|
Live
Wire Links
|
PCs:
The Hope and The Hype
Now
here's a news flash: A recent British survey
showed that employees lose up to three weeks
of working time a year to PC malfunction
or misuse. Also not surprisingly, seven
out of 10 surveyed said that when their
PCs fail to work ... they're frustrated!
In
October, we began talking with Dr. Ben Shneiderman
about how today's complicated, feature-heavy
computers and software make a mockery of
the term "user friendly." Dr. Shneiderman,
a champion of simplicity, is head of the
Human-Computer Interaction Laboratory at
the University of Maryland, College Park.
This month, we conclude the conversation
by sorting out the industry's hopes from
its hype.
Q.
Last month we talked a lot about how things
are not as good as they could be. It's not
all bad news. What do you think has improved
in PC ease of use over the past few years?
A.
The major improvement is the advent
of the World Wide Web, which provides
remarkable access to remarkable resources.
As for the computers themselves, machines
are faster and screens are bigger
... but the user interface is not
much improved over what appeared on
the Mac of 10 years ago [it was Apple's
best-selling desktop PC which introduced
the icon-laden graphical user interface].
In
the future, the keyboard will stay with
us. We'll perform more tasks by selecting,
dragging, pulling and grasping, but the
keyboard is good for originating textual
materials. And in displays, we'll see larger
but flatter screens.
Q.
What advances in the computing field are
encouraging?
A.
I've been happy to see: and help spur: the growth of information visualization
as a topic. [Presenting information better]
can help people find information in digital
libraries, help them understand financial
information more fully. I'm talking about
presenting thousands of data points, not
just tens. At the HCI Lab, a project called
"Lifelines" is the most exciting thing we're
doing now. It presents medical histories
as graphical records: and we think it can
change the nature of medical delivery. Today,
doctors rarely look at a patient's history;
it's hard to follow. A system such as Lifelines
can make it accessible.
Q.
What advances do you find overly hyped or
discouraging?
A.
First, "agents," the supposedly almost magical
software that will carry out your tasks
for you perfectly. Then, speech interaction.
Voice recognition is important, but it's
not as useful as visual interaction. Voice
recognition has limited application and
only partial accuracy, and also, speaking
interferes with problem solving. When we
look at things visually, we use a different
part of the brain: and we can continue
to solve problems.
Q.
In a recent essay, you said that it's silly
to try to make computers lifelike: that
what's important is to make them more useful.
Can you explain this more fully?
A.
It's the fantasy of many designers to make
PCs lifelike, but it's a dead end. First
these machines or devices are cute, then
they're silly, then by the third time it's
an annoying distraction. People want to
be in control; they don't want a relationship
with a machine. History and research show
us that people reject these things.
Q.
Is your field, Human-Computer Interaction,
valued in the computer industry, or do companies
merely pay lip service to it?
A.
It's increasingly valued. We see more people
coming out of programs such as ours getting
jobs in the field. It's still a struggle,
but our department continues to add courses.
I see the field as a healthy adolescent
that still has a way to go.
Q.
How might human-computer interaction research
change the way we measure what's good and
bad in our computers?
A.
When we look at the ad, we see that
hardware developers push speed. But
users: consumers -- want shorter
learning times and faster task performance,
measures that compare, say, to a car's
mile per gallon or trunk space. In
software, we might see newer measurements
such as "minutes to learn" or "seconds
to accomplish."
The
field offers a golden opportunity for psychologists: never before have we had such complex
problems amenable to study, and the chance
to influence important commercial practices.
Q.
How would you like to see the computer integrated
into everyday life? What would be your dream
setup in terms of input and output devices,
processing and functionality?
A.
I'd like to see something more graceful,
and I'd to see more "interoperability" among
different networks: so they can handle
different information-transfer speeds, and
offer end-to-end privacy and security at
low cost. It's still frustrating getting
things into and out of machines, and across
a network of machines. It should be easy
to get stuff back and forth.
|
