1. Interaction Design Process COMPSCI 345 S1 C and SOFTENG 350 S1 C Lecture 19 Lecturer: Jim Warren Based on Heim Chapter 3

2. Interaction Design Process Outline –Iterative Design –User-Centered Design –Interaction Design Models –Overview of Interaction Design Models 2 © Heim 2008

3. Iterative Design Interaction design is an iterative process –Two steps forward, one step back –Similar to drafting and revising a paper The knowledge path is constantly moving forward 3 © Heim 2008

4. User-Centered Design (UCD) The objective of UCD is to develop a design framework that enables interaction designers to build more usable systems Design should emerge from the user’s –tasks –goals –environment Focuses on human-centric issues –cognition –perception –physical attributes and conditions user environment 4 © Heim 2008

5. User-Centered Design The basic tenets of user-centered design: –Early focus on users and their tasks –Continuous evaluations to determine ease of learning and ease of use –Iterative design KNOW YOUR USER 5 © Heim 2008 User Participation Walkthroughs Focus Groups Ethnographic Observations Questionnaires Expert Evaluations Usability Testing

6. User Participation in UCD Users should actually be part of the design team –Constant contact with the target audience –However, if the project is long, then any particular user may lose their perspective and start to relate to the problems of the designers instead of their user community 6

7. Interaction Design Models Waterfall Model Spiral Model Dynamic Systems Development Method Prototype-Based Models Discount Usability Engineering Contextual Inquiry 7 © Heim 2008

8. Waterfall Model Was a great advance on earlier approaches that had just analysis and then straight to programming Strength of well-defined progress, clear deliverables and easy to plan (great for government contracts) But very rigid for interactive applications Back-tracking (dashed arcs) isn’t really integral to the model (hence the name!) 8 © Heim 2008

9. Waterfall Model Advantages of the waterfall model: –Highly disciplined process of documentation –Easily observable development progress –Easy to create budget –Consistent review process Disadvantages of the waterfall model: –Document / deliverable centered rather than user centered –Requirements are inflexible 9 © Heim 2008

10. Spiral Model More flexible Centered on risk reduction Incorporates prototype development Encourages iteration Breaks project into subprojects identifying specific risks 10 © Heim 2008

11. Spiral Model Advantages –Appropriateness for large-scale enterprise systems –Flexibility in terms of its sensitivity to the dynamic nature of the software industry –High sensitivity to risk at each stage of development Disadvantages –Complex nature makes it difficult for customers to grasp –Perhaps excessively focused on risk and risk assessment rather than usability 11 © Heim 2008

12. Iterative design and prototyp-ing is ihherent 12 In-built opportunities to stop project

13. Dynamic Systems Development Method (DSDM) A flexible, rapid application development (RAD) method Time-sensitive –Timeframe and allocated resources are fixed Functional requirements are flexible Business-centered –Main focus on on-time delivery of high-quality software for current business needs 13 © Heim 2008

14. Dynamic Systems Development Method (DSDM) The DSDM framework recognizes nine principles: 1.Active user involvement is imperative. 2.The team must be empowered to make decisions. 3.The focus is on frequent delivery of products. 4.Fitness for business purpose is the essential criterion for acceptance of deliverables. 5.Iterative and incremental development is necessary to converge on an accurate business solution. 6.All changes during development are reversible. 7.Requirements are baselined at a high level. 8.Testing is integrated throughout the life cycle. 9.Collaboration and cooperation among all stakeholders is essential. 14 © Heim 2008 Very enabling for usability engineering

15. MoSCoW’ed requirements DSDM requirements are prioritised on the MoSCoW taxonomy: –Must Haves – system is not useful without –Should Haves – there’s a workaround in short-term; system of some use without –Could Haves – can be left out of increment under development –Won’t Have – waiting list of valuable features: won’t include until later development 15

16. Prototype-Based Models Prototypes are used to develop, demonstrate and test design ideas Appropriate for small-scale projects Enable discussions of: –Look and feel - Scope –Information flow - Product concept Throwaway Prototyping Model –Used to elicit requirements and then disposed of Evolutionary Prototyping Model –Prototype is semifunctional and evolves over course of project 16 © Heim 2008

17. Prototype-Based Models Advantages of prototyping include the following: –Easy for users to give feedback –Reduced development time and cost –Involvement of the user in the development process Disadvantages of prototyping include the following: –Can be viewed by client as the final product –May lead to insufficient analysis due to the ease of (and emphasis on) development –Difficult for developers to discard and start creating the final product from scratch 17 © Heim 2008

18. Discount Usability Engineering (DUE) Jakob Nielsen (1994) argued that the benefits derived from even small amounts of user testing would have a significant impact on the usability of the design. DUE is based on the use of the following three techniques: –‘Scenarios’ (by which he means tightly scoped prototypes for getting feedback; word is used in slightly different sense in modern Interaction Design) –Simplified thinking aloud (don’t need a fancy laboratory) –Heuristic evaluation (with whatever heuristics, but especially his 10) 18 © Heim 2008

19. Discount Usability Engineering (DUE) Nielsen suggested that the number of problems that could be identified from a usability test with n users can be calculated according to the following equation: 19 N [1-(1-L)^n] where: N total number of usability problems in a design L proportion of usability problems discovered with a single participant (typical value for L is 31% based on their research) © Heim 2008

20. Discount Usability Engineering (DUE) Nielsen proposed a set of 10 usability heuristics that could be used by designers to investigate and document usability problems. –DUE requires some experience –Should be done by a few reviewers to avoid personal bias Will help to indicate issue frequency –Should be done early in the design process 20 © Heim 2008

21. Contextual Inquiry Involves –Observation, Inquiry and Interpretation It is based on four main principles: –Context how and why people use software products –Partnership partnering with a typical user in a master/apprentice relationship. –Focus Observations are focused on collecting information on tools, artifacts, terminology, sequences, methods and interactions –Interpretation Put observations on post-its and create groupings (e.g. as per http://baran-systems.com/Products/Affinity%20Diagram%20for%20Excel/index_concept.htm ) Create models of communication flows, information sequences, physical environments and corporate culture 21 © Heim 2008

22. Overview of Interaction Design Models Elements that appear in many of the standard models –Cost and risk analysis –Observation –Task analysis –Requirements assessment –Conceptual design –Physical design –Prototyping –Evaluation –Usability testing –Implementation –Maintenance 22 © Heim 2008

23. The Design Process Model Discovery phase questions –What are the components of the project? –Who is involved? –What are the current work space and work flow like? –What are the contextual and extraneous factors that affect the work flow? 23 © Heim 2008

24. The Design Process Model Design—The design phase has two parts: –Conceptual Design—What are the possible ways in which the design can address the needs of the problem space? Personas, Scenarios and Use cases, etc. –Physical Design—What are the possible ways that the conceptual design can be realized in the real world? Low-fidelity prototypes Wireframes Functional prototypes 24 © Heim 2008

25. The Design Process Model Evaluation questions –How can we determine the relative merits of one design over another? –How can we measure the success of a proposed design? –How can we get real users to give us feedback about a proposed design? –How can we incorporate usability testing at the early stages of the design process? This is documented by the results of formal and informal usability testing 25 © Heim 2008

26. Summary Interaction design development process –Basic stages are discovery, development, evaluation and implementation User centred development frameworks help us to design more usable systems –Taking into account the user’s abilities, limitations, goals and environment. 26