HCI Fall 2002 Qualifier Exam

Format for Qualifier:

Each student should answer four (4) questions in total.
Two (2) questions must come from Part 1 (Process and Theory).  The two (2) remaining questions must come from your two declared areas of specialization; one question from each area.

Please start each new question on a new page.  When writing your answers, place relevant references that are cited in your answer at the end of the question. 

Part 1: Process and Theory

1. A video game company is designing a game around the premise of the movie Toy Story (i.e., the toys in a child's room are alive with their own personalities, ideals, and agendas - the toys just “play” dead when humans are present).  The genre of the game is to be a detective “Who Done It?”  The toy Mr. Potato HeadTM has had his nose stolen, and the toy community is concerned with discovering who stole it to avoid any more theft.  The player takes on the role of a pair of toy binoculars that is hired to solve the case.  The player is presented with a 360 degree, very highly detailed view of the room.  While the toys go about their lives, the player can zoom in on specific scenes in “infinite” detail to search for clues but can not move from the center of the room.  The player can view multimedia evidence, collect clues (e.g., a ransom note), and arrange interviews with specific toys. Random events (e.g. news of another “nose”-napping) may interrupt the story.  

Choose two of the four following theories to assist in the game design process.  Justify your selection of theories and give examples of applying the theory to the design process. (No need to define the theories.) Are your two techniques orthogonal, complementary or in conflict?

a) Human Model Processor / GOMS family
b) Situated Action
c) Activity Theory
d) Distributed Cognition

2. Pie menus never really “caught on”, that is, they have not been used widely outside particular niche systems and domains.

a) Provide a description of the basic pie menu and two variations on the pie menu such as flow menus and marking menus.
b) Pick two different theoretical/evaluative models of usability and critique the pros and cons of pie menus with respect to each.
c) Why, despite many good arguments in favor of pie menus, have they not become prominent in the desktop graphical user interface?

3. It is often difficult to produce informative and generalizable results from empirical laboratory studies and more qualitative field studies to inform the design of complex human-computer interfaces.  Clearly, they can have benefits too, however.

a) Pick some example domain such as air traffic control or electronic voting, and provide examples and justification for two limitations and two benefits of these techniques.
b) Using your own research experience, discuss the limitations and benefits in applying these techniques (both lab and field)?

4. Creating interfaces on mobile devices can be particularly challenging.  In particular, emulating the context in which the mobile device would be operated can be difficult in a laboratory environment.
For three of the following terms, discuss key strengths and weaknesses of this approach/technique/theory when applying it to the mobile interface domain. (These are separate short answers; you do not need to try to compare or contrast.)

a) Heuristic Evaluation
b) Cognitive Walkthrough
c) Fitts’ Law analysis
d) Wizard of Oz prototyping
 
Part 2: User Interface Software and Technology

5. In the 1980’s two major architecture models were proposed for UI design: the Seeheim model [Pfaff 85], and the Model-View-Controller (MVC) [Krasner 88, Goldberg 84]. Each of these represents a different conception for the organization of a 2-dimensional graphical user interface.

a) Briefly define the basic organizational ideas of the Seeheim and MVC models.  What are the major differences between Seeheim and MVC?  What are the main strengths and weaknesses of each?
b) Both Seeheim and MVC have had influence on toolkits and architectural models to support the development of UI software.  Give one example of a UI toolkit in common use today and discuss how either or both of Seeheim and MVC influenced its development.  Similarly, give one example of a more modern conceptual architecure for UI development and discuss how either or both Seeheim and MVC have influenced its development.
c) What are constraints and how are they used in UI development?  Provide some examples of toolkits or systems that support constraints. What is the relationship between constraint specification and management and the architectures and toolkits discussed above?

6.  Imagine the construction of a classroom for computer-enhanced education.  The room has an interaction surface, or electronic whiteboard, at the front that is 8’ high and 24’ wide, with an output resolution of 3072 x 768 pixels. In addition to accepting traditional mouse and keyboard input, the input technology on this large interactive surface allows for smooth pen-based input (for handwriting), gesture recognition (both on and away from the surface), multiple simultaneous input (i.e., for two-handed input or for multiple person interactions) and recognition of physical objects at or near the surface (such as magnets used to tag spots on the surface.  In addition, camera technology will allow for the recognition and interpretation of paper placed on the surface, such as a posterboard or post-it note.

a) With all of this input technology, there is a potential for many different modes of interaction.     For example, when making marks on the electronic whiteboard, the pen strokes may be recognized into text or special graphical shapes, or recognized as special commands (e.g., to create whitespace on the board) or just left as is as digital ink for humans to read.  Discuss the challenges and potential solutions to minimizing the number of modes and of handling smooth transitions between various modes of interaction with the large interactive surface.  Where possible, provide examples of solutions to these challenges with examples from the research or commercial communities.
b) Discuss the advantages and challenges of simultaneous input on the large interactive surface, as briefly described in the first paragraph. Where possible, provide examples of solutions to these challenges with examples from the research or commercial communities.

Part 3: CSCW

7. Shared text editors and shared drawing editors have been popular application themes in CSCW and the technical challenges of architectures for groupware.

a) What are the similarities and differences between the two domains, when considering the technical challenges of constructing systems to support shared text or shared graphics editing?
b) What solutions have been proposed to address problems of consistency of the underlying shared data across distributed authoring machines?
c) What solutions have been proposed to support awareness of the state of the shared document and the activity of other users as they change the document?

8. The CoWeb has been mildly successful at Georgia Tech.  Students voluntarily use it in many classes, and we've measured learning benefit of at least one kind of use in an English Composition class.  Teachers from Architecture, Math, ChemEng, and CS use it daily.  Yet, it's not been overwhelmingly successful.  Students in some Math, Engineering, and CS classes avoid using the CoWeb.  The perceived sense of competition in these classes makes it irrational behavior to give up one's “edge” by talking with others.  Some faculty avoid the CoWeb because it doesn't fit in with their practices.  Over-generalizing, faculty who aren't looking to create a Web presence for their classes and/or aren't interested in web-based collaborative learning activities can't find the time to find a place for the CoWeb in their classes.  The interesting observation to make is that none of these problems has anything to do with usability.  Every study, questionnaire, and interview about the CoWeb points to a high degree of usability.  Simply put, some people choose not to use it anyway.

a) Can you envision a collaboration tool that might address the students', faculty, or both sets of concerns, and yet would provide some advantages of groupware/collaboration?  What kind of tool might work here?  
b) What other efforts, in addition to or in place of modifying the CoWeb, would you recommend to encourage wider adoption?
c) How would you evaluate the tool to determine if it is addressing the concerns?  Can you evaluate prior to deployment, or would you have to have it deployed to know whether or not it would work?

Part 4: Ubiquitous Computing

9. In designing an “intelligent” meeting room, one challenge involves the coordination of activities across a number of public and private displays. There may be multiple display surfaces on the walls and tables that are viewed and potentially controlled by any person in the room.  In addition, individuals may possess handheld or laptop computers that they bring with them to the room and they would like to share information and control among their devices, the shared displays and other personal devices in the room.

a) Describe one potential software solution for facilitating this coordinated control of multiple devices in the intelligent meeting room.  It is important to take into consideration the dynamic aspects of this problem, that is, you should not assume that the number and identity of the public or private displays/devices is known in advance or remains constant. For the purposes of this question, please ignore the issue of private side-conversations/interactions between the small, private displays.  Provide references to existing research or commercial results that relates to your suggested solution or provides an alternative solution to a similar problem.
b) How would you assess the effectiveness of your coordinated control scheme from the perspective of the humans in the room?  Discuss both formative and summative evaluation strategies.

10.    Two approaches to ubiquitous computing are to place the infrastructure in the environment or to place the infrastructure on the body. This applies to decisions on where sensing occurs, where information is stored and where information is processed.

a) From a technical perspective, discuss the advantages and disadvantages of environmental versus body approach on the basis of sensing, storage and computation.
b) From a human perspective, discuss the advantages and disadvantages of environmental versus body approach on the basis of sensing, storage and computation.
c) Give one concrete application example from the literature in which the environmental approach would be preferred over the body approach because it provides a feasible technological solution and is also appealing from the human perspective.
d) Conversely, give one concrete application example from the literature in which the body approach would be preferred over the environmental approach because it provides a feasible technological solution and is also appealing from the human perspective.