Facsimile of the Voynich Manuscript Now Available to Citizen Cryptographers

Copyright of copies of public domain artworks is odd: the original artwork is not under copyright, but a photo (and possibly a scan?) of the artwork can be. The US legal theory (I don't know about elsewhere) is that a photo of a public domain artwork can itself be an artistic creation, deserving of separate protection.

For example, the photographer might have made an artistic choice to use a diffuse white light without glare, to have the entire frame in focus, and to crop the photo at the edge of the artwork. It is commonly thought that a mechanistic scan fails to meet this ridiculously low threshold of creativity, but this defense has not been well tested in the courts.

So in the absence of a contractual agreement to the contrary, if you make a copy directly from the original, you are likely legal. But if you made a reproduction from a reproduction that could be considered itself an artistic work, you could be infringing on the copyright of that reproduction.

I don't know what the rule would be for making a copy of an authorized "slavish" physical reproduction, but given the low bar for copyright, I'd guess it would still be considered infringement? Anyway, here's an overview of the legal situation in the US: http://blogs.library.duke.edu/scholcomm/2013/06/14/museums-c...

Here's a link to someone who went ahead and created his own physical version from the published scans, printed on a color laser printer, then 'aged' by dunking them in coffee: http://scienceblogs.de/klausis-krypto-kolumne/2016/08/27/pub...

Here's a link with a little bit more detail about the arrangement between the publisher and the library: http://www.thelocal.es/20160821/tiny-spanish-publisher-clone...

Finally here's the page about the Manuscript on the publisher's website, which might have more information if you read Spanish: http://voynich.siloe.es/manuscrito/

To give an example, for a written natural language, https://en.wikipedia.org/wiki/Zipf%27s_law should generally apply to letters, words, and, I think, to n-grams as well. This is not true for encrypted or compressed data, or for data produced by a random process like throwing dice, so you can distinguish the two pretty well with histograms.

The simplest tests like this generally went in favor of the Voynich manuscript having real linguistic content, but there are other statistical tests too and I'm not familiar with what they've shown. :-)

Have you read Stephen Bax' paper on partial decode based on botany?

https://stephenbax.net/?page_id=11

I find it quite compelling.

I've come down to three possibilities:

(1) it could be a copy of a much older, now lost document, which puts the timeline in question.

(2) Colonialism and imperialism really did a number on indigenous people and culture. This could be an ethnography of sorts of am indigenous people's knowledge in their now lost language.

(3) We could be fundamentally "reading it wrong". I've extended this question as to "Well how else would this be done?" beyond mechanical cypher machines.

Maybe glyphs we've bucketed are actually distinct or maybe some aspect of it such as the length of a stroke have semantic meaning. It could be an extremely poorly designed and confusing script.

I know it's been looked at by people likely smarter than me, but I really don't believe in gods, whether human or otherwise - so I'm not discouraged ... although really, finite time, finite time.