resolution - decoding mystery

[Wendybird]

If coding mystery was based on simple substitution, why wouldn't the computers have de-coded/solved this? Even if capital letters were coded differently as well as a few other letter combinations, it seems to me that this would have been easily deciphered given that lots of phrases and sentences would be revealed after so many powerful computers had run through millions of sequences? Wouldn't have required much since there's no way the typeface could have included more than one hundred indentations.

[Renee]

SPOILERS!!!




_______________________
Because the code was in tiny details in the Font, and in the digitization process the font was removed from the analysis. The person decoding had to see the message in the original typeface in order to decode it. When the book was scanned and then made searchable using Optical Character Recognition (OCR), the final digital document was in a generic font, not the original, so the message was stripped.

[Niklas Mehner]

Renee wrote: "SPOILERS!!!




_______________________
Because the code was in tiny details in the Font, and in the digitization process the font was removed from the analysis. The person decoding had to see t..."

Since always the same letter has the same "tiny details" this is still a simple substitution. It would only be different if you had e.g. 24 different copies of the letter "a" each with different "tiny details". But this does not seem to be the case.

So I was very disappointed as well by this resolution.

Also the "making money from font copyright" does not work. The first copyright law in the USA is from 1790 ( http://en.wikipedia.org/wiki/US_copyr... ). And "it did not prohibit copying the works of foreign authors".

And: Fonts cannot be copyrighted: http://en.wikipedia.org/wiki/Typeface
"In Eltra Corp. v. Ringer,[16] the Court of Appeals for the Fourth Circuit held that typeface designs are not subject to copyright. However, novel and non-obvious typeface designs are subject to protection by design patents.[17] Digital fonts that embody a particular design are often subject to copyright as computer programs.[18][19] The names of the typefaces can become trademarked. As a result of these various means of legal protection, sometimes the same typeface exists in multiple names and implementations."

And: Copyright expires after a few years.

So while I found the story a nice read, the book really needed a lot more research :(

[Son]

I don't see why real law has anything to do with a fictional story. I personally don't worry about actual copyright laws or when they expire, I enjoyed the story for exactly what it was, a story. Something tells me there isn't a Gerritszoom font or Unbroken Spine cult either! Or, for that matter, a room with books written in code from those who were in the cult under NYC, or even a realy Mr. Penumbra's 24-hr Bookstore OR if Google has a big box...on and on we can do.

Also, USA laws would not even matter considering the original font was made in Italy (or somewhere in that area of the world, I can't remember exactly now).

The OCR with a new font for the original Gerritszoom makes perfect sense, you had to see the original font in order to see the code placed onto the actual letters not the words themselves. My thinking on this matter was Sloan is ingenious to come up with an interesting code only caught by a human being. Here we are again with computers vs. humans when a human wins over anything (even Google) a computer could come up with. Like it says, a computer can only do what a human tells it to do.

[Nadia]

I just finished the book and had the same concern as WendyBird and Niklas (I searched the web high and low for other comments/questions on this topic, and this appears to be the only relevant thread). Based on Sloan's description, I simply don't see how Geritzsoon would end up being anything other than a simple substitution code. Moreover, if lining up all the punches reveals the final message to Manutius (on p. 277), then how did Geritzsoon code the letters, numbers, and punctuation marks that weren't included in that message? You can't have a code if the only things you can encode are the letters ABDEFGHIKLMNORSTUVY. Am I missing something?

[Random]

Its more than a simple substitution code. I believe a couple examples of this were mentioned, but I only remember one off hand.

The type was more than just an A or a B, etc. Not only do you have uppercase and lowercase differences, but he mentions other special items such as a specific block for ff (which would then be considered as its own character). So you have to differentiate between two single f or the ff.

In order to properly print the book in the old method, they would need to have the blocks for all the punctuation, numbers, etc.

But, as Casey said, I wouldn't take this so seriously. I mean, do you really think that if all the servers at Google were busy it would bring the entire Internet to a screeching halt? Or that Google is researching life extension?

[Son]

Random wrote: "Its more than a simple substitution code. I believe a couple examples of this were mentioned, but I only remember one off hand.

The type was more than just an A or a B, etc. Not only do you have..."

That's funny Random - they might be researching life extension. More than anything else, this was an enjoyable read with a wonderful twist in the end. I see it as the human mind reigning over technology. I really enjoyed that Sloan added in about the smell of books, because that is one of my favorite things about using the library or getting a new book from BAM or Amazon - that paper-y smell of cracking up a book, whether it's new or old!

[dan]

Sloan managed to suspend my disbelief for most of the book (even through the Hackers-esque 3D hacking session), and i give him kudos for that, but the substitution cipher did bother me.

The Unbroken Spine should have been able to solve it pretty easily, even without noticing the tiny notches. They surely would have considered the difference between uppercase and lowercase, the ff ligature, and the alternate glyphs (three P's, two C's, a truly epic Q, etc.). After that, they only would have had to do some frequency analysis, which has been used since Manutius's time.

[Sherri]

I really enjoyed this book. Sometimes you can't overthink stories and pick apart the resolution; you just go along for the ride. The resolution was definitely not the high point, but it didn't take away from the aspect of the story I loved most, the unique characters. I particularly loved how all the characters came to life and their stories wove together. So many fascinating details/people: Mat's crazy creations, Neel's unique business, Deckle, the evening clerk, the Booty business, the knitting gal, Kate's red t-shirts, etc. My take-away was not the logistics of the mystery of the Gerritzoon font; it was about relationships. If more people in the real world were as keen and loyal as Jannon, the world would be a better place.

[Courtney]

What did the "25,000 miles" text to Kat mean?

[Random]

The circumference of the earth.
Its in reference to a conversation he had with Kat earlier in the book.

[Nadia]

All -

I reached out to the author to ask him about this issue, and he responded! Here's his response:

"Let me know if this makes it any clearer (or more believable):

https://github.com/mattmills/penumbra...

The difficulty isn't in the subtlety of the code; it's in the information loss that occurs when you mis-transcribe the encoded text. Clay's insight was to see the encoded text differently. There's more detail & an example on the other side of that link above.

Of course, if this wasn't clear to most readers—and I don't think it was—that's my fault!"

I'm still not 100% convinced that either Google or the Unbroken Spine wouldn't have been able to pick up on this, but it's at least a better explanation. Your thoughts?

[Wendybird]

Thanks for posts Nadia.
I was truly puzzled after reading the conclusion of this story. I'm not a math or digital whiz, so maybe there's something I've missed. I'd never even thought of the punctuation symbols as you suggested.

Why rely on OCR knowing the possibility of sampling error? I never considered that as a method. Analyzing thousands or millions of samples of manually-inserted symbols with even a simple program could figure this out quickly. Less quickly with frequency analysis as Dan suggested. It's still a simple substitution with an unchanging key, no?

[Licha]

I was not a fan of this book. Speaking of codes, was there anything to the code names for some of the characters? I thought this would be explained but it never was. Or did I miss something?

[LindaJ^]

This was a gem of a book - fun to read and a happy ending! I did not worry about the reality of the law or code breaking. I loved the people interactions, especially the two guys living out their 6th grade fantasy! I felt good when I finished it.

[J.S.]

*********SPOILER*********

Regarding the code: The notches on the different fonts were used to decode the codex vitae using substitution. The "final words" or secret message was decoded using the words "hidden in the colophon."

http://www.merriam-webster.com/dictio...

Does this help clear things up?

I loved this book.

[Rob]

Because the book was awful and lazy.

[J.S.]

Mark wrote: "Thank you Nadia, for finding that link.

It's still a bad piece of writing. At one point they're talking about measuring the levels of ink as a possible clue to the cipher, but then we're supposed..."

I'll have to go back and check, but I thought the notches were on the top of the punches, not on the "business end."

[Taryn]

Sherri wrote: "I really enjoyed this book. Sometimes you can't overthink stories and pick apart the resolution; you just go along for the ride. The resolution was definitely not the high point, but it didn't take..."

Sherri has a marvelous idea!

[Amy]

I must be able to suspend disbelief better than others and just enjoy a piece of fiction for being fiction. I really liked this book. As for the code, it seemed plausible to me that it was not easy to break. All the code breakers were trying to find a pattern with the letters standing for another letter. They didn't look at the formation of the font itself and factor in the typesetting "ff" for example. I thought it gave a nice message of the power of the human mind over technology (the "OK") in a world where technology is taking over. Of course the "Googlers" would think it was unbreakable with no message there - they believe their technology is infallible.

[Steven]

Random wrote: "Its more than a simple substitution code. I believe a couple examples of this were mentioned, but I only remember one off hand.

The type was more than just an A or a B, etc. Not only do you have..."

Google researching life extension? Absolutely.

It's called a moonshot and Project Calico was featured in Time Magazine.

So yes, they are.

[John]

dan wrote: "The Unbroken Spine should have been able to solve it"
Clark Moffat was a member and he did solve it.

[Cynthia]

The point of the code was that it was something physical like an actual book made of paper, not etherial like the internet. No matter how much computers change, grow and can do…there are those who will still want to hold (and smell) a real book.

[Yashvardhanbandinliu]

Renee wrote: "SPOILERS!!!

I get that but the font itself was the same .. all throughout the world including all the supposed code in it ..


_______________________
Because the code was in tiny details in the Font, and in the digitization process the font was removed from the analysis. The person decoding had to see t..."

[Jan]

Regarding the 25,000 miles:
Random wrote: "The circumference of the earth.
Its in reference to a conversation he had with Kat earlier in the book."

Would you mind to help me? Which conversation? I figured out the circumference thing... but I do not know what he is hinting at

[Cynthia]

Sorry Jan, I just saw your question. It's been 2 years since I read the book, so no clue about the answer.

[Jan]

Cynthia wrote: "Sorry Jan, I just saw your question. It's been 2 years since I read the book, so no clue about the answer."

Thanks anyhow =)

[Julie]

I just don’t understand the point of it, or the idea that “ the message was in plain sight.”..... thank you for being my friend??????

[Normandy]

Jan wrote: "Regarding the 25,000 miles:
Random wrote: "The circumference of the earth.
Its in reference to a conversation he had with Kat earlier in the book."

Would you mind to help me? Which conversation? I..."

Before the Google decoding attempt, Kat was speculating about the great secrets that might be revealed and Clay responded that medieval people were not very advanced technically. Kat replied that people could calculate the circumference of the world 1000 years before printing and she challenged Clay to do that. Kat is probably referring to Eratosthenes whom you can Google to understand the method.

[Normandy]

Julie wrote: "I just don’t understand the point of it, or the idea that “ the message was in plain sight.”..... thank you for being my friend??????"

I think the message means that immortality comes from thoughts of you remaining in the minds of people with whom friendships were developed. Hidden in plain sight means that when the original Gerrittszoon type was digitized the notches weer copied to and everyone missed it except Moffat.

[David]

As someone who's actually studied cryptography at university, Sloans explanation still doesnt matter, this would have been incredibly trivial to crack. That really shouldn't matter tho, it was a delightful story and its fiction. There really isnt any way Sloan could have come up with through more research and kept the beauty of the story and have thel ciphertext be something that was uncrackable. Maybe its easier for me to suspend disbelief since I am aware of that, but i feel it should actually make it more "cringe worthy" because I am fully aware of just how trivial a problem this would have presneted. Basically in short, if I can get over, so should you ;P

[Normandy]

David wrote: "As someone who's actually studied cryptography at university, Sloans explanation still doesnt matter, this would have been incredibly trivial to crack. That really shouldn't matter tho, it was a de..."

You are absolutely correct if the Gerritzoon type set was small. But as the number of glyphs in the typeset go up there are multiple ways the the substitution cipher becomes more difficult to crack. As an extreme example, I read that a complete Times Roman typeset has 1674 glyphs. I think I told you that I found a resource that indicated there were 1674 glyphs for Times Roman. Assuming the entire type set were notched to produce the 26 letters of the alphabet that means there would be some 64 ways to represent to each letter (1674/26 = 64). Thus for the four letters of the name Dave there would be 64^4 (over sixteen million) ways to encode it.

[Ian]

I’m afraid I must agree with Wendybird. The implausibility of supercomputers being unable to solve a simple substitution code was just too much of a clanger for me. The entire plot revolves around this issue and so the plot simply fails. I was so perturbed I had to search for this thread in order to vent my frustration. I feel better now!

[Thottumkal]

We need a "How It Should Have Ended" video!