Suddenly, Mysteriously, Again - The Toilet Stopped Working
I finished my article, and sought refreshment, my assignment finished,
saved, backed up in case lightning should strike. Life was wonderful.
I wandered back to my desk and started a game of solitaire - my
tiny little diversionary reward.
In the course of play, I slowly became aware of a continuous, undiminished,
rushing whooshing sound of the toilet, filling and filling and filling.
Suddenly, once again, the stupid thing is broken.
It started about a week ago, maybe two. I ignored it, just jiggled
the chrome thing, hoping the problem might go away. I got that idea
from Lewis Thomas, who probably - if he were alive today - would
not appreciate my application. He wasn't keen about magical thinking,
but he acknowledged its existence.
In an article called Medical Lessons from History, Lewis said,
"The long habit of medicine, extending back into the distant
past, had been to treat everything with something, and it was taken
for granted that every disease demanded treatment…."
Who among us hasn't tried to fix something, and made things worse?
Lewis went on to describe the discovery that diseased people got
better faster and more reliably when they were not subjected to
the treatments of the time. That many treatments of the time - giving
patients heavy metals, or taking blood - are now known to kill or
severely damage almost anyone - diseased or not.
Somehow in my sometimes magical mind, I connected ideas. What did
not work in medicine might also not work in other technical fields.
In the real world, computer and other types of technical problems
sometimes - quite often actually - just go away. In a busy world,
doing nothing is sometimes the simplest thing to try. "First,
do no harm" gets translated into "First, do nothing."
One side of my brain rejects this thread of thought. I am an engineer,
the son of an engineer. I believe every system failure must have
a real, physical, scientific cause. Get to the bottom, find the
rotten core, and the problem will be solved.
But I discover I have a problem with such a purely rational view
of problem solving; for when I consider many past experiences, it
just isn't true. For example, when I was in the U.S. Air Force,
surrounded by dedicated, competent, quality minded people, sometimes
problems with aircraft simply went away. They drifted in, caused
distress, and disappeared forever, like ghosts in the machine.
One could argue, "It had to be something. There was a wire
that was crimped improperly, and the right amount of G-forces occurred
to cause a failure. Or there was moisture in a plug that evaporated.
Something had to go wrong." It's impossible to argue. The potential
for failure in complex systems is endless, possibly infinite. When
things are at a certain complexity level, it's amazing they work
at all. Maybe we have become so good at making complex things work
so well, most of the time, that we have ceased to be properly amazed.
What I'm suggesting is that there is a practical viewpoint, grounded
in reality, but difficult to propose. Based on this idea, it can
be rationally argued: "So what? Who cares? The problem went
away, the system now works, and it is not and never will be worth
the trouble to find out WHY it happened in any model of efficient
productive behavior. Go fly the now functioning machine and blow
up something menacing to children or downright Un-American."
This is NOT something you'll find in any U.S. Air Force manuals,
I can assure you. Unless things have changed drastically in the
past few decades, nothing would ever fly unless some brave soul
wrote, "Problem solved. Connected wire A to Point B. OK to
fly the machine." And signed their name.
There is something within our so-called modern, scientific culture
innately unable to deal with totally in-your-face reality-based
ambiguity when discussing technical matters. We are open to almost
any theory except one: that there is no answer within our present
grasp. In my experience, however, this is sometimes the only honest
answer.
I can imagine the totally left-side-brain dominated part of my
audience going absolutely ballistic right about now. With good reason,
I am sure. Still, I feel sure of my turf.
I would love to see the US Air Force establish a research project
that attempts to find the root cause of every single system failure
- without fail. Even if this means taking the whole contraption
apart and scrutinizing every molecule on every IC chip with an electron
microscope, in real time.
I can't prove the outcome, but I submit the project will fail.
There are too many parameters. They won't have enough time and money
to get to the bottom of all things. Not in complex systems like
we have today. It might be possible to do a project like this on
a computer mouse and a pad, if children are not allowed in the laboratory.
I'd post guards, just to make sure.
My interest is not noisy, cranky toilets, or the aircraft that
fly to protect those who use them, but in computers. And in all
the situations I have been in, the most difficult to explain is
when the system crashed, we simply don't know why, we have checked
everything we know how to do, and it is very unlikely to ever happen
again. It went away, mysteriously. And chances are - it won't happen
again.
We hate that. But it happens.
By David Pickens 1/27/00
