The Voynich Ninja

(22-06-2018, 01:59 PM)bi3mw Wrote: You are not allowed to view links. Register or Login to view.I mostly use Bash scripts for searches. Finding strings across word boundaries is possible, but it's difficult to test the hits against the percentage ratio inside / outside the word boundaries. I will think about it.

Uhm, re-reading what I wrote, I think I have been particularly obscure. I guess my reference to a percentage shift was unnecessarily confusing.

I meant that you are now matching for perfect position alignment.
Comparing these two single-line pages:

Page 1: x a b f x x
Page 2: e a f e y y


If I understood correctly, you currently match the two words in position 2 (the two 'a's).
You could also allow the two 'f's to match, since their position difference is just 1. Just an idea, anyway. There always are so many interesting options to investigate!

I have run some more experiments limited to exact matches between You are not allowed to view links. Register or Login to view. and f84v.

In the basic comparison, I find the same 9 matches as bi3mw.

I have run 100 tests, randomly changing the order of f84v. Stats for the number of matches:
min:0 max:10 average:3.98
This confirms that 9 matches in the basic comparison is high, but it also says that it could still be casual.

I also tried adding a single word at the beginning of each of the two files, in order to compare words with a 1-position shift in both directions.
With 1 word added to f84r, I find 4 matches:


With 1 word added to f84v, I find 10 matches:

These two results are inconclusive: 4 is the average number resulting from a random sort, but 10 is the absolute maximum I have observed.