Experiment 57
Experiment 23 removed filler letters at regular spacings. Here the filler is chosen by letter instead: every copy of one letter, or of a pair of letters, is struck out wherever it falls outside the two known stretches, the known letters move to their new places, and the repeating-secret-word test is run at every length.
What the method does with the 24 known letters.
A message can be padded with filler letters that the reader is told to ignore, for example every Q. Experiment 23 tried filler at regular spacing. Here the filler is every copy of one letter, or of two letters, outside the known stretches. Striking them out moves the known lettersthe known letters Sanborn has confirmed that positions 22 to 34 of the message read EASTNORTHEAST and positions 64 to 74 read BERLINCLOCK. to new places, and we ask again whether a repeating secret wordthe secret word The word or string a cipher needs besides the method. of any length could fit them. Random strike-outs of the same size are the controla shuffled control The same experiment run on the carved letters shuffled into a random order..
Words in the margin are defined on the glossary page.
The carved letters go in at the top; the reading comes out at the bottom.
A filler letter is a letter Sanborn might have scattered through the message to be ignored. Striking it out shifts every later place, so BERLINCLOCK lands earlier.
Every letter, and every pair of letters, was tried as the filler, struck everywhere outside the two known stretches: 351 choices, then the usual repeating-key test under every length.
The shortest secret word each choice allows is spread just as for random strike-outs of the same size: nothing about one or two filler letters helps.
This experiment produced no candidate: the search ruled everything out before a full text was built. Its product is the bounded negative in section 7.
Redo the method on a few letters with pencil and paper.
The same steps run by the code on the real letters.
Q at [26, 27, 39, 42]; struck outside the crib ranges: [39, 42]; BERLINCLOCK moves from 64-74 to 62-72
Family, key space, the constraint under test, and the clues it reads.
C4 masking and C18 ('a bit of stego', 98 characters) allow nulls; a null letter that is a real letter chosen by rule is listed open in families.md. Model: every occurrence of a null letter (each of 26) or of a null pair (325) is removed from the carved text outside the crib ranges 22-34 and 64-74 (a null inside a range would contradict Sanborn's positions), shifting the cribs' plaintext indices by the nulls before them; the rest is a periodic tableau text. Test periods 1-52 under the plain and cipher alphabets standard and KRYPTOS and three modes (the key alphabet drops out of consistency, 0006), exactly. Control: for each mask, ten masks removing the same number of letters at random positions outside the ranges. Bounds: complete for 351 letter masks and periods 1-52.
Written before the run; the result decides it.
No letter mask makes any period below 26 consistent more often than random strike-outs do; the smallest consistent period stays in the high twenties. Zero candidates.
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.
351 letter masks; smallest consistent period per mask: {24: 18, 25: 117, 26: 168, 27: 45, 47: 3}; random strike-outs of the same size (per 10): {24: 17.1, 25: 104.6, 26: 174.3, 27: 48.4, 45: 0.4, 46: 1.0, 47: 5.2}. Masks below 26: 135.
What was searched, as the harness recorded it.
How this experiment sits among the others is on the lineage page; its family's open hypotheses are on the families page.