Good idea… I’ll show you mine, will you show me yours?

(not yet prioritized, except that I’m reading the top one right now)

PS: What a pity that the next decent bookstore is in Bern. Or Portland, but that’s hardly around the corner.

It’s a good, intelligent plot, and yes, it’s riveting like hell. Moreover, he always introduces yet one more twist in the tale, characters above any suspicion suddenly look like prime suspects, and the guys where you thought they’re guilty as hell suddenly prove their innocence. And then, 20 pages on, it all reverses.

This method of “confusing the audience” reminded me quite a bit of Robert Shea’s Illuminatus! trilogy. Surprisingly, there even is a book by Dan Brown called Angels & Demons, which to some extent is about the same secret society. So, I took the bait and ordered the “special deal” at Amazon, Illuminati and Digital Fortress, Dan Brown’s first work for a ridiculously low price (something like 8 € both together, I think).

Illuminati, to me, was a disappointment, but that may be my fault. It’s kind of well written, it is definitely a thriller, but I think reading it a few months after the Da Vinci Code spoiled it for me. While the story is different, some elements seem awfully familiar, not only the main character (who indeed is the same person in both books). Still, it’s a good book for a long train ride.

But my ultimate mistake was to read Digital Fortress, Dan Brown’s first book, right after Angels & Demons. Now, this you shouldn’t do. Never. Don’t read it, if you’re into cryptography (which it purportedly is about). Read something else, like Neal Stephenson’s Cryptonomicon (which I can highly recommend).

Digital Fortress shows us all the plot techniques that Dan Brown uses, but in an unrefined way (granted – it’s his first book, give him a break). But having read the other books, the way he weaves his plot becomes quite predictable (but still, I won’t include spoilers). About 30-40 pages in, as the main theme is exposed, I immediately knew who the bad guy would be, and hell, I was right. Not in all details, such as his motivations, but still, the plot was clear. What a disappointment!

Moreover, it’s a book about the NSA and about cryptography. I am not a cryptography expert, but maybe a little bit a crypto buff. I read and (mostly) understood Applied Cryptography, and a couple of other books on the history and current trends in cryptography (such as the excellent book by Simon Singh).
In the preface, Dan Brown thanks two anonymous employees of the NSA for sharing with him information about modern cryptography. I think he was fooled by some anonymous wannabees, or the NSA successfully kept crypto “secrets” from him. Everything that is about crypto in this book is exceedingly unbelievable and uninformed.

Examples abound: One of the main characters is a beautiful female NSA codebreaker, and thus major parts of the book deal with the way that NSA decrypts email. According to the book, they possess a humongous computer for this, a veritable shrine to technology, so big that they built the crypto building around it. This machine contains 3 million CPUs and thus is capable of brute-forcing encrypted emails within 2 minutes per email. This – according to the book – enables them to read any encrypted electronic conversation. In reality, it enables them to read only a selected portion, since 2 minutes is a long time, considering how many encrypted emails they would collect globally in a day. This fundamental error is repeated several times in the book.

In another place in the book, a new encryption algorithm is discussed, “Digital Fortress”, which kind-of holds NSA hostage. They possess the algorithm source code, but unfortunately, it is heavily encoded, with a 64 bit key, equalling 64 characters. Close, but no cigar. And while we cannot measure cryptosystem strength purely by key length, 64 doesn’t sound so impressive. At least the author could have chosen something that sounds cooler.

The errors go on and on, email messages which when decrypted are suddenly executed as code on the code-breaking computer violates the distinction between data and code … or maybe the NSA’s computer runs Outlook?

If you write about crypto, collaborate with an expert (like Neal Stephenson, who collaborated with Bruce Schneier for the trickier parts of Cryptonomicon). Or at least, read a few public articles, before you make a fool of yourself. Digital Fortress left me angry at actually finishing the book.

(1. Grab the nearest book. 2. Open the book to page 23. 3. Find the 5th sentence. 4. Post the text of the whole sentence in your journal along with these instructions).

Back in the eighties, when GEB came out, my father bought the book (or maybe got it as a present). At some time, probably about 10 years ago, I decided to read it, and while it struck me as brilliant, it also struck me as a bit too heavy, and I moved off to another book when I was half-way through. That decision is usually the death sentence for any of my books. Such a book can’t be picked up again easily, for where would you start? Where you left off? You’d be confused. From the beginning? You’d be put off, since it’d sound boring to you now … So the book is condemned to bookshelf limbo.

But for GEB, the 10-year pause may just do the trick. I’ll keep you updated on how far I get.

Update: March 2nd, 2004
As of today, I am still reading GEB. That means two things: First of all, I haven’t finished it yet, which is not surprising, considering that it’s about 800 pages long and not at all the kind of reading you can breeze through in a train or on a plane. Second, however, the fact that I’m still reading it means just that – I haven’t given up yet, even though the thought has crossed my mind once or twice.

In a first glance, GEB is a book all about number theory, logic, and self-referential systems. Now, admittedly, this doesn’t sound very exciting at first. Douglas R. Hofstadter, the author, however, broadens the subject considerably, linking these fundamental aspects (and what could be more fundamental than the properties of whole numbers and logical relations?) to a broad variety of topics, ranging from music and art to the self-replicating nature of DNA and to the way our mind works. This breadth and depth of the subject matter makes for a fascinating read, and keeps you hanging in there when you hit upon a stretch of pages that are not unlike my maths textbooks from university, i.e. utterly unpleasant to read.

These dull moments, of which GEB has quite a few, are nicely contrasted by the introductory preambles to each chapter. In these preambles, the author introduces the next topic in a playful and whimsical manner, sometimes in dialogues, sometimes in one-sided dialogues (one end of a phone conversation), and sometimes in more intricate forms like canons. The ever-recurring, slightly off-key, protagonists, Achilles, the bravest of all warriors, and his friends, the turtle and the crab, allow the author to introduce plently of humor, puns and mind games into a book that otherwise may have died of boredom.

GEB takes us from simple first principles to complex subject matter, explored topics include: meaning and form in mathematics, the dichotomy between figure and background, or between positive assertions and the negative assertions derived from them, consistency and completeness in a system, recursive processes, levels of description and how brains and thoughts may be represented in terms of symbols and meta-symbols, paradoxical statements, which predict their own falsehood, but only if they’re true, and many more wondrous things. Again, this is not a book that’s easy to read on a quick train ride, every page presents a certain intellectual challenge that must be pondered upon. However, don’t be alarmed, the book is not as dull as the quick overview of its topics above may make you believe.

In the introduction to the 20th anniversary edition, the author shares with us some of the toils and trials he had to go through in the making of the first edition, as he essentially typeset and printed the galleys himself, getting excellent first-hand experience about what can go wrong in the printing process. For the 20th anniversary edition, he decided that the book should come out in a form true to the original, no alterations made and nothing updated. Thus, how well has the book stood the test of time?

One item that Hofstadter mentions in his introduction, which strikes him as old-fashioned nowadays, is that almost all of the characters in GEB are male. This sexist, if you will, viewpoint seems to him as something that would offend nowadays, and which he wouldn’t have written that way. Personally, I couldn’t care less. To me, these skin-deep concessions in order not to seem sexist (alternatingly writing “he” and “she” for instance), have no deep relation to the thought that the sexes are equal. Characters in a story are just that, characters, and thus, they should have the sex that fits their character. There are, however, one or two things that I found surprising and a bit out-of-date. When Hofstadter speaks about DNA as a self-replicator, he describes its function in a idealized, mathematical way, which explicitly is not the way things are done in nature. However, later, when he describes how DNA works in real life, some of his statements are simply wrong^*. I cannot tell for sure, but I believe that this is the author’s fault, not the fault of current knowledge 20 years ago. In such cases, I would be a proponent for making a change to the book, since it’s not a sign of the times that is brought up to date, but rather an author’s error that is fixed.

Another strange omission is the extremely short biography of Alan Turing, one of the greatest minds of the 20th century. His whole career in code-breaking during WWII, and thus his interest in information theory and communication systems, is omitted, likewise his suicide at young age, brought upon by his being ostracized as a homosexual, is described as “apparently (…) an accident with chemicals. Some say suicide.”. However, it could well be that the facts surrounding his secret career and his early demise were not publicly known in 1984, when the book was written.

I will write up some concluding remarks when I am finished with the book, so stay tuned …

^*One wrong statement is that the distinction of DNA (deoxyribonucleic acid) versus RNA (ribonucleic acid) is that one of them is lacking an oxygen in the phosphate group. In reality, the oxygen is lacking in the ribose sugar. Yes, I am nitpicking… but then, I am a biologist.