This is important,
and brilliantly written. If you want to understand what's going on in
the mind of Al Qaeda's leaders, and what's going to happen next, please
read it carefully. Thanks to Rayne for the link.
Since I wrote about the work that blackboxvoting
is doing to unearth and investigate possible fraud in the recent
election, quite a few people have sent me links to other investigative
work on this subject. While awaiting detailed information from election
officials, some mathematicians have done some fascinating analysis that
shows, among other things, that while the hated touch-screen voting
machines produced results in Florida not inconsistent with voter
registration, the Diebold Op-Scan-Precinct machines in that state
produced results, at least by the time their data was uploaded electronically to the central vote tabulation machines, that are wildly different. In many counties
80% of the registrants were Democrats while 80% of the votes were cast
for Bush. The most obvious explanation for this is the Dixiecrat
factor: Many older, conservative Floridians registered as Democrats
when Southern Democrats were frequently politically to the right of
Attilla, and never bothered to change when they were outflanked on the
right by state Republicans. The same thing occurred in 2000 in many of
these same counties, though that may just mean the fraud has been going
on for years. The Diebold CEO, after all, is a devout Bush supporter
who promised to do whatever he could to deliver votes to Bush in his
home state of, you guessed it, Ohio. Although there are real concerns
that the lack of a paper trail will never allow the suspicions of
massive voter fraud to be resolved, there is some hope that when
election officials produce the requested information, an investigation
will be able to put the fears to rest. After all, we're not talking
about a few hundred votes difference this time, but over a hundred
thousand.
Of course the best answer is nationwide standards that require
hard-copy verifiability by both the individual voter and by the
auditors of the tabulation. But in the absence of that, why not just ask the people?
Get the pollsters to blitz some of the counties in Florida and Ohio
where the results appear most suspicious, and just ask people (a) did
you vote and (b) who did you vote for. In fact there are some
overwhelmingly Democratic counties in Florida where the total number of
votes was under 2000, where Bush got 80% of the votes. You could
probably canvass the whole county in a day and know beyond a shadow of
a doubt if the Diebold machines were hacked or not. My guess is that
not only would this put to rest the conspiracy theories, it would allow
the Democrats to learn first hand why suburban voters, who make up 50%
of the US voting public, overwhelming rejected their message last week
(Bush's largest plurality in the country was in Orange county, California).
And if I'm wrong and such an intensive poll does indicate widespread
fraud, we've got the makings of a revolution. Either way, the Dems
can't lose.
Last week's New Yorker
featured an article by Elsa Walsh entitled "Learning to Spy". It was
about Maureen Baginski, the woman who's been brought in to clean up and
modernize key functions of the FBI. She puts the FBI's intelligence
mandate simply:
"You need something, you go get
raw material and you add value to it. You put out a product and you
keep adjusting, based on the feedback that you get. That's really all
it is."
So why isn't the FBI doing this? According to Ms. Baginski, most FBI
agents were trained and instructed to take 'intelligence gathering' too
literally. They aggregated data just in case it was useful or needed,
often without doing anything with it. Her motto for reform: Hunt, Don't Gather; Disseminate, Don't Just Aggregate.
Because they focused on gathering intelligence, the FBI ended up with
mountains of information, so much that they didn't have the time or
resources to transcribe it all, so some of it just got erased or thrown
out without even being examined. And this overload of information
wasn't shared or coordinated with other intelligence groups, so there
was no context for assessing its meaning, its importance. That was one
of the reasons for the failure to prevent 9/11. The information was all
there, but no one distilled and organized and analyzed and integrated
it, the various intelligence groups didn't distill and share what they
knew, and no one put two (sharp increase in enrolment in US commercial
flight training schools by people from Arab countries) and two
(increase in frequency of visits to the US by known associates of
CIA-operative-turned-radical Osama bin Laden) together.
But there was a second behavioural problem. Too many of the agency's
employees were content to just gather intelligence, not "go get" it. In
business circles, the gathering of information from readily available
sources -- the Internet, published documents, recordings, files etc. --
is called secondary research. It's what librarians used to do, and what
in many organizations can now be done by anyone. Going and getting
information, by interviewing people directly or obtaining it from
physical observation, is called primary research. It's harder, and
takes what I call 'know-who' (awareness of who to interview or get
permission from, and how to contact them, and a 'great Rolodex') plus
persuasive and interviewing skills (in business, the more extreme
method of conducting primary research -- interrogation -- is thankfully
not called for).
In today's world, there is so much information out there that settling
for secondary sources is tempting. You could spend hours studying Bin Laden's latest tape, for example, to get clues as to his whereabouts, his intentions, his motivations. Without leaving your easy chair.
Hunting for intelligence on these things from primary sources is
harder, more dangerous, and more time-consuming. Contrast the
armchair/laptop intelligence gathering with the intelligence hunting of some of the New Yorker's own fine journalists: Lawrence Wright, who talked
with Saudi princes, journalists, religious leaders, opposition forces
and the man on the street in Riyadh; Jeffrey Goldberg, who
criss-crossed the new wall between Israel and the West Bank of
Palestine to speak
with settlers, pacifists, leaders and angry adversaries on both sides
of the intractable conflict; and Seymour Hersh, still digging into
inconsistencies, interviewing Iraqis from all factions face-to-face and using contacts, guile and dogged persistence to unearth the truth about what has happened in Iraq.
When you read the first-hand
accounts of these investigative reporters, you really begin to
understand what is happening in the Mideast, and why. These reporters,
courageous and competent and connected, but no more so than what we
would expect from professional intelligence agents, surely have a deep
appreciation for what is happening and how to resolve it. But no matter
how much technology and psychological study and fact-checking you apply
to Bin Laden's tape, you really can't do more than guess what's
happening in his world from studying it. That's the difference between
primary and secondary research.
I'm not suggesting that today the FBI and the CIA (unlike the National
Security Agency, whose mandate appears more specifically focused on
analysis of secondary source information) just sit around reading stuff
on the Internet, listening to wiretaps and watching surveillance videos
all day. But Elsa Walsh isn't the first to suggest that a
disproportionate amount of intelligence agencies' time is spent just
gathering intelligence, and an insufficient amount hunting for the
really valuable intelligence that you can only get directly from the
horse's mouth, adding value to it by distillation, analysis, providing
insight and context, and then sharing, coordinating and disseminating
it to others who can help add meaning to it and make it actionable. Knowledge is information that you can do something with.
The same could be said for intelligence-gathering in business. Ms.
Baginski's quote at the top of this post could just as easily apply to
the job of business researchers and analysts. In fact, the discipline
of 'competitive intelligence' prides itself on tapping many more
primary sources than most other types of business research. But 'CI' is
usually narrowly focused, as its name applies, on getting information
about the employer's competitors. Likewise, the 'primary research' done
by investment brokers is conducted through their target's PR department
and other selected executives coached on what to say and what not to
say, rarely much more informative than reading the company's press
releases. Overall, business is pretty lousy at primary research, so
it's not surprising that they're usually out of touch with both
employees and customers, and hence make so many bad decisions.
The real truth about what is going on in a business is best obtained by
talking to front-line employees and to customers, using a technique
that Knowledge Management guru Dave Snowden
calls 'cultural anthropology'. This involves employing a series of
methods to earn, legitimately, the trust of the people you are
interviewing or observing, so you hear and see what is really
happening, not what the employees or customers think you want to hear. That's knowledge you'll never find in any database.
Tapping into that knowledge can be done in two ways: By conducting
one-on-one interviews with front-line employees, perhaps during the
process of providing those employees with Personal Productivity Improvement assistance, or, by canvassing the employees or customers for their suggested answers to business problems to aggregate The Wisdom of Crowds.
Whether the focus is on solving global political or security problems,
or solving business problems, the lessons are the same: Hunt, don't
just gather; Share, collaborate and disseminate, don't just aggregate;
and "go get" intelligence direct from the people who know what's
happening and why, and who can provide context that is missing from
secondary sources.
The cost of doing so is higher, and it's harder work that requires
considerable 'know who' and 'know how' skill. But it's the only way to
effectively reduce the cost of not knowing -- a cost that, as evidenced
by the failure to prevent the 9/11 attacks or to detect the fraud that
led to the collapse of Enron, can be catastrophic.
MADE IN CANADA
trust your
instincts 
Last week's 
