Which other manuscript contain the most Voynichese glyphs?

[crezac]

I'm going to say the same thing I've said many times on other threads and sites...

I don't think there's any doubt that the source for the MAJORITY of the Voynich glyph-shapes is medieval Latin letters, numbers, and abbreviations. A few of them show Greek influence, but this is not out-of-place if you consider documents written in Latin characters (in a variety of languages) frequently included loanwords written in Greek, glosses in Greek, and also pages with the Greek alphabet written out. But the majority of the VMS glyph-shapes are Latin.

I really have no doubt about this. Why? Because they are not only the same shapes, but many of them are in the SAME POSITIONS in the Voynich tokens as we find them in Latin (in particular, the ones based on abbreviation shapes).

A small subset, mainly "gallows" glyphs, look like combinations of Latin ligatures and abbreviations and after an exhaustive search (10 years) of almost every alphabet on the planet, I suspect they are invented and loosely based on Latin and Greek scribal conventions, rather than borrowed from some esoteric source.

When I use the term linguistic isolation I don't mean to suggest that the author was unaware of languages besides his own.  I'm not even suggesting that the manuscript encodes an unknown language.  Whatever is encoded can't be read because the encoding isn't understood.  The "alphabet" being used may have similarities with known alphabets, but not have intended functional overlaps.  Some of the similarities could be the result of convergent evolution too.  I'd refer you to the Cherokee syllabry which borrows the shapes of many Latin/English alphabet letters for some phonemes -- and not always the ones the letters encode in English -- and syllables, and has others that were invented for Cherokee (some of which encode phonemes that already have an associated letter in English's alphabet).  Some of these choices probably came about because Sequoyah liked the way they looked.  I won't disagree that many Voynich glyphs appear to have similarities to Latin characters.  But that's not proof that the glyphs originated as an appropriation of anything Latin or that they will follow the same rules.  

By linguistic isolation I mean we have no linguistic history for the development of these glyphs the way we do for Cherokee.  We don't know what sounds they represent or even if they encode phonemes, syllables, or full words.  We can base some of our guesses about how the script works  upon what information the script has to encode to be functional (nouns, probably adjectives and verbs, maybe more).  We can base some more possibilities on what the script needs to be efficient.  That still leaves multiple possible strategies for implementation at various decision points; so no clear path for interpretation and no practical framework to develop a tool set to help with that either.

Whether it's encoded Latin, the personal shorthand of a small guild from the hinterlands, or a devilishly clever fraud to make people waste time and come up with novel theories, I'm just suggesting that you can't filter out the static if you don't know what part of the signal carries the message.  I find it equally probable that the alphabet is the result of one individual's efforts as I do that it was an alphabet that was used and developed by a large group but lost for one reason or another.  We need a better reason for assigning probabilities that "this is what I feel".  Your work with pattern recognition in the manuscript gives better results that a "my gut tells me" approach (although I do find the abbreviations very compelling).  But I don't see a path from "most of this look like Latin to me" to "I can read this" or even "I can give you some idea of who the author of this manuscript was or at least where he studied".

That's not intended as a criticism of anyone's efforts, just an explanation of my point of view. And I'm not disputing your claims about the Latin origins of the script, because for the purposes of my argument it doesn't matter how the alphabet might have originated if you can't read the message. When we can read it, we'll likely have a linguistic context that will mean we don't need to speculate or to assign probabilities to guesswork.  I'm just saying that if you approach the problem wrong it can become much more difficult to solve -- and decades of effort to date indicate it's not easy now.