8 hours ago
(This post was last modified: 8 hours ago by npcompl33t.)
(9 hours ago)Jorge_Stolfi wrote: You are not allowed to view links. Register or Login to view.What do you mean by "differently" and "same position"?
Yes, sorry, for position, i meant in the same paragraph initial line that seems to have a different register. By differently, i mean the js divergence between the words. So from an information level it is measuring how different the words are from one another. Its just pairwise JS distance of the transition matrices / markov representation of the words.
So my issue is that the transitions of P words don't seem to resemble the transitions of the other gallows words at all.
It sounds like what you are saying, is the overall statistics are different because some classes of words are less desirable targets for puffing, and the desirable targets are mostly depleted, causing the mismatch?
(9 hours ago)Jorge_Stolfi wrote: You are not allowed to view links. Register or Login to view.Assuming that a puff can possibly replace some of the ke/te as well as some of the k/t, the glyph frequencies after a puff should not resemble either those of k/t or those of ke/te. But the frequencies of the glyphs before a puff (excluding the line-initial puffs) should resemble those of k/t, whch should resemble those of ke/te. However, in Herbal-A, Herbal-B, and Stars, the words at the beginning of a paragraph are expected to be "special" -- which may mess up the glyph statistics.
If there is some subset of desireable targets, we should be able to isolate that group and see that the JS divergence of the transition matrices are similar -- that would tell us that they are capable of forming roughly the same types of words at the same frequencies, and help confirm a match. If we can't find one -- then that would help us identify which type of puffing is occurring. Would you agree with that?
Quote:I have a strong feeling -- but not certainty yet -- that qo- means "and", and is used as "wa-" in Arabic, "ve-" in Hebrew, and "we-" in Ge'ez. That is, an enumeration like English "A,B, and C" in Voynchese would be "A qoB qoC".
in this case, would that restrict the amount of words that it could be applied to over regular gallows?
The above chart seems to show that words that can have an o prefix act different from words that do not. And labels mostly carry o prefixes.
I would love your opinion on this. These are the first lines of You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view. -- these would be adjacent interior sheets on the same bifolio. Look at this, the words are very similar.,
f20v: faiis,ar.okoy.shy,pofochey.opchy.qopy.choldy.opydy.cphy
f21r: pchor.o,eeockhy.o.fychey.ypchey.qopcheody.otaiin.chan
Look at the o's , the spaces, and the gallows here though.
The words are very similar, and the places where one is missing a gallows, the other will have a space. And the middle words are nearly identical, but the o's have turned into y's. , or ol->eo
Do you think this could be relevant? Or is it just coincidence? They are adjacent pages, so maybe it is some kind of a shared template, but those are odd parts to chage, right? In the bottom case, it leaves a very odd voynich word, eeockhy
Quote:Whereas I have no good idea of what a prefix o- means. One conjecture is that [aoy] denote pitch levels, which would explain why they seem to be so "fluid". But they could also be letters, and then "okedy" and "kedy" would probably be unrelated words.
You mean for example, if Stars was a part of a larger section that was organized, and maybe the two stars sections i've identified are two different pieces of a greater missing pool?