|
Post by catherwood on Mar 2, 2019 13:12:42 GMT -5
Here are a few prior posts about the blocks of color around the borders of pages 19 and 47 to get us started: Some of the multi-colored border shapes are a solid color. I took the letters that were touched by the solid colors. I included the "U" even though it was barely touched. And, I would think it should be a more clear-cut decision on whether it was touched. Also, there is a couple of solids that don't have letters (a thought just occurred- use them as spaces between words?). Anyway, I came up with this as a possible anagram..."eye on road spur" Glad to see so much discussion on this hunt. I tried (unsuccessfully) to decode the U shaped blocks many times which I firmly believe represent the ‘hidden treasure riddle.’ A few observations/thoughts: -each U block likely represents a letter -a pixel count of five for most blocks would make sense. Noting some blocks are all comprised of one color, these would decode to some of the most common letters in the alphabet using simple substitution (E=5, O= 15, T=20) -simple substitution implied by “ACADIA” on license plate -EACH COLOR MUST BE ASSIGNED A NUMBER (numbers open the riddle) -the total number of u blocks has a similar count of letters to the riddle in Masquerade -the code location in the border definitely qualifies as “hidden”; pages 19 and 47 -“1947” year of the fire on the island given in painting -the pixel colors (roughly 24?) are each seen as different blocks of color throughout important pages of the book (books, sign posts) -colored piano keys implies colors are the “key”, Harlequin definition is multicolored Never have been able to correctly assign a color with a number. Important numbers seem to exist throughout the entire puzzle, most notably the ISBN number on the orange book, the mileage on the signs, and numbers on the grid (?puzzle).
|
|
|
Post by catherwood on Mar 2, 2019 13:23:44 GMT -5
If the blocks were always in a consistent 3x2 cluster, with the same divider square to make a 5-block U shape, we could easily make the case for a glyph alphabet cipher. However, the pattern of divider squares shifts in a few spots, making some glyphs with 5x2 (minus 2 for an 8-bit data point of a single color) and some are just a bar of 2 squares (always a single color). It's not insurmountable, but it is a problem.
The paintings show a careless regard for precise color-matching, which is frustrating. Is light green the same data point as dark forest green? How many shades of blue did he use? I've tried frequency analysis, based only on the pattern of arrangements and not on the colors themselves, but didn't get a good distribution for a short block of (what we assume is) English text.
Are the 2- and 8- bit clusters a mistake? A well-formed glyph alphabet could be made all within the 5-bit U template, so why would more glyphs be necessary?
|
|
|
Post by morpheus221 on Mar 2, 2019 13:59:49 GMT -5
Yes. The 8 clusters could pose a problem. By this methodology, the yellow 8 cluster would yield a 24 (if yellow = 3). Letter X. Not sure what to make of them otherwise. Might he have added them (2 and 8 clusters) due to spacing in the painting since the circle cuts off a portion of it?
I am convinced that there are multiple shades of color here which unfortunately makes deciphering more difficult. Perhaps with a complete key it will be more obvious.
|
|
|
Post by catherwood on Mar 2, 2019 14:06:03 GMT -5
Sorry, found some more older posts with observations: ...The border codes on p.19 and p.47 strike me as the most likely to hide a key sentence. I'm trying a brute-force approach using a computer program, assigning random values to each color, adding them up, and assigning the values to letters. So far I've run half a million permutations, and thought something interesting would turn up just by chance, but have had no luck. I'll share if anything meaningful comes up. How do you think the code might work? I've read the polybius theory but there seems to be a dearth of 4-runs of a color. I, too, believe the key hint is hidden on one of those two pages' borders. However, what's got me stumped are the aberrations on pg. 19. (i.e. the "E-shaped" blocks with 8 units and the "I-shaped" blocks with 2 units.) What do you make of this? I also have been using the brute force approach thus far, but with a pig pen cipher variant instead of polybius. Let me know if anything interesting pops up; I'll do the same. I still like the thought of a Polybius cipher using pairs of colors, but we have too much variation in the border sequences, no matter how we take pairs. PigPen is certainly a creative approach, which is just another way of using a simple-substitution after assigning a value to a unique pattern or group of "pixel" squares.
|
|
|
Post by henrycarnauba on Mar 2, 2019 16:27:02 GMT -5
I believe the inner borders are an alignment factor, and not a code in and of themselves. There are 2 pages that have these borders, they must be adjacent when pages are aligned in a grid. Too random to represent an actual code.
|
|
|
Post by catherwood on Mar 2, 2019 17:06:07 GMT -5
The inner border as alignment would make sense. My data tells me that on page 19 all of the inner/outer pairs match (ignoring the alternating alignment squares of orange which matches the painting's border). However, on page 17 (which also has an orange outer edge), there are 6 instances where the inner and outer squares do not match (not counting the yellow alignment squares).
These are located: on the left side under the second R, on the top under the O, on the top under the U, on the right side under the first A, and both legs of the block between TC on the right side.
darn, it looked like it was about to spell something. Nevermind. But why couldn't this "random" bunch of colored squares mantain any consistent pattern?
|
|
|
Post by morpheus221 on Mar 2, 2019 19:30:01 GMT -5
Can you guys explain “inner/outer” with regard to the border on pages 19 and 47? Not sure I am following correctly.
|
|
|
Post by catherwood on Mar 2, 2019 20:20:39 GMT -5
oh sorry, i thought it was self-explanatory. Maybe 'bands' of squares is a better term than 'rows', as the sides of the page make columns instead of rows, technically. By thinking of the colored squares as just two bands of "pixels" which wrap around the inside of the border frame, one row band is inside and one row band is outside. The corner 4 squares of 2x2 are consistent in marking the rows: white squares align with the outer rows in all 4 directions on both pages, while yellow on page 47 and orange on page 19 align with the inner rows. Note that the color of the inner corner square matches the color of the alternating squares along the inner band, which is what leads to the illusion of what we're calling the U-blocks.
|
|
|
Post by morpheus221 on Mar 2, 2019 21:25:12 GMT -5
Okay. Gotcha. Thanks...regarding frequency and color saturation wondering if a good starting point may be to identify how many actual ‘colors’ we are talking about for both pages 19 and 47. This list probably needs some more scrutiny. I will also add similar color blocks in the paintings. Feel free to add/amend:
1) Red - 3 shades (?) -books, grid puzzle, and signs that spell out yellow 2) orange - 1 shade -grid puzzle, ITLOTW Book, Jordan pond sign 3) yellow - 1 shade -grid puzzle, yellow book on shelf 4) green - 3 shades (?) -grid puzzle, “wind” book, Jonesport sign 5) blue - 2 shades -coast and treasure island books, machias and Sieur De Monte springs sign 6) purple - 2 shades (?) -grid puzzle, signs that spell red, lands and “wh/go” book 7) pink - 1 shade -grid puzzle, ?down under book 8) white - 1 shade -signs that spell out “boonk”, down east book 9) orange/yellow (p 19) - 1 shade -bass harbor sign 10) gray - 1 shade -myth and legends book 11) gold - 1 shade -Montreal sign, book on top shelf on side (no spine showing)
|
|
|
Post by morpheus221 on Mar 19, 2019 17:45:25 GMT -5
Here is a thought. Do you think the colors will make more sense once you have figured out the key? Agree, the color shades are a bit ambiguous but they might be clearer if you identify the proper color/number combinations.
|
|