I have just reread The Design of Everyday Things by Donald Norman, a recommended read for anyone interested in design. As Donald Norman points out, the book was originally titled The Psychology of Everyday Things (from POET to DOET) and the book is full of insights into the psychology behind user experiences, and the failings of many designs to accommodate the experience and learned practices of users (Norman was himself a cognitive psychologist). The book is a great argument for user-centred design.
Most famously, Donald Norman introduced the design concept of ‘affordance’ in the book, building on the work of James J. Gibson who had written about affordance in terms of human perception. The book is also an argument for user-centred design.
What is an affordance? A flat plate on a door or small finger-sized push buttons ‘afford’ pushing while a long and rounded bar or handle ‘affords’ pulling. The use of the object is communicated through its appearance and form, aligning the design of the object with the goals of the user. This thinking has much in common with semiotics (read about the semiotics of design here). Donald Norman popularised the term user-centred design, arguing how design should be based on the needs of the user, with aesthetic considerations important but secondary to this. User-centred design as argued by Norman involves simplifying the structure of tasks, making things visible, using the right mapping, working with constraints, designing for error and explaining affordances.
The book starts by discussing how many objects are difficult to use. As Donald Norman says, “When you have trouble with things, whether it’s figuring out whether to push or pull a door or the arbitrary vagaries of the modern computer and electronics industry – it’s not your fault. Don’t blame yourself: blame the designer. It’s the fault of the technology or, more precisely, of the design.”
He argues that when we first see anything, we rely on the design to give us clues as to how it works. Knowledge of objects exists both in our heads and out in the world. Design must convey the essence of an object’s (or service’s) working, the possible options for action and what it is doing at any time (through feedback). Designers need to understand people as well as technology.
The affordance of something are the perceived and actual properties of that thing, and especially the fundamental properties that determine how it can be used. A chair affords sitting in (as it can support someone’s weight and is shaped to be comfortable for humans – at least usually). Most chairs also afford carrying, making them moveable. Flat, porous, smooth surfaces afford writing on (as does a well designed computer keyboard). Likewise, knobs are for turning, slots are for inserting things into, balls are for throwing (and bouncing depending on the material), etc.
When a designer uses affordances, then users know how to use something just by looking at it. They don’t need to read an instruction manual or talk to another user, but can trust their implicit understanding of how it works. By contrast, many objects don’t present a clear conceptual model of how they would work, making them difficult or impossible to use.
There are four key principles of design in the book. The first is that by looking, a user should be able to understand the state of a device/object/experience and the alternatives that are available for action. This is the principle of Visibility.
The second is that there should be a good conceptual model, with consistent use of features to achieve desired outcomes.
This is related to the third principle of Mapping which is that it should be easy to determine the relationships between the actions that a user performs and the results that are achieved (read more here), so that any controls have predictable effects and the state of the system is clear from how it looks. When mapping is natural, based on physical analogies or cultural norms, then it becomes much easier to use something. For example you move the control up and the object moves up. So brightness, loudness, weight and length are additive systems where the perception is directly related to the amount.
The fourth and final principle is that of Feedback, where the user receives full and continuous feedback about the results of their actions (an important concept in the science of control and information theory). For example, when you press a button there is a tactile feel (rather reduced in digital controls) but this can be reinforced if you also hear a sound which tells you that the button has been pushed appropriately. Similarly, there is a very fascinating sound on my laptop when I move something into the trash can.
Importantly, Donald Norman discussed the trade off between the knowledge that exists inside users heads and the knowledge that is readily available in the world (which generally is more readily available and easier to use immediately). Culture often places constraints on objects (traffic light colours have to be in a certain order and much of our knowledge is tied to specific contexts. In addition, many of the tasks that are designed for are everyday ones, and ones which are not necessarily studied by psychologists (which is why designers and researchers need to understand such behaviours at a much deeper level than would normally be necessary).
Much poor design comes from what Donald Norman calls ‘featurism’; the tendency to keep adding more and more features beyond the core tasks that a device performs. He suggests focusing more on how the task can be made easier, and less on what other tasks can be added to the features (he suggests that complexity increases at the square of the number of features). In contrast to good design, bad design:
- makes things invisible
- uses arbitrary mappings
- is inconsistent
- makes operations unintelligible
- is impolite (treats errors as the user’s fault)
- makes operations dangerous (a single error can destroy previous work)
I am sure we can all think of designs in our life which commit one or more of the above sins.
Therefore good design should always:
- Make it easy to determine what is possible (using constraints)
- Make everything visible including the conceptual model of the system
- Make it easy to understand the current state of a system
- Follow natural mappings between intentions, actions and outputs
The Design of Everyday Things by Donald Norman