Donald Norman is a professor of computer science at Northwestern University in Chicago. From distance learning systems to advaced digital TV, he has been involved in most kinds of interface design work. Norman set up his own consulting ground to help companies make 鈥渉uman-centred鈥� products. He is also a fellow of various august American bodies, including the American Academy of Arts and Sciences. Among his books are Things that Make Us Smart and The Invisible Computer. He is working on a book about the future of everyday things.
What鈥檚 wrong with the new BMWs?
They got rid of a lot of the knobs and controls on the dashboard and replaced them with a device taken from video editing called a jog wheel and a display. They say it鈥檚 just as easy to use as a computer mouse 鈥� with a single control you can select any one of 720 possibilities. It has been universally condemned. It鈥檚 a step backwards. Yet when I talked to two designers, one from BMW and one who claimed he gave them the idea, they both defended it but admitted they鈥檇 never really used it.
Advertisement
Are you disappointed that there has been such slow progress in making computers and gadgets usable?
I鈥檝e been worried about this for some time now. Why do we have so many unusable things when we know how to make them usable? I think it has to do with the fact that the usability advocates don鈥檛 understand business. Until they understand it and how products get made, we will have little progress. In the field of design, people come from three very different backgrounds. They come from art and architecture schools and they know how to make attractive things. Or they trained in computer science and psychology and they know how to make usable things but they don鈥檛 know how to build anything, they鈥檙e just good at finding flaws. Or they come from ethnography, and they are superb at understanding what people really need, but don鈥檛 know how to translate that into products. So all this has to come together, otherwise no decent products will result. I鈥檝e been trying to understand why usability people are left out of the game, and I think it鈥檚 because they appear to have nothing to contribute.
In your best-known book, The Design of Everyday Things, you say that when we rented phones, it was in the phone company鈥檚 interest to make them reliable and long-lived. Now we buy them, the interest in making long-lasting design is gone. Is economic interest the major driver of design?
It鈥檚 one of them 鈥� that and the mad rush to release products. A product released every six months doesn鈥檛 give you any breathing time. So I feel the solution is that people in the field of usability must understand both design and business better. We must be part of the solution, not critics finding the problems. The model from the software industry is that it鈥檚 very important to have flaws in the software to give you a reason to upgrade and buy the new version 鈥� though my friends in the software world will deny that they ever do it that way.
Do you think there is a lot of 鈥渇orgetting鈥� in the design world? In computing, for example, it seems like whenever companies embark on something new, they never look at what has been done in other fields
One of the things I have noticed is that in the computer world a lot is known about how to make things usable, but now as computer technology moves into other fields, they鈥檙e repeating the same mistakes. Each time, people think everything is new. It takes about five years to sort that out, and then the fields do talk to each other.
Could 鈥渙pen source鈥� solve some of this 鈥� in the computer industry at least?
You don鈥檛 do good software design by committee. You do it best by having a dictator. From the user鈥檚 point of view, you must have a coherent design philosophy, and I don鈥檛 see how that could come about from open source software. The person who鈥檚 done it best is Steve Jobs, and he鈥檚 well-known for being a tyrant.
And yet the Internet was built on open standards but is usable by all kinds of people鈥�
The Internet has been successful, but it could have been designed better. What鈥檚 successful is the interface 鈥� the graphical user interface really made a big difference. But we know it doesn鈥檛 work on the cellphone. I believe that having information available on a cellphone will turn out to be more important than on the computer, but it will have to have a different format. And the situation in which you use it will also be different.
It鈥檚 very interesting that SMS has been such a success when it has such a miserable interface compared with a keyboard鈥�
The cellphone is a very interesting development 鈥� it鈥檚 really an emotion machine. The very success of SMS, despite the difficulty of using it, is that it lets people sort of touch each other and keep in touch throughout the day. It鈥檚 not about sending factual information, it鈥檚 about keeping the social structure going.
How did you get into design and usability?
When I was a kid, I was a nerd, a tinkerer. I took things apart and put them back together. That got me interested in electronics, so I started at MIT with a degree in electrical engineering. I cared about computers, and in those days, 1957, there were no universities or departments specialising in the technology, so I went to the University of Pennsylvania, because that鈥檚 where computers were invented in the US, and got a master鈥檚 degree in electrical engineering. Then the psychology department opened up something they called mathematical psychology, and I was perfect for that and moved over. At first, my psychology was no different from engineering. Instead of trying to build intelligent machines, I was trying to understand them. But over the years, I moved more and more towards the study of cognition, and how people do things, and the errors and accidents that people make. And then there was that wonderful year I spent in England at the Applied Psychology Unit at Cambridge University.
What was so great about it?
By then, I had already started to work on the complexity of computer systems, and I had done some work on nuclear power systems and had a little bit on aviation. But in Cambridge I became so frustrated with British water taps and switches and door handles 鈥� those awful sideways handles on many British doors that catch your sleeves. They don鈥檛 exist in the US. The most frustrating thing about it was that no one seemed to care. I realised that the same principles I had been working on with exotic electronic machines could be applied to everyday things. So I wrote The Design of Everyday Things (which was, interestingly, called The Psychology of Everyday Things in its first edition). It became obvious to me that if I wanted to have any impact in the world of design I had to move from academia to industry. People in the design world started talking as if I were a designer, and I thought, 鈥淲hy not?鈥�
In your books you advocated that VCRs should have on-screen programming way before they adopted it, and back in 1993 you were describing something very like the Web as it is now. Do people tell you they were directly influenced by you?
No VCR designer has come to me and said that, but I鈥檝e been very pleased with the impact the books have had. Often I don鈥檛 know about it until I go to a conference and meet somebody who has designed a product I鈥檝e bought and used, and who tells me they were influenced by me. Outside the US the two countries that have been most influenced by me are Japan and Italy.
What are your big priorities in usability now?
Emotions. Trying to build emotions into systems. I did some work for a Californian company called Evolution Robotics that was making a home robot and trying to understand how to prevent it from getting trapped in the corner or falling down the stairs. It seemed to me that the way to do it was for the robot to be frustrated in the corner and give up what it was doing and do something else, and also for it to be afraid of heights.
How does that relate to other work in this area, such as Rosalind Picard at MIT鈥檚 attempts to get computers to interact with us in a more human way?
Picard wants machines to detect emotion and figure out how to respond appropriately, and that leads to them appearing synthetic or artificial, often with condescending results. Picard鈥檚 website has an item on it about tutoring systems that can tell whether the student is frustrated. That鈥檚 fine on paper, but what do you do if the student is frustrated? Maybe you should do nothing because maybe being frustrated is an important step. If there鈥檚 anything more annoying than a machine that won鈥檛 do what you want, it鈥檚 a machine that won鈥檛 do what you want and has been programmed to behave as though it likes you. I鈥檓 not trying to make computers empathise with you. Emotions are best used to make systems more secure. When a computer loses power and uses back-up power you want it to start being anxious so that it becomes more cautious.
How would that work?
It should look at the tasks it鈥檚 about to do, and if they鈥檙e time-critical or important, it should send them to another computer. For example, you鈥檙e about to start payroll computation so you ask: 鈥淎re you feeling good?鈥� The system might say: 鈥淣o, I鈥檝e noticed a few memory errors.鈥� So emotions will be used to help machines survive, not to mimic human beings or try to make people feel good.
Why have we paid so little attention to emotions till now?
We took our lessons from studies of human cognition because we thought that cognition 鈥� understanding and trying to make sense of the world 鈥� was a higher state of human behaviour and that emotions must be some relic from our evolutionary past. But in fact emotions came first precisely because they are all about survival. Modern computers are pretty autonomous and run 24/7, performing a lot of tasks. With machines this powerful, survival is important, so putting emotions in the machine makes a lot of sense.
You don鈥檛 have what I think of as the stereotypical MIT engineer personality.
Actually, I do fit that profile. I feel hypocritical that I鈥檓 writing about the importance of emotions and interactions, when with my family I am poor at these things.
A lot of people are talking about facial or iris recognition as a security measure. Could they work?
Facial recognition not only has technical flaws, it could change the whole way we live. And what is so scary is that someone could compromise the data and masquerade as someone else. An iris scan is unique, but the records are in the computer database and all I have to do is make my iris point to your record, and from now on I鈥檓 Wendy Grossman, and nobody can dispute it. That鈥檚 where it鈥檚 really worrying, because people will think the security is foolproof 鈥� and it isn鈥檛.
Is there anything that could have been done in design terms to stop 11 September from happening?
I don鈥檛 think so. As far as I can tell, no mistakes were made. There were no practices in place that weren鈥檛 followed. I have spent a lot of time thinking about this. The US Academy of Science has just released a report on anti-terrorism, and I was one of 100 scientists on it. I was on the IT subcommittee, and the section I wrote was about social engineering. You know how you still break into most computer systems? Call up and say: 鈥淚 lost the password, can you remind me what it is.鈥�