You must answer 4 of the first 6 "core" questions, 1 of the next 3 "specialization" questions, and then 1 more question of your choice of any of them. This makes a total of 6 questions answered.

Modeling & Evaluation

1. The expression "recognition over recall" is used to describe the advantage of what Norman calls "knowledge in the world" versus "knowledge in the head".
1. How have modern graphical user interfaces used recognition over recall? Give examples of 3 interface elements that support recognition over recall.
2. Pick a particular application domain and discuss how recognition over recall is used (or could be used) to assist a typical user in that domain. Some example domains might be software development, personal productivity, shopping assistants.
3. Give two examples of evaluation methods that can be used to test the effectiveness of recognition over recall.
2. Briefly compare the treatment of goals or plans in the following theories of human cognition: situated action, activity theory and distributed cognition. If you were designing a kitchen interface to help people cook, what are possible ways that potential interfaces could support each theory of cognition?

UI Software

3. There are many WYSIWYG graphical user interface construction tools like Visual Basic, Delphi, or a current Java IDE like Jbuilder. Consider the task of constructing an interface to a relatively complex application, such as a chip design tool. What do the GUI builder tools do best to support this kind of application, that is, how do they help the most in the interface creation process? Conversely, what do these tools do worst, that is, what are they least helpful with? Provide examples to help your explanation.
4. The Model-View-Controller (MVC) Paradigm has been the guiding light of UI software for the last twenty years. It describes how to build systems that are loosely coupled, can handle multiple interfaces on the same data, and are easily maintained. It also describes how to build systems that are inefficient and hard to understand. Pick TWO alternatives to MVC that have been actually implemented and used in some UI toolkit (can be just a research toolkit as long as it's actually been used), and contrast them with MVC. Describe how the alternatives are different, and how they achieve the same goals. Are the alternatives any easier to use and/or more efficient? Are they just as easy to maintain and can they still do multiple interfaces on the same data?

Design

5. Describe and contrast the use of common data gathering techniques (e.g. interviews, observations, beeper studies, diary studies, focus groups) for different phases of an HCI design research project, in particular the difference between exploration and validation.
6. This question asks you to critique the design of either Windows (any version from Windows 95 on) or MacOS. Use the same OS for all parts. In all parts of your answer, make sure to cite the literature to justify your analysis.
1. Assume your user is a ten-year-old child.
1. Describe one feature/design decision of the OS that you feel is badly designed for this audience. Explain why.
2. Suggest how you would change the design decision you discussed in i).
2. Now assume your user is a professional system administrator.
1. Describe one feature/design decision of the OS that you feel is badly designed for this audience. Explain why. (Do not pick the same feature you used in part A.
2. Suggest how you would change the design decision you discussed in i).

7. Usability evaluation of CSCW is hard. Laboratory testing isn't very effective without having lots of users involved, and it's certainly not environmentally valid. Even beta testing isn't that useful, since the real tests of usability are when there are lots of users playing different roles. Imagine, for example, trying to evaluate the usability of E-Bay with only a few dozen users -- you won't run into any of the problems that really occur when you have hundreds or even thousands of users.

   Now imagine that you really do have this problem. Let's say that AT&T and Bellsouth have merged, and one of their big new projects is to create the next generation E-Bay which will be accessible via cellphones or the Web. Your job is to evaluate the usability of the system. What will you do? Explain what conditions you'll test under and what data you'll collect, and what things you will and will not be able to test for.

8. Two popular, general techniques for visualizing multi-variable data sets are parallel coordinates and a Table Lens- style display. Compare and contrast the two techniques, highlighting the relative strengths of each and identifying both attributes of the data set and user tasks that might make one technique a better choice in some situation.

9. Many ubiquitous computing interfaces leverage implicit sensing of people manipulating real objects in the environment. Briefly describe some well-known examples from the literature. Assess the directness of these interfaces in terms of articulatory and semantic distance. Are these direct manipulation interfaces as defined by Hutchins, Hollan and Norman?