Interactions - March/April 2010

Shaowen Bardzell and
I are interested in your
opinion. Write to us and
let us know what you
think: Are we as design-
ers morally bound to
understand and agree
with the messages that
are embedded in the
products and technolo-
gies we work on? If so,
how do we ensure that
we are even aware
of them? Or submit
a paper to a special
issue of Interacting with
Computers entitled
“Feminism and Human
Computer Interaction,”
due out in early 2011.

March + April 2010

interactions

gets a balanced experience means interface and interac-tion-design changes, but it also has implications for the ways in which the machine learning behind the recommendation algorithms is “tweaked”—and thus gender differences in information processing get inscribed into the engine itself.

Organizational and institutional factors. Some argue that this is an institutional problem, and I am wont to agree with them. The lack of women in powerful product/technology-design positions partially maintains the dominance of mainstream (or “malestream”) design practices that assume products are gender neutral, while considering women users as an afterthought. Certainly, most product-development teams are largely male. So, even without intending to, design decisions for an application will likely skew toward that which is more understood, comfortable, or familiar for the team—that is, a male view of how the world works and how the product will fit into that world.

This may seem far-fetched,
but it is not. Early voice-
recognition software did not
recognize women’s voices
because there were no women
available when training the
recognition algorithms, and
the all-male team members
did not think to make sure to
find women during the train-
ing process. Sarah Jain’s study
of car airbags points out how
the initial design of airbags
assumed an adult-male form.
As a result, women and chil-
dren were routinely injured
when airbags deployed. In her
2006 book Injury, she points out
that “airbags provide an excel-
lent way to see how bodies
are built into both social and
technological systems. The air-
bag…was calculated to fit men
of average height and weight
without seatbelts on.” She goes
on, “Consequently as a cor-
relate of height, 42 percent of
women compared to 24 percent
of men received facial injury
from the airbags.” For smaller
people, including children,
the injuries were more severe,
including amputated hands,
broken limbs, and fatal head
injuries. No one intended this;
they simply designed for a “one
size fits all” strategy—the one
size just happened to be a man
of average height and weight,
and that assumption has not
been questioned. Seatbelts don’t
do so well in this assessment
either—many are not adjustable
and are not designed for small
people. Even though unintend-
ed, these are indications of neg-
ligence born of a setting where
to consider gender is akin to
being a bore, or adding work,
or being a whiner. And no,
abstracted personas are usually
not compelling enough to give a
team of male designers a really
good feel for what being female
is really like.

ABOUT THE AUTHOR Elizabeth Churchill is a principal research scientist at Yahoo! Research leading research in social media. Originally a psychologist by training, for the past 15 years she has studied and designed technologies for effective social connection. At Yahoo, her work focuses on how Internet applications and services are woven into everyday lives. Obsessed with memory and sentiment, in her spare time Elizabeth researches how people manage their digital and physical archives. Elizabeth rates herself a packrat, her greatest joy is an attic stuffed with memorabilia.