The Voynich Ninja

Voynich text as Base20
EMS suggested to report failed experiments.
After 5 days of work, this fell on its face, so here it is.

Decoding vms text as Base20 ran into a leading zeroes problem.

Idea as follows:
:: Decode text as Base20
-eva has 26 chars (TT only uses 22-Fail 1)
-assume 6 lowest frequency eva chars represent something else, like planets, constellations, stars, whatever
-leaves 20 letters, map those letters onto the 'symbols' for a Base20 number system (vigesimal)
-convert newly created Base20 num into a decimal number
    Now you have a nice set of decimal numbers that you can do what you want with ( indices in a look-up table == unnecessary-Fail 2).


:: Encode text as Base20
-take some normal text, strip puntuation etc
-allocate each unique word a random number in decimal ( this step is for easy computation (and i had an indexing idea but...)
-convert that number to base20
-convert the base20 number into eva
  ie map Base20 symbols onto the commonest 20 eva symbols

:: Result
Courtesy of project gutenberg : A Scandal in Bohemia by A.conan-doyle

plaintext:
to sherlock holmes she is always the woman i have
seldom heard him mention her under any other name in
his eyes she eclipses and predominates the whole of her
sex it was not that he felt any emotion akin
to love for irene adler all emotions and that one
particularly were abhorrent to his cold precise but admirably balanced
mind he was i take it the most perfect reasoning
and observing machine that the world has seen but as

encoded:
ap oqsgcki oinflyx oie ook frn aled aktkcr esk hhy
adq as rypp tdkg triqya oe hod i omm hy
ha iednsk oie yc olxc odt aled dshg oq triqya
e yets yoci ar ce fm qm hod ydglhy parfgm
ap dhtn ony aye ocr ykl ond olxc ce nkaspk
oysf osc tok ap ha hfcn ya oqc hr gnl
hm fm yoci esk yf yets aled hr eoyh n
olxc rmi hkii ce aled he ohtpxhe hg oqc ae

Conclusion::
idk. 

The leading zeroes problem occurs when mapping letters to digits, but this can be circumvented by not using 0 to 9 but 1 to T where T stands for 10.
So you get:  6 , 7 , 8 , 9 , T , 11 , 12
and :  98 , 99 , 9T , T1 , T2
and :  T9 , TT , 111
This is of course a rather cumbersome system.

However, you are increasing entropy while converting the plaintext, so this won't work.

Do I understand correctly that the main drive behind this approach is to reduce the character set to 20? And that there is nothing inherent to the mechanism that would create the positional rigidity we see in Voynichese? 

I think if there is any converting words to numbers involved, the solution is probably closer to Roman numerals.

Regarding the entropy, yes, i didn't / don't understand it well and had not properly taken it into account - lesson learnt.

The main drive was to create numbers,

Base20 is just more common than >Base20 and with 20 chars there are only 6 chars to ignore.

Regarding the positional rigidity, i was naively ( lazily? ) hoping some would pop out when converting.

I took a 'do it and see philosophy' rather than a 'think about it first' approach.   More thinking required in future  

Thank you for your comments.

(31-03-2022, 07:12 PM)RobGea Wrote: You are not allowed to view links. Register or Login to view.I took a 'do it and see philosophy' rather than a 'think about it first' approach.   More thinking required in future  

There is nothing wrong with that of course, it is quite possible that the eventual solution to the VM will be born out of such an experiment.  And I do appreciate the intentional sharing of failed experiments, you never know if it might inspire someone else.

(31-03-2022, 07:12 PM)RobGea Wrote: You are not allowed to view links. Register or Login to view.Base20 is just more common than >Base20 and with 20 chars there are only 6 chars to ignore.

You can get an alphabet with even less than 20 chars, first assign 1 char to 1 letter, when all chars are assigned do the same with groups of 2 chars along with a way to distinguish when a char form a letter by itself and when it takes part of a group of 2 that conform a letter.

Rob, are you sure you didn't post this a day early?