A while back, I proposed this idea that symbol phrases were written as meandering routes in the z340. To compose the phrases, each symbol could be rotated to find a literal representation of a letter. For example, you can see below 2 of these phrases hypothetically self-describing the encoding of the z340:
I wanted to see if the z408 exhibited the same behaviour.
It does. Here are 2 of the routes that can be found in the z408:
Here are the transcribed phrases:
The first phrase "Cipher route tip" seem to indicate there is a tip written as a cipher route. The second phrase could possibly be that tip, hypothetically naming the author of the cipher.
I find it really hard to imagine that in both the z408 and the z340 the same type of meandering routes self-describing its respective cipher occur by chance alone. My impression is that in the z408, the author used this technique to hide some information in the cipher in form of routes; in the z340, these phrases I believe ARE the message and, combined, make up a null cipher.
_pi
As for "cipher route tip" or "looped null cipher", one could write a program to look for these phrases in shuffles of the Z340. Though the results may be that they are unlikely, because the phrases you found are probably "local". Perhaps the same could be said for period 19 bigrams!
My thoughts are that these phrases are spurious, but I see no way of knowing for sure.
Though in the Z408, if any of these phrases you found have a symbol that is involved in a proper cycle then I would deem it as highly unlikely.
A short 3rd route…
Transcribed here…
Seem to refer to the first 2.
_pi
As for "cipher route tip" or "looped null cipher", one could write a program to look for these phrases in shuffles of the Z340. Though the results may be that they are unlikely, because the phrases you found are probably "local". Perhaps the same could be said for period 19 bigrams!
My thoughts are that these phrases are spurious, but I see no way of knowing for sure.
I agree with this assessment.
If a deciphering method produces an astronomically high number of possibilities (which this one does), then the chances are very good for finding phrases with high specificity.
I studied these kinds of searches previously here:
http://www.zodiackillerciphers.com/?p=370
http://www.zodiackillerciphers.com/?p=420
It would be fun to run another search based on samples from corpora (Project Gutenberg, reddit comments, etc), and try to find the longest coherent phrases. It wouldn’t get us any closer to a solution, in my opinion, but it’d be a fun exercise.
