Fermilab Calls For Code Crackers

atrocious cowpat passes along a call for help from symmetry magazine, the joint publication of Fermilab and SLAC, noting: "Could be just plain gibberish, it could be something like those wonderfully weird letters to the Mount Wilson observatory, or it could be a message from aliens who just happened to have gotten their hands (tentacles/exoskeleton) on a fax machine." "A little over a year ago, the Fermilab Office of Public Affairs received a curious letter in code (4.4-MB image here). It has been sitting in our files all that time and we haven't had much of a chance to look into breaking the code, nor are we particularly expert at this!"

Comments

Clearly..

[Anonymous Coward]

It's Dr. Emma Russells formula for cold fusion.

Re:Clearly..

[sammyF70]

Though I seriously can't believe she missed her error in the 2nd part. Too bad, she was really doing well until that.

Re:Clearly..

[Anonymous Coward]

The message reads "I should have used ROT13"

Re:Clearly..

[CrazedWalrus]

I think it's Digital Fortress.

Fortunately, the Slashdot Hive Mind has a safety shutdown when the story rolls off the front page.

Well obviously...

[clonan]

The Roswel aliens that were stranded on earth brainwashed some NEC employees who planted the code in all fax machines to send the fax to fermilabs who then posts it on the internet (which would be invented in 40 years) and thereby transmitted to mars...

Makes perfect sense!

I'm not positive about my translation

[Eevee]

But it seems he's from Nigeria and wants help tranferring money out of the country.

Re:I'm not positive about my translation

[Anonymous Coward]

And he wants you to "Drink More Ovaltine"

Re:I'm not positive about my translation

[Hinhule]

Really?
I got an "ASCII" version of goatse :-/

Re:I'm not positive about my translation

[spazdor]

me too! And apparently we suxor?

Re:I'm not positive about my translation

[sveard]

But why do they call it Ovaltine? The mug is round. The jar is round. They should call it round tine.

Re:I'm not positive about my translation

[fishbowl]

>But why do they call it Ovaltine?

Because it is made from eggs.

Re:I'm not positive about my translation

[somersault]

Shouldn't they call it Ovarietine then?

Well, obvious stuff:

[morgan_greywolf]

Three "stanzas" maybe.

Starting with the "middle" stanza, that appears to be some sort of "key" perhaps. Each of the different symbols correspond to a different hexadecimal digit.

In the first stanza, each grouping of lines has 1, 2 or 3 lines.

In the last stanza, each group of lines is only 1 or 2 lines.

Maybe the last stanza is binary?

And maybe the first stanza is base 3?

Anyone else care to wager a guess?

Re:Well, obvious stuff:

[spydum]

Agreed, first grouping is probably base-3, 2nd is key with the index being hex, and 3rd grouping is base-2.. and if I convert it out.. it ends up drawing an image of the goatse.cx guy... damnit!

Re:Well, obvious stuff:

[morgan_greywolf]

Mirror here. I think Google's servers can probably handle the traffic. ;)

Re:Well, obvious stuff:

[Anonymous Coward]

The middle stanza is unlikely to be a key as it has many signs repeated. F, D, 6, 3, 9 etc.

Re:Well, obvious stuff:

[azzuth]

That's funny, when i converted it and then etch it onto a vinyl i got Rick Roll'd

Re:Well, obvious stuff:

[mhall119]

I don't think it's necessarily a key. If it were just a substitution key, they wouldn't need to repeat characters. Maybe it's a message in both glyphs and hex, like a Rosetta stone?

Re:Well, obvious stuff:

[moderatorrater]

Another possibility: a blank represents a 0, a mark indicates a 1, giving us binary. This fits better with the hex in the middle of the page, since three binary go into one hex. The symbols don't make any sense, however, so if that's a key, then it's going to just translate it into another code.

If I weren't at work, I would try translating the hash marks into their hex equivalent, storing that, then translating the hex to the symbols that were given and storing that separately. The symbols may mean something to someone other than me (some of them look hebrew, but some of them definitely aren't; perhaps astrology symbols or something else?), or it may be that the key in the middle is a red herring and that the hex itself codes for something (ascii being the most likely generally, although a quick glance seems to indicate that some of them would code for non-display characters).

Most likely a hoax all things considered. A (accidentally?) clever hoax, considering the hex in the middle and the many interpretations of the vertical lines, but most likely a hoax nonetheless.

Re:Well, obvious stuff:

[Dachannien]

I'm thinking the last stanza is also base 3, with digits then represented in unary, but with || indicating a division between the base 3 digits.

So || | | || is "2" with a separation marker on each end.

This would put an anomaly at the end of the second line which could be interpreted as the || being broken across the two lines (so the second line would end with 3 and the third line would start with 3).

Re:Well, obvious stuff:

[Sechr Nibw]

The only problem with that is twice in the last chunk the sequence "1 1 1 1" appears, not counting possible removal of line breaks

Re:Well, obvious stuff:

[azzuth]

Anyone else notice there are 8 digits if we break it into binary verticaly? such as from top to bottom left to right the first being 10100111 and alternatively bottom to top (as if the page were rotated 90 degrees clockwise) 11100101...

The Binary can then be converted to Hex and then perhaps deciphered...

Working on it atm..

Re:Well, obvious stuff:

[Fantome]

My transcription of the first block:
32323333 1112132
33323132 212331
21113311 32312233
33321212 3213113
31133331 3331111
21133332 3232211
23231333 1121231
33231312

My transcription of the second block:
11121211 21212121 21121212 12111212 1121
11211211 21211121 21121112 12112111 21111
11112121 21121121 21112121 21211121 11211
21112121 12112111 21112111 21112111 21112
11121121 11211121 21121112 12221112 1211
12121121 11211121 11211211 12121211 12111
21121121 11211211 12112111 21211211 1212
11212121 1

Any see errors in this (parts were hard to read).

Re:Well, obvious stuff:

[baffled]

You're thinking waaay too deep.

Turn the page 90 degrees clockwise.

It says 'Hi'

Re:Well, obvious stuff:

[Unending]

ugh this thing is so poorly done it's really hard to read also if the "aliens" know hex then they could just write a message this is obviously a hoax or a recruitment thing

Re:Well, obvious stuff:

[DRAGONWEEZEL]

Can someoen please put this to music? it's killing me, I like to think that the |||'s are longer notes, and the |'s are shorter. Maybe backwards though too.

I just have a weird feeling about it.

Then I just got another Idea that's too good!

It's the intro to statistical analysis. Imagine (P.S. I have no effing Idea what the center part is) that this is a guy counting something. Maybe something like how many buffalo show up to the lake at a given day day one, he sees 3, 2, 3, 2, 3, 3, 3, 3, 1 1, 1, 2, 1, 3, 2,

Mom, Dad, baby.

Dad dies, and now he only sees 2 at the lake.

Re:Well, obvious stuff:

[funaho]

The world's first rickroll-by-fax?

Re:Well, obvious stuff:

[Bob9113]

Two quick observations:

The frequency of 3's in the first stanza is disproportionately high.

The pattern '1,1,1,2,1' appears several times in the third stanza.

Re:Well, obvious stuff:

[Sechr Nibw]

Err, disregard my first comment, you already covered that. The issue with taken lines as broken though, is that if you do that with one, how do you know others aren't supposed to be taken that way too?

If you take all of them as broken, and include the spaces that *might* be there on lines 2,4,7,8, then you end up with two sets of |||. Periods maybe? Some other delimiter?

Trying what you suggested with chunks of ||| being line breaks, you end up with:
3 1 2 1 1 1 1 2 1 1 1 3 1 2 3 2 2 1
1 2 3 1 2 3 3 3 1 1 2 2 1 3 1 1 1 3 3 2 3 1 2 2 3 3 3 3 3 3
2 2 3 3 1 2 3 1 3 1 1 1 2 3 3 3 2 3 1 1 3 3 2 2 3 2 3 2 3 1 2 3 1 2 1 1 2

Don't know where this is headed, but could be a breakthrough for someone else! (clearly a slow day at work)

Re:Well, obvious stuff:

[AoT]

All of the symbols I recognize are mathematical or logical notation. i is imaginary numbers, the character under it could be an empty set or a zero. B is a minus sign. 6 is the negation sign. in logical notation. 7 is the greek letter phi, used for a number of things, most importantly the golden ration. D looks familiar but I can't place it right now, same with 4.

Re:Well, obvious stuff:

[goodmanj]

Trinary/hex/binary Rosetta Stone seems likely on the face of it....

However, if Section 1 is trinary, its information content is equal to 180 bits (113 symbols times log2(3) bits per symbol). The second section, in hex, has 96 bits, and the third, if in binary, has 266 bits. Unless one symbol set has a huge amount of redundancy, they're not the same length.

----------

A casual glance at the "binary" third section suggests it's unlikely to be any sort of ASCII-like binary substitution cipher, and possibly not a binary-encoded language of any kind. A "1" can be followed by either another "1" or a "2" (equal probability), but you never see "22" except for one string of "222" on line 5.

This pattern is way too regular to represent a binary encoding of any large symbol-set like an alphabet.

-----------------

Section 2 is absolutely a hexadecimal code: the symbols are a simple substitution for the hex characters below them. The symbols look a little bit like the "pigpen" cipher, but only vaguely.

I can't find any correspondence between the hexadecimal numbers in section 3 and the "trinary" in section 1.

Current best guess: this is a joke, with no actual information content.

Re:Well, obvious stuff:

[BrainInAJar]

I got a piece of spam that read exactly like that!

Re:Well, obvious stuff:

[NeutronCowboy]

I think that's an interesting approach. Statistically speaking, if | and || would be two different symbols, the chance that one of them would never appear consecutively would be zero. As such, I agree that the || is probably a demarcation. Furthermore, the last character at the end of each line of the last stanza is probably also a demarcation, drawn before actual code series was written out. The reason I say that is because the last characters are the only ones that are actually beneath each other - the slanting comes from a right-handed writer slanting the page as they write. This in turn means that the places were 4 | are next to each other would actually be 3 | plus one | that signals end of line.

The dots

[personalo]

My question is: Are the tiny dots in the background a dirty fax or photocopier artifact or are they, in fact, the code.

Google recruiting?

[spydum]

Perhaps Google is targeting Fermilab scientists for hiring.. Don't they have a history of using strange riddles and puzzles for hiring purposes?

Re:Google recruiting?

[BigJClark]

EA posted billboards all over Vancouver, BC awhile, that had char *msg[] = { "10", "43", "14" } (not real values) and essentially it was the ascii decimal equivilant of "Now Hiring".

I thought that was pretty trick.

Re:Google recruiting?

[snaFu07]

Picture here: http://images.fastcompany.com/blog/121503billboard.jpg

Re:Google recruiting?

[2names]

If you MD5 the whole thing and divide by the binary representation of the date they found it, it comes out as:

J O H N T I T O R

Hmmm...

[Omicron32]

4.4MB image link on the front page of Slashdot? I sense a great disturbance in the force...

Re:Hmmm...

[mapkinase]

Well, right now I do not sense no nothing except "Loading..."

Re:Hmmm...

[Sechr Nibw]

For the price of 100 gold, I will identify this cryptic item.

Interesting...

[wile_e_wonka]

Very odd indeed! I received this cryptic message (which it seems I've seen before):

Error!
Could not connect to remote server

You tried to access the address http://www.symmetrymagazine.org/breaking/wp-content/uploads/2008/05/fnalcodeletter.jpg, which is currently unavailable. Please make sure that the Web address (URL) is correctly spelled and punctuated, then try reloading the page.

Re:WTF slashdot?

[Anonymous Coward]

I clicked on this article and a popup advertisement came up. What the hell?? Means you started Internet Explorer instead of your usual browser. WTF were you thinking?

Mathamatically speaking....

[hengdi]

There are 113 symbols, each of which is either 1, 2 or 3 strokes. So it is essentially a 113 digit base 3 number.

This limits the amount of information that the message is trying to pass.

For example, using base 26 - all the letters - means we could convey the same information in ??? digits.

Oh damn it. I'm too drunk and Google ain't working for me. Perhaps someone could give a value for ???

But I'm betting it won't be very many digits. I.e. this message is very short.

Re:Mathamatically speaking....

[whitneyw]

Considering that the ticks are vertically aligned, I would consider the interstitial spaces important parts of the message. Also note that the top block contains no more than three ticks consecutively, while the bottom block contains no more than two. Neither contain consecutive spaces, and both appear to start with a tick (not a space).

Part of it bears a striking resemblance to UPC code for "8200019288".

Re:Mathamatically speaking....

[mr_mischief]

Jose Cuervo Especial, 750 ml, 12-pack case.

Mystery solved!

Damn, Google has us spoiled.

Re:Mathamatically speaking....

[Impy the Impiuos Imp]

The interstices would not represent zeroes, unless the message just happened to have no more than 1 zero in a row. Hence | represents 0, || represents 1, and ||| represents 2, or some rotation of that.

By the way, this is apparently some kind of "unary" notation, though what zero it is remains to be seen. Similarly, the bottom set, having no more than two | in a row, might be another unary notation of something in a base 2.

As for the little dots, some, like the symbol for E, are repeated, so it's safe to assume they're part of the symbol. Others, like 6, are repeated without the dots, so it's probably a stray mark. Still others, like 5, 0, and 8 are not repeated, so we can't tell for sure whether the odd marks are stray or part of the symbol.

It's also possible they are an additional notation applied to the base symbol, giving it a slightly different meaning, or adding more info.

Re:Mathamatically speaking....

[SpeedBump0619]

I'm not certain your statements are entirely correct. I attempted to create a binary pattern to start with and there are some ambiguities in the coding positions. For instance, on the left hand side of both code blocks some of the "first" ticks don't align with other first ticks. There are also some places in the code where the vertical alignment is ambiguous. If there cannot be consecutive spaces and there *can* be spaces in the first and last columns then I'd read the first block as:

Line 1) 11101101110110111011101110111010101011010111011 (47 digits)
Line 2) 01110111011101101110101110110110101101110111010
Line 3) 11010101011101110101011101101110101101101110111
Line 4) 0111011101110110101101011011101101011101010111
Line 5) 0111010101110111011101110101110111011101010101
Line 6) 1101010111011101110111011011101101110110110101
Line 7) 1101110110111010111011101110101011010110111010
Line 8) 1110111011011101011101011 (25/26 digits)

Consecutive spaces allow the coding to always hit 47 characters, but dramatically increases the complexity of the problem. I'm wondering if the whole code is continuous, meaning that the last tick on line 5 and the first two ticks on line 6 are actually one code three ticks wide (also 6-7). The spaces in the first and last columns of the first two lines seem to imply this...

It's a trap!

[Mortiss]

Crackers do not fall for this trap. It is an obvious attempt to spread the Snow Crash!

Strange...

[BlueStrat]

It seems to be some sort of construction/zoning notice. Something about a hyperspace-bypass being put in...

Cheers!

Strat

Okay, let's do this!

[The Insane One]

Just let me get a pencil and some paper.

"All your base are belo"

Oh crap.

I've seen this before

[Crash McBang]

The translation comes out as:

My name is Kosh Naranek.

I am writing this brief letter to bring to your attention a business offer which we believe you might find attractive.
Mrs Maria Garibaldi; wife of one a wealthy Martian executive (Late Mr. Michael Garibaldi) seeks a business assistance from a reliable and reputable businessman to invest and manage funds to the tune of 15 Million Credits...

Source?

[aleph42]

Any hint on the source, or at least why they consider it important?

Not to be harsh, but if I send some random code letter to some lab, I guess (hope) it won't make the news, even on slashdot.

Re:Source?

[esocid]

My plan finally worked. One long year of waiting has paid off. It's just nonsense I made up to get some people to try and crack it.
Cheers.

Re:It says "Welcome to the Slashdot effect."

[Anonymous Coward]

Try this link:

http://filebin.ca/skpzc/fnalcodeletter.jpg

Mirrored Copy

[LaptopZZ]

I mirrored it on my server as well as made a scaled down version which is just as readable as the original (unless you're making a poster).

http://www.pixabug.com/aliens/fnalcodeletter.jpg

http://www.pixabug.com/aliens/fnalcodeletter_sanesize.jpg

Happy Cracking

I've managed to decipher the final section...

[DirtySouthAfrican]

It reads "Use only Genuine Interocitor Parts". Seems gibberish to me.

Myst - Riven

[aarenz]

This is the code from the final puzzel in Myst episode called Riven. Not really, but it looks quite similar in function. Someone who was able to master that game without a guide book would be able to figure this one out in a flash.

So

[Shivetya]

this is how to beat Guitar Hero IV on hard?

other than being a freak or having a pet spider to help you out?

I've seen this before....

[Anonymous Coward]

It's obviously Woodstock. He's telling Snoopy about encryption.

I think it says...

[Uncle Focker]

'PC Load Letter'? What the fuck does that mean?

the strokes are music (?)

[Ralph Spoilsport]

To me, the strokes look like a clapping pattern. I sat and clapped it out at my desk here, and if done at a fairly brisk pace, the top section is an interesting and often asymmetric rhythm.

Parts of it clap out to sound like "apocalypse in 9/8, (co-starring the delicious talents of Gabble Ratchet)" by Genesis from Foxtrot

But the whole thing is scattered enough that it comes out like more of a one handed improv or approximation of Steve Reich's "Clapping Music".

The bottom section is less rhythmically active, but sounds more "even", kind of "rock and roll" ish.

The middle part is a dull cipher, similar to Nugsoth.

That's all I've got.

RS

B E S...

[mcmonkey]

U R E T O D R I N K Y O U R O V A L T I N E

--
"Lameness filter encountered. Post aborted!"

Everyone's a critic.

Context?

[Z00L00K]

To be able to decrypt a message it is usually necessary to know the context to where it belongs. Without context it's hard.

And is it really a message, it can be other things too:

1. A data sampling done by hand with two samplings, the first has three alternatives the second has two. The code in the middle is just thrown in for good measure by a joker.
2. It's someone's way to compose music. The first part is for a simple instrument with three tones, the last part is another instrument with two alternatives (bongo drums?). The middle part is just markers used to remind the performers about what to sing.
3. Someone's idea of a prank to pull.

Re:Context?

[Garridan]

All this reminds me of my algebra prof's (apparently) favorite poem: "One-one was a racehorse, and One-two was one too. One-one won one race, and One-two won one too."

Score two points for the sender

[memnoch37]

Obligatory http://xkcd.com/356/

Re:The raw numbers

[Iamthecheese]

If I change the 1s to 0 and the 2s to 1, I get some interesting bits.
"J)IEQ)"""$curren;\QHquot;$)T

Re:Not Cheere..

[arthurpaliden]

But rather, "So long and thanks for all the fish".

Coral cache

[irae]

Here's coral cache version, works for me fine. Link provided by slashdotter, firefox plugin.

I'm pretty sure it's for my broken office light

[SomeoneGotMyNick]

It's a tempo guide for the person who's messing with the fluorescent ceiling light in the corner of my office. I swear the light is flickering to the same cadence as the tick marks on the page.

DNA sequence?

[fahrbot-bot]

1. Four DNA characters: A, C, G, T
2. Four character sequences: |, ||, |||, (space)
3. Determine character mapping.
4. ...
5. Velociraptor.

Re:The raw numbers

[Anonymous Coward]

Well this is obviously the handiwork of a drunk perl coder trying to be minimalistic.

I'm pretty sure that's a stargate address...

[DangerTenor]

Seriously though, what the frack are these symbols? I think most of them fit into extended ascii, but not D and 4 (at a minimum, i'm basing this on knowledge from 20 years ago...)

Digitized

[panthro]

323233331112132
33323132212331
2111331132312233
333212123213113
311333313331111
211333323232211
232313331121231
33231312

f0be58f2fd63
6c79d2e493e6

S f c

111212112121212121121212121112121121
1121121121211121211211121211211121111
1111212121121121211121212121112111211
2111212112112111211121112111211121112
111211211121112121121112122211121211
1212112111211121112112111212121112111
211211211121121112112111212112111212
112121211

Re:Digitized

[el_gato_borracho]

I independently transcribed the same as you, so I confirm the same digits. Also, assuming the first blob is ternary (I=0, II=1, III=2), the second is hex, and the last is binary, I get the following numbers in decimal:
ternary = 720113244210716512990341782103795379056660114385796527
hexadecimal = 74506518313470710988407084006
binary = 4785997412726154595979555835418260996622867313584208882680343839351760783444564
Of course, this assumes also that these are supposed to be read left-to-right, top-to-bottom, and that the most significant digits come first. Can anyone else confirm these values?

Analysis!

[DRAGONWEEZEL]

It's Caveman analysis.

Someone is counting the number of something. Day one, it was 3. Day 2, 2. on day 4, 2 show up. Probably animals at a watering hole. Then one of them dies... or maybe the hunter kills it.

Then he starts analysis again, to see how long it takes for a 3rd to show again.

Re:4th stanza

[morgan_greywolf]

What a coincidence! That happens to be the combination to my luggage!

Sorry, had to do it. :-D

Re:The raw numbers

[Goaway]

No, Slashdot is just broken. It doesn't get you exactly that. There's no "curren" or "quot" in it.

Tad more to add

[azzuth]

the assumption in other posts is that the spaces are singular. It seem to me that it can't be true, if you count the | marks and the spaces as singular you get:
47
46
46
46
46
46
46
25 not counting final space.
Tried to give | || representation of the bits, but /. demands fewer junk characters...

This leads me to believe there is a double space in there, and with the crooked lining up of the |'s it is hard to tell exactly where... but i believe it is in the bottom most 5-6 lines.

Binary Hex Dec
1 11100101 E5 229
2 11111111 FF 255
3 10011011 9B 155
4 01111110 7E 126
5 11000001 C1 193
6 11111111 FF 255
7 10001010 8A 138
8 01111111 7F 127
9 11100001 E1 225
10 10111111 BF 191
11 01011110 5E 94
12 11111111 FF 255
13 11100001 E1 225
14 10111110 BE 190
15 01011111 5F 95
16 10110101 B5 181
17 01101011 6B 107
18 11110110 F6 246
19 11011011 DB 219
20 10111101 BD 189
21 01100011 63 99
22 11111100 FC 252
23 01010111 57 87
24 10111111 BF 191
25 11101011 EB 235
26 01110100 74 116
27 01001111 4F 79
28 00111011 3B 59
29 01111101 7D 125
30 00010110 16 22
31 01101111 6F 111
32 00111000 38 56
33 01110111 77 119
34 01011000 56 88
35 00100111 27 39
36 01111110 7E 126

This is where i lose my confidence on lining up the bits.. But in total there should be 48 rows... Oh, and I don't believe i mentioned this is the first stanza from the top of the page. Hope something from this helps,
Cheers

Real problem solvers comment here

[Jhan]

(trying to move the interesting stuff to the top) The top and bottom part of the code code the same data. The little indentation at the beginning of the line is important and means that the previous line continues. The indentation in the bottom bottom part is of, perhaps because of writing conditions. The top part consists of five trinary numbers of lengths 29, 46, 14, 14, 8 digits. The bottom part consists of five binary numbers of lengths 75, 110, 37, 36, 8 digits. My best transcription, probably with errors:

char trinary[8][40]={
"323233331112132", // 15
"33323132212331", // 14 29
"2111331132312233", // 16
"333212123213113", // 15
"311333313331111", // 15 46
"211333323232211", // 14 14
"232313331121231", // 14 14
"33231312"}; // 8 8
char binary[8][40]={
"111010110101010101101010101110101101", // 36
"1101101101011101011011101011011101111", // 37 75
"1111010101101101011101010101110111011", // 37
"0111010110110111011101110111011101110", // 37
"111011011101110101101110100011101011", // 36 110
// Should have been more clearly to the left?
"1010110111011101110110111010101110111", // 37 37
// Should have been two steps to the left?
"011011011101101110110111010110111010", // 36
"110101011"};

Re:Real problem solvers comment here

[bodan]

Not a critique as such and only vaguely on topic: Does anyone else find it interesting that parent found it natural to represent ternary using 1-2-3 and binary using 0-1?

That was actually my first instinct too when I was "reading" the thing...

Re:Real problem solvers comment here

[ThreeGigs]

The first and last parts *DO NOT* look like terniary and binary to me.

Instead, they look like an RLL encoded pattern, similar to what you'd see on a hard drive. Maybe NRZI.

In fact, the first looks surprisingly like (1,3) MFM RLL.
Also reminds me of the old Apple Floppy drive "between any two ones there can only be a maximum of one zero" data writing rule.

Re:Real problem solvers comment here

[wirelessbuzzers]

We've mostly solved it further down the thread. The top section is in ternary (with 0 replaced by |||), 000-space, 001->a, 002->b, etc. It's strangely wrapped: the spaces and non-spaces at the beginnings and ends of lines count. It says "FRANK SHOEMAKER WOULD CALL THIS NOISE".

The second part we haven't deciphered yet. It's possible that we'd need a Fermilab insider for this.

The third part is RLL. Once you decode the RLL (number of | between consecutive ||), you get the same code as part one, which says "EMPLOYEE NUMBER BASSE SIXTEEN", or so I've been told. This suggests that at least one part of the puzzle requires help from Fermilab people.

My uninformed guess is that once we solve the middle section, we'll get someone's name. His or her employee number at Fermilab spells something out in base 16, a coincidence which Frank Shoemaker would call "noise". Another possibility is that the middle section is hex-encoded employee ID numbers, which would mean that we can't solve it.

Re:Real problem solvers comment here

[lemnar]

The string of 3 symbols between the second and third parts may be simply SFC => Shoemaker, F. C. => Shoemaker, Frank C. (F. C. Shoemaker via)

It's a receipt

[Bogtha]

See those three symbols on their own? An S, followed by a triangle, followed by a three-pronged character? Well if you look in the table directly above those three characters, you'll see that the triangle translates to F and the three-pronged character translates to C, giving S.F.C. altogether.

Clearly the Roswell Greys were on their way home from picking up a Spacetucky-Fried Chicken takeaway when they crashed here.

Reminds me of the movie Zodiac

[thomasdn]

This reminds me of the movie Zodiac. It was based on the true story of a serial killer in US that sent code letters (similar to this one) to the press. Actually I just noticed that tomorrow Zodiac the movie was released precisely one year ago. The sender of this letter could be inspired by the movie.

All your base...

[marcus]

...are belong to us.

Even I can see that.

Obviously, these Fermilab folks don't get to go outside at all.

writing tendencies and charset

[Anonymous Coward]

while it may not be terribly relevant to the solution, it's worth noting a few oddities about the author's writing style. first of all, the second set of vertical lines is likely to be read right to left. notice how well justified the right side of the message is, while the left side terminates at various locations. we also know he writes the individual clusters from left to right based on the strokes left at the bottom of the bars.

secondly, i assume the author is writing with a felt tip pen and is transcribing this from another source. notice how some bars will have a larger dot at one end or the other. i postulate that the author has set down his pen and is checking another sheet to determine what to write, while his felt pen bleeds a little too much ink onto the paper. this, however, doesn't happen with a ball point pen. perhaps this will be of assistance in determining the order strokes were written in, giving us some insight into the author's intent.

thirdly, notice that the author writes the number 0 with a slash through it, but not the number 7 with a dash through it. i'm not an expert on handwriting identification, but certain groups will often be more keen on the selection of particular stylistic numbers, so this may help us to understand something of the author's background.

finally, note that many of the chars used in the second section are letters from other char sets. the letter i is obvious to the english readers, but less apparent is the instance of one that looks suspiciously like a greek capital phi. also, several bear a striking resemblance to chars that appear in a masonic cypher alphabet. http://www.odr.org/anonymous/fam-code.htm can the other chars be identified as belonging to specific alphabets?

as if...

[whopub]

4.4MB image link on the front page of Slashdot? I sense a great disturbance in the force... as if the whole world was about to be slashdot...

FRANK SHOEMAKER WOULD CALL THIS NOISE

[wirelessbuzzers]

The first part is ternary, with 3 substituted for 0. It's somewhat miswrapped, but it appears to say "FRANK SHOEMAKER WOULD CALL THIS NOISE".

Re:FRANK SHOEMAKER WOULD CALL THIS NOISE

[Anonymous Coward]

It's true:
remove the linebreaks from the first block, interpret # of lines as trinary digits 0,1,2 (|->0, ||->1, |||->2) and you get

212 122 220 001 021
222 120 211 012 201
000 220 021 201 122
222 101 012 102 002
200 222 202 220 002
002 222 121 211 022
120 222 001 012 022
120 201

Converting to alphabet by 000->a, 001->b, ... , 221->z, 222->' ', we get:

xrybh pwftayhtr kflcs uycc qwip bfipt

Write a script to check possible letter substituions against a dictionary, and you find that the substitution cipher
abcdefghijklmnopqrstuvwxyz
nl  o ki wu   strdec hfa

converts this to "frank shoemaker would call this noise"; maybe a coincidence, but looking likely.

Re:FRANK SHOEMAKER WOULD CALL THIS NOISE

[Danny Rathjens]

mod these folks up:

perl -we'$_="323233331112132333231322123312111331132"
. "312233333212123213113311333313331111211333323232"
. "21123231333112123133231312"; @m{glob "{1,2,3}"x3}=("a".."z"," ");print $m{$_} for m/.../g'
xrybh pwftayhtr kflcs uyaj xqjquydpzu

I'm not getting anywhere with the 3rd stanza, though :) e.g. quadruplets for hex codes?

perl -we'$_="11101011010101010110101010111010110111011011010111"
. "01011011101011011101111" . "11110101011011010111010101011101110110111010110" . "110111011101110111011101110" . "111011011101110101101110100011101011" . "101011011101110111011011101010111011101101" . "1011101101110110111010110111010110101011"; @m{glob "{0,1}"x4}=(0..9,"a".."f");print $m{$_} for m/..../g'
eb556abaddb5d6eb77fab6baaeedd6dddd
ddddbbadd1d75bbbb75776ddbb75bad5

or a two level using triplets and then the first method gave me:
wsanqptssqpuzoewoqubugugu fqpggfqxgcgptxgfoa

Re:FRANK SHOEMAKER WOULD CALL THIS NOISE

[femtobyte]

Last stanza:
consider the || separators between trinary digits |, | |, | | |; decode as for first stanza, gives:
tadcfmtt blaztr zyppt pioqttb ->
"employee number basse sixteen"

thus, the central numbers are probably the employee # of the prank letter writer. Someone at Fermilab could probably check this (maybe Frank Shoemaker?)

I think we've solved it!

Re:FRANK SHOEMAKER WOULD CALL THIS NOISE

[Starlet Monroe]

Final cipher I had was:

abcdefghijklmnopqrstuvwxyz
mnlp o k wuy xstrdec hfab

in case anyone wants to repeat the steps like I did.

Re:FRANK SHOEMAKER WOULD CALL THIS NOISE

[Anonymous Coward]

Yes, I've emailed them a short description of the decryption. I also tried to post the results to the "discussion" section on the Fermilab page, but it looks like they stopped displaying additional posts early afternoon (probably once they got wind of the Slashdot onrush...). Whoever is reading Symmetry Magazine's email is probably inundated by quite a rush of wild speculation and irrelevant questions, so it may take them a while to find the actual decryption among all the noise. I hope in a couple of days they will be able to post a followup story explaining the backstory behind this --- it looks like an inside joke of some sort that is probably now obvious to the people involved. There is one "Frank Shoemaker" associated with Fermilab that Google can find, a professor from Princeton working on the BooNE neutrino detector. Given that neutrino detection is all about struggling to detect very small, rare signals above background noise , this is a likely candidate for the "noise" comment.

Oh NO!

[Satanboy]

It's a Cookbook!

If you break this up into tuples of 4...

[tlambert]

If you break this up into tuples of 4 and Google it, you get some interesting matches from geomagnetic observatory data.

-- Terry

Re:Solved!

[fbjon]

Mod parent up as interesting solution. Relevant thread here.