Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Breaking the Fermilab Code

Posted by kdawson on Tue May 20, 2008 03:22 AM
from the paging-frank-shoemaker-white-courtesy-telephone-please dept.
Saiyine sends word that the mysterious code received at Fermilab, which we discussed last Friday, has been mostly decoded, inside of two days, by two separate people. The poster at the second link seems to have constructed a more complete rationale for the message.
+ -
story

Related Stories

[+] Fermilab Calls For Code Crackers 392 comments
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!"
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • by everyplace (527571) on Tuesday May 20 2008, @03:36AM (#23472172) Homepage
    One always conceptually understands the power of numbers, but in this case it is amazing, considering that this problem went unsolved for an extended period within fermilab. The second it is asked to the correct audience though, the gears start going and the answer exists!
    • by CarpetShark (865376) on Tuesday May 20 2008, @04:34AM (#23472494)
      Yes, and if anyone needed proof that open source is better than closed source for finding bugs or fixing security vulnerabilities, this is yet more evidence.
      • No, not really. This only shows that a lot of people will try to solve interesting problems, and some of them eventually will. It does not say anything about open source software and finding bugs or security vulnerabilities, which involves (among other things) reading tons of "boring" code.

        Note: I did not say that open source is bad for finding bugs and vulnerabilities, I just want to mention that breaking this code does not say anything about open source software.

        • by CarpetShark (865376) on Tuesday May 20 2008, @04:58AM (#23472602)
          Well, the fact that you don't see the connection does not preclude the link. Many people would say that coding is boring, and yet others find it interesting just to browse code. The fact that people on slashdot (mostly coders and other IT people) are interested in these codes suggests an overlap. In fact, I doubt many people would argue with that (although I'm sure it'll be the few who would argue that will reply ;). I stand by what I said.
    • by QuantumTheologian (1155137) on Tuesday May 20 2008, @05:45AM (#23472806)
      From the original release by Fermilab, it seems to me like they had this sitting in a drawer somewhere. Sure, technically it went 'unsolved', but no one was really looking for a solution.
  • by adam (1231) * on Tuesday May 20 2008, @03:39AM (#23472184)
    Just want to point out that 1st and 3rd stanzas were cracked WITHIN the slashdot thread. See wirelessbuzzers' post here [slashdot.org] and femtobyte's post here [slashdot.org]. Either of these two individuals may be the two people whose sites are linked in the summary for this current story, but since I can't be sure, I wanted to make sure credit was given to them as well. (The first stanza was cracked within 7hrs of the /. story going live)

    Also, based on the "employee number" speculation in the second link especially, I want to point out that although I am the furthest thing from a "codecracker," I do believe the BASSE misspelling of BASE is intentional and is a clue. Likewise, the FRANK SHOEMAKER WOULD CALL THIS NOISE stanza may be a reference to his work for fermilab (detecting signal that often hides amongst noise), but is probably a double entendre of some sort. If someone is methodical enough to encode this text and mail it to Fermilab, they wouldn't misspell such a simple word (BASE), unless for a good reason. Along these same lines of thought, I believe the "noise" comment is also a clue with multiple meanings. Also, from what I gather, the middle stanza can be assumed to be hex, so that makes the third stanza fairly insignificant, unless it has other meaning (hence looking at "BASSE" for a clue as to some other meaning).
    • by bobdotorg (598873) on Tuesday May 20 2008, @03:54AM (#23472244)
      they wouldn't misspell such a simple word (BASE), ... (hence looking at "BASSE" for a clue as to some other meaning).

      Clearly,
      All your Basse are belong to us.

      Sorry.
    • by adam (1231) * on Tuesday May 20 2008, @04:25AM (#23472440)
      okay, now i'm beginning to become obsessed here, haha. My lack of mathematical background precludes me from decoding the stanzas (2 of 3 already done, and "peer reviewed").. but the psychological clues feel more within my grasp. If we examine the explanation at the first link on the story...

      With my initial interpretation of the top part of the coded message I got the following output: (021) FRANK@SHOEMAKER@WOULD@CAMV@FTVTCAPSBC
      The second link does a better job explaining, but basically one of the "words" in ternary was "wrapped" and due to the lack of hyphen, this was misintrepreted by both crackers. What I find interesting is not that once you actually solve the stanza, you get "FRANK SHOEMAKER WOULD CALL THIS NOISE," but rather that CALL THIS NOISE was the obscured part of the message. The signal that was hidden amongst the "noise" of a missing hyphen. The first cracker (John) speculates that he missed an indentation that indicates this (although he permits the possibility that it may be random), but I think there was no indentation, and the author wanted you to see the significance of this hidden word phrase (regarding "NOISE").

      Again, just as I believe "BASSE" is significant because it is misspelled (when nothing else is), I believe this wrapped word is significant (when no other words are wrapped). It's possible the encoder did this just to make things a bit harder, but if you look at the fact that it happens exactly at the part of the sentence referring to "noise," I believe you must be more inclined to lend it significance.

      Regarding BASSE, again, I am not a mathematician or a cracker, so I may be at a strong disadvantage here. If the significance of BASSE is taking the "extra" S and incorporating it into the middle stanza, I will be of little help to this collective effort. That said, if we attack the problem from a psychological/wordclue aspect... Googling "basse" doesn't help much, but google: fermilab basse ...and the second link [fermilabtoday.com] talks about Wilson Hall, and the Beauvais Catherdral, "occupied by the Romanesque church known as the Basse oeuvre," This page also talks about the fermilab logo, so I spent a while thinking that logo might have sixteen points, or sixteen intersections, etc.. nothing. But if we google image search "wilson hall fermilab" -- images of wilson [flickr.com] hall [fnal.gov] seem to show that it has sixteen stories when I count them. A quick googling reveals, "The 16-story Robert Wilson Hall is named after Fermilab's first director and was inspired by a French Gothic cathedral" --the cathedral occupied by the Basse Oeuvre-- Coincidence?

      In summary, BASSE SIXTEEN is (possibly) a sixteen story Fermilab building, named Wilson Hall. The significance of "NOISE" is still lost on me, and I believe the middle stanza should help with forward momentum. I am now going to review both explanations linked from the /. summary and attempt to parse something from the hexidecimal decoding(s) of the middle stanza.

      Perhaps more now than ever I wish /, posts could be edited, as I am *NOT* done with this, but I want to post it now so others can expand on my thoughts, or perhaps save me from heading down some pointless passageways of reasoning. Further posts to come. Oh, also, if you attempt to edit your previewed post more than three times, slashdot barfs on you and you have to re-write it. Could have saved 10 mins had I known that :(
      • by MoriaOrc (822758) on Tuesday May 20 2008, @04:47AM (#23472556)
        Just to help you stay on track, a note about the odd line breaks from the link to the "thought process" from below:

        The odd breaks occur because the way it's written is in a fixed-width row format. Each row contains an equal number of columns, and each column contains either a '|' or a ' ' (dash or space). The correct interpretation of the message removes the line breaks and translates the sentence as a single line.

        The first stanza has 47 columns per row. The 5-6 and 6-7 breaks occur because the last column in line 5/6 is a '|' but the first column in line 6/7 is also a '|'.

        The third stanza uses the same notation, but now each row consists of 85 columns. The 2-3 break has the same problem as in the first stanza, the row ran out of columns and the gap character had to be continued on the next row.

        If you're looking for significance with those gaps, instead consider the number of columns per row, and the fact that both stanzas have 7 complete rows and an 8th partial row.

        Misc numbers that may or may not be helpful:
        25 columns in the last row of Stanza 1
        21 columns in the last row of Stanza 3
        • by adam (1231) * on Tuesday May 20 2008, @05:14AM (#23472660)
          Expanding upon your "significance of 7 rows for both third and first stanza" theory, I immediately notice that the Wilson Hall building has 7 columns (count them: here [flickr.com] and here [fnal.gov]. Your suggestion appears helpful.

          If the orientation of the columns is rotated 90deg to make them rows, the stanzas may map to the columns in the building. If we assume the messages are significant, and the correlation to building "rows" is significant, and the left over "8th rows" from stanzas are significant.. we could derive all sorts of possibilities for the mapping of the remaining rows to a position in the building. Again, seeing how others here are much better at finding mathematically significant aspects than I am, I will throw this theory out and see if you or someone else can parse it.. because I believe the "25 columns in the last row of Stanza 1, 21 columns in the last row of Stanza 3" will need to be parsed somehow.

          Also, speaking of my lack of math background-- can anyone post something useful for the second stanza? I know John and Geoff (linked crackers) have decoded the three character string below the second stanza, as being "508 (0Ã--1fc) or 2812 (0xafc)" but what about the second stanza itself? If it's base sixteen encoded can someone work on decoding it? We are really working with 2/3 of the available information here, and I think the remaining third will provide a lot of momentum.

          also, as I expect this will continue long after this story is no longer at the top of the page, anyone who wants to collaborate via e-mail, may feel free to contact me. my email address is encoded as follows ;) ... myslashdotusernamewhichisfourcharacters.slashdot at gmail. Now I really wish I'd looked closer at the original story, instead of glancing and thinking, "wow, lots of math and the letter is probably a prank.. what else is there to read on slashdot today.."
      • Left Hander (Score:4, Interesting)

        by FlatWhatson (802600) on Tuesday May 20 2008, @04:55AM (#23472588)
        The message seems to have been written by a left-handed person. Analysis of the vertical lines in the two partially decrypted stanzas show a consistant skew a few degrees to the left which increases towards the right side of the page.

        Another clue on the psych path to decoding the SEKRIT MSGS !?
      • by nacturation (646836) * <nacturation@gmai l . com> on Tuesday May 20 2008, @05:16AM (#23472668) Journal
        I think the noise refers to the many black dots found on the page itself. Just below the vertical bottom of the first section, if you scroll all the way to the right you'll see a cluster of three dots and, going down-left from there, another two dots, another two dots, etc.

        Or look at the symbol section. You'll see the first symbol for 6 looks like a horizontal bar with a vertical hook and a dot under the bar. The second symbol for 6 has no dot. And to the right of the second symbol for 6 is a vertical cluster of three dots.

        Maybe they're nothing, but I get the sneaking suspicion that it's the dots (noise) that's the real puzzle here. Potentially with the symbols indicating the relative geometric arrangement of the dots that then map back to the letters/numbers.
         
  • An additional link. (Score:5, Informative)

    by legutierr (1199887) on Tuesday May 20 2008, @04:07AM (#23472318)
    Some of the thought process that went into the second solution. http://www.gmilburn.ca/2008/05/16/fermilabs-strange-code-letter/ [gmilburn.ca]
  • BASSE (Score:5, Interesting)

    by byennie (1126011) on Tuesday May 20 2008, @04:24AM (#23472434)

    After some Google work:

    Wilson Hall has a connection to ""Basse oeuvre". See this [fermilabtoday.com].

    Wilson Hall has 16 floors, and you must have an employee badge to access the 16th floor.
    • Re:BASSE (Score:4, Interesting)

      by textstring (924171) on Tuesday May 20 2008, @05:00AM (#23472612)
      The top floor of Wilson Hall has a lounge area and a lots of windows and it's where they take visitors for the view. You can however get to the floor above that but it's all concrete and DANGER signs, it is very noisy though: A clue!
  • by kidsizedcoffin (1197209) on Tuesday May 20 2008, @04:53AM (#23472580)
    I got an answer of 42.
  • by shungi (977531) on Tuesday May 20 2008, @05:46AM (#23472814)
    Basse Donnée system is a system that generates all the allowable solutions for given bass tasks within triads There are a few abstracts to the effect of the above that come on on google if one searches Basse and Science. I have no idea what it means, but i note: 1. It mentions triads - 3 is important in problem 2. It has something to do with music - sound, noise! 3. There is some sought of algorithm around it. ... Might be a trap though... (http://ci.nii.ac.jp/naid/110003111379/en/ )
    • Re:Regarding TFB(A) (Score:5, Informative)

      by Nazlfrag (1035012) on Tuesday May 20 2008, @04:35AM (#23472504) Journal
      First part is ternary code, I->1, II->2, III->0

      1) FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE

      Second part unknown

      2) ?

      Third part ternary code, II= seperator, same mapping as 1)

      3) EMPLOYEE@NUMBER@BASSE@SIXTEEN

      It is assumed the three hex symbols are the employee number 0xAFC,

      So lets assume the single "word" in the bottom middle of the page is an employee number. If we decode it using the symbols, we get (something)FC. (something) is an undefined symbol, and the only undefined numbers are 1 and A. So the "employee number in base 16" that "frank shoemaker would call noise" is either 1FC or AFC. My guess? Itâ(TM)s AFC (employee number 2812), who works on the AFC (Absorber Focus Coil [ox.ac.uk], a component of a "neutrino factory" current being studied at Fermilab) - a coincidence Frank Shoemaker would call noise. The employee number is reasonable and fits with the established pattern at Fermilab, see this Fermilab newsletter (page 5) [fnal.gov] which states "At 802, with only three digits, Matthews' employee number reflects the length of his 25-year tenure at the Lab".
      Hope that helped.
      • CHAR MAP (Score:5, Interesting)

        by stupidflanders (1230894) on Tuesday May 20 2008, @06:06AM (#23472920)
        When I saw the second paragraph, my first thought was to fire up Char Map.

        Whereas other people kept thinking that the middle section was supposed to be substituting the hex numbers for the symbols above, I had to wonder if the symbols were trying to tell us something. After all, as was pretty clearly pointed out by the people who have solved paragraphs one and three, each section contains only five lines. (In fact, the middle paragraph was your clue to this one -- it was just obscured by the fact that it was in two different codes -- but still, only five lines).

        Anyway, I realized that many of the symbols in the middle paragraph were in Charmap. AND each of these has a corresponding UTF code, which could be translated in to hex
        For instance:

        "Not Sign", U+00AC
        "Inverted Exclamation Mark", U+00A1
        "Greater Than Sign" (duh), U+003E
        "Single Right-Pointing Angle Quotation Mark", U+203A (note in the code they are two different sizes)
        "Greek Phi", U+03A6

        (Unfortunately, slashdot does not support these extended characters, as I found out. So I could not display all of them.)

        Not sure about the rest. The triangle COULD be a Greek Delta, but usually that is represented as a triangle with its base flat, not turned sideways. I have no idea what to make of the squiggly-"8"-like symbol. The three-pointed symbol could be a Greek Lambda, and possibly the top line is a Greek Tau. For the rest? You guess is as good as mine. I don't have the patience to go through CHARMAP symbol by symbol. Hopefully someone else just KNOWS this stuff. :-p

        I'm not a genius, so I'll leave this to the board to ponder some more. But the way I figure it, once you have the whole middle paragraph in hex, you should be able to translate it easily enough.
      • by zwei2stein (782480) on Tuesday May 20 2008, @06:55AM (#23473186) Homepage
        Obligatory:

        1) FRANK@SHOEMAKER@WOULD@CALL@THIS@NOISE

        2) ?

        3) profit

        (Filter error: Don't use so many caps. It's like YELLING ... jeez)
      • Re:Regarding TFB(A) (Score:5, Interesting)

        by xSacha (1000771) on Tuesday May 20 2008, @09:35AM (#23474446)
        Guys!

        The symbols in the middle are:
        s F C
        Correct? You worked out F and C were the other two symbols but F and C is undefined.

        Did you ever think it was: Shoemaker, F.C.?
        That is the initials Frank Shoemaker (mentioned in code)!
        In the website, using google, you will find only 7 occurances whereas you will find 107 occurances if you search for F. C. Shoemaker! Ding ding!
        • by WindBourne (631190) on Tuesday May 20 2008, @10:12AM (#23474990) Journal
          I am guessing that it is one of 2 things
          1. Seeing how we will do with alien messages. It is possible that we are trying so hard to see OUR information in there that we have missed the real message. Keep in mind that all of ource text is related to our experience. An alien's world will be different, so different context. IOW, we are seeing what we want to see and missing the real message.
          2. The other is that this is an experiment by the NSA to see how humans solve an issue. I suspect that if this is not one, then we have 2 messages embedded in this. That is, there is an innocuous message designed to throw us off i.e. noise. The real message is hidden as steganography i.e. a telephone number or an email. tele is going to be easier to embed in there.