Calgary engineer believes he's cracked the mysterious Voynich Manuscript

[Ahmet Ardıç]

Ahmet Ardıç, the purpose of EVA is described You are not allowed to view links. Register or Login to view., section "Purpose and Terminology" in particular. Also it is worth to note that recently EVA was redefined as being transliteration, not transcription.

Transliteration converts handwritten words into symbolic form, suitable for processing. The key property of the transliteration is that the conversion is one-to-one. Given word in the manuscript you can unambiguously write its machine-friendly representation and then process it or use in interchange with colleagues. Conversely, given the symbolic form you can learn how it is written in the manuscript. This is clearly technical feature and says nothing about underlying language. Mapping some curve to symbol k doesn't mean the curve should be read as "k".
EVA meets all these conditions*. If I understand correctly, it was originally named European because the symbolic form it produces use Latin letters. You can use another set of symbols (say Cyrillic) and create another transliteration (say CVA-даиин for EVA-daiin), but in terms of processing they will be equivalent or isomorphic.

*ok ok, there are some minor inconveniences with EVA (s/sh-issue, rare symbols).

Please put side by side together that transcriptions suggested by EVA and ATA, and look them again? Which do you think is more likely to help us?

However, despite equivalency EVA is obviously better and more helpful than any other existing transliterations, but this is offtopic.

Transcription is conversion of handwritten words into written form of some language. Both language and alphabet should be known in advance. I suppose ATA is transcription (not following the discussion properly, sorry), so it cannot be directly compared to EVA. But there should be mechanism to convert EVA to ATA then.

With EVA, you read multi-syllable characters as alphabet signs in VM texts usually. What kind of overlap do you expect in this case? With using EVA system, read the any sign with any chosen alphabet sign sound value you want, the result does not change. In other words, most of the syllable signs and multy-syllable-signs are words on their own and there is basically no usefull structure that can correspond to this in your EVA.

With using EVA, you can map a sign to different alphabet characters or a syllable but you can't read a multy syllable single-sign.

You have made the same mistake many times when analyzing the statistical structure of words. Even many combined sign in VM not be a single letter or single syllable, but they are independent word with three or four syllables in one sign. This is exactly why EVA didn't work, and it won't. Change the sounds in endless variants as you wish, it doesn't matter. If multiple sounds corresponding to a word are represented by a single sign, you cannot read it with a system such as EVA. So this kind of the details are fixed with ATA.