* * * * *
If this is realistic, I'd hate to see what an unrealistic techno-thriller
would be.
> “_Digital Fortess_ [1] is the best and most realistic techno-thriller to
> reach the market in years … A chilling thrill a minute.”
>
> –The Midwest Book Review
>
My friend Bunny had mentioned reading _Digital Fortress_ [2] and finding it
quite good. I had my doubts, given my feelings [3] towards Dan Brown's The Da
Vinci Code [4] but I said I would read the book (which I borrowed from Bunny)
and take notes.
Well, I'm about a quarter the way through the book and boy, do I have notes.
Aside from the poor writing style, radical shifting of viewpoint, choppy
sentances and short chapters (I'm on page 123, in which Chapter 29 starts),
I'm alternating between laughter and screaming at the book.
Often times simultaneously.
His recounting of recent computer history is anywhere from five years too
early (widespread interest in public key cryptography started in 1992 with
the release of Phil Zimmerman's PGP (Pretty Good Privacy) and not in 1987) to
nearly 15 years (he has the public Internet as being available in the early
80s when it was 1994 when the commercial restrictions were lifted and the
Internet because publically available on a national scale) and at the same
time, he has the NSA (National Security Agency) [5] with quantum computing,
which is still in its infancy (and speaking of the NSA (National Security
Agency), one of his major characters supposedly never heard of the agency
even though it was outted in 1982's _The Puzzle Palace_ [6]—sheesh!).
Then there's the little fact that Dan Brown has absolutely no clue about
programming. The NSA's prize computer, TRANSLTR, a custom built machine with
3,000,000 processors to brute force encrypted messages, has a virus checker!
As if! There's no way a virus could run on the thing unless it was
specifically written for the computer and this thing doesn't officially
“exist.” Even if the thing was built using off-the-shelf processors like the
Intel Pentium, a Windows based virus wouldn't get very far, not finding
Windows on the machine (I suppose—it's never stated what operating system it
runs, but I can't see Windows scaling to a 3,000,000 processor system).
Then there's the email tracer a main character uses to track where an email
messages finally arrives. I'm sorry, but if I received such an email it
wouldn't “automatically” detect the IP (Internet Protocol) address and scurry
back to the NSA without a trace; my email client doesn't execute anything
coming in from email (doesn't even understand HTML (HyperText Markup
Langauge)) and there's certainly nothing in SMTP (Simple Mail Transport
Protocol) that even mandates a return receipt.
Sorry Dan.
And the whole premise of the book, that the NSA's little computer TRANSLTR
can't crack a message because of a newly found “unbreakable” encryption
system, is bogus because there already exists such an ecryption system!
Obviously Dan Brown never heard of one-time pads (or if he did, ignored them
outright).
A one-time pad where you generate purely random data (preferably from some
external source like neutron decay or static from a dead television [7]) and
use as many bits of said random data as the encryption key as there are bits
in the messsage. What you get out is a stream of random data (obviously the
recipient has to have the randomly generated key to decrypt the message but
that's inherent in all non-public key encryption systems).
So I'm having fun with this book, taking notes in order to rip it apart.
More on this when I finish.
[1]
http://www.amazon.com/exec/obidos/ASIN/0312995423/conmanlaborat-20
[2]
http://www.amazon.com/exec/obidos/ASIN/0312995423/conmanlaborat-20
[3]
gopher://gopher.conman.org/0Phlog:2006/05/20.1
[4]
http://www.amazon.com/exec/obidos/ASIN/1400079179/conmanlaborat-20
[5]
http://www.nsa.gov/
[6]
http://www.amazon.com/exec/obidos/ASIN/0140067485/conmanlaborat-20
[7]
gopher://gopher.conman.org/0Phlog:2006/02/01.3
Email author at
[email protected]
