Kryptos Workshop Transcript

Richard Bean provided a transcription of the Kryptos workshop on 24 October 2015.
THANK YOU VERY MUCH!

3:55

I look at cryptography as building things … there’s other aspects, and maybe from your perspective … in the sense of … if you had a cryptographic paradigm is there a way to break it or understand it or what’s the schema that’s involved in that.

so there’s two different kinds of things … in process, or times, we talked about times many years ago … some of the folks at Langley had a gathering of people like about 300 and they invited me as a non speaker I had to sit in the audience and that was always interesting it’s sort of like sitting in this type of an audience which is a small audience but a large scale audience who have looked at Kryptos from _lots_ of perspectives including the NSA and others [5:03] and it was interesting and again I wasn’t acknowledged … looking at some of the work that the next organization [looking at Frank Corr, ACA] does is really interesting also. [5:31]

7:09
some of the things that you see or you should have seen is the impact of some kind of math function on the English language
7:23
and now because the tools that you have been using or would be using are tools [that’s] based on [periosity? periodicity? imperiosity??] of letters and things like that within existing codes
and so there’s a mask of some type now now you don’t have that as your initial tool … you’ve gotta remember your process as well as what you’re trying to do
and most folks their process is let me see how many trials (??) I have and the weighting of Es and things like that (8:03) and that would give you some insight
now added to that is the math associated with the cryptography and that’s some substitution transposition … but you carry (moves both hands L to R) still you’re carrying the weight of the language when you’re doing that
and the other intent was is there a way to make the challenge so that the first step is that thing and now that gets to be hard (8:36) and that’s what you all are (waves hand at audience) facing
is how do you do that [8:41]

11:30
was Jim Sanborn’s dismissal of ID BY ROWS as a mistake correct, did you play a role in embedding any other intentional misspelling
11:55
when you’re doing these types of things and I’m going to say in an abstract as opposed to answering your direct question
when you’re doing these things you look at the intent again … the intent was and still is if I had to communicate to … any of you in a foreign country, a phone call away and I wanted to do it over a period of time
but I don’t know _when_ … [12:26] when I’m saying all of these kinds of things, if I knew when it gets kind of easy you know I’m going to call you at 12 o’clock tomorrow
… but if I don’t know when that gets a little bit more tricky and then I have to do something that you could come back years later and recreate.
and that’s one of the challenges you are being put into the posture of recreating without having prior knowledge of it [13:00] …
the person I would be talking to would have prior knowledge that you know one of these days somebody like an Ed is going to talk to me. [13:10] and if so how do I do that. now this type of question here lends itself to
what happens if I put a gun to his head, and now he wants to tell me there’s a gun there without getting himself shot, I know how to do that … and so it’s a serious business playing spy… that’s what you’re talking about.
and if you do it wrong he’s dead, so you try not to be wrong and at the same time in essence it’s an escape valve for that person, now we all understand we, I mentioned earlier
the cryptography is analogue but the person is also analogue … [???] … you’ve got feelings, you’ve got stress all kinds of things takes [sic] place
when somebody puts a gun against your head, and now the question is do you say I give up, or do you say or rationalize that well I’m not but I’d like to tell somebody I have a problem,
and this type of methodology is one way of doing it.
14:39
many many many years ago there was a group of us at about 10 or 11 o’clock at Langley and what it was was some tapes
from overseas and it was actually a captured person and that person was trying to tell, or at least the belief was, you know … to say that tell my wife I’m all right, or things like that,
but how do you do that when you have a gun on this side and a gun on that side, in the tape, and that gets to be interesting, because the analogue part of the person [15:31] [taps table] I start doing this, or I do that now I can start to [15:32] [correlate?!?!] … I used to make those codes, for captured people,
so it’s an another avenue of what you all are looking at, that was taking advantage of the human side of this paradigm … in other words just [gestures at Frank] asked about [???????] you know [16:01]
that’s the digital side, the human side of it is about 10%, the human side of this type of thing is a much higher percentage of what goes into looking at what that can be to a person [16:16]
the person from _your_ position or the person who knows the answer, or knows the answer in the sense of knows how to use it that can be better … this is a form of saying well I’d like to be able to talk back to you
and then when I’m talking back to you I realize I may have a gun against my head or under oppression so I don’t want to make a mistake …. anyhow … that another facet of it.
[16:55]

18:32
the fun part is to see how the folks approach the problem (18:41) … talk mathematics with people and half the problem is how do you approach the problem as opposed to figuring out all the paradigms and algorithms and things like that …

19:37
we’ve tried to introduce things as we go along that would show up in my dialogue to him (gestures to someone on other side of table) and him is somewhere out in that world and …

22:30
does Kryptos in general simulate multiple overlapping signals?
the technology, and I’ll answer it this way, can show up in other kinds of things because the intent from my perspective now you’ve got an artist and you’ve got the cryppie over here (gestures to himself)
or the person with the math and the math that can be applied to things it tends not to be applied to everyday instances like these machines (taps something) because this is digital [23:05]
this is analogue (hits paper) so you don’t see a lot of analog answers in cryptography (23:13) in usage today.
but at the same time, the principles or the concepts are there to be used and people have used them [rubs Frank’s shoulder] these guys use them every day and uh … pointing to the Cryptogram Association
in the sense of substitution transposition use of things in that context … I’m reading a lot into this work … could lead to puzzles as you said earlier … and it’s always interesting to see that …
the K4 may not be something as such so that’s sort of unique in that sense (24:07) but the others have all the intrinsics of classical cryptographic puzzles
so the answer can be yeah I think you would see it in other non-Kryptos type of scenarios … perhaps not all and perhaps not in the same fashion
how you did the mix in terms of other things like transposition and things like that
as the math model allows you to do it different ways … and you may even get a convoluted way of doing it because it lends itself to building layers or you could mix and match which is a certain amount of that took place
as such (25:04) as principles … the thing that I had thought of and I was asked not to do… this is based on the English language … I could have moved to a different math base (25:28) in other words this is base 2 we could move to base 16 and then evolve the problem … different folks thought that’s not fair Scheidt … seeing the driveway that’s Mayan now it’s a two-dimensional Mayan it’s not a pictorial Mayan but there’s a code out there. Interesting…. Now there is simple substitution you know the Mayan language and you know how to handle base 60 then you’ve figured out the answer. now taking that and throwing it into, adding transposition and some of these other techniques it would give you another [xxx] headache for sure (26:23) so those are the kinds of things that could take place because my example earlier he’s from Slovolia can I map my schema to his language or if I pick an [extinct] Roman language that’s not on everybody’s tongue [or database] that that helps me [narrows it down] … how are you going to get to that first step [27:16]

[29:57] how does the human characterise information? … ok Ed, how would you approach this problem? and I said well, I believe the human uses a base number … base 2 is basically the computer and we understand how to
add logic to that … so if we don’t use base do you believe it can be replicated in math. and I said yes, there were 2 out of 8 that said yes. …. [story about Lockheed, F-35] … that’s a tangent, but that’s part of the puzzle.

[34:37] when was the last time you worked out for yourself what K4 says? I honestly haven’t gone back in many many years to be honest I guess I can use the answer I get consumed with all this other stuff
audience Q: does that mean you’ve done it at least once many many years ago
A: yeah, once, a couple of times …
audience: that’s interesting because Jim seems to claim he did something else on top of this
A: he may have and that’s up to Jim to talk about that but the part that I worked on … [35:30]
audience: so I guess I thought the question was poking at have you taken the actual ciphertext the bottom of K4 …
A: never …
Jim: you never decrypted K4 yourself?
Ed: *coughs* never … like you say in the endgame … I stayed away …
audience: I got really confused there
I have not done the last … K4 …
Jewlee: K4 what we call K4 is that what you’re calling K4 or is it just the last part?
Ed: … the 97 … I intentionally did not do this stay away because if Jim was going to do something then OK Jim does something and then I can honestly say well I only know what I did [36:37] and that’s it.
so I answered your question in that context. Remember, we try not to deceive you.
Jim: so if Jim made an encoding error you would not have known that he did so? [36:53]
Ed: no
Jim: ok
did Jim write the plaintext or did you produce that?
no, he produced the plaintext … that was part of his contract, … this is a, if you read the thing, there’s a story, if you read … it reads like it’s disjointed but there’s a story that goes with it …
if I’m going to be talking to this gentleman again in never never land and then he divulges the code and everything else then is there any other level of secret I can have with him
that only him knows and now that gets you into a depth that hasn’t come out in any of the discussion because it’s not apparent there but at the same time there are other things you can do with these types of things
… this is only an example, it’s not in … I could put the word coca-cola in a text which was encrypted and then you decrypt it and now if coca cola took a meaning a dual meaning then that gets to be interesting
in itself. because you would be led along the lines of it’s a drink as opposed to it really means you should use a black pen or something like that so there’s codes that can take place within the content
and that can be part of how you deal with this type of correspondence.

audience: so the text that’s on the sculpture at some point you went in and you looked at the ciphers and you worked backwards to the plaintext? [39:06]
yeah, except for K4 I didn’t. I did all the others, because the others is … it has things that in the early stages I wanted to ensure that the math was still there as such
that was I come to you and say ok I can show you how to do substitution and you say I understand and you do some sample and I say well ok it looks all right
(trust but verify) yeah, that’s right and now if there’s a tricky part, or tricky part in the sense of you added some things to as you’re asking here on the K4 or another step or whatever it may be called
then do I want to go back and look at that [40:07] or do I say well Jim you have all the tools that you need to do that now go do it … and then if I have to verify everything that gets into these kind of questions
here (hand moves in circles) of [40:33] whose project is being driven … we agreed a long time ago it really is Jim’s project … I’m sort of the back person that says 2 and 2 _is_ 4 not 5 that type of thing …

[Sanborn’s favourite topics to talk about when being taught encryption systems]
[42:00] background audience talk about “the prototype before the sculpture”

[42:31] The intent was to create a secret and then the secret could have different levels [42:44]
of being able to break or different kinds of math and that’s where we had a discussion on here’s some of the things that you can use that are pretty well understood as opposed to
I think I had mentioned earlier using some extinct language which ok that becomes unfair for a lot of folks or to pick a language that may not show up in databases today and that’s kind of unfair too [43:25]
so part of the discussion was well and this was in line with Langley that this is not a one-time system – the difficulty would be forever [43:43]
but at the same time can you make it as a challenging type of thing [43:49] and then you get into the supposition which is what we have talked about earlier [43:54]
that is you know three five seven years short period of time [43:59] that would that be possible [44:02] and then
to what extent would this be a pencil and paper design versus using [44:11] a computer …
are you talking paper for executing it or for breaking it?
executing it [44:22] remember I only know how to executive _he’s_ (Frank) on the other side of the …
I’ve had others say you definitely have a convoluted mind [44:43] and things like that … but there were other things talked about other than just the cryptography and at the same time
the sensitivity of the project was part of keeping the secret [45:12] …
what’s a period, because there’s more than one sentence in there [46:28] and how do you characterise that and
do we have apostrophes, special characters … special characters don’t work [46:40] you have to take that into account …

*cut*

Q: I’ve got K1, 2 and 3 are one-stage encryption – is that also true for K4 [47:06]
aha – now yeah you’re closer to that answer than [saying???] is at one stage [47:16]
no, I would consider it more than one stage [47:19]
I’ve mentioned earlier and I have mentioned in public there’s a masking technique that applied so that the English language wouldn’t be available [47:35]

…
[48:16] Jim’s giving us clues without really a cryptographer’s or a cryptanalyst’s sense of what is really a clue [48:25] and so he gave us that and we didn’t break it immediately
I think most of us realized that we could get another word another word another word and it doesn’t really give us anything else than just a copy of what we’ve got already as this matches up …
so if he gives us ASPARAGUS next year, and OZYMANDIAS five years from now that doesn’t give us anything more than BERLIN did I think would that be a [48:49] correct …
that’s where I would need to defer, because what you are doing is from an analytical side breaking down the process
(Bill?) I’m afraid that Jim’s just giving us clues when he’s really just giving us additional plaintext until it’s all out there [49:24]
and then we still don’t know how to reverse it
if it’s a process that has nine different encryption steps you’re not going to be able to reverse it from the plaintext
or he made you solve it so that it has two solutions

Denny: did you advise him as to which [classical systems] to use or was that his choice? [49:58]
what other systems did you show him?

I just don’t remember how he selected …
but it was his selection?
at the time, I believe it was [50:35]
so you showed him the how and he did the what?
right because that was part of the discussion as such because he’s a sculptor and he wants to put his touch on things and so he could do that either through the metal or he could do that in the content and which is what we talked about.

Denny: why would he use the same system for two different parts – if there were a number of things available, why would he use the same one twice [51:17]
I can’t answer the question directly, because that’s a Jim’s call
[something about a polygram and a polynomial] [51:48]

which came first, Cyrillic Projector or Kryptos [52:42] Cyrillic

[what is clue in plain sight / duress code] [53:45] [Jim] has said that something that was key is no longer there.
Jim questions … [no clear answers]
[masking first then encryption? or the other way around] [no clear answer]