University of Maryland visit, March 28-29, 2007
Page last updated March 19, 2007













Talk: University of Maryland Human Computer Interface Lab; March 28, 2007
Interaction Design as a Branch of the Humanities:
A healthier fit than technology or computer science?

Interaction design centers on the ability for a person to absorb, react to, and act upon information inside computers. It is clear that the manipulation of bits is well served by the strategies and techniques of computer science and electrical engineering, but close scrutiny of the human side of that gap might help us fit minds as well as we optimize algorithms. The last good decade has seen much more focus on "human I/O" in the form of perception, but we can take even better advantage of the eye if we know how it feeds "upstream" sense-making processes. Red may be a visually salient danger warning--but not if it's a tulip in a field in Haarlem, or red type and ornaments in a centuries-old codex. We understand objects in context and reinterpret our perceptions in radically different ways depending on that context.

The center of my research and practice for the last five years has been the exploration of higher and higher mental processes--beyond sensation and perception to object recognition and cognition. Necessity has lately driven me even farther: all the way past semiotics to psycholinguistics, and finally the humanities. It has become an axiom for me that many interfaces are best driven to be consistent not with other interfaces, but the way people think about a specific task. And the best doors into that realm have been explorations in the plastic arts (how people use visual richness to differentiate among categories of information), poetry (using metaphor to tame abstract information), and perhaps even storytelling (to understand process and aid long-term retention). These human endeavors may have more to offer us on the level of technique than we realize.

This talk will structure these observations, support them with references from source materials outside computer domains (such as psycholinguistics and hermeneutics), and show examples of antique information visualization techniques which in many ways are much more sophisticated that what we do today. I will deconstruct as an example recently-completed work for the 2006-7 transition of the New York Stock Exchange to a half-human, half-electronic "Hybrid Market;" a design that has sped up brokers in their most fundamental task from 7 seconds to less than a second--a 15:1 speedup directly resulting from an application of these principles.

Workshop: UMD Maryland Institute for Technology in the Humanities; March 29, 2007
Domain- and Task-Specific Tools for the Humanities:
We'll explore what's needed now, what's attainable

This workshop is an initial exploration of how KWIRQI design principles (Knowledge Work Interfaces: Reality-Quoting Interfaces) might be applied to the domain of research and criticism in literature. The heart of a KWIRQI design is the way it adopts jargon and metaphors from a very limited community of practice, "quoting the reality" of that community. Here, we will be trying to isolate and understand what ideas are common in this domain, and what tasks might best be supported by the development of new computer-based tools for literature.

I will lead the group in an enumeration of bits of jargon and discovery of common concept spatializations. For example, in the financial domain a "high price" is a large number; there is no a-priori reason to assume that a large number is "up." But that pairing of a concrete adjective with an abstract noun tells us that traders have mapped numbers into space: specifically a vertical dimension in their minds. Remarkably, this mapping seems to happen in all knowledge work domains I have studied. We will search for these domain-specific implicit metaphors shared by critics and text analysts; for this reason it will be extremely valuable to have "pre-digital" researchers who have been practicing for many years: they carry the wealth of this local culture in their minds.

We hope to justify their time by developing a listing of tasks that manipulate these concepts and metaphors then doing an economics-style tradeoff; ranking which tasks will help support the field the most, yet be easiest to implement and take best advantage of local university resources and expertise. We want to take the most tedious work they do and offload it to a computer, then present the results in exactly the way they would sketch them on a napkin to relate them to a student. The overall goal will be the creation of scope definitions and perhaps initial design sketches for tools that can be designed, built, and put into use in the immediate future.