Yes there are numbers in the text. I usually box them with a red lines. I have also written a theory in the presentation that numbers, like in numbers in a specific line/setup, could have symbolic meaning, like 123 = accession.
This is only quesses so far. If we can prove that my model seem to work, a number translation model would be a next side project for me.
I will take a look at the folios you mentioned. Expect som new knowledge soon.
I would have some questions.
So you say that Voynich Manuscript is written in some language that is not Indo-European. How do you know it? What is it then? Would it be some well known other language like Hebrew, Turkish or Arabic or something totally unknown? Do you think that people ever talked in this language somewhere or is it an artificial, constructed language?
Can these words be read aloud or not? Does this language have any grammar?
Then I got totally lost. You give many block diagrams and tree graphs, mention Turing machines and some "models" but it all feels like some scientific mumbo-jumbo to me. Could you explain in some simple words what is your method and were you able to read any longer passages with it?
Don't kill the pianist. I haven't written the script. I'm just making a model to understand the structure and word building. I can't say who wrote it or when. All that I know is that it's too unique too resemble indo European languages so far.
Using a method for machine learning, I started to code every word with a individual number. Doing that I discovered that the language uses a simple main category system. Then, after some work, I also discovered sub categories. Meanwhile, using a method called "brute force" and intuition, I also discovered that its also possible to translate words and bend them.
I also think we should be able to read many long sentences when more words are coded. Its just a early draft of my method.
Some questions will be answered when smarter people than me have a working translation model.
(19-03-2025, 01:37 PM)project963 Wrote: You are not allowed to view links. Register or Login to view.I think the limitations of words and a given multiple word combination should show it the translation model works or produces gibberish. So far, after coding 670 individual words, no gibberish have been produced.
Hello,
Not being gibberish is not good evidence of the correctness of the translation - many translations have been proposed before - especially on a short sample of 6 pages and assuming that Voynichese words have meanings, which is extremely unlikely given the many abnormal statistical features.
Also, if you are using some transliteration as a source, you are seriously underestimating the number of missing spaces and errors.
[
attachment=10185]
Maybe your method would be better suited to translation problems that are more constrained, like the Rohonc Codex, where the meanings of many words are reasonably well identified and the subjects of most pages are known.
Quote:Don't kill the pianist. I haven't written the script. I'm just making a model to understand the structure and word building.
Okay
So if I may ask, what are you exactly doing? Are you using some computer techniques, some algorithms or neural networks?
Quote:Doing that I discovered that the language uses a simple main category system. Then, after some work, I also discovered sub categories.
Could you tell more about it? Actually categories and subcategories ring a bell for me. Have you read about philosophical languages? You are not allowed to view links.
Register or
Login to view.
From what I know all these known early attempts were half-baked but they often assumed some hierarchy of words, grouping similar words into categories and subcategories.
For example in real languages words meaning colours aren't really similar to each other but in constructed languages they could be some kind of regular.
(19-03-2025, 02:36 PM)nablator Wrote: You are not allowed to view links. Register or Login to view. (19-03-2025, 01:37 PM)project963 Wrote: You are not allowed to view links. Register or Login to view.I think the limitations of words and a given multiple word combination should show it the translation model works or produces gibberish. So far, after coding 670 individual words, no gibberish have been produced.
Hello,
Not being gibberish is not good evidence of the correctness of the translation - many translations have been proposed before - especially on a short sample of 6 pages and assuming that Voynichese words have meanings, which is extremely unlikely given the many abnormal statistical features.
Also, if you are using some transliteration as a source, you are seriously underestimating the number of missing spaces and errors.
Maybe your method would be better suited to translation problems that are more constrained, like the Rohonc Codex, where the meanings of many words are reasonably well identified and the subjects of most pages are known.
About the examples you are using, I don't think it's one word. The space is a bit too big between the first glyph and the rest.
According to my model this word is a "shape" word. I think the first glyphs are numbers.
So the first line could be "number" + "shape" word. The other line could then be "a different number" and "the same shape word".
What advantage does using a Turing machine give you ?
A Turing machine is a mathematical model used in theoretical computer science.
According to computer science stackexchange they have no practical use.
Practical importance of Turing machines ->You are not allowed to view links.
Register or
Login to view.
(19-03-2025, 03:45 PM)project963 Wrote: You are not allowed to view links. Register or Login to view.About the examples you are using, I don't think it's one word. The space is a bit too big between the first glyph and the rest.
According to my model this word is a "shape" word. I think the first glyphs are numbers.
So the first line could be "number" + "shape" word. The other line could then be "a different number" and "the same shape word".
This is all very complicated. It is my belief that people are thinking too hard about this manuscript, proposing methods of composition that would be too awkward for anyone in the 15th century to be able to undertake. The answers to the why and how about the manuscript will probably turn out to be simple. In much the same way when Linear B was deciphered everyone was a bit surprised to learn that the solution was logical and easy.
I agree. The text I have read so far doesn't indicate special deep new knowledge. But I suspect that the structure of the language has some tricks. Like how an angel on a objectiv is written. There are some mysteries we can unlock yet.
Well, I wish you luck with your dechiperment!