Follow Slashdot stories on Twitter


Forgot your password?

Decrypting Kryptos 235

angkor writes "Kryptos is a sculpture located on the grounds of CIA headquarters in Langley, Virginia. Installed in 1990, its thousands of characters contain encrypted messages, of which three have been solved (so far). There is still a fourth section at the bottom consisting of 97 or 98 characters which remains uncracked."
This discussion has been archived. No new comments can be posted.

Decrypting Kryptos

Comments Filter:
  • Crack? (Score:5, Funny)

    by theluckyleper ( 758120 ) on Sunday January 23, 2005 @04:26PM (#11449516) Homepage
    Dynamite oughtta crack it...
    • Re:Crack? (Score:5, Funny)

      by slAckEr Of dOOm ( 818662 ) on Sunday January 23, 2005 @04:34PM (#11449567) Journal
      Typical gamer - "if you don't understand it, blow it up"
      • But a slacker of doom, on the other hand, would spend years working out a way to decipher it? :P
      • To be fair to the typical gamer, we've been teaching them that if it's weird and incomprehensible, it should be shot for years. Most games follow this formula. Puzzles are obvious and understandable with very clear lights and meaning, whereas bosses are those big things with funny faces that you shoot.

        Things that are not to be shot have to be comprehensible to the players, or else they won't figure out what to do with it.

      • Re:Crack? (Score:5, Funny)

        by david.given ( 6740 ) <> on Sunday January 23, 2005 @07:18PM (#11450734) Homepage Journal
        Typical gamer - "if you don't understand it, blow it up"

        Now, now. The full saying to the shmup motto is, IIRC:

        "If it moves, kill it. If it doesn't move, shoot it until it does. Then kill it."

        Unfortunately I can't find a source --- anyone?

        • That sounds a bit like the "Roadie Rules".

          * If its wet, drink it.
          * If its green, smoke it.
          * If it moves, fuck it.
          * If it doesn't move, throw it in the back of the truck.
        • > "If it moves, kill it. If it doesn't move, shoot it until it does. Then kill it."

          The way I heard it (for old RPG's) was, "If it moves, shoot it. If it doesn't move, shoot it. If it's still there, pick it up."
          • The LARP I used to play in it was "If it moves, hit it until it stops. If it doesn't move, hit it in case it starts."
  • by mr_burns ( 13129 ) on Sunday January 23, 2005 @04:29PM (#11449533)
    "Stop goofing off. Get back to work"

  • by Anonymous Coward
    And if you decrypt it all, dark suited men show up at your door and demand a kidney, or worse.

    Mercury Rising anyone?

    El Saladhead
  • What is the methodology behind code cracking? Many codes I figure eventually get solved due to something like the Rosetta Stone, but how is even that point reached?
    • by Anonymous Coward on Sunday January 23, 2005 @04:39PM (#11449605)
      It depends on the code. Often it involves looking for patterns, knowledge of letter frequencies, a bit of luck, and a while lot of skill. It helps if you know the type of cryptography being used. For example, if there is a known algorithm being used, such as RSA, one can do a brute force attack and test every possible decryption key (a very computationally intensive task, but feasable for small, older keys). Since many codes are based upon "hard math problems" such as RSA whose security is derived from the fact that it is easy to multiply prime numbers, but hard to factor composite numbers into primes, some codes are broken by find fast solutions to hard math problems. This is kind of the idea behind Shor's Quantum Computer algorithm which allows one to quickly factor large integers and thus could, if implemented, completely screw over RSA.
    • by The Only Druid ( 587299 ) on Sunday January 23, 2005 @04:43PM (#11449632)
      As a wonderful sci-fi connection, I suggest reading Cryptonomicron, by Neal Stephenson. Besides providing an excellent story (like all his books), it provides an extensive discussion of how code-breaking works, and how historically it evolved.
    • There's a classic book on the history of codes and codebreaking called The Codebreakers, by David Kahn []. A revised edition came out recently. There is even a Wikipedia article [] about him.

    • by 0racle ( 667029 ) on Sunday January 23, 2005 @06:31PM (#11450414)
      My girlfriend took one look at it and saw an image embedded in the way the characters are laid out. She figures that a real message would be too obvious and since its art, the real purpose could simply be to see what is not plainly there.

      Maybe the boys at Langley are being too literal at trying to solve it.
    • The Rosetta Stone has nothing to do with code breaking. The Rosetta Stone is a stone with writing on it in two languages (Egyptian and Greek), using three scripts (hieroglyphic, demotic and Greek). It is important, because the identical passages in Greek, allowed scholars to understand Egyptian writings.
  • Kryptos? (Score:5, Funny)

    by MikeXpop ( 614167 ) <mike.redcrowbar@com> on Sunday January 23, 2005 @04:30PM (#11449542) Journal
    Superman must be shaking in his boots right about now.
  • Is there a solution? (Score:5, Interesting)

    by mistersooreams ( 811324 ) on Sunday January 23, 2005 @04:31PM (#11449551) Homepage

    Forgive me for being cynical, but how can we be sure that this final piece is actually crackable? I mean, it would be a cruel joke by Jim Sanborn (the structure's creator) to just include a load of junk, but who's to say he didn't? The fact that the other parts have been deciphered suggests that this last part will give way eventually, but maybe that's just to make the final joke even crueler?

    I hope I'm wrong because that would make this story a lot less interesting, but I just thought it should be mentioned.

    • by ncurses ( 764489 ) * on Sunday January 23, 2005 @04:49PM (#11449679)
      of course not, it's just some random chars generated by the Arethusa cipher, seeded with the string COMSTOCK.
      • by EvilSporkMan ( 648878 ) on Sunday January 23, 2005 @06:57PM (#11450563)
        Actually, the characters you're referring to were NOT generated by Arethusa, they were generated by a Riemann zeta function; this was the whole point of that part of Cryptonomicon. Parent should be right where he is, +3 for Effort.
        • by Anonymous Coward
          You said "by a Riemann zeta function?" No sir, there is only ONE Riemann zeta function and it is THE Riemann Zeta Function.

          However, the Riemann Zeta Function is part of a family of functions called Dirichlet L-series.
          • Well, I was just going by the wording Stephenson used. From a search on
            3. on Page 880: "... found that the Arethusa intercepts did not represent coded messages at all.They were simply the output of a particular mathematical function, a Riemann zeta function, which has many uses-one being that it is used in some cryptosystems as a random number generator. He proved ..."
    • Yes, he could've used ye olde book of random numbers [], for a laugh.

      If so, he'll get what's coming to him in the geek afterlife: An eternity of this [].
    • by iabervon ( 1971 ) on Sunday January 23, 2005 @05:41PM (#11450040) Homepage Journal
      It's certainly possible that the area people call "part 4" is actually random junk stuck in to fill up that section of the sculpture (the width of that portion is determined by the first two parts, and the height by the table on the right), and that the final solution to the sculpture uses entirely different elements.

      If the end of the text is part 4, there's probably at least a part 5, because there's a bunch of other stuff in the installation not yet used in any solutions (including, for example, a compass that points the wrong way).

      I wouldn't be surprised, from the known decryptions, if you had to stand with a candle at a certain point and notices the letters cut in the shadow cast by the sculpture at particular points. It seems to me like the bottom rows would be most likely as the letters for this, so it may be that some of the letters are important, but that there's no way to solve it by just looking at a transcript.
    • Sanborn is probably having a laugh at the pros trying to decrypt what is really only jibberish. It is just like the joke Magic Eye [] poster that doesn't have a hidden image.
    • OK, here we go. From the FAQ []:

      Q: Are you sure that Kryptos part 4 is solvable?

      Yes. Both Jim Sanborn and Ed Scheidt have repeated over and over that it's solvable. Sanborn has also been quoted in interviews as saying he was surprised that it hadn't been solved yet. And when Elonka Dunin, co-moderator of the Kryptos group, asked him flat out in mid-2003 whether or not part 4 was solvable, his answer was: "Yes. It ain't easy, but it's solvable!"
  • 97 or 98? (Score:5, Funny)

    by Anonymous Coward on Sunday January 23, 2005 @04:36PM (#11449586)
    If you can't tell how many characters it has, no wonder it's so hard to crack.
    • Re:97 or 98? (Score:5, Informative)

      by Zocalo ( 252965 ) on Sunday January 23, 2005 @05:06PM (#11449787) Homepage
      From the FAQ:

      All of the characters on the ciphertext side of Kryptos have been solved, except for 97 characters at the very bottom. There is also a question mark between parts 3 and 4. Some say the question mark is part of part 3 (since it ends with the question, "Can you see anything q"). However, it's possible that the question mark is part of part 4.

      So, there you go. Having a question mark at the beginning is a little odd, unless it's a hint that you are meant to read it backwards like some of the other Kryptos puzzles.

  • Wouldn't the key to cracking it lie in like a key inside of the code? That somewhere in the code, it explains how to decode itself. Also, by looking at the artwork itself, it looks like two semi-circles.
    • One idea, and I'm sure others have thought of it, is to connect the two semi-circles then draw lines. How the lines are drawn, don't know. We could try common letters, vowels, sentences, alphabets, etc.
  • by The Journalist ( 844669 ) on Sunday January 23, 2005 @04:38PM (#11449601)
    There's a "W", then an "e", a space, "a", "r", "e"...

    "We are"..."

    There's more! "s", "o", "r", "r", "y"

    "We are sorry"...We knew this!

    "f", "p", "r" - "We are sorry for" - "t", "h", "e", - "We are sorry for the what!?!

    "i", "n", "c", "o", "n", "v", "e", - almost got it - "n", "i", "e", "n", "c" "e".

    "We are sorry for the inconvenience"? You bastard!!

    *Apologies to DNA, who orignially used this idea as God's final message to the univers.

  • damn (Score:3, Funny)

    by 2MuchC0ffeeMan ( 201987 ) on Sunday January 23, 2005 @04:41PM (#11449618) Homepage
    where that's kid from mercury rising when you need him.
  • I got it !! (Score:5, Funny)

    by Dolphinzilla ( 199489 ) on Sunday January 23, 2005 @04:42PM (#11449625) Journal
    Be Sure to drink your Ovaltine.
  • The article says that the NSA failed as well, but, honestly, the NSA wouldn't say if they had.

    They probably have, long ago, and just don't tell anybody. They're probably chuckling about it right now.
  • by Anonymous Coward
    stupid decoder ring
  • It's a confession that Elvis isn't dead; that he's been working for the CIA all this time, in the disguise of an Elvis impersonator.
  • I saw this article from Wired yesterday and looked at the image that was provided. They said it was given to them by Elonka Dunin. The thing is, if you look very closesly at the letters in an image window under magnification, there are intermittent red pixels occurring on certain letters on what should be a black and white only image. Could this be another code?

    Here is the image []

  • by Linker3000 ( 626634 ) on Sunday January 23, 2005 @05:11PM (#11449808) Journal
    There seems to be a pattern emerging in Slashdot posts - it can be outlined by the following pseudo-code:
    Program GoSlash;
    CDate = Today - 1;
    ..For I = 1 to NumArticles(Fark(CDate));
    ....If SubjectScore((FarkArticle[I,CDate]) >= MildlyTechnical then
    ......NewArticle = Rehash(FarkArticle[I,CDate]);
    ......SubmitToSlash (NewArticle);
    ....end; I;
    • Actually the program you've discovered is run every other day. On the other days they use:
      Program GoSlash;
      CDate = Today - 1;
      ..For I = 1 to NumArticles(BoingBoing(CDate));
      ....If SubjectScore((BoingBoingArticle[I,CDate]) >= MildlyTechnical then
      ......NewArticle = Rehash(BoingBoingArticle[I,CDate]);
      ......SubmitT oSlash (NewArticle);
      ....end; I;

  • by Weaselmancer ( 533834 ) on Sunday January 23, 2005 @05:12PM (#11449809)

    The people who solved the first 3 are currently in jail for DCMA violations. ;^)

    • Mod this man to the top!!!

    • No, they're actually enjoying the Good Life on a year-long world cruise. Haven't you seen the commercial on TV before "Hockey Night USA" on TechTV?

      We just need someone to bust into the CIA to find the cells where these three have since become mummified husks, chained to their cell beds...
  • by krypt0s ( 72886 ) <krypt0s&yahoo,com> on Sunday January 23, 2005 @05:12PM (#11449812)
    Someone could've just asked me. I'm not THAT complicated.
  • by BierGuzzl ( 92635 ) on Sunday January 23, 2005 @05:15PM (#11449827)
    Weird. This guy apparently metaphorically solved section 4. But he got an email from Jim saying "Dear John, this is not the way" Did he cheat? :) l
  • Sanborn may be referring to something he buried on the CIA grounds, though he's not saying. The decrypted text mentions a burial and gives latitude and longitude coordinates (38 57 6.5 N, 77 8 44 W), which Sanborn said referred to "locations of the agency."

    GlobeXplorer [] shows some parkland adjacent to a collection of buildings (presumably CIA HQ), but no "X".
    It's probably only a cache of Iraqi WMD, or an alien spacecraft... Move along, nothing to see here.
    • by Stiletto ( 12066 )
      You must not have put in the correct coordinates. In decimal degrees:

      -77.1455555555555555 lon
      38.95180555555555555 lat

      which is a courtyard right smack in the middle of CIA headquarters.
    • Looks to me to be on the edge [] of a courtyard.

      Zooming out a bit, and overlaying the 1984 topographic map on top of the 1988 B/W USGS photo, I get this view [] which looks like that point used to be in a field next to the building, but when it was expanded (sometime between 1984 and 1988) that location was enclosed in a court yard.

      The nearest benchmark is HV4826 [] which is reported as being destroyed in 1984 - probably during the expansion of the building, so I would say it happened early in that timerange.

  • by gelfling ( 6534 ) on Sunday January 23, 2005 @06:07PM (#11450234) Homepage Journal
    My God it's a cookbook.
  • Plagiarism (Score:5, Informative)

    by Osty ( 16825 ) on Sunday January 23, 2005 @06:10PM (#11450262)

    From the Slashdot summary:

    angkor writes
    "Kryptos is a sculpture located on the grounds of CIA headquarters in Langley, Virginia. Installed in 1990, its thousands of characters contain encrypted messages, of which three have been solved (so far). There is still a fourth section at the bottom consisting of 97 or 98 characters which remains uncracked."
    And from the actual page:
    Kryptos is a sculpture located on the grounds of CIA Headquarters in Langley, Virginia. Installed in 1990, its thousands of characters contain encrypted messages, of which three have been solved (so far). There is still a fourth section at the bottom consisting of 97 or 98 characters which remains uncracked.
    So, unless angkor is the author of the page over at, he's plagiarised the article for his summary. Now, I understand that this can be a difficult call to make, since the article is clearly cited. However, the language of the summary ("angkor writes ...") and lack of explicit citation ("the article says ...") leads one to believe that angkor is writing an original summary of the submitted article, which is clearly not the case.

    Sadly, this is not the first time this has ever happened on Slashdot (in fact, it happens in nearly every posted article). Come on, people. If you're going to submit a story, either summarize the article in your own words or attribute your summary text to the article. And editors, pull your thumbs out of your asses and actually edit your site once in a while. In a case like this, it's pretty damned obvious that the article summary is just part of the first paragraph of TFA, and so rather than attributing the summary to the article submitter ("angkor writes ..."), use other language that makes clear the quoting ("angkor quotes from the article ...").

    • Re:Plagiarism (Score:2, Informative)

      by frakir ( 760204 )
      Not to mention "its thousands of characters" are actually just below 900...
    • It probably comes under "Fair Use", but a very good point all the same - the whole F/OSS movement is very strongly based on correct attribution, and just as the code has to be developed carefully, it wouldn't hurt if the most prominent website about F/OSS took just a little care about proper attribution.
      Better still, (no, please, don't laugh, just hear me out...) Slashdot should be setting an example - raising the bar of properly accredited journalism.
      Okay, you can laugh now. I know you've been holding it
      • NOT Fair Use (Score:3, Interesting)

        by new500 ( 128819 )
        . . .

        In the UK at least, an author has stautory provision against false attribution. Fair use itself does not usually take consideration of the effect or intent. No new work was created in which a fair use rule can be applied. The effect is redistribution in a database, for which there is a ton of case law saying the incident is actionable.

        Although to go into the grey area here would take too long, the person who "writes" is attributing material to themselves.

        Ah, but it's Slashdot who writes "writes". S
    • Given the dodgy anaphor of the second sentence (which looks more like a cataphor), it's distressing that anyone would plagiarise it directly.
  • /. ings (Score:2, Funny)

    by McBeer ( 714119 )
    It is interesting to note that Elonka Dunin (one of the most prominent people involved with the cracking of Kryptos and the Cyrillic Projector) put a slashdotting in her timeline of important Kryptos events. []
  • Dupe (Score:3, Funny)

    by morningstar8 ( 234758 ) * on Sunday January 23, 2005 @07:29PM (#11450841)
    This story is a dupe... The original was posted 5 1/2 years ago. (!) Here's the original story. []
  • by new500 ( 128819 ) on Sunday January 23, 2005 @10:10PM (#11451898) Journal
    . . .

    George Smiley,
    Asst. Attorney to
    Director, National Security Agency
    Chief, Central Security Service
    (NSA Information Assurance Department)
    Date As Decrypt Key

    Re: Unauthorized Use and Disclosure of Intellectual Property


    Dear Cryptanalyst,

    I serve as legal counsel to the NSA Information Assurance Department, owner of extensive intellectual property rights and trademarks pertaining to the use, distribution and deployment of intelligence worldwide. In fact you may have heard of us. To make you fully understand our concern and the reach of our recognised brands throughout the world, let me put it this way, we do what RIAA only dreams they could.

    It has recently come to our attention that John Doe, in personam, i.e. youself, the only possible recipient of this message has sought to circumvent our intentional copy protection of classified communications, thereby exposing our proprietary materials, name, marks, trade dress, intellectual property and good will to possible illegal misuses including but not limited to commercial exploitation or karma whoring on Slashdot.

    By reading this message you have violated federal laws, including (among others) the Digital Millennium Copyright Act, the Economic Espionage Act, the Electronic Communications Privacy Act, the Wiretap Act, the Legal Lobbyist Retirement Protection Act, and the Consumer Fraud and Abuse Act, as well as State of Wisconsin Natalie Portman 3D Redistribution Act (HP Amendment). (We're the NSA, we know about that one too.)

    Therefore we require that you immediately CEASE AND DESIST from any and all activities causing, leading to or which might be construed to result in the actual or potential dissemination of the proprietary information and excellent legal drafting contained herin. Under the terms of the DMCA, inter alia, we inform you that henceforth your knowledge of this text will be deemed to be a Circumvention Device, and as such we are required to place restrictions on your person. Kindly call me on the number below and await instructions. Do not move, do not try to escape, do not pass go and do not collect $200 (that's all we have left after spending $20Bln on the Great Monument to ourselves you see before you.)

    Failure to comply with these requests may expose you or your organization to an action for injunctive relief or monetary damages, and any other relief permitted under state and federal law, including court costs and attorneys' fees. You may also wish to consider and examine the potential criminal consequences, under theories of aiding and abetting and conspiracy to denigrate the agencies elite avant-garde sculptural skills.

    If you fail to comply with these requests we will have to invoke recourse under the Homeland Bitchslap Act of 2001.

    George "W" Smiley.

    P.S. Son, you should have just applied through personnel. Way back when I was a junior we dreamed up this sucker distract the Russians who'd waste all their time drinking vodka and analysing it just to get one over us. Don't worry, I'll tell your Ma it was friendly fire.

  • No use in trying to decypher it. The aliens didn't put it in English when they gave it to the CIA...

    It's obviously written in the standard galactic alphabet...

  • by 222 ( 551054 ) <> on Monday January 24, 2005 @01:06AM (#11452975) Homepage
    I got into the project about a year ago, and I can tell you that we (Yahoo Kryptos group) have some brilliant people trying just about everything imaginable.
    I cant tell you how much time i've spent in the middle of the night arranging the letters in x,y grids... and even more bizzare.
    I reconstructed the statue in 3D Studio Max so that i could tinker with the idea of folding the statue on itself, etc. One of the vigenere keywords in an earlier section hints at the reuse of the message, so its just an angle i've been tinkering with. []
    If anyone would like a copy of the max files to tinker with, shoot me an email at minus food :)
  • but unfortunately the margin of this post is too small to contain it.

    - Jim

Experience varies directly with equipment ruined.