Pattern Posted March 5, 2014 Report Share Posted March 5, 2014 And yet you consider a nonprefix code which is an order of magnitude harder than prefix one XD The answer to that is quite simple: I don't know how to work a prefix code, so I tried substitution which looked feasible enough. Assign a key to a number and there you go. Unfortunately it turns out to be more complicated... 1 Link to comment Share on other sites More sharing options...
Gary Singer Posted March 5, 2014 Report Share Posted March 5, 2014 The problem is that it's just numbers and we don't know how they split up and what they relate to. That's why I made the assumption they must be related to the only other paragraph from the Book of the Second Ceiling Rotation. I did some research on the net with ciphers when I started looking at it, and most of those don't seem to use numbers. They use letter for letter substitution. Or recursive algorithms using strings against other strings. Yes, ultimately the strings may be ASCII or other binary representations with XORs and other mathematical operations, but that's not exactly what we're dealing with here. There's been some really clever and interesting speculations on this thread! Link to comment Share on other sites More sharing options...
Satsuoni Posted March 5, 2014 Author Report Share Posted March 5, 2014 (edited) Ok, I am trying to decode the first 60 symbols for now, so that my program would run faster. As mentioned, it splits string semi-randomly into prefix codes, assign letters to codes and try to solve substitution using tWoK as a pattern for letters and words. It is still probably futile. This is what it came with for now: acheandirtseabethadneangullowaed ringhasnyothermenowahauldicres shinetomeaaspacksteasedullfrsat And other like it Incidentally, can any people in this thread run Python scripts? If so, I may consider releasing randomized version of script for others to play with. Edited March 5, 2014 by Satsuoni Link to comment Share on other sites More sharing options...
Gary Singer Posted March 5, 2014 Report Share Posted March 5, 2014 I could probably read it. I wrote a simple piece of a Perl hack code to break up the numbers in pairs and index into paragraph 15. The interesting thing was that arrays are zero based, so I put a dummy character in position 0 to offset the string so 01 would index into the first letter. I'll try the different number splits that have been posted a bit later. I wish there were some hints about what epigraphs are important. Maybe someone can ask a Brandon for hints at a signing. I'm going to one this weekend, and I certainly intend to! 2 Link to comment Share on other sites More sharing options...
Satsuoni Posted March 5, 2014 Author Report Share Posted March 5, 2014 Asking him would be nice Maybe we can at least narrow down the type, and eliminate possibility of transposition, etc ... As for the frankenscript: reading it is less important than having the ability to run it... Not all people can use Python and commandline. Link to comment Share on other sites More sharing options...
Pattern Posted March 5, 2014 Report Share Posted March 5, 2014 A ciphertext attack without knowing the encryption algorithm and key. That's what we are doing. Mapping into epigraphs could mean a mapping to letters or even to words. Stays the problem of splitting our 154 digits. So, I've got some real work to do, I'll be back in the evening. Perhaps someone finds out which algorithm has been used, at least ;-) Link to comment Share on other sites More sharing options...
Pattern Posted March 5, 2014 Report Share Posted March 5, 2014 (edited) Asking him would be nice Maybe we can at least narrow down the type, and eliminate possibility of transposition, etc ... As for the frankenscript: reading it is less important than having the ability to run it... Not all people can use Python and commandline. At the moment I don't know how to run the script, but I can learn - what do I need? I was wondering how the code would look in translations of WoR (e.g German)... Edited March 5, 2014 by Pattern Link to comment Share on other sites More sharing options...
Satsuoni Posted March 5, 2014 Author Report Share Posted March 5, 2014 At the moment I don't know how to run the script, but I can learn - what do I need? I was wondering how the code would look in translations of WoR (e.g German)... The script doesn't seem to use any advanced libraries, so a barebones installation of Python 2.7.6 and some knowledge of command line... Ok, what kind of things do you want it to ba able to do? Randomly split the code and give a decoding? Be able to fix some codes? Or just as is (which is probably very confusing)? Link to comment Share on other sites More sharing options...
Fabrial Posted March 5, 2014 Report Share Posted March 5, 2014 (edited) Fascinating watching you guys work on this. Have you considered that perhaps the message you're trying to decode isn't one message but two? The two translated sections: AhbuttheywereleftbehindItisobviousfromthenatureofthebondButwherewherewhereSetoffObviousRealizationlikeapricity TheyarewiththeShinWemustfindoneCanwemaketouseaTruthlessCanwecraftaweapon and TherehastobeananswerWhatistheanswerStopTheParshendiOneofthemYestheyarethemissingpiecePushfortheAlethito destroythemoutrightbeforethisoneobtainstheirpowerItwillformabridge are both translated from "Floorboard 17: paragraph 2" but the first one used "every second letter starting with the first" and the second uses "every second letter starting with the second". Do we first need to split the 4 line decimal code into two separate messages before you start trying to match letters to codes? I'm going to continue having a play, I've not done any cryptography before but will see if I can spot anything regardless. EDIT: Using RegEx I found that would give us our messages as so: every second letter starting with the first: 112511719121014115711012138310515434914993215410511019113455522155121012510634 every second letter starting with the second: 181012245101110211121117441171421410611441221012271510121211515711341191211534 This assumes that letters are made up of number pairs. I didn't edit our the 111s you were talking about. Here is the site I use to quickly try regex replaces, I've saved my search in that link: http://regexr.com?38ec3 EDIT 2: Just seen you mentioned this in post #26 Pattern. I think you're on to something. Definitely. Edited March 5, 2014 by Fabrial 1 Link to comment Share on other sites More sharing options...
Gary Singer Posted March 5, 2014 Report Share Posted March 5, 2014 Here's a slightly cleaned up version of what I had. OSX or Linux already have Perl. You can download Strawberry Perl for a PC. I couldn't upload it as code1.pl, so I renamed it code1.txt. You can run "perl code1.txt". Or rename it to code1.pl on a pc and run code1.pl. For unix flavors, you have to change permissions on the file after you've copied it - "chmod 755 code1.pl". code1.txt Link to comment Share on other sites More sharing options...
Satsuoni Posted March 5, 2014 Author Report Share Posted March 5, 2014 Hm, I think I am going to do something else for some while, and go to sleep for today. My mind is rather blank... Link to comment Share on other sites More sharing options...
Aminar Posted March 5, 2014 Report Share Posted March 5, 2014 Fascinating watching you guys work on this. Have you considered that perhaps the message you're trying to decode isn't one message but two? The two translated sections: and are both translated from "Floorboard 17: paragraph 2" but the first one used "every second letter starting with the first" and the second uses "every second letter starting with the second". Do we first need to split the 4 line decimal code into two separate messages before you start trying to match letters to codes? I'm going to continue having a play, I've not done any cryptography before but will see if I can spot anything regardless. EDIT: Using RegEx I found that would give us our messages as so: every second letter starting with the first: every second letter starting with the second: This assumes that letters are made up of number pairs. I didn't edit our the 111s you were talking about. Here is the site I use to quickly try regex replaces, I've saved my search in that link: http://regexr.com?38ec3 EDIT 2: Just seen you mentioned this in post #26 Pattern. I think you're on to something. Definitely. I don't think there are enough numbers for this, unfortunatley. Nor do I think splitting it straiht at pairs of 2 works. Link to comment Share on other sites More sharing options...
Fabrial Posted March 5, 2014 Report Share Posted March 5, 2014 (edited) We don't know that this is as long as the other two. It can't be... Okay how about this.... Assuming that the two parts of Floorboard 17 paragraph 2 were found by splitting the letters then decoding, we know that before they were decoded, the length of the two paragraph two pieces in decimal format were the same. Right? However, after they were decoded, the lengths are different. Does this confirm that there are stop or start bits for sentences or capitalisation? Because: Every second letter, starting with first. AhbuttheywereleftbehindItisobviousfromthenatureofthebondButwherewherewhereSetoffObviousRealizationlikeapricityTheyarewiththeShinWemustfindoneCanwemaketouseaTruthlessCanwecraftaweapon Total characters: 183 Capitals: 12 Every second letter, starting with the second. TherehastobeananswerWhatistheanswerStopTheParshendiOneofthemYestheyarethemissingpiecePushfortheAlethitodestroythemoutrightbeforethisoneobtainstheirpowerItwillformabridge Characters: 170 Capitals: 10 Edited March 5, 2014 by Fabrial 1 Link to comment Share on other sites More sharing options...
Harakeke Posted March 5, 2014 Report Share Posted March 5, 2014 A ciphertext attack without knowing the encryption algorithm and key. That's what we are doing. 3 Link to comment Share on other sites More sharing options...
Voidus Posted March 5, 2014 Report Share Posted March 5, 2014 I find the number of occurences of 1's and 5's next to eachother disconcerting since it's the 15th pattern, wondering if there might be something to that, but then the patterns might not really be sequential and so could just be randomly named. Rrrg, time to study up on cryptography. Link to comment Share on other sites More sharing options...
Luke.spence Posted March 5, 2014 Report Share Posted March 5, 2014 Someone just message me when you smart people figure this out, lol. 1 Link to comment Share on other sites More sharing options...
Aminar Posted March 5, 2014 Report Share Posted March 5, 2014 I find the number of occurences of 1's and 5's next to eachother disconcerting since it's the 15th pattern, wondering if there might be something to that, but then the patterns might not really be sequential and so could just be randomly named. Rrrg, time to study up on cryptography. That means it's probably a vowel. Vowels show up quite a bit more than Consonants. Link to comment Share on other sites More sharing options...
Harakeke Posted March 5, 2014 Report Share Posted March 5, 2014 Someone just message me when you smart people figure this out, lol. Aha! I got it! The cipher is actually a syllabic code derived from Elantrian Aons! If you split the number into triads, and assume that Aon Aon is already drawn, then the three numbers in the triad tell you how to add additional lines, starting in the upper left and working clockwise. Primes are straight lines, and nonprimes are curves. Then you mirror that shape along the diagonal axis (which makes sense, since Rosharians have such a thing for symmatry), and you wind up shapes that match the known Aons. Each Aon stands for a syllable, and if you write down the sequence and sound it out phonetically, you get: BesuretodrinkyourOvaltine! 16 Link to comment Share on other sites More sharing options...
ddleg Posted March 5, 2014 Report Share Posted March 5, 2014 lol Link to comment Share on other sites More sharing options...
Pattern Posted March 5, 2014 Report Share Posted March 5, 2014 At least you still have humour... Considering Floorboard 17: I don't think we can deduce anything from it - the shorter message could be cut, only gibberish after the readable stuff, etc. Perhaps Capitals can be used as a key to 2nd Ceiling rotation, Pattern 15. Link to comment Share on other sites More sharing options...
b4dave Posted March 5, 2014 Report Share Posted March 5, 2014 (edited) So far we have seen every second letter starting from the first, and every second letter starting from the second…..has anyone tried every second letter starting from the third? Edited March 5, 2014 by b4dave Link to comment Share on other sites More sharing options...
Pattern Posted March 5, 2014 Report Share Posted March 5, 2014 (edited) The script doesn't seem to use any advanced libraries, so a barebones installation of Python 2.7.6 and some knowledge of command line... Ok, what kind of things do you want it to ba able to do? Randomly split the code and give a decoding? Be able to fix some codes? Or just as is (which is probably very confusing)? Just as is could be ok - since I don't know what it does. A short instruction would be nice. Otherwise I ran out of ideas. Helpful would be, if we could guess some pieces of plaintext. Therefore, what has been written about in all epigraphs and what is still missing? I don't beliebe, the code is just a repetition of the other stuff. Keywords could be tried to decipher the code. So far we have seen every second letter starting from the first, and every second letter starting from the second…..has anyone tried every third letter starting from the second? The problem we have is to split the code into coded letters (btw, could also be syllables). This is possible in many ways and each possibility has to be decoded. But - feel free and give it a try ;-) Edited March 5, 2014 by FeatherWriter Please don't double post! Link to comment Share on other sites More sharing options...
DragonReborn Posted March 5, 2014 Report Share Posted March 5, 2014 This would be great if Brandon purposefully did this to make everyone on this site start going crazy and figuring it out. Link to comment Share on other sites More sharing options...
Luke.spence Posted March 5, 2014 Report Share Posted March 5, 2014 This would be great if Brandon purposefully did this to make everyone on this site start going crazy and figuring it out. I wouldn't put it past him or Peter for a second. Link to comment Share on other sites More sharing options...
Aminar Posted March 5, 2014 Report Share Posted March 5, 2014 I'm pretty sure this code is aimed straight at us. As in if he didn't have us as a fanbase he'd never put in a code like this. Link to comment Share on other sites More sharing options...
Recommended Posts