Software Usability Course notes for CSI 5122 University of Ottawa

1
Software UsabilityCourse notes for CSI 5122 -
University of Ottawa

• Section 1
• Course Outline and Background
• Timothy C. Lethbridge
• lttcl_at_site.uottawa.cagt
• http//www.site.uottawa.ca/tcl/csi5122

2
Themes of the course

• Main theme How to develop software systems that
  are highly usable
• Sub-theme Adapting the software engineering
  process to produce more usable software
• Sub-theme Enhancing your skills at design and
  evaluation of usability
• The original proposal was that the course be
  entitled Software Usability Engineering
• Secondary theme How do do good research in
  software engineering, HCI, and Usability
• You will learn how to critically evaluate and
  write papers
• Subtheme How to design a good experiment

3
This is not a pure HCI course

• It is an applied HCI course
• Ecourse in OCICS
• There is also a separate A/S course going on
• We will focus on engineering practicalities
• Less consideration of HCI theory

4
Background required

• Just being a grad student in CS or SE should be
  enough
• All grad students will have had some undergrad
  background in software engineering
• No HCI course is assumed as background
• So far this is not required in undergrad CS
  programs
• Those who have an HCI course or other HCI
  background will have only a small advantage
• Come to class, but 20 of the material may be
  review

5
Learning about each other

• Who am I
• I have taught
• CS since 1985
• SE since 1990
• Usability at the undergrad/grad level since 1993
• Main current research topics
• Software engineering tools, including their
  usability
• Software Engineering Education
• Research projects with several companies over the
  years
• Worked at Nortel for 2 years in the 1980s
• Current co-ordinator of the Software Engineering
  undergraduate program here

6
Who are you

• Name?
• University?
• Program? (MSc, PhD, OCICS, OCIECE, Systems
  Science, etc.)
• Year of grad study?
• Why are you interested in this topic

7
Topics for the course 1

• Not necessarily covered exactly in this sequence
• What is usability
• How it compares with other qualities
• Usability in the Software Engineering
• SE methods to improve usability
• The Users and Usability Maturity Model
• Economics of usability
• Justifying an investment in usability
• Measuring usability
• Setting realistic usability objectives

8
Topics for the course 2

• Design for usability
• Task analysis
• User centred design
• Evaluation techniques
• Heuristic Evaluation
• Evaluating usability by analysis of videotapes of
  users
• Conducting formal experiments to validate
  usability
• Internationalization and localization
• Case studies in usability

9
Pedagogical method

• I will lecture about half the time
• We will study user interfaces together and
  perform evaluations live
• You will read research papers and we will discuss
  them together
• Everybody will study all the papers
• One student will start the discussion by
  summarizing the paper
• You will do projects, and present the results
• Topics include, UI design, UI evaluation,
  experiment, research

10
Evaluation of your work Your grade will be
tailored, based on a plan you will prepare

• 20 Everybody writes a final exam unless you
  already have A (gt 90) from the other components
• In which case multiply the other components by
  10/8
• Beyond this, your choice of some combination of
  the following summing to gt 80
• a) 5 Leading discussion of a paper in class
• b) 5 Leading discussion of a systems UI in
  class
• c) 25 Research (literature review) paper (20
  pages)
• d) 25 Conducting a formal experiment with 5
  users, and writing up - 20 pages
• e) 15 Evaluating a user interface and formally
  writing it up - 15 pages
• f) 10 Presentation of results from c, d, and e
  in class - 30-45 minutes, including discussion

11
Evaluation

• Guidelines and details will be provided later
• We will learn
• What constitutes a good research paper
• How to run an experiment
• How to do evaluation
• How to do a presentation
• The weights for each component can be varied
• (e.g. gt 25 or lt 25 for a larger or smaller
  experiment)
• I will give you the weights when you propose to
  me what you plan to do

12
Evaluation Your work plan

• I must approve each work item before you start
• You can prepare a full or partial work plan at
  any time
• Add to it as your ideas become more detailed
• Send plans to me by email
• For a and b, every does one, but some of you may
  do 2-3
• I will give you topics, and you can volunteer, or
  I will assign them
• For each of c, d, and e
• Email me a detailed outline of the proposed work
• I will give you feedback before you start

13
Ideas for finding topics for research papers

• Browse the literature
• I will give you some techniques for this
• Pick a topic we have discussed in class.
• The following are a few examples
• Metrics
• Design techniques
• Usability guidelines
• Evaluation techniques
• I reserve the right to deny a topic if other
  students have already picked it.

14
Research paper basic criteria

• Must be written in the same style as if you are
  submitting a paper to a journal or top conference
• E.g. SIGCHI, ICSE
• We will be reading many papers in the course, so
  you can learn from the style of those papers
• But I will also point out some bad things to
  avoid
• You must have at least 10 peer-reviewed
  references (from good conferences or journals)
• You must analyse what you have read
• Just giving me lots of facts is no good
• At all costs, avoid plagiarism

15
Evaluation project basic criteria

• The system can be open source, commercial, your
  thesis topic or something from a work environment
  (past or present)
• It cannot be so confidential that students from
  the class cannot see it
• I will later on be bringing some ideas to class

16
Deadlines You propose your own deadlines for
items c, d and e

• I will revise them for you if they are
  unreasonable!
• Clearly you have to finish your work before you
  can present it.
• I also dont want all presentations right at the
  end.
• I also dont want you to submit 2 major work
  items in the same 3-week period
• Once deadlines are set, you are held to them!
• 5 marks lost per day for each item

17
Interim evaluation of all materials

• Optional Send me a draft of your work 6 or more
  working days before it is due
• I will skim it and give you some ideas for
  improvement
• Also optional Send me your presentation slides 3
  working days before your presentation date.
• Mandatory Send me a 3-4 page summary the design
  of any experiment 4 working days before you plan
  to start executing it

18
What is usability One of many possible models

• System acceptability
• Social acceptability
• Practical acceptability
• Cost
• Compatibility
• Reliability
• Usefulness
• Utility (functions provided)
• Usability
• Easy to learn
• Efficient to use
• Easy to remember
• Few errors
• Pleasing

19
Another model of usability
20
Disciplines that contribute to the study and
improvement of usability - 1

• Cognitive psychology
• Capabilities and limitations of human senses and
  thought processes
• Ergonomics
• Hardware and software efficiency, safety and
  reliability
• Linguistics
• Syntax and semantics of commands
• Speech I/O
• Artificial intelligence
• Speech I/O
• Intelligent 'guessing' what the user wants to do
• Knowledge representations of users and tasks

21
Contributing disciplines 2

• Sociology and social psychology
• Assisting people to work in groups with software
• Ensuring software works in different cultures
• Industrial design
• Aesthetics
• Storyboarding etc.
• Engineering
• Economic analysis, cost-benefit, alternatives
  analysis
• Standards-based approaches
• Integration with other qualities

22
Dual-processor metaphor

• A good way to think about the user interface

23
Key ways of improving usability

• Focus on understanding users
• Their motivations
• Their tasks
• Their problems
• Design using effective guidelines
• Evaluation
• Repeatedly, regularly and involving users
• Heuristic evaluation
• Observation
• Setting measurable objectives, and working
  towards their achievement

24
Five levels at which you can analyse usability
issues

• 1. Task Level
• What is to be done by the user
• 2. Conceptual Level
• User's model of the system
• 3. Interaction Style Level
• Command-driven, menu-driven, direct
  manipulation, hypermedia
• 4. Interaction Element Level
• Windows, dialogs, commands, menus
• 5. Physical Element Level
• Bitmaps, characters, data structures, callbacks

25
At each of the five levels

• Design must be done
• Problems can occur
• We can think about the aspects of usability
• Learnability
• Efficiency
• Memorability
• Error handling
• Satisfaction

26
Some psychological background Attention
27
Attention - 2

• People can be readily distracted
• Provide cues about what to focus on
• People get lost in complexity
• Structure information so it is easy to browse
  through
• not too many items
• not too few items
• grouped logically

28
Attention - 3

• People multitask
• Make the 'state' clear so users can jump
  backwards and forwards
• Some mental processes are automatic, or become so
• (contrasted with controlled processes)
• These processes are very hard to unlearn
• Watch out for conflicting or changing aspects of
  the user interface

29
Some psychological background Memory
30
Memory - 2

• Short term memory can contain 72 'chunks'
• Avoid situations where users have to remember
  more than this
• Logically group things so users can chunk them
• The more meaningful, the more easily remembered
• (familiarity, imagery and consistency contribute
  to meaningfulness)
• Use effective names and icons (even) animated
  ones
• Combine icons with words
• Icons can be analogies, examples or abstract, but
  not arbitrary
• Watch out for cultural differences (e.g. washroom
  symbols)
• People can more easily recognize than recall
• Use menus, icons, quick lookup