Three Great Challenges for Half-Century-Old Computer Science

FREDERICK P. BROOKS, JR.
University of North Carolina at Chapel Hill

1. Quantification of Structural Information

Shannon and Weaver [1949] performed an inestimable service by giving
us a definition of information and a metric for information as
communicated from place to place, negentropy. This metric,

      H = - Σ p_i log(p_i)

and the associated concept of noise, have proved rich sources of
further theory and of applications galore.

We have no theory, however, that gives us a metric for the information
embodied in structure, especially physical structure. We know that an
automobile is a more complex structure than a rowboat. We cannot yet
say it is x times more complex, where x is some number. Yet we know
that the complexity is related to the Shannon information that would
be required to specify the structures of the car and the boat.

I consider this missing metric to be the most fundamental gap in the
theoretical underpinnings of information science and of computer
science. Recent developments, however, make it timely to address
it. The fundamental biological structures are rich enough to repay
study and yet simple enough that there is hope of making real progress
on an information theory of structure. (Rowboats and automobiles are
much too hard.) The coding of genetic information by DNA is apparently
simple enough that it can be handled with existing communication
theory. The folded structure of proteins is not. Entropic and
energetic considerations are necessary, but not yet sufficient, for
explaining and predicting that structure, even after the amino acid
sequence is known. Yet proteins are relatively simple structures.

A young information theory scholar willing to spend years on a deeply
fundamental problem need look no further.

2. Software Estimation

Given specific functional, reliability, and performance specifications
for a software system, we do not yet know how to estimate the effort
required to build it. Nor, given the product specifications and a
quite specific description of the skills of a particular team, can we
reliably estimate how long it will take to grow the product.  The
first book on computer software, by Wilkes et al. appeared in
1951. Now, a half century later, there have been many, many advances
in concept and technique.  Not the least of these is Boehm's
monumental Software Engineering Economics [Boehm 1981] the COCOMO
model set forth therein, and the subsequent software economic models
stimulated by that work. Nevertheless, we still don't know what we are
doing, unless it is very similar to something we have done before.

The challenge is to make software engineering as predictable a
discipline as civil or electrical engineering. I still do not expect
any radical breakthrough, any silver bullet, to solve this problem
[Brooks 1986]. But the accretion of many contributions has already
made much progress, and I believe continued careful research, ever
validated by real practice, will bring us to that goal.

3. User Interface Design for Computer Systems

Today, the design of the user interface for an operating system or an
application program is still an art, not yet an engineering
discipline.

Much research has been done on human perception, human cognition, and
on the human factors of output from the mind to the bodily
effectors. On the other side of the human---computer interface, much
research and development has advanced interface technologies: computer
graphics, sound synthesis, speech synthesis, speech recognition, and
haptics.

Substantial as these research corpora are, we seem still to lack any
systematic or disciplined way of integrating them when designing a new
hardwaresoftwareuser interface. We do not have reliable ways even of
predicting whether a proposed specific interface design will be good,
that is,

---Intuitive for the novice;

---Efficient in perception and motion for the expert;

---Robust under misuse;

---Facilitating in recovery from cognitive or manipulative mistakes;

---Helpful in diagnosing errors and suggesting corrective action;

---Rich in incrementally learnable functions, like the alphabetical
   shortcuts on the Mac interface.

The challenge is to integrate the relevant but disparate bodies of
knowledge into a discipline of design.

ACKNOWLEDGMENTS. Conversations with colleagues over the years have
stimulated these ideas. I am especially indebted to Professor Jurg
Nievergelt, whose thoughts on these challenges especially resonated
with mine.

REFERENCES

BOEHM, B. 1981. Software Engineering Economics. Prentice Hall,
Englewood Cliffs, N.J.

BROOKS, F. P. 1986. No silver bullet---Essence and accident in
software engineering. Information Processing 1986. Reprinted as
Chapter 16 of Brooks, F. P. 1995. The Mythical ManMonth, Anniversary
Edition, AddisonWesley, Reading, Mass.

SHANNON,C .E.,AND WEAVER, W. 1949. The Mathematical Theory of
Communication. University of Illinois Press, Urbana, Ill.

WILKES,M .V.,WHEELER,D.J.,AND GILL, S. 1951. The Preparation of
Programs for an Electronic Digital Computer, with Special Reference to
the EDSAC and the Use of a Library of Subroutines. AddisonWesley,
Reading, Mass.