+- 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: Meta EVA (/thread-2123.html)
Meta EVA - Emma May Smith - 27-09-2017
I wonder how useful it would be to have an agreed set of meta characters as part of EVA.
I have seen others use capital letters to stand for groups of characters which are similar in some ways. For example [E] can be used to stand for [e, ee, eee], ignoring exactly how many [e] characters are in a row. So [okEdy] could stand for [okedy, okeedy, okeeedy]. This is useful in talking about more general and abstract word structure.
Some proposals for meta EVA:
[B]: any bench [ch, sh]
[E]: any [e] sequence [e, ee, eee]
[G]: any non-bench gallows [k, t, f, p]
[I]: any [i] sequence [i, ii, iii]
[N]: any sequence ending [n] with any number of [i], so [n, in, iin, iiin]
[R]: any sequence ending [r] with any number of [i], so [r, ir, iir]
Thoughts? Would it be useful?
RE: Meta EVA - MarcoP - 28-09-2017
Hi Emma,
I guess that something similar is necessary if one wants to statistically analyze the text. In particular, the obvious digraphs (e.g. ch) should be mapped into a single character. Of course, there are several ways of doing this and I don't know how to choose the optimal one (I think that someone suggested on the forum that maximizing entropy could be a possible criterion).
Anyway, what you are proposing here is something different. The term "meta" seems to suggest that you mean it as terminology to exchange ideas rather than substitutions for data processing. If, so the extended descriptions you propose seem clear and functional to me.
"any sequence ending [n] with any number of [i]" is something I can understand
On the other hand, there are transcriptions that use N for in and M for iin, so I am afraid I might be confused by [N].
RE: Meta EVA - ReneZ - 28-09-2017
I think that this is a very useful 'tool' for analysing the text, but that they can usually be defined 'ad hoc' depending on the case being analysed. What you did in your most recent blog post is a good and clear example.
RE: Meta EVA - davidjackson - 28-09-2017
Quote:Thoughts? Would it be useful?
It would be useful for human discussion. Actually tranliterating it into core EVA would mess up the computers
There is also the feat that widespread implementation of such a feature would cause users to start skipping over the actual glyphs and start considering the meta characters, possibly causing them to miss crucial details.
I think Rene is quite correct when he says that ad hoc implementation is probably best, when the terms are clearly defined on a case by case basis depending upon the authors intention; rather than being a core feature of the transcription alphabet.
RE: Meta EVA - Searcher - 28-09-2017
Hi!
Maybe, for both, computer statistics and human analysis, we need a few variants of EVA (EVA A, EVA B, EVA C and so on). I can't mention all my and others' suppositions, but will place here some of them. I think, these possibilities may increase the word entropy and change the other properties of the VMs text.
1) We have combinations ckh, cth, cph and cfh in EVA as three characters, indeed, they can be ligatures of three or two letters. It is the first thing that our logic suggests us. We are inclined to set similar values to similar forms, characters psychologically. I'm not an exception. But, from the other side, I don't see any reason not to think that those combinations can substitute only one, quite another letter. For example, t is b, cth - m, k - c, ckh - f, etc. In this case, we will get + 4 letters/characters in the text.
2) As well, I have a version that value of a character can change depending on the place in a word, as 9 in Latin abbreviations, but with single letters. Let us suppose that k, t, p, f stand for certain letters at the beginning of a word, but, even if one character stands before any of those glyphs, it changes its value. This also can add letters of the underlying text.
3) i, ii, iii, e, ee, eee -sequences both may stand for a single, but particular letter, in the same time, they can mean three different letters (for example, e - e, ee - i, eee - y).
4) so, i, ii, iii-sequences can substitute one single letter or three different letters for every case. But my favourite idea exactly for this group is that n, in, iin, iiin, ir, iir, iiir, etc, all represent different suffixes/endings without separate meaning of the character, besides the cases, when they are in the middle or the beginning of a word, i. e., are not suffixes/ends.
5) q can be a null-character.
It is difficult for a human to go over all these options, but easier - for machine. Did someone ever tried to do something similar? I mean computer statistics of the VMs text with changed conditions.
RE: Meta EVA - -JKP- - 29-09-2017
I'm trying to think this out in terms of human searches and machine searches.
When you use languages and utilities that are good for parsing text, one can simply use "and" "or" and other qualifiers to include or exclude, and since these can be saved, the searches need only be defined once.
For those who don't have a programming background, Emma's suggestion might be useful. A modified version could still be searched with regular search tools that have a few wild card options, without writing any code, and generating those optional files is trivial (for someone with parsing utilities or coding skills).