University of Glasgow

June 2002

CS1Q Exam (sample solutions).

Time allowed: 2 hours and 15 minutes

Answer ALL questions.

  1. a) Please provide BRIEF answers to the following questions:

    1. What is a logical input device?

      A logical input device is an abstraction of a number of physical input devices. For instance, text entry is a logical input function that can be implemented by a keyboard, speech recognition, cursive recognition on a stylus based system etc. This distinction offers considerable support for the development of device ?independent? interfaces.

    2. Name two differences between formative and summative evaluation.

      Formative evaluation helps to form design decisions while summative evaluation is often used at the end of development to demonstrate that requirements have been met. Formative evaluation is often low cost with the assumption that resources will be available to address any problem identified. Summative evaluation techniques involve a higher cost and there may not be sufficient resources or time to correct major design problems. Formative evaluation often relies on only a partial implementation while summative techniques may require a full system etc?

    3. Briefly describe two situations when you might use a tracker ball rather than a mouse.

      In cluttered or confined spaces you do not need to move a tracker ball, it?s footprint is smaller than that of a mouse. Also, in areas in which there are vibrations or movements that might dislodge the mouse.

    4. Name two differences between long and short term memory.

      Long term memory is often characterised by the ease of recall. It is highly associative and persistent. In contrast, short-term memory requires conscious effort to ensure that particular items can be recalled. It is also more vulnerable to interference effects. Short-term memory can also be thought of as a form of working memory - hence any processing of information in long-term memory can also be vulnerable to the problems that affect short-term memory etc. The capacity of short-term memory will be more limited than that of long-term memory. Some solutions may refer to the 7 + or - 2 heuristic for non-associative capacity of short term memory.

    5. What is internal consistency?

      Internal consistency refers to an important property or principle of interface design. Similar operations should be performed in a similar manner, similar information should be displayed in a similar format. Internal consistency refers to similarities within the same interface. This can be more easily ensured than external consistency when any single development team may not be in a position to ensure that all other applications follow this principle.

    [2 marks per answer, 10 marks in total]

    b) Why can it be particularly difficult for designers to obtain an impression of an expert user?s mental model of an interactive computer system?

    [3 marks]

    [Seen problem/bookwork]

    First of all, it can be difficult to gain an accurate impression of any users mental model. It is difficult to verbalise knowledge and information about complex technologies. Expert users pose particular challenges because they will have become ?habituated? to many aspects of an application. Complex aspects of a system can become so familiar that they lose their complexity. Important knowledge may become held within long-term memory so that explicitly recalling it in a form that can be discussed may become difficult. In Rasmussen?s terms actions become skills and the procedural rules of intermediate users will be lost.

    c) User interface design often centres on identifying an appropriate abstraction for an underlying computer system. For example, the waste-paper basket on the desktop of most operating systems provides an abstraction of the delete function. Files are deleted by placing them in the waste-paper basket.

    What are the dangers associated with hiding the underlying complexity of computer systems in this way?

    [5 marks]

    [Seen/unseen problem]

    The most obvious danger is that users will have little idea of what to do when the intervening abstractions break down. For instance, if a disk error occurs they may lack the necessary insight to be able to diagnose a file format warning or disk head error. Alternatively, they may extend the metaphor in inappropriate ways. For instance, they may continue to place files and folders in the trash can without ever emptying it. The physical objects support a form of behaviour that makes little sense in their virtual counterparts. Stronger answers may also refer to the work of Winograd and Flores on the notion of objects being invisible ?to hand? until breakdown occurs.

    d) The Johnson Corporation have won a contract to develop a web-based system that is intended to help elderly users find friends and relations that they have known in their childhood. Write a brief technical report for the manager of the project describing how you would go about conducting requirements elicitation for the design of an interface to this application.

    [7 marks]

    [Unseen problem]

    A variety of solutions are possible here?

    Firstly, there are relatively few applications that target this user group. This means that great care must be taken. We cannot assume that interaction techniques, which have worked for other groups, will work here. The lack of previous applications also creates considerable risk - it is unclear what particular aspects of the proposed system will appeal most to the intended users. I it would be necessary to observe these users to see how such contacts are formed and reformed without information technology. It would also be necessary to see how such facilities are provided to other age groups to see whether ?best practice? can be propagated to this particular group.

    A key problem here is that the potential users may not have any idea of what is possible given current technology. This would need to be tested, however, because there is a danger that this attitude is a cliché that cannot be supported by the recent uptake of Internet services. If it were upheld then it would be necessary to mock up a system that would provide potential users with an idea of what the system might do. This creates a danger that the users will simply request an extension of the ?mock up? rather than thinking more broadly about how the proposed system could satisfy their requirements.

    It would be important to understand the particular perceptual and cognitive characteristics of these individuals. Some time must be spent looking at guidelines on accessibility for elderly users. Any requirements that are elicited from these sources must also be a focus for validation and evaluation at the end of the development process.