(6 hours ago)ololololo Wrote: You are not allowed to view links. Register or Login to view.Damn graphics... 
However, for users, it will be enough to explain the following:
1). What will I get if I don’t turn the text around?
2). Why does Voynichese look so good, given that it’s flipped (I mean rules like CLS)?
Let me be the one who makes your theory clearer to others 
In the forward-writing system:
When writing the “sh” in “shedy,” the writer would already have to know that the next word begins with 51 % = “qo.” That is what is peculiar: a glyph in a word “knows” something about its successor. That is strange—a kind of long-range effect.
Words have such connections; the best-known example in English is probably “neither … nor.” Here, when you write “neither,” you know that “nor” is coming. But a glyph within a word shouldn’t have any knowledge of which glyph the next word begins with. And unlike a word-to-word relationship, this strange remote effect between “sh” and “qo” exists in most of all tokens (as I’ve shown in the posts above).
When reading backward:
If you reverse the direction of reading, the dependency is reversed. The word with “qo” now comes before the word with “sh.” The writer has therefore already written “qo” by the time he reaches the “sh.” Consequently, he doesn’t need to know anything in advance; he simply has to look back. This is the normal situation for any writer: what I write next depends on what’s already there. That’s how language works.
The regularity is the same in both directions. What changes is what one must assume in order for it to arise. Moving forward, one needs a writer who, upon encountering the first letter of a word, knows the next word. Moving backward, one needs a writer who remembers what they just wrote. The second model is therefore simply the easier one.
But that’s not the only effect that becomes simpler.
For example, there are the 7 rules that explain/define almost all the spaces in the test. If you define the system as a cipher—since that’s rather unusual for a language—then in the Reva version, these become the first letters of the tokens, and they serve, so to speak, as an entry point into a tabular format that builds logically, step by step, through the entire VMS token. When read forward, this is not nearly as simple, logical, or feasible to follow step by step.
Or, for example, the “qo.” “qo” almost always appears together—in nearly 99% of cases. So when read from the front, it is always a single unit—an atom.
When read backward, many words end in “o” and “oq”—here, the “q” is simply a possible extension of the “o” ending, and thus no longer a fixed unit. This, too, is the simpler interpretation.
These are the reasons why reading it backward “simplifies” some of the VMS statistics. This is NOT proof, without a doubt. But it is a clue.
