Experiment 192
F11.4. Friedman, Military Cryptanalytics III par. 35b (report reports/Scheidt patents and field ciphers.md, issue #9): a keyword's numerical order is used as a columnar key on a rectangle holding the normal alphabet repeated an agreed number of times; the column read becomes the key stream. Here: the standard or KRYPTOS alphabet repeated 1-8 times, every distinct column order of the dictionary and vocabulary words of 2-32 letters, the stream used cyclically from every start, 8 table rules (Vigenere, Beaufort, variant, Bury's reciprocal on standard or KRYPTOS numbering); each setting checks all 24 known letters (compiled). K4 against 10 shuffled copies; plants. A free table (the report's partition test) is not run.
What the method does with the 24 known letters.
Friedman taught a way to make a long key from one word: write the alphabet several times into a grida grid Writing the message into rows of some width and reading it out in another order, column by column for instance, which shuffles the letters without changing them. as wide as the word, then copy the columns out in the word's alphabetical order. That jumbled alphabet is the keythe secret word The word or string a cipher needs besides the method.. Only the word and the number of repeats need remembering. We try every dictionary word.
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.
Friedman taught how to make a long key from one remembered word: write the alphabet several times into a grid as wide as the word, copy the columns out in the word's alphabetical order.
Every column order a dictionary word gives, the standard or KRYPTOS alphabet repeated 1-8 times, every starting letter, eight tables; all 24 known letters must agree.
Nothing fits K4 or any scrambled copy in 2.45 billion settings; planted texts are found. F11.4 closed.
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.
BERLIN, the alphabet four times in rows of 6, columns read in BERLIN's order: AGMSYEKQWCIOUAGMSYBHNTZFLRXDJPVBHNTZEKQWCIOUAGMSYEKQWDJPVBHNTZFLRXDJPVFLRXDJPVBHNTZFLRXCIOUAGMSYEKQWCIOU
Family, key space, the constraint under test, and the clues it reads.
A9, F11: a keyword-only key stream from Friedman's training texts, the kind of rule a CIA cryptographer of Scheidt's generation learned; 'a matrix, then the table' (C1). Model: stream s = columnar_read(alphabet^n, keyword order), k_i = s[(o + i) mod 26n], c = T(p, k). Statistic: fitting settings, K4 against copies.
Written before the run; the result decides it.
No setting fits all 24 known letters on K4 or on the copies; plants are found. Zero candidates.
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.
163,726 column orders; fits K4 0 (copies 0-0); plants [62, 26]
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.