+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Analysis of the text (https://www.voynich.ninja/forum-41.html)
+--- Thread: Discussion of "A possible generating algorithm of the Voynich manuscript" (/thread-2790.html)
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - ReneZ - 15-09-2020
Perhaps yet another short summary:
Torsten's input is something that is like a correlation: similar words tend to appear near each other.
This correlation is converted into a causality: this happens because the writer took previous words, modified them a bit and wrote them down again.
One cannot go from correlation to causality without extra data/information.
The only extra data/information presented is the output of Torsten's App. We don't know what it does exactly.
On top of that, it has been demonstrated that a simple method of converting a meaningful text to unreadable text can have exactly the same property as seen in the MS. This shows that the conclusion that the text is meaningless is not a valid conclusion.
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - Stephen Carlson - 15-09-2020
(15-09-2020, 05:58 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.On top of that, it has been demonstrated that a simple method of converting a meaningful text to unreadable text can have exactly the same property as seen in the MS. This shows that the conclusion that the text is meaningless is not a valid conclusion.Is this process reversible?
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - ReneZ - 15-09-2020
(15-09-2020, 08:08 AM)Stephen Carlson Wrote: You are not allowed to view links. Register or Login to view.(15-09-2020, 05:58 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.On top of that, it has been demonstrated that a simple method of converting a meaningful text to unreadable text can have exactly the same property as seen in the MS. This shows that the conclusion that the text is meaningless is not a valid conclusion.Is this process reversible?
Yes.
It is a word-by-word translation of the original text, so basically one would have a dictionary that can be used in both directions.
I am not trying to argue that this is how it was done.
This method suffers from some of the same problems as the auto-copy theory, for example all the aspects of the 'Line as a functional unit' observation of Currier.
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - Torsten - 15-09-2020
removed
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - ReneZ - 15-09-2020
(15-09-2020, 09:44 AM)Torsten Wrote: You are not allowed to view links. Register or Login to view.Actually I say that there is a relation between word similarity and context AND ALSO that there is a relation between word similarity and word frequency. This means that it is possible to demonstrate "a deep correlation between frequency, similarity, and spatial vicinity of tokens within the VMS text" (Timm & Schinner 2020, p. 3). This means that similar words depend on each other. (Timm 2015, p. 14).
I don't know what is a "deep correlation". However, it is not demonstrated that similar words depend on each other. They may be found near each other. To some extent. It is another leap from correlation to causality that I consider unproven.
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - Torsten - 15-09-2020
removed
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - ReneZ - 15-09-2020
This is getting too heated and too personal.
I am not denying observations. I am critical of the conclusions drawn from them.
That is not something specific to this hypothesis, but is general to all hypotheses that I have been confronted with.
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - RobGea - 15-09-2020
@Torsten,
just a small question regarding this:
"when we look at the three most frequent words on each page, for more than half of the pages two of three will differ in only one detail" (Timm & Schinner 2020, p.3)
Perhaps i have made an error but i think the question is valid. When the frequency counts are equal , how are the top 3 then decided ?
for instance for You are not allowed to view links. Register or Login to view. i get a Top5 of : [('sho', 5), ('daiin', 4), ('chor', 2), ('sheo', 2), ('shody', 2)]
i ordered 'chor', 'sheo, 'shody' alphabetically as a secondary sort when frequencies were equal.
We do not know the alphabetic order of the VMS text
so here 'chor' is in the top 3 but 'sheo' also has the same fequency and if i use that i.e'sheo' as the 3rd word then sho and sheo have a levenshtein of 1 and then this folio matches the 2/3 words.
What did you use for a secondary sort ? and why ?
It's only a small methodological issue and really changes very little in the overarching scheme, I was just curious.
Thanks.
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - geoffreycaveney - 15-09-2020
Torsten, I have a sincere question for you: Is there any colleague of yours, whether a co-author or some other colleague, who agrees with your papers and your conclusions, who could possibly join this forum and help explain your arguments and conclusions, and answer the questions and comments that other participants have about your work?
I can see that there may be some validity to many of your observations, points, and arguments. That is not the same thing as agreeing that your papers and all their conclusions are necessarily 100% correct. But frankly it is difficult to work one's way through understanding all of your points and arguments, when almost every paragraph you write on this forum begins with "René Zandbergen didn't respond to the argument", "René Zandbergen denies the observation / conclusion", "René Zandbergen argues this way", etc. (If we just took your posts here and coded a cipher with "denies" = daiin, "Rene" = okain, "argued" = olchedy, etc., we could probably get some very Voynich-like statistics.)
Perhaps a different person, or just a second person, explaining the ideas behind your arguments and conclusions, perhaps in a different style and in a different way, could be helpful to the progress of this discussion on this forum? A more user-friendly approach from another angle might be able to help some of us see some of your points more clearly that way. Just a thought and a suggestion.
RE: Discussion of "A possible generating algorithm of the Voynich manuscript" - Torsten - 15-09-2020
(15-09-2020, 05:16 PM)RobGea Wrote: You are not allowed to view links. Register or Login to view."when we look at the three most frequent words on each page, for more than half of the pages two of three will differ in only one detail" (Timm & Schinner 2020, p.3)
Perhaps i have made an error but i think the question is valid. When the frequency counts are equal , how are the top 3 then decided ?
In such a case I have chosen a word that is similar.
This is the list I had used to count the words:
Code:
f1r daiin (7) dain (6)
f1v chol (5) shol (4)
f2r saiin (4) daiin (3)
f2v chol (5) chor (5)
f3r chor (7) chol (7)
f3v chor (3) chom (3)
f4r
f4v sho (5) sheo (2)
f5r
f5v chol (3) chor (2)
f6r
f6v
f7r daiin (2) dain (2)
f7v dol (4) dor (3)
f8r
f8v chol (8) shol (5)
f9r
f9v chor (3) chol (3)
f10r
f10v daiin (8) dain (3)
f11r
f11v
f13r
f13v
f14r daiin (3) dain (2)
f14v
f15r
f15v chor (5) chol (4)
f16r daiin (3) dain (2)
f16v chol (4) chor (3)
f17r chor (2) chol (2)
f17v ol (4) or (4)
f18r
f18v qokchy (4) qotchy (2)
f19r
f19v
f20r
f20v
f21r chol (4) chor (2)
f21v
f22r
f22v
f23r dal (3) dar (3)
f23v dol (2) dor (2)
f24r
f24v
f25r cthain (2) chain (2)
f25v
f26r
f26v
f27r chy (6) shy (5)
f27v sho (3) shy (2)
f28r
f28v
f29r
f29v chol (4) cho (4)
f30r chey (5) shey (3)
f30v sho (3) chy (2)
f31r okedy (4) qokedy (4)
f31v al (2) ar (2)
f32r daiin (7) dain (3)
f32v daiin (12) dain (4)
f33r
f33v
f34r
f34v chedy (5) chdy (4)
f35r
f35v daiin (8) dain (2)
f36r chol (2) cthol (2)
f36v
f37r
f37v daiin (16) dain (2)
f38r
f38v daiin (8) dain (3)
f39r
f39v or (9) ar (4)
f40r ar (4) or (4)
f40v okain (4) qokain (3)
f41r
f41v daiin (3) aiin (3)
f42r shol (11) chol (8)
f42v
f43r
f43v chedy (6) chey (4)
f44r
f44v
f45r daiin (3) dain (2)
f45v
f46r daiin (6) aiin (4)
f46v chedy (6) shedy (4)
f47r chol (9) shol (4)
f47v
f48r
f48v
f49r shor (4) chor (3)
f49v chor (5) chol (5)
f50r kar (5) ar (4)
f50v okain (3) okaiin (3)
f51r
f51v
f52r oty (3) oky (2)
f52v
f53r oty (3) qoty (2)
f53v
f54r or (6) ol (5)
f54v
f55r
f55v daiin (7) aiin (6)
f56r chy (5) chey (3)
f56v chol (9) shol (3)
f58r
f58v qokal (9) okal (9)
f57r cheody (4) chedy (2)
f66r
f66v dal (5) kal (3)
f75r
f75v ol (15) qol (13)
f76r shedy (28) chedy (23)
f76v chedy (17) shedy (12)
f77r
f77v shedy (18) chedy (16)
f78r
f78v ol (22) qol (13)
f79r shey (10) shedy (10)
f79v
f80r
f80v qokain (14) qokaiin (9)
f81r shedy (9) chedy (9)
f81v
f82r qokeedy (14) qokeey (11)
f82v shedy (11) chedy (10)
f83r shedy (15) chedy (13)
f83v shedy (15) chedy (11)
f84r
f84v
f85r1
f85r2 or (7) ar (6)
f86v5 ar (10) or (9)
f86v3
f87r
f87v
f88r chol (6) cheol (6)
f88v
f89r1
f89r2
f89v2 daiin (7) aiin (5)
f89v1
f90r1
f90r2 daiin (3) saiin (3)
f90v2
f90v1
f93r
f93v chody (3) cthody (2)
f94r
f94v
f95r1 chdy (7) chedy (4)
f95r2 or (3) ol (2)
f95v2 okal (3) okar (2)
f95v1 otar (2) otal (2)
f96r sheol (2) cheol (2)
f96v cheor (2) chor (2)
f99r
f99v qokeol (5) okeol (5)
f100r chol (8) shol (3)
f100v
f101r1 ol (7) or (6)
f101v2
f102r1
f102r2 dol (7) dor (3)
f102v2 ol (5) or (5)
f102v1 cheol (5) cheor (3)
f103r shedy (18) shey (15)
f103v shedy (15) shey (14)
f104r
f104v
f105r al (7) ar (6)
f105v aiin (14) daiin (13)
f106r qokaiin (8) qokain (7)
f106v
f107r
f107v okaiin (14) qokaiin (12)
f108r qokeedy (16) qokedy (16)
f108v qokeedy (28) qokeey (21)
f111r qokeey (21) qokeedy (15)
f111v chedy (17) chey (16)
f112r
f112v
f113r qokaiin (9) okaiin (8)
f113v okaiin (11) otaiin (11)
f114r daiin (11) aiin (9)
f114v qotaiin (9) otaiin (7)
f115r
f115v
f116r