A virtual box puzzle and garbled datastreams
From Intimation
Contents |
It takes teamwork
The following part of Sys’ twitter feed introduced today’s puzzle:
>>Sys:Log//Safeguards in place. Supplemental precautionary processes established. Initiating Stonewall crack.
>>Sys:Log//A tentacle has breached the crack. It is not hostile. I have attached messages for S. Now analyzing signature.
>>Sys:Log//The tentacle contains this data: "I've got a new game for you! I guess my last one was too hard, so I'm re-hiding some pieces!"
>>Sys:Log//Continued: "Not all of them though, only a few. This game needs teamwork, but I bet you won't finish it and find me too! :@)"
>>Sys:Log//Diagnostics report much activity utilizing this network tentacle. This has happened before. What has S done this time?
>>Sys:Log//S's tentacle has remained still for some time now. Can she sleep? Or is she waiting? Curious.
These tweets were also relevant:
>>Sys:Log//S's tentacles are extensions of her personality, but there may still be much to learn about S and the their nature.
>>Sys:Log//Observing this tentacle I've found a very faint anomalous datastream. I've extracted the data, but am unable to make sense of it.
>>Sys:Log//Garbled datastream: gpKKCQkNDQ0 H+AfeODC7Pg CQkPB2Ji5KQ QUFCgoOCgIA 5AQIECDAAAA AAADBAgQICA +AcABAYDAQE AADAIBAIBIQ
>>Sys:Log//Garbled datastream: HX39+fn6+vI AAMe+IAAB/g gICAgIBDR0Q c39/fnzwAAA gAA8yJgwMDA AIDA4T8A4B8 qDAhAQMDAwE JicTCQQDAAA
Direct messages were also sent by Essy to several different twitter users. Each of these DMs was of this form:
- Shhh...don't tell sis! http://thewaytheirworldende... Now find the others, then come find me! :@)
Each of these DMs linked to an image file on the main website. It was quickly discovered that not every twitter user received the same link. Essy stated “This game needs teamwork” in the public tweet for a reason: we had to make sure we compared the links we’d each received to gather together all the image files needed to solve the puzzle, then continue the puzzle-solving from there.
The puzzle pieces
Koreth summarized an IRC discussion of the various images and the puzzle in general thus:
As Nyst pointed out on IRC, the puzzle block images follow the form of Bn... where n is a number, and the fill-in-the blank images are Fn... where again, n is a number.n can have values from 1 to 6, each value corresponding to one of the six sides you can view a cube. Now, our wooden block puzzle is not a cube in shape, but could be considered to fit inside a clear cube of the same size, giving us 6 different angles from which to view the completed block puzzle, and thus, fill out the boxes and get the words for the next part of the puzzle.
Here is a summary of the images we had to work with:
Blocks:
- http://thewaytheirworldended.com/BOXES/B1BVND6253KHOJ8767.png
- http://thewaytheirworldended.com/BOXES/B2CVTD3577OGKB0977.png
- http://thewaytheirworldended.com/BOXES/B3DGFH8237MBJS8735.png
- http://thewaytheirworldended.com/BOXES/B4HGYO2775NVBB9275.png
- http://thewaytheirworldended.com/BOXES/B5PALK6629UIIU6736.png
- http://thewaytheirworldended.com/BOXES/B6UYVC5430UVYL5243.png
 |  |  |
 |  |  |
Solutions: (note are still missing F6)
- http://thewaytheirworldended.com/BOXES/F1BZYR9374PALY5344.png
- http://thewaytheirworldended.com/BOXES/F2GFTI8765BCHO9238.png
- http://thewaytheirworldended.com/BOXES/F3PAJG5456KKRN0190.png
- http://thewaytheirworldended.com/BOXES/F4PAYV6298JJQT0993.png
- http://thewaytheirworldended.com/BOXES/F5PSUH1084BZUI4785.png
 |  |  |
 |  |
Solving the block puzzle
Cortana noted:
Judging by the two kinds of images, I think once we have all the pieces, it's a "use these letters, make them into words" kind of puzzle. The ones like this:http://thewaytheirworldended.com/BOXES/F1BZYR9374PALY5344.png
Are telling us where to find the letters, while the ones like this:
http://thewaytheirworldended.com/BOXES/B4HGYO2775NVBB9275.png
Are telling us what the letters are.
anaerin added:
And I know the puzzle, or at least the basic premise. I had one in wood, as I mentioned. Problem is, we need to "Make" (in some form) this puzzle to solve it.As another clue on assembly of this thing, the X's denote where pieces cross.
Koreth suggested a solution method: “IIRC, the most basic form of this kind of wood block puzzle has six pieces, and the Fn solutions look like the 6 piece kind. Does anyone have an actual one of these wood block puzzles and a sharpie, perchance?” It turned out, however, that Nyst was already creating a model of the pieces with styrofoam.
Koreth also found a video that goes step-by-step through how to assemble a wood block puzzle of this type at http://www.youtube.com/watch?v=7IoH2ESLLto .
Nyst was able to assemble his model and provide us with diagrams of the finished product (see image link at right). We then had to figure out which Fn solution went with each face of the puzzle.
Koreth worked out that F1 corresponds to the upper middle shape in Nyst's diagram, leading him to find http://thewaytheirworldended.com/DEEP-BREATH/ . This page contained part of a cartoon picture of an excited young girl. This was true of the solutions to the other sides we were able to discover as well.
anaerin found that F2 matched up with the bottom-left diagram (leading him to http://www.thewaytheirworldended.com/ALWAYS-KNEW/ ) and that F4 matched up with the one in the bottom-right corner (leading to http://thewaytheirworldended.com/FALLEN-HARD ).
Nyst found that F5 led to http://www.thewaytheirworldended.com/SEE-THROUGH .
drizjr found that F4 matched up with the upper-right diagram and led to http://thewaytheirworldended.com/READY-OR-NOT/ . This page is also found to be where Essy is hiding, and some text and a link to a tune Koreth composed for her (which she liked) are also found there. (Note: when we first found this page, the message there used to end after the line that reads "♫ doo-be-doo-be-doo ♫.")
Shad0 found that f3 matched up with the bottom-middle diagram and led to http://thewaytheirworldended.com/GO-TOGETHER
A diagram showing most of the puzzle solutions we were able to find is linked to at the right here.
drizjr also discovered that as we found each cartoon piece, a copy of it appeared at the bottom of http://thewaytheirworldended.com/BROKEN-PIECES/ .
Do we tell Sys?
Recall that at the end of the last puzzle (Hostile tentacle attack 2), we found an image seemingly scrawled by Essy that read, "Don't trust her!" We were not sure what to make of this message, but we were concerned that it was telling us that Sys was untrustworthy and not to tell Sys where Essy was hiding. We decided that until we could decode the "Garbled datastream" that was part of the twitter feed for the puzzle, we could get away with not telling Sys where Essy was hiding.
Eventually, however, the following lines were added to http://thewaytheirworldended.com/READY-OR-NOT/ (note that these were added one line at at time, over time, except the section from "I wonder" to the end was added as one chunk):
I sure hope the rest of my hidden pieces are found soon!
Boxes, boxes, boxes. That was a fun puzzle! I think I'll make a harder one sometime.It really is quiet here...
When I yawn, I do this: 001111000011101001001111I wonder what's taking so long this time for sis to find me. Maybe this will get her attention!
00000000 00111100 00110000
00011000 01000010 00011000
00011000 10011001 00001100
00000000 10100101 00000100
00000000 10100101 00000100
00011000 10011110 00001100
00011000 01000000 00011000
00000000 00111110 00110000
hee hee!
Based on the "I wonder" line, we assumed that Essy actually did want to be found, and sent her location to Sys. This caused this message to appear at the READY-OR-NOT page:
- >>SYS.S3:MSG//CONTROL POINT LOCATED/TENTACLE RETRACTED/CRACK REPAIRED
Decoding the binary puzzles
enaxor found that the binary string at the end of the "When I yawn" line on the READY-OR-NOT page translated to <:O.
Koreth discovered that the blocks of "binary" on the same page are actually an ASCII drawing, but you have to get rid of the zeroes to see it. anaerin drew up the solution:
1111 11
11 1 1 11
11 1 11 1 11
1 1 1 1 1
1 1 1 1 1
11 1 1111 11
11 1 11
11111 11
This turned out to be a hint as to part of how to decode the garbled lines of text.
Decoding the garbled datastream
After all of this, we still had not decoded the garbled datastream from the twitter feed before the next section began. Here is a repeat of the relevant tweets:
>>Sys:Log//Garbled datastream: gpKKCQkNDQ0 H+AfeODC7Pg CQkPB2Ji5KQ QUFCgoOCgIA 5AQIECDAAAA AAADBAgQICA +AcABAYDAQE AADAIBAIBIQ
>>Sys:Log//Garbled datastream: HX39+fn6+vI AAMe+IAAB/g gICAgIBDR0Q c39/fnzwAAA gAA8yJgwMDA AIDA4T8A4B8 qDAhAQMDAwE JicTCQQDAAA
Koreth and snorkle256 posted that they had a hunch it was base64 encoded, but could not make sense of the results of doing base64 decoding on the datastream. anaerin confirmed, “Each section is 1 character too short to be a base64 stream in itself, but you can pad with ‘=’ on the end. The sets of 8, however, are complete base64 phrases,” but was also unable to make any sense of the decoded datastream.
It was at this point that Essy posted her binary yawn on the READY-OR-NOT page, and, a little later, her pseudo-binary ASCII pignose drawing.
This gave drizjr a brainwave:
- Maybe it's a hint to what to do with the garbled text... ASCII drawing?
Several forum members then gamely tried to make an ASCII drawing by converting the garbled datastream to binary blocks similar to Essy’s pignose drawing, then pulling out the ones, but nothing sensible seemed to emerge (despite a large number of attempts to pick out large block letters from the resulting drawings). MrToasty described the process this way:
- “If you binary encode them such that each group of characters (don't encode spaces) gives you one line, then for each tweet you get 11 blocks of 8x8. You can start to see what look like patterns, possibly letters, but nothing immediately jumps out. Maybe need to swap some rows?”
Here is an example of what was generated through this attempt:
gpKKCQkNDQ0 H+AfeODC7Pg CQkPB2Ji5KQ QUFCgoOCgIA 5AQIECDAAAA AAADBAgQICA +AcABAYDAQE AADAIBAIBIQ 01100111 01110000 01001011 01001011 01000011 01010001 01101011 01001110 01000100 01010001 00110000 01001000 00101011 01000001 01100110 01100101 01001111 01000100 01000011 00110111 01010000 01100111 01000011 01010001 01101011 01010000 01000010 00110010 01001010 01101001 00110101 01001011 01010001 01010001 01010101 01000110 01000011 01100111 01101111 01001111 01000011 01100111 01001001 01000001 00110101 01000001 01010001 01001001 01000101 01000011 01000100 01000001 01000001 01000001 01000001 01000001 01000001 01000001 01000100 01000010 01000001 01100111 01010001 01001001 01000011 01000001 00101011 01000001 01100011 01000001 01000010 01000001 01011001 01000100 01000001 01010001 01000101 01000001 01000001 01000100 01000001 01001001 01000010 01000001 01001001 01000010 01001001 01010001 11 111 111 1 1 11 1 1 11 1 11 1 1 1 11 1 11 1 111 1 1 1 1 1 11 1 1 1 1 11 1 1 11 11 11 1 1 1 1111 1 1 1 11 11 111 1 1 11 111 1 11 1 1 1 11 1 11 1 1 1 1 11 1 1 1 1 11 1 1 11 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 11 1 11 11 111 11 1111 1 1111 1 11 11 111 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 111 1 1 1 1 1 1 1 11 1 1 1 1 11 1 1 11 11 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 HX39+fn6+vI AAMe+IAAB/g gICAgIBDR0Q c39/fnzwAAA gAA8yJgwMDA AIDA4T8A4B8 qDAhAQMDAwE JicTCQQDAAA 01001000 01011000 00110011 00111001 00101011 01100110 01101110 00110110 00101011 01110110 01001001 01000001 01000001 01001101 01100101 00101011 01001001 01000001 01000001 01000010 00101111 01100111 01100111 01001001 01000011 01000001 01100111 01001001 01000010 01000100 01010010 00110000 01010001 01100011 00110011 00111001 00101111 01100110 01101110 01111010 01110111 01000001 01000001 01000001 01100111 01000001 01000001 00111000 01111001 01001010 01100111 01110111 01001101 01000100 01000001 01000001 01001001 01000100 01000001 00110100 01010100 00111000 01000001 00110100 01000010 00111000 01110001 01000100 01000001 01101000 01000001 01010001 01001101 01000100 01000001 01110111 01000101 01001010 01101001 01100011 01010100 01000011 01010001 01010001 01000100 01000001 01000001 01000001 1 1 1 11 11 11 111 1 1 1 11 11 11 11 111 11 11 1 1 11 111 11 1 1 1 1 1 1 1 1 11 1 11 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1111 11 111 11 111 1 1 1 1 11 1 1 11 111 1 1 1 1 1 1 1 1 1 1 11 1 1 1 11 11 11 11 111 1 1 1111 11 11 11 111 1111 1 111 111 1 1 1 1 1 1 11 111 1 1 1 1 111 1111 1 1 1 1 11 111 111 111 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 111 1 1 11 1 1 1 111 111 1 1 1 1 1 11 1 1 1 1 1 1 1 11 1 1 1 1 1 111 111 1 1 1 1 1 1 11 1 1 11 11 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1
In response to some inquiries by Kilination, Sys sent a DM with another clue:
- >>Sys:Msg>@Kilination//Being relatively subconscious in nature, perhaps the segments are encoded in a method familiar to, even favored by S.
0N1MUSH4 mentioned to Sys that we'd tried various things, but had no idea what order the segments of the datastream were supposed to be in. Sys replied via DM:
- Sys:Msg>@0N1MUSH4//I will attempt another reading of the faint datastream, and attempt to determine its most likely segment order.
Decoding continues with clues given next active session
New clues were found in part of the next active session's twitter feed.
For one thing, Sys suggested an order for the segments:
>>Sys:Log//Continuing from previous session. There has not yet been success decrypting the oddly garbled datastream found in S's tentacle.
>>Sys:Log//The datastream was very weak, barely detectable, and retrieved in 16 segments. These segments have a meaning. Continuing analysis
[snip!]
>>Sys:Log//Though the garbled datastream is intriguing, I am also curious to continue my research.
[snip!]
>>Sys:Log//Preparing to initiate crack.
>>Sys:Log//Safeguards in place. Supplemental precautionary processes established. Initiating Stonewall crack.
>>Sys:Log//S has breached the crack as usual, the tentacle now displaying signs of heavy activity. Analyzing.
>>Sys:Log//Analysis shows there is more common order to the garbled segments. Datastream segments in their most likely order to follow.
>>Sys:Log//Possible order: AAADBAgQICA H+AfeODC7Pg +AcABAYDAQE AADAIBAIBIQ QUFCgoOCgIA gAA8yJgwMDA CQkPB2Ji5KQ gpKKCQkNDQ0
>>Sys:Log//Continued: gICAgIBDR0Q c39/fnzwAAA qDAhAQMDAwE HX39+fn6+vI JicTCQQDAAA AIDA4T8A4B8 AAMe+IAAB/g 5AQIECDAAAA
A further clue was provided by Essy:
>>Sys:Log//I also retrieved another prominent signature: "Yay! Part of my story was found! But there's still so much more to go!"
>>Sys:Log//Continued: "c I've 3 hidden R some y more Z for X you d to u find, I so G maybe F more i people b can 3 help V! 0 :@)"
>>Sys:Log//Due to extended analysis of this tentacle, I am postponing hibernation 120 minutes.
>>Sys:Log//Crack stability holding. There is no more clear data signature in this tentacle. If S is hiding, her clue is visible.
Pulling out the extraneous characters from the second tweet gave us c3RyZXduIGFib3V0.
halo422mjk found that this was base64 encoded, and upon decoding it found http://thewaytheirworldended.com/STREWN-ABOUT/ .
Indeed, another tweet from Sys noted:
- >>Sys:Log//S has hidden an additional set of data to be recovered by her friends at [STREWN-ABOUT]. I do wish I could help; were it so easy.
This gave us some confirmation that the garbled datastream was indeed in base64; we just needed to try some other tactics on the decoded data (such as putting our decoded chunks into the correct order).
Koreth drew up the new ASCII diagrams resulting from the new order:
Using the rearranged chunks from the first garbled tweet:
Each segment is a block
AAADBAgQICA H+AfeODC7Pg +AcABAYDAQE AADAIBAIBIQ QUFCgoOCgIA gAA8yJgwMDA CQkPB2Ji5KQ gpKKCQkNDQ0
00000000 00011111 11111000 00000000 01000001 10000000 00001001 10000010
00000000 11100000 00000111 00000000 01000001 00000000 00001001 10010010
00000011 00011111 00000000 11000000 01000010 00111100 00001111 10001010
00000100 01111000 00000100 00100000 10000010 11001000 00000111 00001001
00001000 11100000 00000110 00010000 10000011 10011000 01100010 00001001
00010000 11000010 00000011 00001000 10000010 00110000 01100010 00001101
00100000 11101100 00000001 00000100 10000000 00110000 11100100 00001101
00100000 11111000 00000001 10000100 10000000 00110000 10100100 00001101
11111 11111 1 1 1 1 1 1 1
111 111 1 1 1 1 1 1 1
11 11111 11 1 1 1111 1111 1 1 1
1 1111 1 1 1 1 11 1 111 1 1
1 111 11 1 1 11 1 11 11 1 1 1
1 11 1 11 1 1 1 11 11 1 11 1
1 111 11 1 1 1 11 111 1 11 1
1 11111 1 1 1 1 11 1 1 1 11 1
Using the rearranged chunks from the second garbled tweet:
Each segement is a block gICAgIBDR0Q c39/fnzwAAA qDAhAQMDAwE HX39+fn6+vI JicTCQQDAAA AIDA4T8A4B8 AAMe+IAAB/g 5AQIECDAAAA 10000000 01110011 10101000 00011101 00100110 00000000 00000000 11100100 10000000 01111111 00110000 01111101 00100111 10000000 00000011 00000100 10000000 01111111 00100001 11111101 00010011 11000000 00011110 00001000 10000000 01111110 00000001 11111001 00001001 11100001 11111000 00010000 10000000 01111100 00000011 11111001 00000100 00111111 10000000 00100000 01000011 11110000 00000011 11111010 00000011 00000000 00000000 11000000 01000111 00000000 00000011 11111010 00000000 11100000 00000111 00000000 01000100 00000000 00000001 11110010 00000000 00011111 11111000 00000000 1 111 11 1 1 1 111 1 1 11 111 1 1 1111111 11 11111 1 1 111 1 11 1 1 1111111 1 1 111111 1 1 11 11 1111 1 1 111111 1 11111 1 1 1 111 1 11111 1 1 11111 11 11111 1 1 111111 1 1 1 11 1111 11 11111 1 11 11 1 111 11 11111 1 111 111 1 1 1 1111 1 11111 11111
Putting them both together:
11111 11111 1 1 1 1 1 1 1
111 111 1 1 1 1 1 1 1
11 11111 11 1 1 1111 1111 1 1 1
1 1111 1 1 1 1 11 1 111 1 1
1 111 11 1 1 11 1 11 11 1 1 1
1 11 1 11 1 1 1 11 11 1 11 1
1 111 11 1 1 1 11 111 1 11 1
1 11111 1 1 1 1 11 1 1 1 11 1
1 111 11 1 1 1 111 1 1 11 111 1
1 1111111 11 11111 1 1 111 1 11 1
1 1111111 1 1 111111 1 1 11 11 1111 1
1 111111 1 11111 1 1 1 111 1 11111 1
1 11111 11 11111 1 1 111111 1 1
1 11 1111 11 11111 1 11 11
1 111 11 11111 1 111 111
1 1 1 1111 1 11111 11111
MrToasty was able to do some rearranging(?) to finally obtain the following:
░░░░░░░░░░░██████████░░░░░░░░░░░ ░░░░░░░░███░░░░░░░░░░███░░░░░░░░ ░░░░░░██░░░█████░░░░░░░░██░░░░░░ ░░░░░█░░░████░░░░░░░░█░░░░█░░░░░ ░░░░█░░░███░░░░░░░░░░██░░░░█░░░░ ░░░█░░░░██░░░░█░░░░░░░██░░░░█░░░ ░░█░░░░░███░██░░░░░░░░░█░░░░░█░░ ░░█░░░░░█████░░░░░░░░░░██░░░░█░░ ░█░░░░░██░░░░░░░░░░░█░░██░░░░░█░ ░█░░░░░█░░░░░░░░░░░░█░░██░░█░░█░ ░█░░░░█░░░████░░░░░░█████░░░█░█░ █░░░░░█░██░░█░░░░░░░░███░░░░█░░█ █░░░░░███░░██░░░░██░░░█░░░░░█░░█ █░░░░░█░░░██░░░░░██░░░█░░░░░██░█ █░░░░░░░░░██░░░░███░░█░░░░░░██░█ █░░░░░░░░░██░░░░█░█░░█░░░░░░██░█ █░░░░░░░░███░░███░█░█░░░░░░███░█ █░░░░░░░░███████░░██░░░░░█████░█ █░░░░░░░░███████░░█░░░░███████░█ █░░░░░░░░██████░░░░░░░░██████░░█ █░░░░░░░░█████░░░░░░░░███████░░█ ░█░░░░██████░░░░░░░░░░███████░█░ ░█░░░███░░░░░░░░░░░░░░███████░█░ ░█░░░█░░░░░░░░░░░░░░░░░█████░░█░ ░░█░░██░░░░░░░░░░░░░░░░░███░░█░░ ░░█░░████░░░░░░░░░░░░░██░░░░░█░░ ░░░█░░████░░░░░░░░░████░░░░░█░░░ ░░░░█░░████░░░░██████░░░░░░█░░░░ ░░░░░█░░░░███████░░░░░░░░░█░░░░░ ░░░░░░██░░░░░░░░░░░░░░░░██░░░░░░ ░░░░░░░░███░░░░░░░░░░███░░░░░░░░ ░░░░░░░░░░░██████████░░░░░░░░░░░
Yamato noted that this looks like a scorpion. Brae noticed that this symbol appears on http://thewaytheirworldended.com/SCORPIO/ , a page that was part of an earlier puzzle. The symbol also appears on the follow-up page from that puzzle at http://thewaytheirworldended.com/STAY-AWAY/ . The symbol was on both of those pages at the time of that puzzle, but this is the first time we were able to see it clearly, without distortion.
There did not seem to be any more to this puzzle (recall that Essy's location had already been reported to Sys and the tentacle retracted--see the "Do we tell Sys?" section above for information on this). It seems the purpose of the "garbled datastream" part of the puzzle was to show up a picture of the "bad little scorpion" that Essy said her story would be about when she set up the GPS missions. We consider both the virtual box and garbled datastream puzzles to be complete.
