machine code, and all the symbols are in fact just what Mr Shoemaker calls "noise", we just need to know that phi referes to 7 and S to FC

10:27 PM  
Blogger 6smail said...

I think the code is
F7 BE 58 F2 FD 63
6C 79 D2 E4 93 E6
FC

it looks like some machine code that we can translate to assembly(I did some assembly last year and it seems to be similar to some code I saw)I'll try to translate it tomorrow.

10:46 PM  
Blogger 6smail said...

Could the S be E6(the combination of the two symbols corresponding to E and 6 made a symbol corresponding to S), so could it be :
"employee number E6FC" ?

or S.F.C stands for the Signature of Shoemaker Franck ?

12:20 AM  
Blogger NimbleSquirrel said...

Someone touched on this on Slashdot, but the symbols in the second stanza were interesting. Nearly all have Unicode equivalents (some even multiple possibilities).

Here is what I found:
0=U+0069 (lower case i)
2=U+203A (single right pointing angle quotation mark)
3=U+29A3 (reverse angle)
4=?
5=? (looks like a circle, but with a mark on the lower right)
6=U+00AC (not sign)
7=U+03A6 (Greek capital phi)
8=U+2E07 (Raised dotted interpolation marker)
9=?
B=U+002D (minus sign)
C=possibly U+03BB (lower case lambda)
D=U+10E2 (Georgian letter Tar)
E=U+27D3 (Lower right corner with dot)
F=U+25C4 (white left pointing triangle)

It is odd to have the hex with the symbol. If you are going to hide a message in hex, doesn't that make the symbols redundant? The person that made this isn't that silly. Maybe the message is hidden in the unicode bits all combined, and the hex digit is actually some kind of checksum? Within Unicode, there are many similar symbols with different code. A checksum of some sort would be a way to ensure the reader has the right unicode for the symbol.

12:49 AM  
Blogger Ivan Godard said...

"employee number basse 16" - French female Fermilab employee with id < 16.

7:27 AM  
Blogger Ivan Godard said...

This post has been removed by the author.

7:28 AM  
Blogger Ivan Godard said...

This post has been removed by the author.

7:38 AM  
Blogger Andrew Jackson said...

@Ben:
> Anyone know how to do an analysis
> to figure out the date when the
> northern sky looked like that?
It could be the southern sky just to confuse people.

10:26 AM  
Blogger Ben said...

Andrew, I'm quite confident it's a mirror image of the northern hemisphere's sky, there's a very good correlation, check it out yourself if you don't believe me.

11:13 AM  
Blogger Ben said...

It's possible the things he's referring to are astronomical objects to be found on the image of the sky... eg, wasn't there a comet a few years back called 'shoemaker levy ' or similar?

I'm thinking it's possibly referring to an astronomical event that was captured on the film in question...

2:31 PM  
Blogger thoma said...

if you try to find a corresponding stellar constallation try
16° 0′ N, 61° 43′ E
as the point of view. this is the exact possition of the city basse-terre on the caribian island guadeloupe.

i believe that it is more than an incident that this city is exactly at 16° 0' N. isn't it quite unlikely that the text reffers to basse 16 and this place doesn't have any thing to do with it?

5:21 PM  
Blogger thoma said...

btw. the name of the comet was schoemaker-levy 9
but the full name of this schoemaker was Eugene Merle Shoemaker

5:29 PM  
Blogger Ben said...

Bingo, infrared survey of the sky here http://www.esa.int/images/IRC_AllSky_Cnstellation_E_H.jpg

shows up the 'missing' stars when compared to stellarium.

To the people who think it's fax noise, yes, faxes are noisy, but the correlation of the larger of these noise dots to star positions is way too strong to ignore! I wish I had a scanner here to show you the map and how it compares... Plus, there's the link to 'noise' that is referred to in the first paragraph, which given the almost perfect correlation I'm reasonably confident that's what they're on about.

The question to me is, is there more information to be had out of this star chart, or is the clue about noise just so we don't ignore the background image...

6:08 PM  
Blogger Eric said...

Following the theory of a star map, if it is a mirror view of the northern sky, maybe we need to try decoding the 3 stanza’s mirrored so as to make them right side up when matched to the sky?

8:41 PM  
Blogger Eldad said...

I took these from John Graham's notation:

The first block:

(012) CIBYS@KDUGZBSGI@PUOXH@FBXX@JDRK@YURKG
(021) FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE
(102) JDNXUMEISOZNUODMFSGYQMPNYYMCIVEMXSVEO
(120) PVLBEMUQGK@LEKVMTGSAIMJLAAMWQDUMBGDUK
(201) THYLOZGRKUMYOUHZCKENVZWYNNZFRQGZLKQGU
(210) WQXAGZOVES@XGSQZJEKBRZTXBBZPVHOZAEHOS


The second block:


(012) GZWXUNGG@YOZAGI@ABKKG@KRLJGGY
(021) EMPLOYEE@NUMBER@BASSE@SIXTEEN
(102) OZTYSBOOMXGZLODMLNEEOMEVACOOX
(120) K@FAGXKKMBS@NKVMNLUUKMUDYWKKB
(201) UMJNKAUUZLEMXUHZXYGGUZGQBFUUL
(210) S@CBELSSZAK@YSQZYXOOSZOHNPSSA


Take a closer look at two consecutive lines and see that they tell you how to exchange letters. The letters exchange are stable.
Notice that lines 3 and 4 is very similar in format (notice the stops) to 5 and 6.

Armed with tr, I got variants as: (see the semi-garbled message repeated. you can "fix" this because it is still stable)


(012) JINXUMEDSOZNUOIMFSGYQMPNYYMCDREMXSREO
(021) PRLBEMUQGK@LEKRMWGSADMJLAAMTQIUMBGIUK
(102) CDBYS@KIUGZBSGD@PUOXH@FBXX@JIVK@YUVKG
(120) FVANK@SHOEMAKEV@TOULI@CALL@WHDS@NODSE
(201) TQXAGZORES@XGSQZJEKBVZWXBBZPRHOZAEHOS
(210) WHYLOZGVKUMYOUHZCKENRZTYNNZFVQGZLKQGU


and:


(012) OZWYSBOOMXGZLOIMLNEEOMERACOOX
(021) K@FAGXKKMBS@NKRMNLUUKMUIYTKKB
(102) GZTXUNGG@YOZAGD@ABKKG@KVLJGGY
(120) EMPLOYEE@NUMBEV@BASSE@SDXWEEN
(201) S@CBELSSZAK@YSQZYXOOSZOHNPSSA
(210) UMJNKAUUZLEMXUHZXYGGUZGQBFUUL


There's also the same for the last row.

This could be a leading hint to some sort of mapping you need to get the real message (this is just noise).

10:52 PM  
Blogger Eldad said...

Maybe S252 is really "5252".

This is ASCII for "RR". (who's he?)

In base 16 it is 1484. (was that an employee id?)

10:56 PM  
Blogger yomgui said...

some symbols have been identified as coming from the “standard galactic alphabet”:
“m” for sure and possibly “c” and “s”

from the lineophon script:
http://www.omniglot.com/writing/linephon.htm
I retain “ê” and “r” for sure and "u" as possible match

maybe also “h” from the golem script:
http://www.gregvilk.com/Golem/img/alphabet1.gif

11:15 PM  
Blogger Robert said...

My boss used to work over at fermi lab next to Frank Shoemaker. After I told him about this he said the most likely meaning is that there are some younger less experienced physicists who have preliminary data that they want to publish but an older and wiser one is trying to point out that it could also just be outliers and noise. If you want to decrypt the remainder of the data, it might be good to look into the most “cutting edge” experiments that are going on over there at the moment.

4:00 PM  
Blogger raphael said...

maybe whoever sent this was trying to mean that some random data, which would be called noise by shoemaker, may contain information from outer space begin sent through neutrinos.

8:36 PM  
Blogger Robert said...

It is possible that my point is not clear. Many researchers when they start out are very excited by the initial data. It takes rigor and effort to rule out all possible sources of noise and more experienced researchers like Shoemaker have probably learned this the hard way. The message sounds like it is from an upstart who is upset that the old guard (who would have the low id numbers, since I verified that they are in numerical order--An id of 7xx was given to my boss 25 years ago so one even lower would be associated with someone who had been at the lab for almost 30 years) had put a stop on them publishing what they believed to be great scientific revelation because they simply didn't have the statistical basis to say it wasn't noise. What better way to make an ironic point than with an encrypted message that "looks like noise" but has some clever scientific information embedded in it.

8:48 PM  
Blogger 6smail said...

“Employee number basse sixteen”
let’s decode “sixteen” using the midle stanza:
we can “draw” 16 using the symbols I, I > and ^, (try to put’em one to one)and if we take the hex corresponding to each symbol we have :
8863 or 8836 (6 = I>^ or 6 = I^>)
we have to separe 1 and 6, so we put 0 as separator:
80863 or 80836
but we have got “BASSE” which means in french “low” so we have to start with the lower symbol(which is >, note that we also have 3 lower than 6)
so our employee ID is : 80836N

Please check this ID, to see if I’m wrong or not(I think I’m not) or let me know how to check by myself.thx

Note: the employee ID at the fermilab is encoded on 5 integers and a caracter (C for Contractor, V for Visitor and N for Employee )

1:37 AM  
Blogger 6smail said...

Try to look in the phoenician and and other semitic alphapets(arab, hebrew...) even the greek onevto decrypt the midle stanza
I made some interessting discoveries (the first symbol is the letter 'D' in phoenician alphabet...)

11:07 AM  
Blogger 6smail said...

William Basse (c.1583-1653/4) was an English poet(notice that he lived in the 16th century) so perhaps the name of the author is William.

William Basse was also known as a follower of f Edmund Spenser(c. 1552 – 13 January 1599) ho was an important English poet that contribute mainly to the english poetry by his “Spenserian stanza” a stanza construct as this : 8 lines of 10 syllabus verse and the last line is an alxandrin (12 syllabus)

11:09 AM  
Blogger n0p said...

This post has been removed by the author.

3:40 PM  
Blogger n0p said...

Did anyone notice that both messages are defining a character interchange key cypher (or however is said in english, i'm spanish :P ).
So, we have this:
(012) CIBYS@KDUGZBSGI@PUOXH@FBXX@JDRK@YURKG
(021) FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE
(102) JDNXUMEISOZNUODMFSGYQMPNYYMCIVEMXSVEO
(120) PVLBEMUQGK@LEKVMTGSAIMJLAAMWQDUMBGDUK
(201) THYLOZGRKUMYOUHZCKENVZWYNNZFRQGZLKQGU
(210) WQXAGZOVES@XGSQZJEKBRZTXBBZPVHOZAEHOS

Let's order it:
(012) CIBYS@KDUGZBSGI@PUOXH@FBXX@JDRK@YURKG
(021) FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE

(102) JDNXUMEISOZNUODMFSGYQMPNYYMCIVEMXSVEO
(201) THYLOZGRKUMYOUHZCKENVZWYNNZFRQGZLKQGU

(120) PVLBEMUQGK@LEKVMTGSAIMJLAAMWQDUMBGDUK
(210) WQXAGZOVES@XGSQZJEKBRZTXBBZPVHOZAEHOS

You can use this character interchange key to go from one to another sentence in a pair.
A-B
B-A
C-F
D-H
E-G
F-C
G-E
H-D
I-R
J-T
K-S
L-X
M-Z
N-Y
Ñ- no way you don't have these :P
O-U
P-W
Q-V (from the "garbage" lines, not from the sentences)
R-I
S-K
T-J
U-O
V-Q
W-P
X-L
Y-Y
Z-N


My guess: this is a key reminder, or a key exchange letter. The two "nongarbage" sentences may be a first cypher validation.

Any other thoughts?

You can contact me by leaving a comment in any of my blog articles.

3:42 PM  
Blogger Ben said...

Congratulations n0p .... you found a pattern....

Pity your observation is about a pattern between bit-shifted permutations of the same message...

Which means unfortunately the pattern you noticed doesn't mean anything.

eg,
cipher text is:
khuhvdphvvdjh


I'm going to start rotating through combinations to find the answer:
gdqdrzldrrzfd
heresamessage
ifsftbnfttbhf
jgtgucoguucig

Notice a patter? Making the note that between each of these each char is char+1 isn't really interesting, is it?

7:22 PM  
Blogger n0p said...

shame on me, programming all day long has side effects xD

8:59 PM  
Blogger Al Feersum said...

Just a thought - wouldn't 'FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE' actually /be/ noise? And everyone's barking up the wrong tree?

10:24 AM  
Blogger Nicholas said...

This might be silly but there are small dots, hardly noticeable all over the original picture that have a uniform pattern, could this provide some kind of clue or just wasting your time, at first I thought they were fragments on a dusty scan but they are in some kind of pattern, maybe the person was just doodling, thinking as they taped the page with the pen or it could be significant

4:44 AM  
Blogger Nicholas said...

This may be silly but there are lots of small dots, hardly noticeable on the scan of the original and they for some ki9nd of pattern, maybe the person was just taping with there pen or it might be significant

4:48 AM  
Blogger Ben said...

Nicholas, not silly at all.

It's pretty clear it's a mirror-image picture of the sky, probably in the infrared spectrum (certainly it's not optical as there's a few extra features).

I'm guessing the message was written on a transparent film from an older telescope, but as yet I've been unable to get the image of the sky dated.

I suspect there's an interesting astronomical event in the image, but without a date am at a dead end.

7:50 AM  
Blogger Dr. Wellwonder said...

SFC = Space filling curve: http://en.wikipedia.org/wiki/Space-filling_curve

8:38 PM  
Blogger Dr. Wellwonder said...

An interestingly related book: http://www.amazon.de/Hackers-Delight/dp/0201914654

8:44 PM  
Blogger Jon said...

Basse could also refer to this guy:

http://www.oldschoolhack.de/tutorials/BiW%20reversing%20-%20CryptoTut%20by%20Basse.htm

12:50 PM  
Blogger Hedgecore said...

Assuming 'basse' is French for 'low', what about a literal meaning - - 'low sixteen'. Perhaps it refers to an employee number "in the low sixteens" like 1601, 1602, etc.

The Basse building is interesting as well being that it is 16 storeys tall; perhaps there is some relevance to the lower stories (1 through 7 inclusive)?

I find it hard to believe that this would be a typo given the precision used to create it.

11:20 PM  
Blogger callum said...

0xFC Is wrong.
0xCF is right. S stands for swap, swap F with C, employee ID = 207 maybe?

9:51 PM  
Blogger callum said...

SFC Means swap F with C meaning the hex 0xFC = 0xCF = 207

this also goes with chris's comment.

IF Basse means low in french, maybe it denotes that the Hex is LSB, MSB.. i.e. CF(16) = 207 decimal...

10:01 PM  
Blogger Daniel said...

One thing I have not seen anybody do... is take that middle section, and group it into threes like everything else is.

If you take that, and turn it into trinary code like the rest, you get a really strange pattern. (replacing letters with zero and numbers with 1, or vice versa)

101 100 101 100 010 010 100 010 011 or
010 011 010 011 101 101 011 101 100

Notice a lot of the groupings are repeated... converting these to letters using the already determined code gives me gibberish though:

JIJICCICD or
CDCDJJDJI

Seems far too coincidental to be ignored though...

11:46 AM  

Post a Comment

Links to this post:

Create a Link

<< Home