How many pages?

[Koen G]

Yes. Quality matters, but quality is subjective. If the method produces a word salad, it will not be clear to someone who is not proficient in the language(s) involved. This is why I don't comment on Persian, Turkish, etc. theories. Some amount of ambiguity and interpretation must be allowed in translation. All sorts of special pleading can then be invoked to allow almost anything: the author used words from several languages, words have a lot of undocumented variants, grammar is bad, etc. All possible, but where is the limit?

I think the criteria I described account for this situation. A translation of Voynichese consists of two big steps: convert Voynichese to plaintext, and then translate the resulting plaintext to something a modern audience can understand (i.e. modern English). We (you and I) may not know any medieval Russian for example, but if someone were to propose a Russian theory, he could still explain to us how to convert Voynichese to the plaintext. You and I will then be unqualified to translate this plaintext to modern English, but we should be able to assess if Voynichese can be converted to a linguistically feasable plaintext using the proposed method.

What many proposed translations do is to circumvent part of this process by squeezing Voynichese towards existing vocabulary making use of the dreaded interpretative step. This is why their method is hard to explain to others and near impossible to repeat, and even more impossible to expand.

To put it differently: if your method actually works, I should be able to use it to convert Voynichese to a real language. Whether or not I am well versed in this language should not matter.