Designing Graphic Presentations from First Principles
Doctor of Philosophy in Computer Science
University of California at Berkeley
Professor Robert Wilensky, Chair
This dissertation outlines a first-principles approach to automatically designing graphic presenta-
tions of information. The components of this approach include a conceptual framework for discussing how
presentations encode information, algorithms for determining whether a method of presentation will be ca-
pable of presenting a given type of information, and design principles for ensuring the interpretability and
perceptual effectiveness of a method of presentation.
Compared with previous approaches to automatically designing presentations, the approach out-
lined in this dissertation is more fine-grained and more general. It begins with an extremely general notion of
how graphic presentations can encode information, then develops this into a useful framework by making a
number of explicit assumptions about the types of presentations that people can use. This framework serves
as a basis for analyzing the space of possible graphical languages—i.e., the space of systematic methods of
presenting data. The logical adequacy of different graphical languages for different types of information, and
criteria and methods for composing graphical languages for different data are also explored.
In addition to this logical emphasis, this dissertation also emphasizes the influence of psychological
issues on the design of presentations. It explores factors influencing the interpretability of presentations (i.e.,
how easily viewers will grasp how information is encoded) and outlines some general design principles for
creating interpretable presentations. It also explores perceptual issues in presentation—including perceptual
organization, dimensional structure of visual stimuli, and the effectiveness of perceptual operations—and
outlines design principles for guaranteeing the perceptual effectiveness of presentations.
The last emphasis of this dissertation is on operationalizing the framework and principles—i.e.,
on using them to create graphical languages in a relatively efficient manner. The implementation, AUTO-
GRAPH, demonstrates the flexibility and viability of a first-principles approach.
I have spent quite a few years working on this dissertation, and more generally in grad school. This
experience would have been less fruitful and much less bearable without the support and advice of many
I would first like to thank my advisor, Robert Wilensky, for granting me the independence to pursue
my own research interests and still managing to provide me with generous support and valuable advice. This
dissertation might never have been completed, and certainly would have been weaker, without his insight. I
would also like to thank the other members of my thesis committee, Marti Hearst and James Landay, whose
comments and critiques greatly improved this work.
My graduate school experience was vastly enriched over the years by the various members and
associates of my research group, including (in alphabetical order): Michael Braverman, Anne Fontaine,
Marti Hearst, Narciso Jaramillo, Dan Jurafsky, Peter Norvig, David Palmer, Tom Phelps, Nigel Ward, Dekai
Wu, Jamie Zawinski, and Jordan Zlatev. Most of these people were my officemates at one time or another. I
never got much work done when they were around the office, but they nevertheless made my academic and
personal experiences in grad school infinitely more rewarding.
Outside the office, other friends and colleagues in the department also helped make the experience
of being a grad student not only interesting, but fun, too. Among the students and faculty members whom
I had the pleasure to know were: David Cohen, Eric Enderton, Francesca Barrientos, Diane Hernek, Nina
Amenta, Kim Keeton, David Parsons, Nikki Mirghafori, Terry Regier, Mike Luby, Oliver Grillmeyer, Dana
Randall, Mike Clancy, and Ashu Rege.
Many other friends, near and far, also provided both diversion and support. Without the diversion,
this dissertation might have been completed sooner, but without the support I’m sure it would never have been
completed at all. Some of the nearby friends included Jon Levine and Susan Lin, Carrie Timko, Beth Multer,
Becky Gross, Morrisa Sherman, Pete Chow, Erin Vaca, and Melanie Light. Others, further afield, included
Jan Rivkin and Debbie Kadish, Ben Gillum and Heather Rayburn, Mike Rosenfeld, Tarun Khanna, Rahul
Asthana, Pascale Fung, and Joanna Sadowska. Others, once near but now far, included: Ivelina Zlateva,
Robert Chang, Anna Herreras, John Liechty, Toby Falk, Zenda Kuo, and Shari Cohen. A special thanks is
due Kristin Conradi, who gave much support during the last year of the writing of this dissertation. And I’d
like to add an extra thanks to Jon and Susan, and to Carrie, all of whom helped see me through more or less
my entire grad school experience.
I’ve had a number of housemates over the years I’ve lived in Berkeley, but I’ve been particularly
lucky in the last four or five years. My housemates made life at home fun, and always yielded the kitchen
table when I needed it, even if they did sometimes leave dishes in the sink. Among the people I had the
pleasure to live with were: Shari Rubin, Holly Holmquist, Roger Studley, Carla Savage, Sasa Gabarsek,
Rebecca Gelman, Alex Cuthbert, and Sharon Gibson.
I have felt very fortunate during my graduate school tenure to have had the unfailing support of a
very helpful Computer Science Department staff. Within the larger UCB bureaucracy, the department pro-