HCI in the software Iterative design and prototyping process - PDF document

HCI in the software process • Software engineering and the design process chapter 6 for interactive system s • Usability engineering HCI in the software • Iterative design and prototyping process • Design rationale the software lifecycle The waterfall model Requirements specification • Software engineering is the discipline for Architectural understanding the software design process, or design life cycle Detailed design • Designing for usability occurs at all stages of Coding and the life cycle, not as a single isolated activity unit testing Integration and testing Operation and maintenance Activities in the life cycle Verification and validation Requirem ents specification Real-world designer and customer try capture what the system is requirements and constraints The formality gap expected to provide can be expressed in natural language or more precise languages, such as a task analysis would Verification provide designing the product right Architectural design Validation high-level description of how the system will provide the designing the right product services required factor system into major components of the system and how they are interrelated needs to satisfy both functional and nonfunctional requirements The form ality gap validation will always rely to some extent on subjective means Detailed design of proof refinement of architectural components and interrelations to Managem ent and contractual issues identify modules to be implemented separately the refinement design in commercial and legal contexts is governed by the nonfunctional requirements 1

The life cycle for interactive Usability engineering systems The ultimate test of usability based on measurement of user cannot assum e a linear Requirements experience specification sequence of activities Usability engineering demands that specific usability measures be as in the waterfall m odel Architectural made explicit as requirements design Usability specification Detailed – usability attribute/ principle design – measuring concept – measuring method Coding and – now level/ worst case/ planned level/ best case unit testing Problems Integration – usability specification requires level of detail that may not be lots of feedback! and testing – possible early in design satisfying a usability specification – does not necessarily satisfy usability Operation and maintenance part of a usability ISO usability standard 9241 specification for a VCR adopts traditional usability categories: Attribute: Backward recoverability • effectiveness Measuring concept: Undo an erroneous programming sequence – can you achieve what you want to? Measuring method: Number of explicit user actions • efficiency t o undo current program Now level: No current product allows such an undo – can you do it without wasting effort? Worst case: As many actions as it takes to • satisfaction program-in mistake Planned level: A maximum of two explicit user actions – do you enjoy the process? Best case: One explicit cancel action Iterative design and some metrics from ISO 9241 prototyping Usability Effectiveness Efficiency Satisfaction • Iterative design overcomes inherent problems of incomplete objective measures m easures m easures requirements Suitability Percentage of Time to Rating scale • Prototypes for the task goals achieved complete a task for satisfaction – simulate or animate some features of intended system – different types of prototypes Appropriate for Number of power Relative efficiency Rating scale for • throw-away trained users features used compared with satisfaction with • incremental an expert user power features • evolutionary Learnability Percentage of Time to learn Rating scale for • Management issues functions learned criterion ease of learning – time Error tolerance Percentage of Time spent on Rating scale for – planning errors corrected correcting errors error handling – non-functional features successfully – contracts 2

Techniques for prototyping Design rationale Storyboards Design rationale is inform ation that explains why need not be computer-based a computer system is the way it is. can be animated Lim ited functionality sim ulations Benefits of design rationale some part of system functionality provided by designers – com m unication throughout life cycle tools like HyperCard are common for these – reuse of design knowledge across products Wizard of Oz technique – enforces design discipline Warning about iterative design – presents argum ents for design trade-offs design inertia – early bad decisions stay bad – organizes potentially large design space diagnosing real usability problems in prototypes… . – capturing contextual inform ation … . and not just the symptoms Issue-based information Design rationale (cont’d) system (IBIS) Types of DR: • basis for m uch of design rationale research • Process-oriented • process-oriented – preserves order of deliberation and decision-m aking • m ain elements: • Structure-oriented issues – em phasizes post hoc structuring of considered – hierarchical structure with one ‘root’ issue design alternatives positions – potential resolutions of an issue • Two examples: arguments – I ssue-based inform ation system (I BI S) – modify the relationship between positions and issues – Design space analysis • gIBIS is a graphical version structure of gIBIS Design space analysis supports • structure-oriented Position Argument responds to • QOC – hierarchical structure: Issue responds to questions (and sub-questions) objects to Position Argument – represent major issues of a design specializes options – provide alternative solutions to the question Sub-issue generalizes criteria questions – the means to assess the options in order to make a choice Sub-issue • DRL – sim ilar to QOC with a larger language and m ore form al sem antics Sub-issue 3

the QOC notation Psychological design rationale Criterion • to support task-artefact cycle in which user tasks are Option affected by the system s they use • aim s to m ake explicit consequences of design for users Question Option Criterion • designers identify tasks system will support • scenarios are suggested to test task Option Criterion • users are observed on system • psychological claim s of system m ade explicit • negative aspects of design can be used to im prove next Consequent … … Question iteration of design Question Summary The software engineering life cycle – distinct activities and the consequences for interactive system design Usability engineering – m aking usability m easurem ents explicit as requirem ents Iterative design and prototyping – lim ited functionality sim ulations and anim ations Design rationale – recording design knowledge – process vs. structure 4