BKMILPRJ.RVW   990620

"The Millennium Project", Joseph Massucci, 1998, 0-8439-4460-9,
U$5.99/C$6.99/A$12.95
%A   Joseph Massucci joe@massucci.com www.massucci.com
%C   276 Fifth Avenue, New York, NY   10001
%D   1998
%G   0-8439-4460-9
%I   Dorchester Publishing Co./Leisure Books
%O   U$5.99/C$6.99/A$12.95 dorchedit@aol.com
%P   360 p.
%T   "The Millennium Project"

First of all, let's start out with the positives.  The brief
description of the Y2K problem is not too bad at all.  Unfortunately,
it doesn't get much space, since the plot involves manipulating
financial transactions, and blaming the resultant chaos on the Y2K
bug, rather than being a predicament caused by any actual date related
computing errors.  That's a pretty good idea itself, and one that the
RISKS-FORUM digest has discussed.  The danger of "single source"
creating a kind of single point of failure is quite real.  Chemical
and semi-biological computers with massively parallel capabilities are
considered to be strong possibilities for the future, although they
are only in the beginning stages now.  (They are, mind you, unlikely
to run anything like current operating systems.)  There is even one
piece of common sense in the book: when computers start to go wrong,
an old helicopter is put back into service in order to try and avoid
disasters.

However, the preface tell us that "Unless they're fixed, all computer
programs, everywhere in the world, are expected to stop working at the
stroke of midnight, January 1, 2000."  And, leaving aside the fact
that whatever hiccups might occur will happen at midnight on December
31, 1999, the book goes downhill from there.

First of all, this is pure, unadulterated, pulp.  Lotsa things that go
bang (all with brand names and model numbers, of course), steely eyed
heroes, depraved villains, and universally pneumatic heroines and
villainesses.  (Actually remarkably little sex, but this deficit is
made up for by having all but one episode involve violence,
degradation, rape, or deviancy.)  We also have the stereotypical
secret conspirators with identifying marks literally emblazoned on
their arms.

But this review series is about technology, not literary criticism.

Right off the bat we have a plane without non-computer backups for the
computerized flight controls.  A 767, in fact, so I'm sure that Boeing
would be more than willing to explain that modern aircraft do have
such things as batteries, emergency means of getting the gear down,
and minimal but mandated non-computerised instrumentation.  (Oh, and
when all the power goes off, how come the passengers don't seem to
notice that it's suddenly gotten very dark in the cabin?)

A major factor in the nefarious plan is that one manufacturer has
managed to corner the market on military hardware, specifically the
CPUs upon which all manner of equipment, from satellites to aircraft
to ventilation systems, is based.  There is, mind you, the small
problem that military purchasing is now heavily in favour of COTS
(Commercial Off The Shelf).  But then, this may be a good thing for
the author, since he seems to have changed his mind late in the book,
and jumps from shutting down the military to scrambling the world's
financial markets.  At which point we find that all commercial
computers seem to use these same chips.  Intel, Motorola, Compaq, HP,
and IBM would probably be a bit surprised to learn of this.

Returning to the military for a moment, aircraft, and particularly
prototype aircraft, in transit are not shipped with a full tank of
gas, and most especially not with a full load of armaments.  (And that
is a really stupid way to fight on a train.)  Oh, and just so you
know, the US military has its own top level domain.  The Marines would
have addresses ending in usmc.mil, not usmc.com.

How about the government?  The book states that industry is in bad
shape for Y2K remediation because the government can strip businesses
of necessary skilled personnel by offering more money than industry
can afford.  This proposition is, in fact, precisely opposite to the
true state of affairs: governments have been somewhat hobbled by
trained people leaving to pursue more lucrative offers in the
corporate world.

Using a cell phone to get around restricted landlines is a good hack,
but it is *not* hacking.  And most cell phones don't even work from
inside parking garages, let alone mountains.

Encryption is as seriously confused at usual.  At one point a key is
128 bits, at another it is 128 digits, and at yet another it is
described as being like a "lock with 128 ten-digit tumblers."  We also
have a need for the world's most computationally powerful computer to
perform a brute force password attack: essentially counting.  Of
course, the power of such a computer would be completely wasted in the
proposed situation, where the limitation would be the communications
link.  A T-1 is fast, but it isn't unlimited.

But then, as with most authors of this calibre, Massucci doesn't know
arithmetic, either.  We are told that there are 17,901,787,425,874
attempts in a forty minute period or slightly less.  Using the full
one and a half megabits per second (and not waiting for a response
from the host), you would not quite be able to transmit four gigabits
of information in that time.  Being kind to Massucci, and using the
smallest of the various keys, 128 bits, it would take fifty two years
to send 17,901,787,425,874 attempts.  Our heroes are also phenomenally
lucky: the chances of hitting the right password; again using the
smallest, 128 bit, address space; in 17,901,787,425,874 guesses is
about one to 10,000,000,000,000,000,000,000,000,000 against.  In other
words, if you ran something the size of that entire sequence a billion
(American) times per second, then you might get it right in only twice
the age of the known universe.  Half the time.

And one other thing to which I can personally guarantee and attest:
very few computer manuals are published in hardcover.

copyright Robert M. Slade, 1999   BKMILPRJ.RVW   990620