CPSC 554M
Topics in Human Computer Interaction
Universal Usability, CSCW, and Personalization
UBC Computer Science - Winter 2012/13

Here are some Assignment 0 examples, together with their ratings for some of the initial articles covered in class. Note that these examples were not necessarily taken from this year's submissions.

2. Most readers will be tempted, much like Stuart Card (p. 53), to attribute the reported lack of productivity gain to the limitations of pre-1995 computers. If modern computers do indeed fare better, it would be interesting to consider the reasons why they do so and at which point productivity gains began to emerge. If not, it would be interesting to consider instead whether productivity captures accurately the usefulness of computers and what explains their popularity.

3. I've had an example of a terrible case of the 'two variable problem' burning itself into my soul for most of a decade. My rant usually decays into this: "What the hell kind of world /*is*/ this, that a microwave has more than two knobs?" Like all people I have ever observed (and I do pay attention to this), I learned long ago not to trust or use any of the advanced 'automatic' settings on any microwave. This stems from realizing that my cognitive model of properly cooked popcorn (popped, neither charred nor smoking) is clearly different from that of the system's designers (smoking heap of slag). Two variables: duration and power. It's been more than 60 years, people! Get it right! (Note: the SUB has two machines each of which have two big knobs, one for power and one for time. I believe there isn't even a start button - it's a beautiful thing.)

4. Norman's article was written at the cusp of a great sea change in terms of what desings were being sold and used; 2002 was after all the beginning of the iPod era (Oct. 2001), and the beginning of Apple's great gains in market share over the past decade. Do iPods work better than other digital audio players with similar feature sets and price ranges? Likely not, but they are undoubtedly more aesthetically pleasing than most other devices, thus making users feel good when using them. I'm curious as to how much impact Norman had at this point in time, or was he merely commenting on what he saw materializing in the design community.

5. Discussions on understanding whether a difference between means represents a real effect or not (statistically significant or not) makes me think of other ways of understanding if a difference is caused by chance alone or not. Can we use triangulation instead of it? For example, if the qualitative results support the existence of a difference between two groups, do we really need the difference to be statistically significant to accept its existence?

6. One of the mentioned problems is that reviewers favor one method over others. Is it the matter of preference, or familiarity and ability to evaluate? I believe it is the latter. For example, some may be able to evaluate the validity of a controlled experiment, while not being able to evaluate the reliability of a field study. Isn't it a general problem that the community is so diverse that reviewers cannot be enough knowledgeable to thoroughly evaluate papers that are relevant to them in some way?

2. I found it very surprising that statictics show that computers have actually reduced efficiency - maybe it was true some years ago when there wasn't enough training and computers weren't as common as they are today; but is the situation still the same in today's world, when computers are apparently very popular and have reached almost every area we can think of?

3. Norman went into a great deal of detail around the gulf between the user’s goals and the physical system; in particular, he talks about the psychological variable associated with the user, and the mapping between the psychological goals and intensions to the action sequence that will be required. Indeed these variables will be very detailed and involved for any given task, as Norman discusses; however, *how does one design for a diverse group of users?*  A group of users will certainly be individuals, and have their individual psychological variables.

4. I did not find his conclusion to be anything new to me, I have always agreed that the best interfaces are a harmony of aesthetics and usability as he said. For example, I believe that Google is winning the web application/service war because its interfaces have a good balance of design and usability. Microsoft and YAHOO are constantly struggling to play catch up and they don't seem to understand why, but I think it's because their interfaces don't achieve the same affect as Google's.

5. The author stresses the dilemma in that all methods have their own flaws and advantages so it is important to use multiple methods in empirical research to study the same problem.  He claims that this will ensure that the weaknesses in each method can be offset by the strengths of another in order to avoid any misinterpretation of the results by properly qualifying the conclusions.  However, even though he states this importance several times in this section, he never offers an approach as to how we can actually go about choosing all the appropriate multiple methods that make sense for a particular problem, and how we can properly verify that the conclusions drawn from each of the methods together support the problem being studied.

6. Most of the problems noted in pages 113 and 114 of the paper pertain to the academic scenario (and issues that arise from CHI's bias towards objective methods). Industrial research should not be affected by these issues since getting published is not as important in the industry. What are the issues concerning research and evaluation in the industry?

2. The author presents many statistics on the fact that when computers are applied in the phase two (mostly in services as opposed to manufacturing) applications (i.e., augmentation) they are not as productive as they are supposed to be. However, the data mostly reflect the situation in 80s. I wonder, is it the same in recent years as well?

3. In order to design an interface, at least three kinds of special knowledge is required: knowledge of technology, people and the task that is to be accomplished. Although a team of people can be trained in each specific area, each person should still have some understanding of the other areas in order to communicate.

4. Although the paper says that positive affect tends to make people think in a broader perspective , which might be distracting - still it goes on to state in the end that attractive interfaces work better - I found this conclusion slightly confusing. I think something that distracts the user is more likely to lower efficiency.

5. Computers can log a users interaction with a particular piece of software (eg. mouse clicks/movement, keystrokes, etc). Can this resulting data be considered a trace measure? What are the ethical concerns related to collecting this data?

6. I liked this paper because it handles one of the issues in getting research published. As the authors stated that reviewers most of the time focus on the usability, evaluations, empirical studies, results and significance which should not be the case in every research. I liked their way in giving examples where no significant usage was seen in some researches and they made a huge contribution to many recent products and research. I totally agree with authors that conferences and reviewers should give some space for replications and new ideas even when they don’t have significance usability because they may provide some insights to other researchers. Thus new inventions could occur from those ideas if they are published or would save time and resources from replicating them.

2. This reading emphasizes the lack of productivity that has resulted from phase II computing. Lots of evidence from other studies, statics, and graphs are given as supporting evidence.

3. Author expends the mapping and conceptual model  terms mentioned in previous chapter. Psychological and physical bases of cognition are important as they both require attention. Human factor is mainly the psychological base, however our design forms the bridges between these two bases.

4. The author here makes a very simple point: At first we have to ensure usability then we should go for pleasure or beauty.

5. no example

6. no example