23. Februar 2021, 11:00 Uhr | Harry Schubert
Mann mit Tablet-PC vor Schweißroboter
As the range of functions of modern devices and systems increases, so do the demands on human-machine interfaces. Despite greater complexity, modern user interfaces must be easy to operate. Prof. Robert Oshana from NXP proposes putting the human at the focus of the human-machine interface.
? Since industrialization, humans as users have been forced to adapt to machines. Prof. Oshana, you are now proposing to put the user at the center of the human-machine interface. What does that mean in concrete terms?
! Prof. Rob Oshana: The description »…forced to adapt to machines« is exactly what happened in the early days of user interface design and a shining example of this was the article »How the dumb design of a World War 2 plane led to the Macintosh« by Cliff Kuang . Understanding basic human factors and then designing to support these human factors is essential to successful user interface design.
Key design concepts such as minimizing a user’s memory load, mapping to the user’s mental model, and adhering to Gestalt principles are just a few examples of putting the user at the center of human-machine interaction. This is becoming increasingly important as user interfaces become essential components of wearables, automobiles, medical, and manufacturing. Forester Research has estimated that, on average, every U.S. dollar invested in the user experience brings 100 U.S. dollars in return.
? What are the distinguishing features of human-centric interfaces?
! Prof. Oshana: There are three distinguishing features of human-centric interfaces:
Another way of thinking about this is to remember the 4 Es of a good user interface; »Easy to use«, »Easy to understand«, »Error-free«, and »Effective for achieving the end goal«.
? How should a developer proceed when designing and implementing an efficient human-centric interface?
! Prof. Oshana: From a design perspective, it’s important to understand and leverage the key »laws« of user interface design and experience. A few examples are:
There are many others, and the application of these laws in the design of a user interface leads to more effective user experiences.
From an implementation perspective there are some good embedded tools technologies for UI that can be applied, these includee Embedded Wizard and Light and Versatile Graphics Library (LVGL). LVGL is an open source graphics library that provides graphical elements and widgets, python support, multiplatform, and lots of git-based examples. You can find this in the open source community or bundled with embedded software enablement packages such as NXP’s MCUXPresso.
? How can the function and quality of such a human-centric interface be verified?
! Prof. Ohana: There are some relatively simple techniques to verify the function and quality of a user interface. Jacob Neilson’s Heuristic Evaluation approach uses rules of thumb In other words heuristics – to measure the usability of user interfaces by applying inspection techniques.
I like this approach because it can be applied early in development from the prototyping phase through to final product evaluation in the field. Also, cognitive walkthroughs is another technique where evaluators ask questions about the interface from the perspective of the user performing various tasks or use cases. This is useful in the understanding of system learnability for new or infrequent use?
? Can human-centric interfaces be made abstract so that they can be ported to any application?
! Prof. Oshana: Many human factor concepts can port easily because they are innate properties of humans that never go away. If you are developing interfaces to be easily be applied in a wide variety of application segments such as wearables, medical, automotive etc., then I would recommend leveraging some of the excellent embedded UI technology tools and libraries that can facilitate embedded UI design and development.
As an example, MicroEJ offers simulation technology that supports rapid prototyping and virtual devices in order to get fast and early feedback from users. Furthermore, for embedded devices it's important to optimize memory ROM/RAM and hardware resource usage for the UI. Using tools like this and a common set of graphical widgets can help with portability across different electronics which helps leverage the initial design investment.
Technologies like Crank Software’s Storyboard gives developers the option to select the best OS for a given embedded product, whether it is for a resource-constrained microcontroller or a high-performance microprocessor, and provide a way to easily port to another when the project needs to be changed. So in summary; the human factors port easily, the hardware implementation ports more easily when using good embedded UI tools technologies.
? What examples of good and not so good human-centric interfaces can you name?
! Prof. Oshana: In general »not so good« human-centric interfaces are generally sluggish and unresponsive, complicated to use, and can be confusing to the user which leads to errors and inconsistent design techniques.
In the microprocessor category there are some excellent examples from Ventec Life Systems who have implemented multi-function ventilators with intuitive user interface technology for frontline medical professionals fighting the COVID-19 pandemic.
Ford’s Sync 4 is an excellent example of bringing usability to the car. The Sync 4 will be included in cars such as the Mustang Mach-E electric vehicle and will leverage, among other things, artificial intelligence to learn driver habits and help customize the user interface accordingly. In the microcontroller category there are some excellent embedded user interface examples in wearables from Fitbit – now Google – as well as smart home appliances and metering applications.
 Kuang, C.: How the Dumb Design of a WWII Plane Led to the Macintosh. Wired, 13.11.2019, www.wired.com/story/how-dumb-design-wwii-plane-led-macintosh.
Prof. Robert Oshana
is VP of Software R&D for the Microcontroller and Microprocessor business line at NXP, responsible for software enablement, IoT connectivity, software, middleware and security, operating systems, software services and advanced technologies.
He serves on multiple industry advisory boards and is a recognized international speaker. Prof. Oshana has published numerous books and articles on software engineering and embedded systems. He is also an adjunct professor at the University of Texas and the Southern Methodist University and is a Senior Member of IEEE.