Georgia Tech, College of Computing, PhD Program
HCI Area Qualifying Exam (Written Portion)
Wednesday, November 12. 9am-5pm.

Answer FOUR questions as follows: Answer any TWO questions from Section A (HCI Process and Theory); From Section B (Special Topics in HCI) answer ONE question from EACH of your TWO declared areas of specializations.

Each of the four questions you answer will be given equal weight.

You will be assigned an identifying number and are required to hand in printed or written copies of your answers with each page identified only by that number. This enables us to grade your answers anonymously. You should NOT identify yourself by name on your answer sheets.

Please answer each question starting on a new page, with your answer following the text of the question. To avoid having to type the questions yourself, electronic copies of the questions will be available after 9am from a URL that you should already have been given. You may copy Place any relevant references that you cite in an answer at the end of that answer, NOT in a separate section at the end of the whole exam. If you have any questions or feel it necessary to make any assumptions in your answers, do not seek clarification from faculty or staff. Simply record your assumptions as part of your answer.

 (A) HCI Process and Theory
1.
Telephone-based interfaces to computer systems have become common.  The systems are used by banks, airlines, credit card companies and others to provide customer information.  Some systems require the user to push keys to provide responses; others use voice recognition, some allow either.

Discuss the advantages and limitations of EACH of the following formative evaluation methods with respect to their applicability to evaluation of these telephone-based interfaces:

(a) Fitts law
(b) GOMS
(c) Heuristic Evaluation
(d) Cognitive Walkthrough
(e) Wizard of Oz prototyping

2.
Discuss the role of emotion in user interfaces and HCI.
(a) Discuss two existing systems that make use of emotional expression.
(b) Describe the potential benefits of incorporating emotion into interaction design.
(c) Describe the potential problems.
(d) Speculate about the future:  In your   opinion, will emotion play a greater role in interfaces. If so, how? If not, why not?

3.
Consider the following six statements. Each statement may in your opinion be self-evidently true or outrageously misleading, or somewhere in between these two extremes. In the case of this question, however, the faculty does not want you to give your opinion. Instead, give a short, cogent argument of one or two paragraphs in favor of ANY FOUR statements. Your arguments should make use of the HCI literature, including both the research literature and published evidence concerning practice as appropriate. Be careful not to criticize the statements negatively; give ONLY supporting arguments. [If you can’t restrain yourself, wait until the oral part of the exam to criticize a statement!]

(a) People always interpret interactive systems as if they were interacting with another person.
(b) Eventually all interactive systems will be adaptive, creating an interface tailored to the specific individual user.
(c) Effective summative evaluation of interactive systems is moot.
(d) The ultimate goal of ubiquitous interactive systems is to be invisible.
(e) The ultimate goal of HCI design is to "underdesign" allowing for users to "finish" the design through appropriation and configuration.
(f) Membership in virtual communities erodes participation in traditional, physical community activities.

4.
A never-ending controversy in the HCI community is understanding the strengths and limitations of different HCI theories of human behavior (physical, cognitive and social).  Some researchers characterize their theories as accounts of the "truth:" that is, as unbiased factual accounts of human behavior.  Others question the goal of objectivity in HCI and characterize their theories as valuable perspectives that provide a framework for coherent analysis and bring certain elements of human behavior to the forefront during design.

(a) Give ONE example of EACH type of theory.
(b) Discuss the relative strengths and weaknesses of the two theories you identified. Evaluate each theory on its own terms and in the light of the other type of theory: To what extent is the “objective” theory amenable to scientific testing? Does it provide useful guidelines for designers? Conversely, to what extent can the “perspective” theory be validated or justified? Is it really useful in design?
(c) In your opinion, how valid is this distinction between theories as factual accounts versus theories as valuable perspectives? As an HCI researcher, should you feel obliged to ally yourself with one type of theory or the other? If you were an applicant for an HCI faculty position, how would you deal with the skepticism of a traditionally minded CS professor who questioned during your interview whether HCI had any “theory”?


(B) Special Topics in HCI
User Interface Software
5.
In the earlier days of user interface programming, there was a competition between tools to support manual construction of WIMP interfaces and tools to support declarative or model-based development.  

a) Which class of tools won this competition to prototype two-dimensional GUIs and why?
b) With the proliferation of mobile and handheld devices, there is again a competition between tools for rapid prototyping of interfaces. How is this competition different than the previous competition for UI tools?  Give an example of a research effort that represents the model-based approach and describe how it works, and comment on its strengths and weakness.
c) Predict whether the model-based tools will win the competition to build interfaces for mobile and handheld devices.  Present a well-reasoned argument defending your prediction

6.  
Lenses (magic lenses or toolglasses) were popularized in the research literature in the early 1990’s

a) Give two concrete examples of a lens.
b) If you were to define a design space of lenses, what dimensions or characteristics would you use to distinguish one lens from another?  For example, you might consider the geometric operation performed by the lens.  Give at least FOUR defining characteristics of a lens.
c) How might a lens (or combinations of lenses) support search across a large repository of annotated digital photographs?
d) Discuss some of the issues to consider when implementing the input and output behaviors of different lenses.
e) Using Michel Beaudoin-Lafons Instrumental Interaction model (CHI 2000), analyze the properties of a lens

CSCW
7.
You work for the research division of Colossal Cable Monopoly (CCM, inc.).  CCM Research is currently trying to develop novel consumer applications for fiber to the home.  The R&D team includes designers, programmers, and a customer service team that is charged with offering experimental services to customers and gathering their feedback. Communication among these sub-groups has historically been a bit tense.

CCM Research is about to deploy a new groupware system, designed to increase collaboration both within and among sub-groups and increase over-all productivity.  You have been put in charge of tracking this deployment.  After one year, you will need to make a recommendation as to whether to renew the expensive groupware product license.  You also would like to write up the entire experience for CHI. That is all in the future. Right now, you are at the stage of planning your study; in particular, you need to decide what kinds of data will you collect to help you make this decision. Since there is budget uncertainty, you do not know at this stage whether any study you propose will have to be scaled back in some way, so you are concerned not to propose an “all-or-nothing” study that will have no validity unless it is completed as planned.

In a memo to your boss, describe the different methods you might use, and what kinds of insights each is likely to contribute.  Conclude your memo by suggesting how you might proceed with a limited-resources study if a more comprehensive study is not possible.

8.
Throughout the history of CSCW research and design, there has been a tension between designers who want to encode how users should behave when using their systems and those who do not. For example, the speech-act encoding of Winograd and Flores’ early messaging system Coordinator was justified by claiming that conversations for action describe how people naturally communicate commitments and that the technology facilitated such collaboration by making commitments more explicit. Critics countered by saying that users were made to feel that every communication was a binding contract.

(a) Give a detailed description of these tensions between encoring and non-encoding of patterns of behavior with plenty of examples.  Do no limit yourself to one domain of CSCW but touch on multiple application domains.
(b) Many system designs attempt to do both. How does IBM's Babble system encode social policy yet allow for appropriation?

Information Visualization
9.
Suppose that a treemap is being used to visualize some multivariate hierarchical data set of 200 items.  The treemap display will be running inside a window of size 1024 x 768.  The variable/attribute of each data item that is being mapped to region area is an integer whose value can range from 0 to 10,000.  The designer of the visualization wishes to see every data item, even those with zero value.

(a) How is the naïve attribute-to-area mapping done?
(b) Why is the final objective above problematic/challenging?
(c) Design at least THREE mapping strategies that could accomplish the visualization goal.
(d) Discuss the advantages and disadvantages of EACH of the mapping strategies you propose.

10.
Jock Mackinlay, in his dissertation and subsequent ACM TOG paper (April ’86), developed a system that automatically creates information visualizations based on the user's stated requirements. Since then, others have tried to apply automatic generation techniques to more complex types of information, with mixed results.

(a) In general, how do existing systems in this area approach the problem?
(b) What are the stumbling blocks in automatically creating information visualizations?
(c) How would you approach the problem?

[Note: Mackinlay’s paper is not one of the required readings, although his work is discussed in the Infovis course. While familiarity with his work would be beneficial, this question relies on your knowledge of information visualization research in general.]
Ubiquitous Computing
11.
Several researchers have posited that attention is the new scarce resource: rather than managing our computers' available storage space, processing power, or I/O capabilities, we should instead be managing our users' attention. Weiser's "invisible computer" is arguably a step in this direction, shifting the computer from the foreground of the user's attention to the background. However, changing human-computer interaction so that users no longer explicitly invoke commands has a potential disadvantage: A computer might misinterpret the user's intent and take an inappropriate action.

(a) Potential approaches include doing nothing, attempting to draw the user's attention to relevant information so that the user can decide how to act, interrupting the user to suggest an action, or acting without the user's explicit consent. For EACH of these approaches, mention a system that uses it and list the advantages and disadvantages of the approach.

(b) Instead of choosing a single approach, an alternative is to create a system that can change which action it chooses (e.g. drawing attention, interrupting, acting without permission, or doing nothing). Discuss how you might implement such a system.

(c) Some researchers argue that adaptive systems empower computers instead of users. They suggest that we should instead focus on creating environments that provide a variety of output devices with different attention requirements and allow designers or end users to shift information to those resources when appropriate.
(i) Discuss the affordances of potential output devices for managing attention, and give examples of systems utilizing those resources.
(ii) Discuss the strengths and weaknesses of this approach relative to an adaptive system.

12.
Ubicomp researchers are often dogged by questions and concerns about user privacy.

(a) Privacy in ubicomp systems can range from practically moot to a significant concern. What are the critical parameters that determine where a particular system falls along this continuum?
(b) Starting with Bellotti's framework of "feedback and control" (Bellotti, V. & Sellen, A. Design for Privacy in Ubiquitous Computing Environments. ECSCW 93, 77-92. http://citeseer.nj.nec.com/286107.html), describe techniques for mitigating privacy concerns in ubicomp systems.  Give examples.
(c) Pick ONE of the following domains and propose a novel design for addressing privacy concerns in it:
- Classroom 2000 / eClass
- Digital Family Portrait
- Your research
(Note that a “novel design” may be the application of a known technique to this new domain. In your answer please remember that privacy and security concerns are different and limit your discussion to privacy.)