The Voynich Ninja
[Article] The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - Printable Version

+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: News (https://www.voynich.ninja/forum-25.html)
+--- Thread: [Article] The Voynich manuscript decoding method proposal by Maria Rita Lunazzi (/thread-3673.html)

Pages: 1 2 3


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - MarcoP - 19-10-2021

(16-10-2021, 08:07 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.What worries me is that the system is complex and ambiguous. Also I am not sure if it is possible to decrease the entropy while compressing text.

Hi Nablator,
personally I find your objection totally conclusive; BTW it can be applied to all the "solutions" that keep popping out. The great news is that these delusional efforts have not been published on widespread media as it happened in the past.

Checking character conditional entropy on the Bonaventura passage I transcribed You are not allowed to view links. Register or Login to view., I find the following values:
unabbreviated: 3.13
abbreviated: 3.50

This is in line with You are not allowed to view links. Register or Login to view.'s results:

Quote:abbreviated Latin has higher conditional character entropy than unabbreviated Latin

From their Figure 14, I would say they get a 3.2 measure for the unabbreviated text vs 3.3 for the abbreviated version.

Entropy tells us that Voynichese is not a simple substitution cipher of Latin (or Persian, Old English, Turkish etc), but it is even clearer that it is not abbreviated Latin.

The Latin produced by Maria basically has the same character conditional entropy as the corresponding EVA:
Maria: 1.98
EVA: 1.81

The passage is short and this affects entropy. For a true Latin text of the same length, entropy is close to 2.7.

Maria gains her small entropy increase by reading character sequences in several different ways (e.g. EVA:ar is read "arte", "ati", "atres", "quatre", "autem", "artes" - each occurrence gets its own individual reading).


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - Maria Rita Lunazzi - 19-10-2021

The problem is that its translate in Italian even if don't want to ?


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - Maria Rita Lunazzi - 19-10-2021

[quote="MarcoP" pid='47642' dateline='1634660319']
[quote="nablator" pid='47608' dateline='1634411226']
Quello che mi preoccupa è che il sistema è complesso e ambiguo. Inoltre, non sono sicuro che sia possibile ridurre l'entropia durante la compressione del testo.
[/Citazione]

Ciao Nablator,
personalmente trovo la tua obiezione del tutto conclusiva; A proposito, può essere applicato a tutte le "soluzioni" che conosce a saltar fuori. La grande notizia è che questi sforzi deliranti non sono stati pubblicati sui media diffusi come è successo in passato.

Verificando l'entropia condizionale del carattere sul brano di Bonaventura che ho You are not allowed to view links. Register or Login to view. trascritto , trovo i seguenti valori:
non abbreviato: 3.13
abbreviato: 3.50

Questo è in linea con i risultati di You are not allowed to view links. Register or Login to view. :

[quote]il latino abbreviato ha un'entropia del carattere condizionale maggiore rispetto al latino non abbreviato[/quote]

Dalla loro Figura 14, direi che non disponibile misura 3.2 per il testo abbreviato rispetto a 3.3 per la versione abbreviata.

L'entropia ci dice che il Voynichese non è un semplice cifrario a del latino (o persiano, inglese antico, turco ecc.), ma è ancora più chiaro che non è abbreviato in latino.

Il latino prodotto da Maria ha sostanzialmente lo stesso carattere di entropia condizionata del corrispondente EVA:
Maria: 1.98
EVA: 1.81

Il passaggio è breve e questo incide sull'entropia. Per un vero testo latino della stessa lunghezza, l'entropia è vicina a 2,7.

Maria ottiene il suo piccolo aumento di entropia leggendo sequenze di caratteri in diversi modi (es. EVA:ar si legge "arte", "ati", "atres", "quatre", "autem", "artes" - ogni occorrenza ha il suo lettura individuale).
[/Citazione]

EVA standardizza troppo i glifi manoscritti, durante il mio studio ho provato ad usare i glifi EVA, ho dovuto abbandonarlo perché non si adattava a tutte le variazioni dei tratti. Queste variazioni sono molto importanti per l'esatta determinazione della lettera giusta. 

Infatti, se ho notato nell'appendice del mio lavoro non ho replicato l'EVA, avevo messo i miei.


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - Maria Rita Lunazzi - 19-10-2021

io write in english, but automatically its converted in Italian ?


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - davidjackson - 19-10-2021

(19-10-2021, 05:45 PM)Maria Rita Lunazzi Wrote: You are not allowed to view links. Register or Login to view.io write in english, but automatically its converted in Italian ?
Hi Maria
If you are using Google translate, just click the copy button at the bottom of the translation window and then paste the text in this window.
If you are using an automatic website translator, you should switch it off for this site otherwise it will be changing your replies.

I'm afraid I don't entirely understand your usage of the term "brachygraphy", when you say it is stenography + latin abbrv.

Brachygraphy is a particular dead end of short-hand, which had a brief popularity in the late 16th century and early 17th century, and was developed by a London schoolmaster called Peter Bales. Another popular system, used for a short while by the Royal Court, was Characterie - there is, for example, a reference to it in Shakespeare's Merry Wives of Windsor. There are others. Brachygraphy and Characterie were developed by some other authors, but the dictionary used was unwieldy and newer systems, usually based upon phonetics (stenography), supplanted it. Both systems were based upon the same list of 536 key words. (You combined the "key words" to copy out the whole word, they were common start and endings of words).

It is, in brief, a "proprietary" short-hand system developed by one man which had a brief popularity, and saw several other books develop upon it, but which was based upon a dictionary lookup system and was quickly supplanted by true stenography, based upon phonetics. See, ie, "‘Swifte and Secrete Writing’ in Seventeenth-Century England, and Samuel Shelton’s Brachygraphy" by You are not allowed to view links. Register or Login to view. for more upon the subject.

Latin abbreviations, on the other hand, are a way to quickly compress common Latin words and phrases to speed up handwriting. They are indeed a type of short-hand, but are mixed in with standard text. They pre-date any reasonable attempt at short-hand, especially the ones that were popular around the 17th century onwards. Medieval L.A. are often codified and well known, and indeed many VM glyphs resemble some of them.


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - nablator - 20-10-2021

(19-10-2021, 05:45 PM)Maria Rita Lunazzi Wrote: You are not allowed to view links. Register or Login to view.io write in english, but automatically its converted in Italian ?

There is a setting to disable it in Chrome. By default it is enabled. You are not allowed to view links. Register or Login to view.


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - nablator - 20-10-2021

(19-10-2021, 03:58 PM)Maria Rita Lunazzi Wrote: You are not allowed to view links. Register or Login to view.Inoltre, nell'alfabeto EVA i glifi sono troppo standardizzati.

Per es. Non tiene conto delle varie curvature dei segni diacritici

That's true, but it could be fixed in a better transliteration if we had consistent characteristics to identify. Instead we have a continuum. In your own writing I don't see the differences (=) between those you interpret differently:

You are not allowed to view links. Register or Login to view.


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - Maria Rita Lunazzi - 20-10-2021

Sorry but I'm a boomer ?


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - nablator - 22-10-2021

I've read heavily abbreviated Latin, but this is on another level.
Undecided Confused Rolleyes

You are not allowed to view links. Register or Login to view.


RE: The Voynich manuscript decoding method proposal by Maria Rita Lunazzi - LisaFaginDavis - 22-10-2021

Indeed, Nablator. I'm sorry Maria, but you really can't use Cappelli to read this manuscript, even on f. 116v. Your interpretations of the letterforms are incorrect, and so your abbreviation expansions are also incorrect.