Re: [SG2/DS] Intel

From: "John Atkinson" <johnmatkinson@yahoo.com>

> That's the funny thing.  They've got a lot of really
> good raw data.
> 
> It's the conclusions they reach from that data that
> makes me call 'em weenies.
> 
> Ivory tower idealists, actually.

Much as I disagree with a lot of their conclusions....
Gathering data is relatively easy. Analysing it is the
hard part. Taking 2+2 and making 5.

My own experience has been limited to semi-commercial
military analysis. Basically, given a whole heap of
unclassified data on possible opponents, decide what
equipment is required to beat them, so we can then make
said equipment.

Our proposal is then submitted (along with competing
proposals) to the *real* analysts, who have access to
data we don't have.

This process involves a LOT of talking to serving and
ex-serving military personnel. Otherwise it's far too
easy to make something the Engineers and Scientists
will love, but will fall apart at the seams in combat.

What I'm trying to say is that military anlysis is
hard: it's an art, not a science, with a lot of
heuristics ( ie guesswork and hunches ). The people who
are good at it can't always, or even usually, give a
closely-reasoned analysis of why they came to the
conclusion that they did on such contradictory and
skimpy data: they just point to their track record of
success. Some of them are so good, it's scary - even to
them. But they don't like letting the troops in the field
know just how thin a justification they have for their
conclusions: it would sap confidence. Whereas the Intel
world is full of weenies who have much more plausible
and well-reasoned ideas, that are just plain wrong. And
so people die, because someone was too afraid to say
what the best guess was, giving the one with the best
reasoning instead.