Philip Neal wrote:The regular expression is something like
^[qd_][aoy_][lr_][ktpfKTPF_][CS_][eE_][d][ao_][lrmn_][y_]
(C =ch; S =sh; E =ee; KTPF = the complex gallows)
...
In other words, you chose any one character from each set in square brackets and rewrite the_ as zero, for instance qo_k_Ed__y -> qokeedy. The null character _ can occur anywhere....
If the null character can appear anywhere, should the regex have [d_] instead of [d]?
As it is written, the regex cannot generate Voynichese words that don't contain "d" (actually, the majority of the words).
If I understand correctly, the regex also misses the words in which "ch" or "sh" occur directly or indirectly before one of the gallows (about 20% of the words).
It would be interesting to see how much of Voynichese can be captured with such a simple expression....
In general, I think the problem of defining word structure (similarly to Philip Neal's regex) is contiguous but not identical to the identification of "character classes". For instance, in Latin the phonetically similar "n" and "m" have very different positional statistics. The two letters have roughly the same number of occurrences, but "n" appears as the last letter in about 1% of the words that contain it, while "m" appears as the last letter in about 50% of the words with at least an "m".