The Psychotic State
                              Number 13
                            Sept 30, 1999

             French Fries, Physics, Truth, and Psychosis

My delusions this month are a direct result of a change in job status.
As you probably know, until recently I hovered in the ether, far from
the stresses of clients and billable work.  My recent reinvolvement
in such matters has pushed my addled mind in new directions, no less
crazy, but certainly different.

In a way this is a very personal tale.  To understand it, you have to
know a bit about my life's journey.  Until just a few years ago, I
had a very different career.  I was an academic, a theoretical
physicist.  It is, as you can imagine, a pretty big jump from the
world of equations, experiments, tests, and tenure, to AGENCY.COM, the
epicenter of late 90s e-marketing.  My recent reentry into Real Work
has made me think again about the nature of the divide between my two
careers, and its significance.

Let me put it in the stark terms of a true psychopath: Physics and
Marketing have a different notions of, and uses for, "truth".

The people who I've been working with lately have seen me go into a
mode I haven't really been in for quite some while.  Recently, I've
been in meetings where clients have presented various numbers to me,
results of market research.  Perhaps because figures are something I
understand implicitly, I always start to ask a large number of
questions.  "How was this figure derived?"  "Are you sure that a
correlation coefficient is the right measure?"  "That doesn't seem
right, can you explain it?"

To understand the environment I come out of, you have to picture a
person at the front of the room giving a seminar, with a full audience
of skeptics.  All are mentally asking these questions - and, like me,
they're willing to pipe up and ask them aloud as well.  To outsiders,
it may seem downright hectoring, but it isn't meant that way - or at
least, it doesn't have to be.  There's nothing at all personal in it -
or at least, usually there isn't.  At the end of the day, everyone in
the audience feels that the lecture is at least partly a challenge,
the goal of which is to find something wrong.

In that culture, the "truth" is meant to be sharply delineated.  Every
fact is probed to find out where it begins and ends.  The edges are
sharp.

What fascinates me about science is that what comes out of it - though
intended to be black and white - is often tremendously subtle.  It's
extremely difficult to explain what it is, for example, that makes
quantum mechanics or chaos what they are.  (Buy me a beer some time
and I'll give it my best shot.)  Yet the scientist can have a way of
looking at the world that allows for a tremendous amount of
flexibility and nuance, that are often arrived at only _as a result_
of the sort of relentless challenging that goes on.

As a sort of aside, let me give an example, one that was relevant to
the work I used to do.  What I am about to describe is called the
"renormalization group", and the ideas of it were developed more or
less in the 60s and 70s.  Thinking about it had a pretty big effect on
my career. (Another subject for a beer.)

Suppose we want to explore the surface of a brick wall, and this we'll
do by bouncing rubber balls of different sizes against it.  We start
with tiny ones, a bit bigger than grains of sand.  What we'll discover
is that the wall is quite rough, everywhere you send the ball, it
bounces off at a random angle.  Next we try it with balls the size of
marbles.  What we learn this time is different: the wall will seem to
mostly flat, with regularly spaced curved sections.  Last we try it
with a basketball, and we find that the ball always bounces off at the
same angle it hits.  To the basketball, the wall is perfectly flat.

The use of the renormalization group shows that there is a sense in
which a physical system described by a theory theory is like that
wall.  Looked at at close range, it seems one way, but from further
away it will seem quite different.

The subtlety is this: to the scientist, the statements "The wall is
flat" and "The wall is rough" can both be true at once, and both have
very concrete (no pun intended) definitions.  To predict the behavior
of the basketball, you don't have to understand the details of the
wall - and in fact, you can't really explain what the basketball does
without "integrating out" (as it's called) the complexities.
Therefore, despite the popular perception, the scientific quest is no
longer an endless reductionism, in which the explanation for
everything is forever sought in more and more fundamental theories.
The physics of the world around us has to be understood on its own
terms.  So even though there are very many ways too look at the world,
all valid, they can all be understood in a well defined, careful way.

My psychosis has taken this essay a bit away from New Media, but it
can in fact be brought back to reality.  I think that you can see that
there is a sort of similarity between the relentless questioning that
scientists do, and that a good programmer must always undertake.
"Stare at it until your eyes bleed", I tell people new to programming
who come work for me.  Code is either right or it isn't, and only by
challenging it from every possible angle can one think through the
ramifications of the program's construction.  "What happens if the
user doesn't check any of the boxes on the form?"  "What happens when
this variable is zero?"  "What happens the second time this function
is called?"

Programming is nowhere near as public an exercise as science, so the
"culture of questioning" I was describing above is somewhat less
prevalent.  Code reviews are rather infrequent, formalized procedures,
so most of the challenging is strictly internal.  Nonetheless, at
least for me, the feeling is the same - only through trying to
understand where "facts" fail can one see one's way forward.

In 1995 I left all of that, and I think that I feel a bit like any
other expatriate.  I've learned to speak the language of my new
country - though I'm quite heavily accented - but I still notice
differences from my homeland.

Suppose a burger chain were to retain AGENCY.COM.  Our client wants to
sell more fries, and it's our job to try to get people to buy more of
them.  We would tell our client to try to get the public to "Live the
Brand Online".  Our documents would discourse on our observation that
there are "Many channels, one customer".

What fascinates me is that many of these statements sound like they're
proscriptive or factual, but their use and meaning actually lie
elsewhere, at least sometimes.  These phrases are just as important
at producing a certain mental state in the people who listen to them
as they are at describing any state of affairs of the world.  It would
be silly to try to set up an experiment to measure how many hours per
week someone "lived the brand".  The notion of a roomful of skeptics
challenging statements like these, probing for the limitations and
edges, is a nonstarter.

Such discourse isn't the point.  It's the job of the marketers to
produce mental states in the people who are hungry, and in order to do
that they have to be thinking a certain way themselves.  We aren't
professors, we're coaches, and we have to get our clients to see the
world a certain way.  Describing the actual world is often secondary.

Now I know what you are thinking.  Physicists are famous for their
arrogance, and their belief that everything in the world is somehow a
subset of what they already know.  "So", you're saying, "here's
another one who believes that what everyone else does is somehow
useless".  _My point is quite the opposite._ The culture of
questioning as developed by programmers and scientists is flat-out
terrible at marketing.  The relentless probing to find what's wrong
with any technical proposition puts one in the mindset to expect that
every business proposition will fail.  It often breeds a certain
conservatism and inertia, which is the opposite of what is needed to
get people to buy the fries.  The skepticism, the literalism - so
useful in developing subtlety in describing the world, and finding
the bugs in the program - can crush the life out of the attempt to
more forward in business.

This land of New Media, to which I moved after leaving academia, is
even more different than I had expected.