A case for Gibberish

[MarcoP]

So what to do to be more focused with Voynich studies, would be to find the exact algorithm which produces random  VMS text. If one finds a 95% accuracy for the VMS text generation then in my mind he/she will have finally solved the Voynich Gibberish case!

I know Torsten Timm has gone down this path, however I don't think he has had a an equivalency of a 95% success rate.

My personal opinion is that, in order to compute a numerical similarity like "95%", you need a formally defined set of quantitative features to compare (e.g. bigram and word histograms, character and word entropies, reduplication rate, Schinner's alpha of F(l) etc.). When you have a formal definition of the text you want to obtain, as well as an example of an actual text conforming to it (the VMS), I believe that creating the software is just a matter of putting enough effort into it. I don't claim it will be easy, but it seems obvious that it can be done.

As I think Koen pointed out when discussing Timm and Schinner's paper, if you define a set of high-level quantitative features for Shakespeare's sonnets, one can write a software generating text with those properties. The output of the software will possibly be sonnets from the metrical point of view, but it will be meaningless. This would not prove that Shakespeare's English is meaningless.

There could be people that would be interested in a closer computational simulation of Voynichese, but many are not. For instance, Torsten Timm explained why writing a more complex software would be a waste of time:

Also, please correct me if I am wrong, but I think the generated text here doesn't feature any clear dependency between the last glyph of a word and the first glyph of the next word?

"Keep in mind that the VMS was created by a human writer who had complete freedom to vary some details of the generating algorithm on the spur of a moment. ..." (Timm & Schinner 2019, p. 16). With other words we didn't argue that the text was created by a computer program and we didn't argue that our program is able to simulate the complexity of a human mind.

Beside line breaks numerous similar rules for describing some kind of local repetition exists. Even if I would try to add each rule it would still be possible to point to an another similar rule. At the same time each additional rule would increase the complexity of the program and it would be argued that the program is too sophisticated. Therefore our goal was to keep the algorithm as simple as possible.

If you really want to give this a try, you can download Timm and Schinner's sources from github and start from there. But you seem to be already convinced that it's gibberish, so I can't think of any reason for doing that.

I have spent several hours browsing through Timm and Schinner's java software. Though it is not huge (the core is about 10,000 lines of code) I have only seen a small part of it. You should expect that nobody will ever look into anything more complex.

---

EDIT You are not allowed to view links. Register or Login to view.

My manuscript, you inspire me to write.
I love the way you disturb and puzzle,
Invading my mind day and through the night,
Always dreaming about the laid mizzle.

Let me compare you to a foul ember?
You are more ancient, random and cryptic.
Past winds shake the leafage of September,
And autumntime has the afraid triptych.

How do I love you? Let me count the ways.
I love your repetitions, words and text.
Thinking of your loathsome words fills my days.
My love for you is the funny pretext.

Now I must away with a sunny heart,
Remember my old words whilst we're apart.