The Voynich Ninja

(01-02-2025, 01:18 AM)pfeaster Wrote: You are not allowed to view links. Register or Login to view.

(31-01-2025, 10:47 PM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.The counterpart

Yes, this is an interesting one!  At first I assumed it was an example of a bar running forward from nothing.  My best guess now is that it represents one of those rare cases in which a curvelet [e] has both a forward "bar" and another flourish associated with it, as we sometimes see with [{oh}].  There don't seem to be any definite cases of [{y*}] that couldn't just as well be interpreted as [q*].  So maybe this was an attempt to write [{yh}] in a way that couldn't be confused with something like [qh].

My personal thought from the work I've done so far is that line-start text is not "Voynichese".
I can't really think of a reason for this sort of stuff (below). Well, not in a natural left to right - top to bottom, "I'm just writing words as I go" way. 

(Most of the mid-page highlights are unsure-spacing, few.. as with most Voynich stuff look more valid, yet extremely rare)

(31-01-2025, 08:23 PM)pfeaster Wrote: You are not allowed to view links. Register or Login to view.

(31-01-2025, 08:52 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.In a word, curvlets (e-sequences) typically precede minims (i-sequences). But later in the post there is a careful argument proposing that word spaces convey very little information, so the concept of precedence within words becomes redundant.

I wouldn't say precedence necessarily becomes redundant according to that argument -- even if spacing doesn't convey essential information, it might still reflect a logical structure, as if we were to write "po ta to" broken into CV syllables instead of "potato."

Hi Patrick,
I agree, I expressed myself poorly. Though there is considerable uncertainty about word spaces, they are still informative and show a number of patterns. Also, there are a number of standalone labels that one can use to check ideas on a more reliable (if much smaller) set.

(01-02-2025, 01:30 AM)Bluetoes101 Wrote: You are not allowed to view links. Register or Login to view.3) Does your system only consider individual glyphs? So, all the rules are like "glyph of type X is followed by glyph of type Y" and never "a combination of glyphs AB is followed by a glyph of type Y".

It only checks pairs until you "switch", then it tracks what you switched from and to and if that sequence conforms. 
You can have an extra switch (following a switch, only) which does nothing functional but is allowed, so technically it tracks up to 4 glyphs. 3 switches is considered a non-conformance, so it can never get past 4 and conform.

Ok, it's more complex than I thought. Let's try to see whether we can express this as a grammar.

What is "a switch"? Is this a specific character (like EVA a) or just a point in the sequence where one type of characters changes to another type of characters? 

(01-02-2025, 01:30 AM)Bluetoes101 Wrote: You are not allowed to view links. Register or Login to view.4) Is it possible to fully describe the rules of your system by using a square table where there is one column and one row for each glyph and the table is filled by Y and N denoting whether the column glyph can appear immediately after the row glyph?

I think so. I best think in pictures to be honest, I see it like a simple electronic circuit diagram. 

Currently there are 4 groups. 
3 can move to 1 other place, or stay and do another of them self (to a limit) then be forced to change or end. 
1 can not do another of itself but move 2 ways, or end.

Pictures will do! I love pictures.

Can this whole system be expressed as a directional graph? Something like this?

[attachment=9901]

I think it will be better to just release it once done, otherwise I'm just explaining stuff that might change  
But basically yes, nothing is forced to end atm though other than 1 camp, which would be "i, n, m".

The absolute basics. 
Take CLS rules; 
Change "a" to "a,o,y" (switch)
Change "l" (line) to "K" (line-curve)
Add "l" back and only assign "i, n, m". 
Words now die with "l" but not "k".

The idea is that we now bounce around text, like a language. It is not all "cccalll" for my system but more "akcakall" 

So a very basic example.
"oral"

CLS
"clal"
This breaks CLS rules at "cl" and "la"

My system
">k>k"

">" "switch" from nothing can't fail as we didn't switch from anything on the page.
The system comes into play with "k>k".
"k" represents "\+c" Line-Curve. We then "switch" from "c". At this point (by my systems logic) there was never any choice to use a curve based glyph next because we switched away from a curve.
Hopefully this clarifies why I chose the word "switch". Lights on, lights off, there's no press it and lights on to on or off to off. 

If the person writing wanted to write another letter on the end, no line based glyph is allowed. As "k" starts with a line it is included, not just "i,n,m".

"k" is my group that can move to 2 places as I mentioned before, but it depends how you want to look at it. In a diagram it can now choose "c" or ">", in another way of thinking (the one I pref) because it ends with "c" it has been teleported to "c" already and continues on its available path, more c's or switch, once it switches it can interact with lines again. 


***Subject to change, T&Cs apply!!!  
The constraining of the idea is taking for more time than the idea itself. I hope this clarifies the process though.

(02-02-2025, 12:05 AM)Bluetoes101 Wrote: You are not allowed to view links. Register or Login to view.I think it will be better to just release it once done, otherwise I'm just explaining stuff that might change  
But basically yes, nothing is forced to end atm though other than 1 camp, which would be "i, n, m".

The absolute basics.

Take CLS rules;

I think we better wait till it's final(ish) then.

If it's possible, in the final version could you state all the rules of the system explicitly, without assuming readers are familiar with other previous systems    I think most people with real interest in the MS are familiar with EVA, anything beyond that I guess should be spelled out.

Personally, I have no idea what "CLS rules" are and I'm not sure whether they were well formalized (quite possible they were, I just don't know) and whether there were several versions of them. All I know is that they describe some patterns in which glyphs follow other glyphs based on their shapes. There are dozens of very intriguing models showing various patterns of Voynichese, I don't think it's reasonable to expect people know all of them by heart 

I think that is probably best, I'll probably move it to the text analysis section, give it a new name and link CLS. 

This is where I am up to currently. I am trying to make it more rigid and restrictive. I think I was probably a bit naive in believing the details were the sprint to the finish, but it is getting closer. Making "in" its own thing felt like a good step today. I'm wondering if I need "\" or "EVA: i" at all currently, if most "i" "ii" "iii" roads can be moved elsewhere like "in", I can remove around half this image. 

I'm finally tackling your work Pfeaster,

Some notes I made as I went. 

[*]On the "fachys" debate and ambiguity with ch/a. In my Currier B notes (for when I tackle it), I have wondered if "ci" could be a bench alteration, which would (if squeezed up like this) end up looking like a ch/a hybrid. 

[*]I personally believe "ydaraishy" is "ydaraii(s?)hy" at the bottom of this paragraph. The clue is provided by another thing you touch on, vertical partial repetition.







I'm going to stop here and pick it up tomorrow, but it seems like an absolute treasure trove for my work! So, really thank you very much for the work and alerting me to it. I would comment on much more but I think you come to many of the same thoughts I had as you went anyway. 
 
You seem to be running into all the same oddities with "curve/e" that I have been banging my head against in the part I am in now (almost at section 3). Looking forward to seeing where that goes and reading the rest of your work tomorrow.

It took me a lot longer than I expected to finish @Pfeaster  
Its full of valuable info and lots of ideas from yourself and others that seems to reflect my findings in places. I think I will note them once I am finished and put something out rather than go into here. I definitely need to go back and look at some parts a couple times.

I did find it interesting that you explored the idea of numbers. I have, rather tongue in cheek, said "numbers" on a number of occasions when people say "well explain how XYZ, does YOUR solution answer that!".. I find "numbers." answers all the questions.. but I'm not sure if it is a good answer or just one that is hard to argue against. 

Regarding the repetition and especially "qo", I have wondered for a while.. maybe it is just "+". I think I am right in saying is almost never starts paragraphs or labels and is pretty exclusive to mid-text. 

As in;




If I were to go really off the deep end.. you could imagine the x shape like EVA y/l (if you start the loop from the top, it is exactly EVA: l), then we have end-y/l start-qo x4 in a row, which feels pretty "Voynich'y"
 -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

As for my systems progress I've spent some time considering if "a" and "y" are Curve-Line, which very nicely wraps up how I think they function.. but I seem to back myself into a corner with "oy" and "oa" where I need to say "o" before "a" or "y" is ignored. I've not entertained a rule like this until now and just said "no it doesn't conform". 
On one hand I describe how things function better and restrict the system further, on the other I add an exception.. which seems like a slippery slope. 

I have also thought maybe "o" is a break and all the spaces are meaningless. Its something I plan on trying out anyway.


Edit: Yeah.. my system doesn't work at all between "o"s, or line end to start. Guess it's nice to see it doesn't work for any old Voynich text. Will have to make a choice on how to represent "a,o,y" though..

I have been taking time away from analysis and code to collect the thoughts of others.
Two of these ideas were;
Lisa's hypothesis that EVA: "f" and "p" may be abbreviated "te" and "ke"
Rene's hypothesis that EVA: "cXh" may be "oXe" (X for gallows)

Using Lisa's reasoning that "pe" and "fe" are mostly not found but "te" and "ke" are common is the same reasoning I would apply to state that maybe
benched gallows are "aXe". Maybe it is where all the "ap, af, at, ak" went to.

"a" in the case of "p,f,k,t,d" seems to have been replaced with "e" in non-benched versions but isn't found in other line-curve glyphs r,l,n,j,m.
Using the same logic, the missing "e"'s preceding r,l,n,j,m may be answered with "sh" in that the different loops represent one of r,l,n,j,m.

In my mind, although I know nothing of linguistics, being stuck with doing "ake" for every "ak" and "ere" for every "er" doesn't make much sense.
There I would lend from another of Rene's thoughts that the bench character is likely a single character not two.

So maybe "cXh" is ap,af,ak,at (or ake, ate, ak, at using Lisa's idea) and "sh" is er,el,en,ej,em. Maybe "m" could be "re" and "ere" when part of "sh", but that's just a passing thought. 

Going further down the rabbit hole, maybe "ch" is a combination of "a" and "e" - "æ".

Some (not all) of these types of hypotheses can be checked to some extent.

For example: ch as a unit is quite frequent. If one were to propose that this stands for some PT character or pair, then one can compare the counts, and see if they are in the same ballpark.

It is more difficult when one tries to identify ligatures or substitutions, which I think is what you mean with
"ch" being "ae" (as a hypothesis). However, one can do other things.

With ch, there is the convenient fact that we also have a character written in Eva as "sh".
Whatever we suppose for ch, we should find a valid equivalent for sh.

Another example is Lisa's equivalence of "f" with "te" and "p" with "ke".
This is based on credible statistics, but it still raises questions.

The first is: should "f" be the prime form and "te" an elaborate (verbose) version, or is "te" the standard and "f" a ligature to be preferably (but not uniquely) used in top lines of paragraphs.

The other is: while "f" and "p" are never followed by "e", this is not the case for "te" and "ke" which are frequently followed by another "e". Does this lead to contradictions or to additional tentative conclusions?