How easy is it to create a cipher which is very hard to break?

[Mark Knowles]

I wasn't talking theoretically, Mark.

Since you say it should be relatively easy, I would like to see an example. I'm sure the others would too. Create four or five pages of text that have the same properties as the VMS text.

I would not suggest taking whole tokens (similar to those in the VMS) and simply repeating them because if you do it that way, it does nothing to demonstrate that VMS "filler" text might be meaningless. It would have to be constructed out of "components" (however you wish to define those) that have Voynichese properties in order to fulfill the assertion that filler would be easy to create.

You might not have been talking theoretically that doesn't mean that I wasn't.

I don't like to be dictated to. I have no intention  of producing 4 or 5 pages of text as that would be quite an effort with even the simplest of rule. I should add that if I go down that avenue I will be expected to provide a full decipherment of the Voynich or the argument about filler text would mistakenly not be considered valid.

You have to understand a simple point that one can never demonstrate, by it's very nature, that filler text is meaningless in any context. That is why one can never prove that the text of the Voynich is nonsense as there is always a possibility of a hidden meaning.

The idea that the filler would have to be constructed out of "components" that have Voynichese properties in order to fulfill the assertion that filler would be easy to create, really over complicates things.

As an example say every 10th symbol is a meaningful character and everything else is filler. The author just has to manually invent some filler that looks like Voynichese, they don't need to implement some component manipulation algorithm.