+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Analysis of the text (https://www.voynich.ninja/forum-41.html)
+--- Thread: A sieve cipher based on f66r (/thread-3963.html)
A sieve cipher based on f66r - Arichichi - 13-03-2023
Undoubtedly, so far, I find You are not allowed to view links. Register or Login to view. the most intriguing page. I think that if there is a substitution cipher that works, it has to be accounted for by the page f66r.
Just to shed some light on oddly working substitution cipher: Arabic readers, most of them, know that when the Arabic script was used to record the Quran at the first time, it didn't have diacritics. So, many letters were represented with the same character. Thus, b, t, thorn th, y were all represented the same, d, the th, were represented the same, and so on. So it's not impossible to find a substitution cipher that attributes many sounds to the same character, therefore attaching the rest of EVA, to perhaps, vowel signs and/or liturgical markings.
One intriguing aspect, is how You are not allowed to view links. Register or Login to view. tends to display, repeatedly, EVA-d symbols that are bolder than usual, as well as some gallows. Could that be a sieve to select letters on other pages, much like a masking aid to decode other pages?
RE: A sieve cipher based on f66r - stembyday - 13-03-2023
This page is interesting. I see 3 columns: Left column of “words”, middle column of “characters”, and right column of “sentences”.
The rows don’t line up though. (I think) 15 rows on the left, 34 in the middle, 32 on the right. And visually no two columns, let alone all three, appear to correlate nicely with one another.
Maybe the right column is just prose. And the two left columns are enumerated lists that don’t necessary line up horizontally.
RE: A sieve cipher based on f66r - Arichichi - 14-03-2023
Maybe the first column describes the categories of sounds, the second presents the characters and the last shows how you are supposed to sound the characters...
RE: A sieve cipher based on f66r - davidjackson - 14-03-2023
It's true that the early Arabic script, known as the Kufic script, did not use diacritics or vowel markings when it was first used to record the Quran. This means that several letters were indeed represented by the same character, and readers had to rely on context and their knowledge of the language to understand the text correctly. This script was eventually modified with the addition of diacritical marks and vowel signs to reduce ambiguity and make the text easier to read and understand.
In the context of a substitution cipher, it is theoretically possible to create a cipher that assigns multiple sounds to a single character, as you've mentioned. This would make the cipher more challenging to decode because the usual frequency analysis used to crack substitution ciphers would be less effective. However, it would also make the encoded message more difficult to read and understand, even for someone who knows the cipher key, because of the inherent ambiguity in the text.
The addition of other symbols, like vowel signs or liturgical markings, could provide additional layers of complexity and obfuscation to the cipher. However, it would also increase the difficulty of encoding and decoding the message, as the system would be more complex and require a deeper understanding of the language and its nuances.
So, while it is possible to create a substitution cipher that mimics the early Arabic script's lack of diacritics and assigns multiple sounds to a single character, doing so would create a more challenging cipher to decode but also make the encoded message more difficult to read and understand.
RE: A sieve cipher based on f66r - Arichichi - 15-03-2023
It can be easier to write, though, if close sounds are substituted with the same symbol like b/p/m or d/t/th or l/n/r and so on.
RE: A sieve cipher based on f66r - zobowiazanie - 15-03-2023
The more I study the Voynich manuscript the more I believe that without a giant dictionary of Voynichese we aren't ever going to be able to read it.
RE: A sieve cipher based on f66r - Anton - 28-03-2023
This is not the only page in the VMS with kind of "index". I think a prospective way of looking at it might be trying to bring forward a reasonable use case that would require introducing some kind of index, to begin with. Then think over the patterns observed in the index, this could narrow the scope.
In this case it is evidently not as simple as plain enumeration/ordered list. Perhaps some type of table content. The number of items is fifteen. What is fifteen? Some characters in the middle column are very rare - why would that be? Next, all are single, bar air. Does that mean that air is to be treated as a sigle glyph elsewhere? And so on...
(To make it clear, I have no ready idea to propose).
RE: A sieve cipher based on f66r - Koen G - 29-03-2023
(28-03-2023, 09:56 PM)Anton Wrote: You are not allowed to view links. Register or Login to view.Does that mean that air is to be treated as a sigle glyph elsewhere? And so on...
I wrote about these lists once: You are not allowed to view links. Register or Login to view.
Basically whenever Voynichese needs single glyphs, it's always the same suspects: s, y, l, o, r, d and gallows. And then there are a number of less frequent items, like rare glyphs and this "air". Like I wrote before, if our assumptions that these are single glyphs is true, then it means that EVA-a is like a first minim, part of a stroke sequence to form larger glyphs.
Another way of looking at it is to see which glyphs are not popular in these lists (and as standalone glyphs in the main text). If EVA-a were a full-blown glyph, it would appear in isolation much more often, and the same goes for eva-n.
Another hypothesis is of course that EVA-a does not appear here because it gets another shape when it's not part of an i-group, like the way a and y are complementary distributed.