Don Norman, Usablity Expert
Design errors in academic design curricula, the 2000 presidential ballot and aviation and nuclear accidents are considered, the concept of human error is called into question, and a call for human-centered design for the fallible user is put forward in this discussion from 2001.
How did you become a usability expert?
I was a Cognitive scientist at the University of California, San Diego studying, among other things, human error. As a result I was called in to look at some accidents (nuclear power and aviation). I soon realized that the accidents were the result of bad design if the designers had wanted to cause human error, they could not have done better.
This led me on a quest to understand why design fails the people who must use it. Thus, when I look at the curriculum of noted schools of design, it is rare to find anything at all about how to understand the people who will use, read, or otherwise make use of the designs. See for example, http://design.schoolofvisualarts.edu/curr_program.html
Why do people make errors? That is a nonsensical question. Human error is an entirely invented term. In the real world, we dont err. We approximate. Language and motor skills are wonderful at adapting to changes, to getting closer and closer to the desired outcome. Do we err when we walk along a weaving path? Of course not that is how people walk. It is only when engineers and designers require us to walk in straight lines that we call the behavior erroneous so too with almost every place where people err.
Do we err at arithmetic? Of course. Arithmetic is an artificial skill, not one well sited for the human mind (thats why I champion calculators to do the arithmetic for us). Do we err at precision? Of course. Thats not a biological necessity - it is only an engineered and designed necessity.
Society has invented machines and designs that require inhuman acts. We require precise numerical precision from the human body that did not evolve with precision. We create fanciful print displays that are delightful and fun especially for the designer but that are difficult or impossible to read.
How did I become a usability expert? I rebelled against the arbitrary nature of engineering, of design. I determined it was time to put the human foremost into the equation. That is why I have written my books, taught my students, and championed human-centered design. I am now a trustee of the Institute of Design in Chicago. Maybe I can have an impact in design curricula. Hurrah hurrah.
Design came to the fore of Americas consciousness with the 2000 election butterfly ballot. Tell me, why are such problematic things designed in the first place? Is someone asleep at the wheel? Or is the design done by non-designers?
No it didnt. Design has long been in our consciousness. Look at automobiles, look at architecture, look at furniture. Industrial design has long been considered to be essential to product success look at the annual Business Week awards.
The butterfly ballot simply points out what happens when you neglect human psychology. The ballot was designed by well-meaning amateurs who never considered testing the resulting ballots (and probably didnt know how to do such a test). The ballot machinery was designed to save money, not to maximize accuracy.
Is the design done by non-designers? Yes. But, most designers would do worse. Designers learn about aesthetics. They seldom learn about human psychology, about the need to do iterative design with a repeated cycle of watching people at work, determining requirements, doing a rough prototype, testing the rough prototype, then refining the design by watching, predetermining requirements, doing a new prototype, and retesting. The cycle can be done in one day. Any design schools teach this? I thought not.
Most designers are just as guilty as non-designers. It is not their fault - it is the fault of the design schools.
You are right, design schools are lax about addressing the end-user and examining how products work in society. There are very few classes that I know about, including my own (which you cite above) that overtly and intensely do what you do, and what you propose. But isnt usability and functionality implicit in any design program where the goal is indeed utility? How can this be taught better?
In my experience, an item does not receive the proper attention and concern unless it is explicit, unless conscious attention is paid to the issue. Things that are not conscious that are implicit are things ignored. If you want to do something right, you have to attend to it. Can design be taught? Of course. Can design for usability be taught? Of course. Can design that emphasizes functionality be taught? Of course. Note, however, these are three different things. Emphasis on usability does not mean that the result provides anything of value. Emphasis on functionality does not mean that the result is usable. Emphasis on usability and functionality does not mean that the result is aesthetically pleasing, or manufacturable, or cost-effective. Every dimension is important. Each must be considered explicitly and consciously. Each must be taught explicitly and consciously.
The butterfly ballot is not the only poorly designed ballot. We have quality standards for so many things, why cant our most valuable right be protected?
Alas, the prevailing opinion is that if someone uses a piece of equipment or software improperly, it is the persons fault. So the designer walks away smugly saying, well, if you cant use it right, you shouldnt be using it. Bad attitude is the real culprit.
Note how many people said that those who voted wrongly in Florida were just stupid voters.
Bad attitude and a lack of consideration for people is the major culprit.
Probably you do the same. When you turn on the wrong burner on your stove, you probably blame yourself. In fact, it is the bad design of the stove controls that should be blamed.
You are an advocate of good design. But what is the definition of good design when applied to everyday things? Is good design a priori simple design?
Is good design a priori simple Of course not. In fact, the simplest looking designs are often the most complex to produce. I own a Michael Graves Nana teapot. It isnt simple, it is bizarre. I love it it is great design.
What do you mean by design"? Aesthetics? practicality? manufacturability? Usability? Cost?
All are part of design.
Can good design be governmentally legislated?
Of course it can. You can mandate anything: good sex, good food, good books, good baseball. Its silly, but governments often do silly things.
If usability and functionality are to be legislated, it should be in the form of performance requirements. Thus, the proper way is not to force design guidelines that unnaturally constrain the designers. Instead they should force performance requirements, such as: 99% of people should be able to perform the task with less than 2 minutes instruction, with an error rate less than X. I favor performance specifications for situations that are critical. I favor hands off for the rest.
I frequent a restaurant that has a door without a push or pull sign, but when pushed (which is the right way to exit the door) my fingers invariably brush against a metal doorstop. Im sure that I am not alone, and Im sure that this occurs in many places. Why cant standards be applied and then monitored?
Why cant common sense be legislated, mandated, applied and monitored?
(But quick does your design school teach designers to test with real people to discover such things? I bet not.)
My design school teaches students to be aware of what the public needs and doesnt need. Do they test their products on real people? Not often. Does this process start in school or should it start elsewhere?
Although this process should clearly start in school, I am bothered by having someone else determine what I need or dont need. Do I need the Graves Nana teapot? Of course not. Do I like it: Yup. Do I need reality shows on television? Of course not. Do I get it? Yup.
In advocating good design as a solution to unnecessary mistakes, where in your equation is user fallibility?
On the side of the angels. Humans are fallible. Learn that. Cherish that. It is a fact of life. We people are creative, exciting, adventurous, imaginative, artistic, emotional, musical. And fallible.
Design for people as they are, not as you would have them be. Design for inefficient users. Design for fallible users. Design for creative, imaginative people who will do things with your design that you never dreamed of, things both good and horrid. Design for people who are tired and stressed, cranky and irritable, sloppy and inattentive. In other words, design for real people.
Can good design actually be infallible enough to mitigate the human element?
Do you believe, as I do, that the designers of instruction manuals make confounding guides? In other words are there built in handicaps for usability, and if so, why?
Those who write instruction manuals know what it is like to be in hell. They are given unusable objects and told to write clear, intelligible instructions. And, of course, they must also heed the legal warnings. They have an impossible job.
Want to know what designers of manuals should do? They should design and write the manual before the product is designed. Make the manual simple and elegant. Then insist that the designers build it the way they have described it. Then we might actually get usable products. and simple manuals.
The best designed products wont even need manuals.
When we talk about poor usability we refer to windows that dont open, VCRs that are too complex, toys that have too many parts. Other than the butterfly ballot, in your experience what graphic designs would say are user unfriendly?
What graphic designs are user unfriendly? Almost any graphic design that wins a prize. Any graphic design that is intended to impress design colleagues and design judges rather than to be used by people.
This raises the issue of web site and interface design. In the past a poster or brochure performed two tasks, to attract the eye (style, aesthetics) and convey a message. Of course, as you imply, sometimes these are at odds in overly designed graphics. Today, on the one hand, designers are applying a lot of quirky typographic styles to web pages. But on the other, many web pages (i.e. Excite.com, Amazon.com, Inside.com) are packed with text. What is desirable? What is most useable? How can the web be made more efficient?
I dont want to judge a website unless I know the goal of that site. Quirky typographic styles have their place. Yes, they are hard to read, yes, they are hard to use, but they display interesting design ethics. They are part of the exploration of art, to push the boundaries of what is possible, to cause people to think anew, to view the world differently.
But if the site is for a business, trying to attract customers, then the rules differ. First of all, we dont need to attract the eye in quite the same way as in a poster. With a poster, you want someone to look at it. With a web page, well, the person is already there. Presumably they went there voluntarily. So you dont need to attract their attention you already have it. Now is the time to provide real value. To answer questions, to make the trip worthwhile. Excite, Amazon, Inside, Yahoo have all discovered that packing the website with lots of information is what their viewers want, just as newspapers cram multiple stories onto page 1. Is this ugly? Perhaps. Is this desirable? It all depends upon the audience you are trying to work with, to keep happy, to cause them to return. Is it efficient? Who cares? People are not efficient, so I do not know what it means to want an efficient website. Mind you, I do want one that does not waste my time, that delivers what I want. But mind you, sometimes what I want is fun, excitement, art, or entertainment.
Exercise: Go back to that website for the design school: http://design.schoolofvisualarts.edu/curr_program.html . Does this excite you, answer your questions, and make it obvious why you are there and what you will learn? Nope. The designers of this website may know about visual design but they do not understand people.
How can we make a website better? Design for the people who will be using it.
If you had all the power in the world to effect universal usability, how would you wield it?
I would train designers, engineers, and computer scientists differently. This way, we might solve the problems right at the source.
Other Articles By Steven Heller
- The Boulevard of Broken Dreams
- Marshall Arisman: Facing the Audience
- Edward Fella: Letters on America
- R. Crumb: Odds and Ends
- Tomi Ungerer, A Childhood Under the Nazis
- The Swastika: Constructing The Symbol
- Art of the Third Reich
- Design History
- Graphic intervention
- The Small Crit
- Zap Comics
- Quick Cuts, Coarse Letters, Multiple Screens:
- Ladislav Sutnar and Knud Lönberg-Holm
- Otto Bettmann: Father of Retro
- Advertising Arts: America Becomes Modern(istic)
- Erik Nitsche: The Reluctant Modernist
- Old Guns
- Graphic Design Magazines: Das Plakat
- The Object Poster:
- Walter Herdeg, editor and founder of Graphis
- Cult of the Ugly
- Paul Thompson, director of the Cooper Hewitt Museum
- Merrill Berman, Design Connoisseur
- Neal Gabler, media critic
- Ladislav Sutnar, Web Design before the Internet
- Typographica Mea Culpa, Unethical Downloading