+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Voynich Talk (https://www.voynich.ninja/forum-6.html)
+--- Thread: Paths to Decipherment (/thread-4244.html)
Pages:
1
2
Paths to Decipherment - Mark Knowles - 19-04-2024
I like to be goal focused and I am interested in the key goal of deciphering and therefore unlocking the Voynich. A question that I always have in the back of my mind is which are the best strategies to move towards decipherment. These are my thoughts:
1)A crib or more likely what Nick Pelling calls a block-paradigm. Such a block-paradigm may or may not exist. That would be a parallel and identical document or piece of text to what we find in the Voynich. The question then becomes what is the best avenue to finding such a document.
2)The discovery of a related cipher. This has been a goal I have been interested in, though it is a challenging one as so many cipher records from the early 15th century are lost.
3)Some AI based approach to decipher the script. Statistical analysis on its own I doubt will be nearly sufficient. If AI is used it will need to be highly sophisticated to find the answer from the solution space.
4)Some lost pages or document which will illuminate the script. Finding anything of the kind seems unlikely.
I daresay there are others that I have forgotten to mention.
RE: Paths to Decipherment - bi3mw - 19-04-2024
(19-04-2024, 06:34 PM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.1)A crib or more likely what Nick Pelling calls a block-paradigm.
I would tend to assume that a crib is most likely to lead to the solution. I don't know Nick's work on this and would use the following simple definition:
Quote:In cryptology, a crib is a plaintext phrase that the codebreaker knows, suspects or guesses occurs in a ciphertext in encrypted form.You are not allowed to view links. Register or Login to view. is a failed and admittedly somewhat clumsy attempt in this direction. In any case, it is a search for a needle in a haystack.
RE: Paths to Decipherment - pjburkshire - 19-04-2024
A.I. has my vote but A.I. is not magic. A.I. is only as good as the training.
RE: Paths to Decipherment - R. Sale - 20-04-2024
Your solution #1 is analogous to a Rosetta Stone solution - having the same text contents in two different languages, one that is readable matched with one from the VMs.
Pelling's block paradigm attempts to make this match based on page formatting of the text, long and short lines etc. in a pattern down the page. Problem is finding the prospective candidates, as (to my knowledge) none have been suggested.
Slightly better is a potential matching based on plant identifications. If a plant ID'd in the VMs can be matched with a plant from a medieval herbal, then perhaps the two texts would also correspond. Some plants have been matched. At least there are potential candidates. No texts have been shown to correspond.
Matching could also be based on other factors, if they are believed to be relevant.
RE: Paths to Decipherment - pjburkshire - 20-04-2024
Question: What would an A.I. program need to solve the puzzle of the Voynich Manuscript?
Microsoft Copilot: "In summary, deciphering the Voynich Manuscript requires not only AI algorithms but also deep linguistic and historical expertise. Perhaps someday, a combination of AI insights and human scholarship will unlock its mysteries! ?2"
(2) Researchers Claim to Crack the Voynich Manuscript Using AI, But Experts .... You are not allowed to view links. Register or Login to view..
Google Gemini: "Overall, solving the Voynich Manuscript would require an AI program with advanced language processing, image recognition, and reasoning capabilities, coupled with a vast knowledge base of languages, history, and culture."
RE: Paths to Decipherment - ReneZ - 20-04-2024
This is somewhat interesting.
At surface level, these are trivial answers that I (and many others) could have come up with, without AI.
Interestingly, though, the first AI does not stay within the bounds of the question, and suggests that AI cannot do it alone, and needs the help of mortals.
RE: Paths to Decipherment - pjburkshire - 20-04-2024
The book belongs to Yale. Maybe someone there would be interested.
You are not allowed to view links. Register or Login to view.
You are not allowed to view links. Register or Login to view.
RE: Paths to Decipherment - Mark Knowles - 21-04-2024
The path that I am most intrigued by at the moment is Number 1.
I doubt that the plants on the herbal pages were copied from another herbal manuscripts along with the text, though it is possible, so there is unlikely to be a "block-paradigm" amongst them, I think. The same would apply to the botanical pages.
The Rosettes Folio, I very much doubt, given my theory of the page, it was copied from another "map" so I don't see a block Paradigm here.
I can't imagine the balneological pages could have been copied as a whole or in part from elsewhere, though it is possible.
Similarly, the so-called "recipes" text looks like a bad place to look for a "block-paradigm".
Amongst the remaining cosmological and astrological pages I think the probability of one of those being largely copied from another manuscript is greater. But which pages are the most likely candidates I am not sure.
RE: Paths to Decipherment - asteckley - 23-04-2024
Regarding AI, I am very doubtful that it can help much, although I would love to see it happen (since it's been my primary field.)
But regardless of my pessimism, its well worth brainstorming to come up with possible approaches to try, and I'd love to hear people's ideas, even if they are just very general conceptually ... even nascent ideas may spawn something clever to try.
When we refer to AI, we really mean the sub-technique of "Machine Learning". That is the basis of all the successful AI you've hear about recently. And ML is in essence nothing more than "pattern recognition", although it can be remarkably good at it -- It can recognize deep patterns in the data that are beyond our abilities as humans to detect. But it requires first having many ... MANY!... examples with which to train it. It's not as simple as just feeding the Voynichese into some existing model (i.e. AI system) and then letting it come up with some kind of answer for what it says, or what language it's based on, or whatever.
One needs to first train the system with a lot of examples from which it can "learn" how to recognize the underlying patterns (like the encryption rules, for example.)
But the challenge is that we have only one example -- the manuscript itself. When it comes to using AI/ML in cryptology, it's what I refer to as the "single sample quandary". We need thousands or even millions of examples, depending on just how deep and complex the patterns are that we are trying to find. But an unknown cipher often involves just one case (e.g. Voynich, Zodiac cipher, etc.)
So you have to come up with some clever way to use the power of the pattern recognition that ML offers -- some framing of the problem for which you DO have a large amount of training data. (Or can generate lots of data in a way that accurately reproduces the deep patterns even though you haven't yet identified them.) Then you can use the AI/ML to solve at least a piece of the problem.
That might all sound a bit hopeless, but it isn't. Like many techniques though, it's only 10% about actually applying the technique, and 90% about framing the problem in a way that can leverage the technique.
RE: Paths to Decipherment - pjburkshire - 24-04-2024
(23-04-2024, 11:43 PM)asteckley Wrote: You are not allowed to view links. Register or Login to view.And ML is in essence nothing more than "pattern recognition", although it can be remarkably good at it -- It can recognize deep patterns in the data that are beyond our abilities as humans to detect. But it requires first having many ... MANY!... examples with which to train it. It's not as simple as just feeding the Voynichese into some existing model (i.e. AI system) and then letting it come up with some kind of answer for what it says, or what language it's based on, or whatever.
Yes, exactly. Correct training material should be identified. I don't think the "380 versions of the Universal Declaration of Human Rights" used by Greg Kondrak was the correct training material.
There really are two groups of people interested in this. There are those who think it is an encryption ( I include constructed languages and shorthand in this) and there are those who think it is a lost natural language, so two groups of training materials need to be identified.
I think the manuscript should be processed in sections, not as a whole. I think to start the focus should be on the pages with one plant each so we may want to review those pages to be sure the data is as clean as possible.
The training material should include 14th and 15th century books about plants and should include information about languages from that time.
Since I think it is a lost natural language and I speculate that "daiin" (daiin) may mean "and", I would like to know about the words that are before and after "daiin" (daiin).