« GCC: "error: extra qualification" Torpedo for C++ Programs in 4.1 | Main | Three Topical Kipling Poems »

Thursday, May 25, 2006

Reading List: Cryptonomicon

Stephenson, Neal. Cryptonomicon. New York: Perennial, 1999. ISBN 0-380-78862-4.
I've found that I rarely enjoy, and consequently am disinclined to pick up, these huge, fat, square works of fiction cranked out by contemporary super scribblers such as Tom Clancy, Stephen King, and J.K. Rowling. In each case, the author started out and made their name crafting intricately constructed, tightly plotted page-turners, but later on succumbed to a kind of mid-career spread which yields flabby doorstop novels that give you hand cramps if you read them in bed and contain more filler than thriller. My hypothesis is that when a talented author is getting started, their initial books receive the close attention of a professional editor and benefit from the discipline imposed by an individual whose job is to flense the flab from a manuscript. But when an author becomes highly successful—a “property” who can be relied upon to crank out best-seller after best-seller, it becomes harder for an editor to restrain an author's proclivity to bloat and bloviation. (This is not to say that all authors are so prone, but some certainly are.) I mean, how would you feel giving Tom Clancy advice on the art of crafting thrillers, even though Executive Orders could easily have been cut by a third and would probably have been a better novel at half the size.

This is why, despite my having tremendously enjoyed his earlier Snow Crash and The Diamond Age, Neal Stephenson's Cryptonomicon sat on my shelf for almost four years before I decided to take it with me on a trip and give it a try. Hey, even later Tom Clancy can be enjoyed as “airplane” books as long as they fit in your carry-on bag! While ageing on the shelf, this book was one of the most frequently recommended by visitors to my reading list page, and friends to whom I mentioned my hesitation to dive into the book unanimously said, “You really ought to read it.” Well, I've finished it, so now I'm in a position to tell you, “You really ought to read it.” This is simply one of the best modern novels I have read in years.

The book is thick, but that's because the story is deep and sprawling and requires a large canvas. Stretching over six decades and three generations, and melding genera as disparate as military history, cryptography, mathematics and computing, business and economics, international finance, privacy and individualism versus the snooper state and intrusive taxation, personal eccentricity and humour, telecommunications policy and technology, civil and military engineering, computers and programming, the hacker and cypherpunk culture, and personal empowerment as a way of avoiding repetition of the tragedies of the twentieth century, the story defies classification into any neat category. It is not science fiction, because all of the technologies exist (or plausibly could have existed—well, maybe not the Galvanick Lucipher [p. 234; all page citations are to the trade paperback edition linked above. I'd usually cite by chapter, but they aren't numbered and there is no table of contents]—in the epoch in which they appear). Some call it a “techno thriller”, but it isn't really a compelling page-turner in that sense; this is a book you want to savour over a period of time, watching the story lines evolve and weave together over the decades, and thinking about the ideas which underlie the plot line.

The breadth of the topics which figure in this story requires encyclopedic knowledge. which the author demonstrates while making it look effortless, never like he's showing off. Stephenson writes with the kind of universal expertise for which Isaac Asimov was famed, but he's a better writer than the Good Doctor, and that's saying something. Every few pages you come across a gem such as the following (p. 207), which is the funniest paragraph I've read in many a year.

He was born Graf Heinrich Karl Wilhelm Otto Friedrich von Übersetzenseehafenstadt, but changed his name to Nigel St. John Gloamthorpby, a.k.a. Lord Woadmire, in 1914. In his photograph, he looks every inch a von Übersetzenseehafenstadt, and he is free of the cranial geometry problem so evident in the older portraits. Lord Woadmire is not related to the original ducal line of Qwghlm, the Moore family (Anglicized from the Qwghlmian clan name Mnyhrrgh) which had been terminated in 1888 by a spectacularly improbable combination of schistosomiasis, suicide, long-festering Crimean war wounds, ball lightning, flawed cannon, falls from horses, improperly canned oysters, and rogue waves.
On p. 352 we find one of the most lucid and concise explanations I've ever read of why it far more difficult to escape the grasp of now-obsolete technologies than most technologists may wish.
(This is simply because the old technology is universally understood by those who need to understand it, and it works well, and all kinds of electronic and software technology has been built and tested to work within that framework, and why mess with success, especially when your profit margins are so small that they can only be detected by using techniques from quantum mechanics, and any glitches vis-à-vis compatibility with old stuff will send your company straight into the toilet.)
In two sentences on p. 564, he lays out the essentials of the original concept for Autodesk, which I failed to convey (providentially, in retrospect) to almost every venture capitalist in Silicon Valley in thousands more words and endless, tedious meetings.
“ … But whenever a business plan first makes contact with the actual market—the real world—suddenly all kinds of stuff becomes clear. You may have envisioned half a dozen potential markets for your product, but as soon as you open your doors, one just explodes from the pack and becomes so instantly important that good business sense dictates that you abandon the others and concentrate all your efforts.”
And how many New York Times Best-Sellers contain working source code (p, 480) for a Perl program?

A 1168 page mass market paperback edition is now available, but given the unwieldiness of such an edition, how much you're likely to thumb through it to refresh your memory on little details as you read it, the likelihood you'll end up reading it more than once, and the relatively small difference in price, the trade paperback cited at the top may be the better buy. Readers interested in the cryptographic technology and culture which figure in the book will find additional information in the author's Cryptonomicon cypher-FAQ.

Posted at May 25, 2006 17:23