"Turning the Clock Forward"
Dr. Rita R. Colwell
Director Designate
NATIONAL SCIENCE FOUNDATION
Computing Research Association Conference
Snowbird, Utah
July 27, 1998
(As delivered)
Good afternoon to all of you, and thank you, Ed for
the kind introduction. It's nice to be introduced
by someone from the University of Washington, my alma
mater. I'm pleased to tell you that Ed Lazowska was
selected to receive the 1998 University of Washington's
Outstanding Public Service Award for, among other
achievements, having helped the Seattle Public Schools
develop technology standards, raise funds, and get
more than one-third of the schools connected to the
Internet. I would also like to thank Bill Aspray for
his long-standing leadership of CRA. It is truly a
pleasure to be with you today. Besides giving me the
chance to escape Washington's murderous heat for such
lovely surroundings, I'm genuinely delighted you've
asked me to come to speak at this juncture. I'm really
just beginning to get my feet wet as NSF director
designate. I can assure you that I'm looking forward
to "plunging in" completely to this exciting position.
This is my first speaking engagement wearing my new
NSF hat, and deliberately so. It is both inspiring
and appropriate that I have this chance to meet with
such a distinguished group of leaders from the computer
science and engineering community. I know a lot of
you from other contexts and I look forward to getting
to know and working with more of you.
In many ways, the history of computing is an astonishing
and very modern tale. So much has been telescoped
into such a short time, compared to the centuries
of stately development that are typical of the more
traditional and older sciences, like my own discipline
of biology. Of course, biology's course is now inextricably
woven together with the path of computer science.
Both are transforming at the same dizzying pace. So
often we hear terms like the "explosion" of the Internet,
and we know that computing has changed virtually every
facet of our lives. (As one bellwether of the times,
even the venerable Mary Worth on the comics page has,
however reluctantly, just gotten her own PC!)
When I think about the history of computing, I'm reminded
of another woman, one with three dimensions instead
of the two occupied by Mary Worth. I'm talking about
that remarkable computing pioneer, the late Grace
Hopper, whose career spanned much of the 20th century
and who now lives on thanks to the CRA symposium series
in her memory. My students and I had the good fortune
of meeting Grace several years before her death. She
was indeed a character. My students had one of the
"Grace Hopper nanoseconds" pinned to the laboratory
bulletin board. She lived through and helped move
along the evolution from primitive programming to
modern data processing.
Among her many accomplishments, we celebrate Grace
Hopper for her vision that the development of high-level
programming languages would open up the power of computing
beyond the refined world of mathematicians. Not only
that, she served as a messenger for technology transfer,
the process that is now the darling of the business
world, persuading business managers to use the new
computer languages such as COBOL.
On the lighter side, Grace is also credited with having
discovered-and captured-the first computer "bug."
This was a moth that unluckily landed inside her computer
and that she taped like a trophy into her computer
logbook!
As I look back on my own beginnings as a researcher
using computers, I can vividly relate to Grace's capture
of that "bug" - the moth. In fact, I did my own Ph.D.
thesis on another kind of "bug"-bacteria in marine
animals-at the University of Washington. I wrote the
"little bug program" for handling bacteriological
data which we used to classify marine bacteria, especially
those living in association with marine animals.
I wrote the program for the IBM 650 computer. We used
it to handle what we thought was a large amount of
data gathered for several hundred bacterial cultures.
This was the first use of computers to classify marine
bacteria, or, for that matter, any bacteria from the
environment. Many other microbiologists have worked
on the problem since then and taken it well beyond
these rudimentary origins. In fact, computer identification
of microorganisms is now standard in hospitals around
the country. However, the coding scheme we developed
for bacteriological data remains in use today. That
IBM machine I used had been installed in the attic
of the chemistry building, Old Bagley Hall at the
U. of W., the only space available at the time...
and we graduate students got to use it between the
hours of two and four a.m. Some of you will remember
that we actually had to wire the boards ourselves
in those days and we wrote in machine language. That
was a time when you did everything yourself - from
collecting the data to running the computer. (And
an IBM 650 is now on display at the Smithsonian!)
To recall Grace Hopper's words, "Life was simple before
World War II. After that, we had systems." Well, my
computer program, written in the sixties, is absolutely
sophomoric by today's standards, even though it was
a really big deal at the time. I still have two of
the punch cards pasted between the pages of my thesis.
Who knows, they may be valuable as a rare antique
some day!
It is inconceivable now for me to imagine doing my
research in the decades since my graduate student
days without computers. My work on environmental factors
associated with cholera epidemics would be impossible
without the power of computing. My students and I
use remote sensing and computer processing to integrate
data from many disciplines: oceanography, epidemiology,
ecology, microbiology, clinical medicine-the list
goes on. We are currently developing models to enable
prediction of conditions conducive to cholera epidemics...that
is, to allow proactive, not just reactive, measures
against cholera-a possibility that could never be
hoped for without the advances in information processing.
Well, I use this example to be instructive as to why
computer science and engineering stands in singular
stead today, as the science of creating, processing,
and transforming information. It's truly breathtaking
to note the speed with which ideas in computer science
spin out into the marketplace-without parallel, if
you'll pardon the expression. This translation happens
in perhaps a third or a quarter of the time the process
takes in most other disciplines. Computer scientists-you--are
a powerful economic force.
There is a rather complicated graphic from a National
Research Council report on high performance computing
that helps to visualize how government-sponsored R&D
stimulates commercial innovation.
I've heard this figure called "the tire-track graphic"
for its intersecting patterns. But it actually illustrates
its point quite effectively. It shows how technologies
such as networking, workstations, and parallel computing
developed. The process was anything but linear. They
evolved through repeated exchanges of ideas and people,
through synergy, among industry, universities, and
government. And several of these "tire tracks" mark
the pathways to billion-dollar industries.
A good example of that is the development of Mosaic.
It got its start at the National Center for Supercomputing
Applications, the NSF-supported center at the University
of Illinois. Netscape Communications soon followed,
and gave us a really good example of how research
can produce unexpected outcomes in the marketplace
with great rapidity.
The information sciences also stand out in their role
as the mortar, the cementing material for the entire
edifice of modern science. NSF, of course, supports
computer science as a fundamental field, along with
the other basic disciplines.
At the same time, its integrative capability is enormously
invigorating. I see possibilities for flagship projects
at NSF that similarly draw upon many disciplines.
We can expect these to ignite the public imagination,
advance our knowledge about the world, and bring societal
benefits. Computing and communications will have a
highly central role in this vision.
NSF has named its current broad effort to derive knowledge
from access to information KDI, for Knowledge and
Distributed Intelligence. As you know, this effort
spans all of the Foundation's directorates. Its aim
is to create networked systems that can make all kinds
of knowledge available to anyone at any time. This
is not only an ambitious objective-it could even be
considered somewhat utopian, not to say perhaps even
rather overwhelming.
Here I'm reminded of science writer K.C. Cole's observation
in her book, The Universe and the Teacup
[quote]: "...with the explosion of information reverberating
in our brains, it becomes harder and harder to hear
the clear ring of truth through the competing facts
and philosophies".
Later in the book, she observes that "nature bestows
her blessings buried in mountains of garbage. Well,
NSF's intent is to develop more effective ways to
create and organize information to glean useful knowledge-extracting
the blessings, or the wisdom, if you will, from the
rubbish.
In my on-going crash course on NSF, I've already learned
about ways that these ideas are taking shape. Our
Partnerships for Advanced Computational Infrastructure-better
known as PACI-are a key part of this. The goal is
to make high-end computing available for all fields
of science, and to ensure broader access to this capability
across the nation. Also exciting is the vBNS-the backbone
network that already links some 55 or so universities
with very high-performance connections. And we expect
this number to triple. Underlying all of this is ongoing
support for fundamental research and education in
computer science and engineering-the wellspring from
which these applications flow.
These are all visionary efforts that will need sustained
backing to succeed. We all need to work together for
our common future, and I am optimistic about our near-term
prospects as we wend through the budget labyrinth.
When President Clinton spoke to graduates at MIT in
June, he pledged to propose "significant increases
in computing and communications research." I'm sure
you're aware that Neal Lane, now OSTP director-designate,
has been charged with developing a plan for the President
to review. And NSF will be helping to formulate this
broad research initiative in information technology.
There isn't any doubt in my mind that we have a lot
to celebrate in computer science and engineering,
but we're confronting tough issues too. Juris, myself,
and others at NSF look forward to gaining the benefit
of your insights on these challenges.
All of us at NSF share your desire to strengthen the
information technology workforce at all levels. We're
facing some really disturbing trends here. We're puzzled,
frankly, by the drop in the enrollment of women in
computer science over the past five years or so.
Awareness of this gender gap in computing has even
reached the level of Ann Landers. I guess that means
it's REALLY serious! She recently ran a letter that
enumerated reasons why men think computers should
be referred to as female. One was that "Your smallest
mistakes are stored in long-term memory for later
retrieval." Well, women had their reasons, too, why
computers should be referred to as male. One was:
"Computers are supposed to help you solve problems,
but half the time, they ARE the problem."
Lightness aside, we need to work on broadening the
spectrum of people who are in the mainstream of the
information revolution. In fact, the picture for minorities
in computer science is even more dismal than for women.
For example, the number of African-American Ph.Ds
in computing has barely increased in recent decades.
From zero or one doctorate per year from the seventies
into the nineties, the number has gone up maybe to
only four or five who graduate now. What does that
say about the future, when the transformation in communications
and computing, the very cutting-edge of science, has
not yet swept major sectors of our population into
the excitement?
The country vitally needs the talents of the groups
underrepresented in the computing field. Just finding
enough faculty to teach the burgeoning numbers in
computer science courses is proving to be a challenge-as
is filling entry-level positions for coders, web-designers,
and so forth. Yesterday's Washington Post
ran a story about "the new underclass" - created by
importation of programmers and computer scientists
from China, India, Latin America, and other countries
to meet the needs of industry. These issues represent
critical challenges. They call for creative solutions.
Even though we don't fully understand the reasons
for these dilemmas, there are success stories out
there worth noting. One NSF-supported mentoring program
matches female undergraduates with faculty mentors.
And from this, an amazing 90-95% of the students have
gone on to graduate school.
There's a program at the San Diego Computer Center
that's aimed at much earlier ages. The program gives
computers to teenage girls, while mentors supervise
their research assignments. The teenagers, in turn,
teach younger girls in the fourth through sixth grades.
The girls learn to network, to use the web, to use
their computers on science projects. And if a girl
finishes enough assignments, she gets to keep her
computer. One girl's mother commented that "It's nice
to have a program like this when a public school can't
do it."
I've touched upon some issues that are pressing, even
critical, for computer science and engineering, but
also for all of science and engineering. I hope we
can see them as challenges that we all face together
employing fresh ideas.
I go back to the rough but honest wisdom of Grace
Hopper, who said, "Humans are allergic to change.
They love to say, 'We've always done it this way.'"
And Grace added, "I try to fight that. That's why
I have a clock on my wall that runs counter-clockwise."
Like Grace's clock, we need to turn some things around.
Grace had the insight that computing had potential
for commercial applications-it seems strange now that
she was considered a prophet for having that vision.
She also understood that computers could be accessible
to everyone-despite the fact, as she said she was
told, that computers didn't understand English.
Likewise, we also need to reach out with clarity,
explain what we do to the public, talk to the media,
try new approaches to achieve a more inclusive workforce.
We at NSF need your help on all these fronts to get
the message out about the importance of federal investments
in research and education, and the extraordinarily
rich returns they reap. CRA has done yeoman's service
in helping to spread the word. Now I think we need
to make this a grassroots effort.
I want to leave you with one final thought. Shortly
I'll be flying home from these beautiful Wasatch Mountains
to a very different geography, back to steamy Washington
and the wide expanses of the Chesapeake Bay, where
my husband and I have been sailing competitively as
a team for a number of years. If I may be allowed,
I'd like to use a nautical metaphor: I want you to
know that I'm genuinely excited at taking the helm
at NSF. But we sailors sometimes describe how the
wind "clocks," or veers. Occasionally, it surprises
us when it "backs" or shifts counterclockwise, sort
of like Grace Hopper's contrary clock. We should embrace
such new directions and make them work for us as we
do in a regatta, to help us reach the finish line
first. I'm very much looking forward to embarking
on this voyage to the future with you.
Thank you very much.
|