Selected Publications
2010 onwards
HFES 2018 User-Centred Design Award
The HFES Product Design SIG awards the Stanley Caplan Award every year and in 2018 Elekta’s MRLinac (Unity) was the winning entry, celebrating work that is documented more completely in other publications.
Design of new radiation therapy machine
Millions of people worldwide benefit from radiotherapy every year, and the treatment cures more people than cancer drugs. DCA and Elekta describe a project in which human factors played a critical role in the design of new equipment that delivers the therapy to patients
Making Things Better
Human Factors and User Experience design have developed relatively independently and from different traditions and yet they both share the all important goal of making better things for people to use. This paper explores how we might bring them closer together and recognise the overlaps – especially in the context of safety-critical systems.
21st Century HMI
The Human-Machine Interface (HMI) went graphical many years ago, but it never quite went fully digital, which makes bringing the enhancements of data analytics and AI to bear much harder. This poster summarises GE Research’s efforts to explore what a digital HMI might be like and how users would respond to it.
Decision-making in a design process
Based on the story of the development of the Unity MRLinac, this paper focuses on the use of human factors methods to improve the decision-making during the design process.
Assessing the design of a new HMI
As industrial plants (e.g. gas-fired power stations) become more and more digital the human-machine interface (HMI) is in danger of being left behind. These studies evaluate modern digital designs for power plant HMI.
Usability engineering for a complex, medical device
This paper summarises the activities required and complexities encountered while undertaking the usability engineering process for a large, complex, and new-to-the-world medical device – namely Elekta Unity, the first high field Magnetic Resonance (MR)-Linac.
Radiotherapy's Future Experience
This paper describes how a suite of research techniques were used to inform the development of a vision for the future of radiotherapy that was to later become Elekta’s MRLinac Unity.
2000 - 2009
Design in Harmony with Human Life
This presents IDEO’s approach to human-centred design as one in which the relationship between people and technology needs to be thought of as musical notes in harmony – not the same note in different octaves, but different notes that come together to make a wholly new sound.
Bringing together the design community at the CHI conference
While most of the HCI literature can be seen as part of an engineering-science practice (with an emphasis on the acquisition and interpretation of ‘facts’), the CHI2006 Design Community focuses on how arts and engineering come together in the construction, study and interpretation of created objects (maybe more like the study of literature and criticism).
HFES 2002 User Centered Design Award
The Evenflo Stroller was full of user-centred innovations that consumers who bought the pushchair loved. The design process was well-documented and was exhibited at the Smithsonian (at the Lemelson Centre, Invention of Play) for a number of years (but no longer).
Assessing the value of a product's experience for a user
Through the 2000s design practice increasingly realised that a great user interface – even a great user experience – was not sufficient if the offering was not valued by users and customers. This CHI workshop held in Florence, Italy, explored the territory of ‘value-centered HCI’ searching for ways to accommodate financial and cultural (moral) values into the equation.
Heat, Fire, Temperature – understanding theory and practice
This paper explores the question as to whether we really know what we mean by usability, user experience and user interface, drawing analogies to pre-industrial revolution understanding of the concepts of heat, fire and temperature – concepts we take for granted now, but which were not trivial to discover.
Getting contextual research supported
In the early days of asking to observe users in context as a key part of the innovation and product design process there was much resistance to the time and budgets required, with companies believing that they knew their user needs perfectly well. This article explores ways in which this resistance might be overcome.
Being alert to good design
Just as with psychology, user experience design has been hampered by a disease model … with arguments focused almost exclusively on the diagnosis and avoidance of damage, rather than on developing our knowledge and understanding of how to create pleasurable (and profitable) experiences. This paper presents an argument for a more positive frame of mind in our approach to our discipline.
1990 - 1999
The nature of expert knowledge in programming
This book chapter summarises the research to date on programming plans and programming expertise.
The Psychology of Programming
This volume brought together the work of the previous ten years or so on the psychology of programming from across the world.
Methodological Issues in the Study of Programming
This book chapter summarises the challenges and solutions to the methodological issues involved in studying the psychology of programming.
Towards a cognitive browser for OOPS
In order to support the reuse of object-oriented programming code, the project documented here aimed to produce a ‘cognitive browser’ to enable software engineers to find (re-find) familiar and re-usable code elements.
Does the Programming Notation Matter?
This book chapter examines the question as to whether the programming notation matters so much once we have more sophisticated software engineering environments and the tools within them. It concludes that notation matters, but recognises that some of the notational problems may be alleviated through the development of tools.
User-Centred Requirements for Software Engineering Environments
An edited volume that resulted from a NATO-funded workshop held in France in 1992. It brings together psychologists, sociologists and software engineers to discuss what is known and understood about the needs software engineers have of their software engineering environments.
Learning graphical programming: An evaluation of KidSim™
This paper presents part of an evaluation of a new children’s programming environment, developed by Apple Computer Inc. for 10–13 year old children. It concludes that it was easier to measure the success in engaging the students than it was in measuring the programming abstractions learnt .
Are Objects That Important?
Whatever the improved effectiveness of object-oriented programming, is it the objects that are providing that improvement? Is it really the case that objects are a more natural way for a developer to think about their code?
Factors influencing the classification of object-oriented code
How do expert programmers approach the problem of thinking about object-oriented code and the task of efficiently and effectively re-using it?
Giving children an experience of programming
As part of studying children learning to program in a new environment (KidSim) we examined how collaboration and discourse changed between code-writing activities and code-testing activities.
Usability is not all that matters
Building interfaces that make a piece of educational software easy to use, may not make it more educational – indeed it may even make it less educational. This paper looks at this phenomenon in the context of moving balloons, boats and ponies around a virtual world.
Breaking the Rules of Direct Manipulation can be a Good Thing
It has been shown that easier interfaces may not always lead to the most transferable learning – this paper examines possible reasons why breaking the rules of direct manipulation might have this effect.
Transporting Honey Bears
This paper examines how interface style can change a user’s approach to problem-solving through that user interface. It raises questions about the appropriate metrics for understanding ‘usability’ (learning and problem-solving success versus speed and button clicks in performance).
Selection through Rejection: Reconsidering the Invariant Learning Paradigm
Implicit learning studies tend to assume that a participant has learnt the intended, unstated ‘concept’. This study shows how it is possible to score as though one has learnt concept <A>, when in fact what one has learnt is the ability to recognise a few “NOT concept <A>” elements.
Before 1990
Flowchart Usefulness
This was my undergraduate final year project, examining claims about the universal value of flowcharts (or structure diagrams) for helping reason about algorithms. In the beginning of a recurrent theme, the results showed that the value of a representation was dependent on the task to be performed.
Are some programming languages harder?
Differences in the usability of programming notations are shown to be task-related, rather than inherent in the notation – perhaps the first part of recognising that usability is related to the intended user experience and the range pf tasks that that encompasses.
Seeing structure
This paper examines the way that visual structural information in a computer program may provide important information to a programmer and assesses that for both novice and expert programmers.
Programming 'Plans' and Expertise
Although program plans are a recognised component of programming expertise, the programmer’s ability to deploy that expertise may be dependent on the particular programming language, with more structured representations making it easier for the expert to see the plan being used.
The application of machine learning to intelligent tutoring systems
A review of ways in which (in 1986) machine learning might be applied to educational tutoring systems.
Programming Plans and Programming Expertise
This paper shows how programming plans do not represent the programmer’s deep understanding of the domain, so much as its expression in a particular language.
How to make a coin appropriately valuable?
Designing new coins requires that they convey an appropriate sense of their value. These studies showed that value was largely interpreted with reference to cultural norms (brass < silver < gold) and also by past experience (older generations being influenced by memories of coins long gone).
An early piece of work in machine learning
This exploration into machine learning found that the representation of the structure of the problem (data) was a more important factor for AI performance than the algorithm used.