Quack, Quack?
By Donald Christiansen
“If it walks
like a duck and quacks like a duck, it must
be a duck.” So goes the popular paraphrase
of the classic quotation attributed to John
Whitcomb Riley.
Wait a minute.
What was very likely true in days of yore is
no longer the case in the oxymoronic world
of virtual reality. Consider the incident
related by MIT professor Sherry Turkle of
sailing on the Mediterranean when her
daughter was eight. Spotting a sea creature
the youngster excitedly remarked “Look,
Mommy, a jellyfish! It’s so realistic!” Her
standard of comparison, of course, was the
many aquatic creatures she had seen on the
computer screen. The real world had become
secondary to its simulation.
In her book Simulation
and Its Discontents, Turkle goes on to
describe situations in which simulation
might beguile but mislead—particularly in
the case of architects and design engineers.
Architectural students in
the 1980s were encouraged to rely on
“defaults,” predrawn architectural elements
that avoided the need to design from
scratch. Even today some veteran professors
and senior architects feel the overuse of
defaults discourages individual ingenuity
and subverts creativity. The process of
moving ideas from the designer’s brain to
the hand (creating architectural sketches
and detailed renderings) and then to the
physical model has been disrupted. Today the
pencil has pretty much disappeared as an
important tool of the architect.
Some experienced
designers decry the black-boxing of analytic
and design elements in computer programs.
Such “cook-book” techniques, they say, mask
problems and challenges that the designer
might otherwise confront and overcome with
ingenious new solutions.
Today’s sophisticated
simulation techniques can emulate reality in
such extreme detail as to sometimes seem
more “real” than that encountered in nature.
Such a “hyperreal” simulated design might
not work in practice. Architectural critics
have noted that some buildings conceived in
this way should not have been built.
Academic
Simulation
In many university
laboratories, hands-on experiments have been
phased out in favor of simulated
experiments. Chemistry students at MIT in
the 1980s were divided on the values of
simulated vs “wet” experiments. Simulations
are less odoriferous, and several students
explained why they preferred the simulated
experiments. Simulation enabled them to see
things “actually happen,” they said. Of
course it does not. A more accurate
statement would have been that simulation
helped them better understand what actually
happens. It does so rapidly, smoothly, and
without the chance of experimental error.
Not all physics
professors are as enamored of simulated
demonstrations, feeling that the very notion
of physics demands direct experience with
physical things, including the intimate
knowledge of laboratory equipment and
instrumentation.
In the 1980s, thought by
some to represent simulation’s dark ages,
professors often identified “simulation
free” zones. Civil engineering faculty
warned of the dangers of trusting computer
programs for structural analyses. Others
cited the dangers of computer pedagogy that
overlook the factors of estimation, scale,
and error.
Engineers and Simulation
While the foundations of
civil and mechanical engineering were based
on physical structures and their graphic
depiction, electrical and computer engineers
have always been much more comfortable with
abstractions—for example the machinations of
the invisible and often elusive electron. I
recall an incident when, in an undergraduate
circuit-design class, as a few of us were
attempting to understand the underlying
physics of a particular circuit action, one
of the top students said “I never worry
about that. If it checks out mathematically,
I’m satisfied.” Historically, EEs could
readily embrace the capacitor, the vacuum
tube, the transistor, and ultimately the
computer chip as building blocks, and thus
the evolution from breadboard to CAD was
accepted as a natural progression. The
identification of simulation-free zones may
never have existed for us as it did for
other design disciplines.
Even so, some of the
concerns relating to design by simulation
may be worthy of our attention. Professor
Turkle observes that simulation produces
paradoxical effects. It offers the
possibility of multiple design iterations,
yet in practice it often turns out that “the
first idea wants to be the last idea.” This
is particularly true in architecture. There
is also a difference between design logic
and computer logic, an MIT student noted in
2005, stating that the codification instinct
intrinsic to computer logic inhibited his
creative (design) thinking.
Contemporary scientists
and engineers have few reservations about
the advantages of simulation, its
shortcomings notwithstanding. Four distinct
areas that benefit from simulation are
design, operations, research, and
entertainment. In the realm of design,
simulation is based on the known physical
world and thus can be exploited with great
confidence. Code can be written that
accurately reflects established design
techniques, avoids re-invention, saves time,
and cuts costs. The same advantages apply to
operational simulation. Flight trainers are
a well-known example. In another, the U.S.
Navy is developing an LCS (Littoral Combat
Ship) training simulation, so that the
actual ship can be actively deployed while
replacement crews are trained ashore, saving
fuel, reducing transit time to operational
locations, and more efficiently deploying
new crew members. In this latter example,
however, one retired Navy captain cautions
that an important “socialization factor”
cannot occur until the new crew is actually
aboard the ship and working together.
In the case of the Apollo
lunar landing, the astronauts had undergone
training in a months-long series of
simulated failures prior to liftoff. During
the actual landing process, contact with the
ground was intermittent, requiring astronaut
Buzz Aldrin to make manual adjustments to
the automatic antenna control system. In the
course of receiving limited information on
Aldrin’s actions, one ground controller
could not resist remarking “This is just
like a simulation!”
The area of simulation in
which much development is still needed is
that of exploratory science. In the life
sciences, simulated structures are
themselves experimental models, not to be
trusted except as steps toward the better
understanding of reality. Therefore young
investigators (and sometimes even more
experienced ones) must be reminded of this,
or they begin to believe that screen objects
are real. They are abetted in this misplaced
confidence because they can so readily
manipulate the structures, and they are no
longer familiar with the shortcomings of the
underlying code.
In the entertainment
field, on the other hand, the ability to
misrepresent the real is widely exploited,
but I will leave that for another time.
Meanwhile, it seems safe to conclude that
avatars, sociable robots, hyperreal
jellyfish, and quacking but featherless
ducks are with us to stay.
Resources
Turkle, S., Simulation
and Its Discontents, MIT Press, 2009.
Baudrillard, J., and J.
Nouvel, The Singular Objects of Architecture
(trans. R. Bonanno), University of Minnesota
Press, 2009.
Baudrillard, J.,
Simulacra and Simulations (trans. S. F.
Glaser), University of Michigan Press, 1994.
Erwin, S. I., “Sailors
Move From Classrooms to Shipboard
Simulators,” National Defense, December
2004.
Ludwick, P.M., “New LCS
Training Process Allows Crew Members to Man
Ship Anytime, Anywhere,” 2006 (U.S. Navy
Official Website).
Mindell, D., Digital
Apollo: Human and Machine in Spaceflight,
MIT Press, 2008.
Pine, A.,
"Simulation: The (Almost ) Real Thing,"
Proceedings U.S. Naval Institute, Dec.
2009.
Holmes, N.,
"The Varieties of Reality," Computer
(IEEE), Jan. 2010.
