Tomas Petricek
email: t.petricek@kent.ac.uk
twitter: @tomaspetricek
office: S129A
How to design a user interaction?
How to do it more systematically?
Model is a simplification of reality. It is useful
if it helps in designing, evaluating or otherwise
providing a basis for understanding the behavior
of a computer system.
Metaphor |
Formula |
\(ID=log_2(A/W+1)\) \(MT=a+b\times ID\) |
Know performance a priori
Make quantitative prediction
Provide framework for thinking
Raise important design issues
Perception |
Motoric | ||
Cognitive |
Social |
How to design interfaces for two hands
Comprehensive, high-level evaluation
Modelling systems with multiple participants
\(ID=log_2(A/W+1)\)
\(MT=a+b\times ID\)
Predictive model of hand moving towards a target
Simple descriptive model of keyboard keys. Is this model correct?
Two-handed interaction paradigm
Sketching using a curve drawing tool
Our hands have different roles & perform different tasks
Nonpreferred hand
|
Preferred hand
|
Evaluating design: Will an operation be easy to do?
Informing design: How to improve usability?
Delete a file
Left: Move to a file icon & select
Right: Press the DELETE key
Scrolling panel
Use nonpreferred hand to set spatial frame of reference
Notations used in computing
User experience questions
Comprehensible broad-brush evaluation
Understandable for non-specialists
Distinguish different user needs
Prompt designers to see more choices
Incrementation - e.g. adding formulas to spreadsheet
Transcription - e.g. copying data from paper
Modification - e.g. changing formula in a spreadsheet
Exploratory design - e.g. designing software structure
Searching - e.g. finding uses of a function
Exploratory understanding - e.g. understanding code
Viscosity - resistance to change
Visibility - ability to view components easily
Premature commitment - constraints on work order
Hidden dependencies - important links not visible
Role-expressiveness - purpose of an entity is clear
Error-proneness - notation invites mistakes
Abstraction - types and availability of mechanisms
Consistency - similar syntax means similar semantics
Diffuseness - verbosity of language
Hard mental operations - high cognitive demand
Case study
Two ways of specifying email filters
Visual rule editor vs. scripting language
vs.
Incrementation
Adding new condition
Viscosity
Not all additions possible
Abstraction
Condition format is fixed
Hard mental operations Everything is simple & clear
Incrementation
Adding new condition
Viscosity
Edit text for any change
Abstraction
Possible via a script
Hard mental operations Understanding code is hard
Cognitive dimensions framework
What is a better notation?
Analysing typical work situation
Multiple people, interacting with multiple computer systems
Analysing typical work situation
Multiple people, interacting with multiple computer systems
Distributed information processing system
Properties of the model
Mission-critical systems
Traditional cognitive science model
Uses the information processing metaphor
Same model for a machine and a user
Information representation and exchange is the key
Mobile unit
Radiology unit working within a large hospital,
processing examination requests from other doctors
How does it work?
What are important practices?
Can computer system help?
Request Form (RF)
Radiology unit practice
Information about request comes before patient
Request not linked to patient details
Request form is token for examination
Computer system only at the end of examination
Perception |
Motor |
Cognitive |
Distributed |
Motor model of bimanual skill
Roles of preferred and non-preferred hands
How to evaluate and improve interactions
Cognitive dimensions of notations
Activities and cognitive dimensions
Evaluating two ways of specifying mail filters
Distributed cognition model
Distributed information processing
Understanding radiology unit
What you should remember from this lecture
Tomas Petricek
t.petricek@kent.ac.uk | @tomaspetricek