[Blog Post] MSI at last!

[LisaFaginDavis]

My friends, I have some very exciting news for you! Roger Easton, who took the MSI images of selected pages of the VMS back in 2014, has sent me the images and given me permission to make them public. You can find them all, including the raw 16-bit TIFFs, here:

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

I've written a lengthy blogpost with my own initial thoughts about some of these images:

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

A few important revelations:

1) On f. 1r, the MSI images make the alphabets in the righthand margin legible. There are THREE columns of alphabets, and Rene and I have identified the hand as Marci's! Much more on this in the blog...

2) Those of you working on the inscriptions on f. 17r and 116v will find these images VERY useful!

3) These images make a very strong proof-of-concept case for imaging the entire manuscript. I'm working on it!

4) Dig in! There is much to see and explore!

[RobGea]

Woo Hoo !  Thanks to all involved for making these available.

[LisaFaginDavis]

A few more observations that weren't really appropriate for a public-facing blogpost:

1) In the Voynichese sequence revealed on 1r, Marci interprets EVA [an] and [in] as single characters instead of bigrams. I'd love to hear from the data-gatherers among you how this interpretation might impact analytics!

2) There's a Voynich-y flower revealed in the upper right corner of 17r. It's barely visible in natural light but is quite clear in MSI. It isn't an offset from 16v or any other leaf. I think it's just a doodle, perhaps added at the same time as the inscription in the upper margin.

[bi3mw]

1) In the Voynichese sequence revealed on 1r, Marci interprets EVA [an] and [in] as single characters instead of bigrams. I'd love to hear from the data-gatherers among you how this interpretation might impact analytics!

Total occurrences of the specified substrings (an, in): 6166
Total number of words in the text: 39020
Ratio of found words to total words: 15.80%

Substring combinations (sorted by frequency):
in, in: 1
Words: opoiiinoin ( f72r3 )

Total sum of all substring combinations: 1

So there are practically no combinations.

[RobGea]

For bigram counts, I get.

EVA [an]  Rank : 89   Count :  122
EVA [in]   Rank : 06   Count : 6044

So, EVA [an] is a very curious choice for collapsing to a single character.
It occurs most frequently at the beginning, on f1r, You are not allowed to view links. Register or Login to view. and close to the end at You are not allowed to view links. Register or Login to view. and f116r.

Pure speculation, but maybe Marci [1] scanned the first and last few pages and thought EVA [an] was more common than it is, 
indicating perhaps that Marci had not proceeded far into his analysis when he wrote that column..

[1] The column writer tentatively identified as Marci.

[LisaFaginDavis]

I agree, because it's so rare! I was really surprised. The idea of Marci refering only to the first and last few pages is a good one.

[Koen G]

Thanks, Lisa! I've been looking forward to these.

Are there any ideas on the purpose of the offset columns? When you offset the alphabet, you're basically making a Ceasar cipher, might it have something to do with that?

[LisaFaginDavis]

I asked my brother (who is a computer science professor and expert cryptologist at the US Air Force Academy), and he said the same thing, a Caesar cipher, but he was at a loss to explain why there would be a second Roman alphabet. I leave that to the historical cryptologists in the room!

[Aga Tentakulus]

Now you can even read Tepenece's name quite clearly.

[Aga Tentakulus]

With two rows + one row, you could assume a grid system (chessboard). But that would be around 400 possibilities.