HCI 17

1
HCI 17 18

• Tom Moran interview
• User Centered Design

2
Tom Moran on Design

• There are three types of design
• Custom design to meet a customers needs
• Generic design to create a generic product
• Technology exploration design to understand the
  properties of technology this is played against
  imaginary scenarios

3
Tom Moran on HCI Design

• Its like an apprenticeship
• Obviously it includes
• Figure out what the real problem is
• Considering the user
• Considering the work environment
• Considering what you are trying to address or
  redress

4
Tom Moran on the Design Process

• Carefully consider the context of the design
• Consider the organizational structure within
  which you are designing
• Big design teams require management of the teams
  focus
• There are resource constraints to deal with
• People
• Money
• Deadlines

5
HCI 17 Principles of UCD

• Design is a process
• It involves developing an artifact
• It involves different ways of representing the
  artifact during the process to test
  appropriateness
• The representation for a particular stage of the
  design process will vary depending on the needs
  of the particular stage

6
Understanding the requirements involves

• Looking at similar products
• Discussing the end users needs
• Examining existing systems to determine problems
  with current designs
• Producing various different representations of
  proposed designs
• Testing those and redesigning until a suitable
  artifact is produced

7
Representations

• Should be appropriate for the particular stage of
  design
• This implies that they will become more complex,
  complete, detailed and realistic as the design
  process proceeds
• Generally it can proceed from sketches, to paper,
  to models appropriate for wizard of oz testing,
  to full mock-ups, to initial functional models,
  to final models.

8
Models and Representation

• Need to reflect the parts under consideration
• Must be appropriate for their intended audience
• Hand drawings for exploration of the problems
  space
• Engineering drawings for functional descriptions
• Physical 3D instantiations for marketing
• Working models for user testing

9
Software system design models

• Waterfall model (seen earlier this semester)
• Assumes a fairly linear process
• Is modified to reflect iteration
• Depends on the vagaries of natural language
• Provides for checks between development stages
• Validation Check to make sure it conforms to
  client requirements
• Verification Checking to see that it does what
  it claims
• Supports keeping to a development schedule

10
Software system design models (cont)

• Spiral model
• Takes into account iterative process
• Can be hard to manage
• Can be expensive
• W model
• Easier to manage
• Allows for only one prototype iteration

11
Prototyping

• Prototyping systems support prototype development
• I use Squeak to prototype systems (you might
  want to think twice before using Squeak in a
  production system)
• Helps deal with the problem of understanding the
  system requirements

12
Discussion of OMS example

• Utilized printed scenarios of use to work out
  interaction
• Scenarios were tested against designers,
  management and users
• User guides were developed and tested resulting
  in 200 iterations on the interaction
• Simulations of the system were built and tested
  against the user groups

13
Discussion of OMS example (cont)

• Data collection
• Tours of the site
• Demonstrations of the system during development
• Interviews of users and Olympic personnel
• Retained an ex-Olympian on the design team
• Informal hallway discussions with passers-by
• Try-to-destroy-it testing
• Try-to-hack-it testing

14
Discussion of OMS example (cont)

• Resultant general design principles
• Early focus on users and their tasks
• Focus on users cognitive, attitudinal and social
  characteristics
• Measure user reactions using prototypes of the
  different aspects of the system
• Design iteratively
• Usability of all factors must evolve together and
  under one control group

15
Discussion of ATC example

• Evaluated the controllers task
• Build an initial system, install it in one
  location and test it which uncovered usability
  problems and resulted in a redesign.
• Took redesign on travelling road show to get
  local reaction to the design which resulted in
  further redesign before final development

16
Interaction design key factors

• The technology that carries the interaction
• People includes individuals and groups
• Work Analyze the tasks and consider constraits
  (time, money, manpower, safety )
• Environment Workspace, organizational...

17
Design space and your projects

• If you are working on or with an existing system,
  then the design space is limited.
• The design space is easily covered so the
  challenge will be to generate a number of
  different design alternatives
• If you are creating a new product, the design
  space is enormous.
• The challenge becomes how to constrain the space
  so that it can be covered by several (rather than
  many) different design alternatives can be
  generated

18
HCI 18 HCI Methods

• Soft Systems Methodology (SSM)
• Cooperative design
• Participatory design
• Sociotechnical design
• Open Systems Task Analysis (OSTA)
• Multiview

19
Soft Systems Methodology (SSM)

• Human actions take place within a wider context
• There is a need to understand the whole
  human-computer system
• SSM addresses this need
• Emphasizes understanding the situation in which
  the perceived problem is thought to lie, not in
  finding a solution to the problem

20
SSM (cont)

• Stages 1 and 2 concentrate on obtaining a rich
  description of the problem situation
• Stakeholders are involved in meetings
• Different stakeholders hold different views of
  the systems purpose, emphasize different aspects
  of the same problem
• These differences must be reconciled

21
SSM (cont)

• Move from the real world to the abstract
• In Stage 3 a precise definition (root definition)
  of the system is developed
• This precise definition is used by Stage 4 to
  produce conceptual models which are abstract
  representations of the system
• These abstract models should be free of real
  world constraints

22
SSM Stage 3 root definition describes

• Stage 3 root definition should cover all aspects
  of the system
• Clients (those who will benefit (suffer) from the
  new system
• Actors (those who are involved in the system)
• Frequently derive little from the new system
• Usually have to alter their work practices to
  accommodate the system (little incentive!)
• Transformation (purpose of the system)

23
SSM Stage 3 root definition (cont)

• Weltanschauung (World view)
• Owners
• Environment
• Environmental constraints placed on the system
• CATWOE model

24
SSM (cont)

• Step 5 moves back to the real world and
  compares the conceptual model developed in Step 4
  to the problem situation expressed in Step 2
• Feasible and desirable changes are exposed in
  Step 6
• Step 7 is how to carry out those changes

25
Cooperative design

• Involving the users in the design process
• Participatory design argues that the users
  (actors in the SSM model) have the right to be
  involved in the development of systems that
  affect them
• Sociotechnical design
• Focuses on developing complete and coherent
  human-machine systems
• Emphasizes consideration of both technological as
  well as social alternatives to the problem
• Open Systems Task Analysis (OSTA--gt )

26
OSTA

• Provides a means of understanding the
  transformation that occurs when a new system and
  its artifacts are introduced into an existing
  work enviornment
• Emphasizes side by side development of
• Technical requirements
• Systems structure and functionality
• Social system requirements
• Usability and acceptability

27
Cooperative design issues

• There must be an expert to guide the design
  process
• Requires what is perceived (by the designers) as
  additional effort
• Only applicable when the organizational and
  political climate is suitable
• Cost effectiveness issues require that management
  be committed to the process
• They are good descriptive methods but they dont
  provide guidance to designers on how to involve
  users

28
Multiview

• More prescriptive (rather than descriptive) in
  nature
• Elements of sociotechnical and soft systems
  approaches are embedded in this approach
• Primary Task Model (PTM) describes the purpose of
  the system, the stakeholders and the perspective
  of the system owner.
• Information analysis deals with conceptual
  modeling and information flows and structure
• (more--gt )

29
Multiview (cont)

• Elements of sociotechnical and soft systems
  approaches are embedded in this approach
• Functional model involves task allocation in the
  form of people tasks, role sets and computer task
  requirements
• After considering the people, the computer system
  is designed

30
Multiview (cont)

• Positive aspects of this approach
• Has explicit HCI design component
• Considers sociotechnical alternatives
• BEGINS analysis with human activities
• Includes entity relationship modeling and
  dataflow modeling (more on this later)
• Develops conceptual model before physical design
  decisions are made

31
Multiview (cont)

• Provides direction for system designers (unlike
  other methods in this chapter)
• Emphasizes setting the order of the activities
• Enables design to be done carefully and logically

32
Star Life CycleAn HCI design approach

• De-emphasizes the ordering of activities
• Derived from observation of actual HCI
• Relies heavily on prototype and evaluation
• Stresses rapid prototyping and incremental
  development of the final product
• Activities are similar to the waterfall model
  (prototyping, implementation, task analysis,
  requirements, conceptual design but includes
  evaluation

33
Star Life Cycle (cont)

• Can be begun in any of the stages
• Any stage can be followed by any other
• Iterative and evolutionary in its approach to HCI
  design
• The conceptual model deals with what is required
• The physical design deals with how this is
  achieved

34
The End!