PhD Qualifier, Spring 2000

HCI Area

Answer any five of the eight questions.

1. In UIST'98, Mankoff and Abowd introduced a novel input interactor for pen-based computing, a word-level unistroke keyboard, called Cirrin (see http://www.cc.gatech.edu/fce/pendragon/cirrin.html for an illustration and further information). The basic idea behind Cirrin is to arrange the letters of the alphabet around the permiter of a circle. The user starts inside the circle with a pen stylus. A single pen gesture (from pen down until pen up) creates a path that intersects with the perimeter of the circle at various points. The letters that are traversed upon these intersections form the word corresponding to the gesture. This simple description actually defines a rather large family of input devices.

Discuss the challenges in exploring this design space, addressing questions of how to analytically and empirically evaluate design variants, how to handle feedback, and how to compare this input interactor with other options for pen-based text input.

2. While we typically worry about building software for users, not programmers, there are cases where the programmers are the users. Consider the case of building a programming environment.

  1. Imagine that you were going to conduct a heuristic evaluation of a programmer's interface. What heuristics would be particularly good to apply?

  2. Use those heuristics to describe the usability for programmers of a UNIX environment (e.g. the symbolic debugger sdb, or command-line tools) vs. any IDE you know (e.g. Java Workshop or Squeak).

  3. Describe how you might conduct an empirical usability analysis of these two environments to further explore what you learned from the heuristic analysis.

3. Designing groupware is hard because of the diverse user population and diverse roles that users may play. Designing educational software is hard because of the novice nature of students and the roles and tasks of teachers. Now, consider the design of a piece of groupware that allow teachers, parents, and students to communicate about the activities in the class (e.g., teachers asking for volunteers to help judge an on-line art fair, and parents doing the judging on-line) and about student learning (e.g., an on-line teacher conference called at the parents' request).

How would you design such classroom-based groupware? Don't design an interface or functionality here: Tell us what your process would be for designing such a tool.

4. Consider the formal user modeling approaches GOMS and the keystroke-level model. What is at the essence of these approaches? For what situations are they appropriate? Do you feel that the techniques are over- or under-utilized in HCI practice?

5. There is much current debate about the design of web home pages and whether they should

  1. be more dense, containing many links to pages throughout a site

    2. or

  2. be less dense and more graphically/visually appealing, but with fewer links to other pages.

How would you design a study to address this question for a particular organization's home page? Include appropriate citations to the research literature in your answer.

6. Describe at least three different ways that animation can be used to enhance a user interface or an information visualization. What service is animation providing in each case?

7. Is "Software for supporting UI construction" a solved problem? Argue either your affirmative or negative response.

8. "HCI research in universities has contributed little to the practice of usability engineering for fielded products." Discuss.