Welcome Back! As we bring this first course towards a close, I have the distinct pleasure of bringing you an interview with Don Norman. As you may know, Don is a thought leader in the field of user interface design. He's the author of Design of Everyday Things and many other books. But Design of Everyday Things being the one that we've worked through many of the principles of the psychology of user interfaces from. He's also principal and co-founder at Nielsen Norman Group, and currently serving as well as director of the University of California San Diego Design Lab. Welcome, Don. >> Thank you. >> So since we have a set of learners here who've been introduced to a lot of the basics of UI design without having stepped through all the details yet, I thought it might be useful to get your perspective, as somebody who has really helped launch this field and has guided it in both industry practice and in research. Tell us about where you see things having changed over time, the evolution of bringing in design practices on top of engineering practices in the field, and where you think we're going. >> That's the big question. So, in the beginning, and this is true with the computers, but it's actually true of almost every single technology, in the beginning, the original technology developed by well hobbyists, tinkerers, engineers, technologists. And they're really very interested in what can it do and it's great potential. And they don't understand people, so that usually these first things are difficult to use. Which is okay, because it proves that the technology actually can be of value to people. The second phase comes when it becomes more popular and people have great difficulty with it. And they have difficulty using it, difficulty understanding it, and in some cases this leads to serious incidents, serious accidents and deaths even. So, at some point the profession wakes up and tries to understand what they might do. At the last, roughly hundred years, little bit less, there has been a profession, human factors, which have been very concerned about the way that people interact with technology. And that interest has now expanded into a number of other areas. For example, the field that we call human computer interaction, which is probably the major basis for the course you're now taking. The real point here is that you have to understand people and the way that people do things. And you have to do this not by asking them, but by watching. By watching you see them what they look at, what they struggle with, what they find easy, what they find hard, and accept what you see. The traditional problem that we have with engineers is that engineers, it gets all very sensible and logical. And when somebody isn't doing things the right way, they say, [SOUND] why are they so stupid? And they really like to go over and say look just do this, cant you see? And I have to explain often to the engineers that the problem is you're being too logical. You have to take people the way they are, not the way you'd like them to be. And people are not logic, logic was invented. It's not a natural way of thinking. So, people are emotional, and that's good, because emotions are where we have our values. So we decide what is good and bad, safe and dangerous. And people do use reasoning and logic. They also use past practices, what they're used to. So, the first step in any industry is, understand the people and understand the task and design for the task and the people as people are, not as you would wish them to be. And in the field of computer technology, I would say that's where we are, that almost every single company that develops products today has somebody from the field. They might call them human factors. They may call them user experience designers. They may call them human computer interaction experts. But they usually have somebody or some small team. [LAUGH] They put the team at the end, saying, hey we're all finished. Now make it easy to use or understand, or make it pretty. Which is the wrong way to be used, because, gee, if we're really watching people, we can tell the company, or yourself, here's a great new product we could develop that nobody every thought of. We discovered this real need. And, I think that's where we are moving. And, so the future is that direction, where we actually base products on what people really need, as opposed to what our technology can do, but there will always be a mix of the two. But that wasn't really the question. I think the real question was, the technologies we use are changing rapidly. There's a good example, last night, I had drinks with a friend who works for HP, the company that makes a new 3D printer. [COUGH] And we talked about the way that this new 3D printer is going to change manufacturing, and more importantly the kind of design tools that we will be using for the 3D printer. Completely different than anything you've ever seen. Today when you design, and what you'll be taught in this class, is actually pretty low level design in the sense that you're going to be designing the controls, the menus, the structure. Every little detail you'll be thinking about trying to understand how that fits. [COUGH] How it provides a right function. And how it makes it easy for people to understand and use. But imagine a design process that's this. I have a smart design tool and I tell it my goals. Maybe I'll tell it cost, maybe I'll tell it size, maybe I'll tell it weight. But also I'll tell it the functions it has to produce and what the goals are. And then I'll sit back and let some kind of genetic engineering, or genetic algorithms, or neural networks, or who knows what, machine learning, [COUGH] develop some possibilities. 5, 10, 1,000, and what I do as a designer is, I look at them, and I say, wow, I never thought of that, or that's horrible. And so what I'm doing is I'm training a system, and I say, go in this direction, or no, don't do that. And I end up me and the device designing in this really wonderful, collaborative way, ending up with something you'd never seen before. Can you imagine what the role of the designer is like then? Are you ready to be that kind of a designer? It's really quiet exciting, but it's a completely different way of thinking. And the reason this came out with 3D printing is because, with 3D printing, you could produce devices that look completely different than anything we've ever done before. We could make things of different shapes that are hollow inside, and therefore, they're much lighter and stronger than anything we do conventionally. So that's the physical side. Now consider the, if you will, the usability side, the cognitive side. What's going to happen there? I think we'll see a similar evolution. So that's one direction we're going to move but the other direction is sheesh, we're moving into virtual reality, augmented reality. New types of haptics, feeling, gestures, presses, sensors in everything, telecommunications. It's going to be an exciting new world. And you're going to play the major role because it's people like you who are just starting. That's going to make the big difference. That's a good introduction, there's more that I can say but I'll let you drive me to it. >> Well I think that's a fantastic vision. I'm going to ask you a couple of appointed stops along the way in that vision. I know when I joined this field there was a dominant strategy, what was called typically usability engineering that said look, we're not smart enough, we don't know enough to get it right, let's just build something and run it through some tests and make it better. And even back then there were people who were pushing against that strategy. They were saying wouldn't it be silly if the way you designed a house was to say well let's build it and see where it turns out that you walk into the wall and then we'll tear it down and put doors there. But the concept of let's be iterative. Pretty strongly caught on, and then there was this wave, maybe lead by folks in the industry, places like Apple, that had designers in more leading roles that said, you know, maybe this isn't about Iterate and perfect. Because iteration only gets you to a local maximum, maybe this is about, how do we get somebody skilled enough that the thing they designed already has built in all of this knowledge about how people do things and what they need and what they don't even know that they need. And we spent a lot of time envisioning how do you create that kind of a designer. And I'm hearing you talk about the idea that maybe we need to think about the tools for those designers that allow them to go beyond. But could you reflect for a minute about this tension or transition between the engineering focused approach to usability and the design focused approach? >> In the end I think that there isn't a tension. That we're going to need all of these different approaches. I'm actually a firm believer in when you're not quite sure what you want to do, just do something. Building, you have no idea what you want to do, build something. See what happens, see what people make of it. See what you make of it, and iterate your way through. Yeah, say that is. Gee, I never realized that people would think of doing this with what I just did. That gives me an insight for a whole new direction. And actually, I designed a house like that, kind of. I mean, [COUGH] we bought a house here in Southern California, it was one story house and we said, it's kind of old, why don't we remodel it? And so we brought in some architects and the architects they came to house and they climbed up to the roof and sat on the roof for three to four hours. And when they came down, they said, you know you better put a second story on this house. And we're going to put the living room and dining room and the kitchen on the second floor because on the second floor you can see the ocean, you can watch the surfers, and the whales and the dolphins, and we never thought of that. So they started with that vision. But then we said, what's it going to feel like? How are you going to design the inside? So, they gave us a rough plan [LAUGH] and we said no. That doesn't fit. So, what we did is we went to the apartment we were living in, my wife and I. And we took tape and we put it on the floor. And we took boxes and so we taped out roughly the kind of area that we thought we would have in the new house. And we put boxes where we thought we might put the table here or the counter here or the sink here. And we made believe we were cooking for quite a while. And we kept moving things around. And we brought these ideas to the architects. Even while they were building the house when it was just framed we went upstairs to the second floor and we made believe we were cooking and we said it isn't going to work and we had to change it and the result was spectacular. It was really wonderful because we tried it out. Now you can't do that with big extensive homes and you can't do that with big tall buildings but actually today with virtual reality you can do that. You can actually build it and make believe you're walking through it and it really feels real and you can change things and I've heard of architects who say they make major changes in major buildings because of this technology. So, that's one step. Second, we have quite a bit of science that we could use, it's the book, The Design of Everyday Things, there's quite a bit of that, and you will probably learn others, because this book doesn't contain all that we know. But as a result of the fact that we know a lot, we don't just start from scratch, we start with knowing a lot about how people work, but they the kind of clues they need to make them understand what is happening. So we start with that. I am also a firm believer in watching people. I don't think we're ever going to have that knowledge about what people really mean because everytime we build something it changes what people need. It changes how people work and what they think is possible. So it's always looking to the future and it's our job as designers to watch people and say, why are you always stretching like that? Or, gee I see you, you tape these two things together, why did you do that? Or why are you bending? Or why did you, that's interesting, we see you put things on the stairs. Why do you put things on the stairs? They're reminders. That's interesting. Why do you need reminders? So I think we're always going to need someone who has this brilliant new idea and says let's try this. No one's ever thought of this. And somebody who's just modifying what exists. Making it better. Maybe watching how people use what exists and suggesting changes. And in all cases in the end I believe that we have to do this prototype, we do something rough. We make it so that people can try it out. We watch what they're doing and we realize that we have to change it and yes that only get, just better and better but as you said it gets you know if you think of there are these hills all around us. And the higher up you go the better it is, and so if you're on a hill is you always go back to the top. Now that's what this iteration does, it gets you to the top of the hill. But it doesn't necessarily get you to the highest hill. In fact if you're guaranteed to get to the top of this hill then there's no way to get to the higher hill because any direction I move makes it worse. I start going downhill again. So, the only way to get to the higher hill is, well actually there's several ways. One is, don't make one product, make six of them. And they're probably all going to be on slightly different hills and you'll see which one seems better. Or second that's where watching and being creative, and that's where the great creativity comes from. That's actually where inventors and tinkerers are so wonderful. They have some idea and they try it and it's crazy and crappy and looks horrible. But it moves you in a whole new direction. And what you have to do is not look at it and say ew, that doesn't work. You had to look at it and say, wow, that's a really interesting possibility. It's not very well done, but suppose we did it well. Maybe we'd be in a whole new space. This happened, by the way, with digital cameras. The first time people were starting to get rid of film. You know what film is? It was that piece of cellulose with chemicals and you expose it to light and then you had to keep it in the dark until you could put it to chemical baths. Then it came out as a picture. No, you don't remember that? Well that's how we used to do pictures. And when someone first said we could do it with a little digital camera. All the pictures were pretty crappy, and so they were rejected for a long time. Even though that's all that we do today, essentially, so you've got to be careful here that the first few products of a revolutionary new idea are often really bad. It's as if, yeah, this is a higher hill. But I'm starting off at the bottom, way down at the very, very bottom, where I'm so much lower than what we can do today. Why would anybody ever want to do that? So here's your challenge. Sometimes those other things are that correct way. And sometime those other ways like flying automobiles are the wrong way. How do you know which is which? And the answer is you don't. >> Neat challenge and neat example. I love the digital camera example because not only were the original pictures not very good. But I think part of the challenge is that people using and building those cameras didn't know yet what they were going to be good for. And, they couldn't have imagined that most pictures would never be printed and would be mailed around and posted in a form where not that good was good enough because they were going to be displayed on a watch or a cell phone. And obviously because they persevered we've gotten to a very different age of photography. >> [CROSSTALK] That's right, but It's interesting because all sorts of experts like me. Said yes I can see why these pictures would be wonderful for social interactions because they're not very good but who cares when I'm just trying to show my new child or my new friend or where I was but they just can't be any good we said. Because look how small the camera is. Look how small the lens is. Look how small the detector is. Physics tells us, they can just never be very good, there's not enough quanta, light quanta hitting the detector. So, see, you're always going to need a big real camera to take really good pictures, ha! >> Well, we were wrong. >> We were wrong. It's amazing how the technology has improved, so even the very tiny little camera in your cellphone takes pictures that are equivalent to what we used to get out of expensive cameras. Now it is still true and if you want to really great professional level picture we'll use a bigger camera. Which has a bigger detector and a bigger lens but it's been amazing than most of, the more we knew about the technology the more wrong we were. It's a real difficulty if think you know too much, because you know the limitations today. People didn't know about the limitations, and therefore, poof! They went past them. >> Yeah. Indeed. Well, let me ask you one last technology question as we close, which is a technology I know you've been following is what's going on with automobiles today? And automotive technology including smart cars and self-driving cars and interfaces for systems in cars. Obviously that could be a topic for a whole other discussion but is there something you've distilled from this? That might be a lesson as we think about what we need to know about people and keep in mind as we think about designing the technology. >> Yes. I think that automobile driving is actually not a good thing for people to do. Now, a lot of you listening to this say, what? I love to drive. What do you mean? I'm really good at it. It's really fun. You know, it's not really fun. Look, I used to, for years, I owned a Porsche. And I really loved driving. And I took driving classes. And I took my Porsche out of the racetrack. And track. Did all sorts of wonderful things with it and with my sons. But that's wonderful, but everyday when you're on the highway and it's rush hour and you're just sitting there and hardly moving and it takes a half hour or an hour to get to where you're going where it really should only take ten minutes. That's not fun. But that's how driving is more and more. On top of that in the United States roughly 30, to 35,000 people die every year in automobile accidents and about a million are injured. And in the world about a million people die every year. That's not something that we should really allow to continue. And, so, fully automated driving has the potential to reduce those deaths dramatically. Well? We're making huge progress. Much more rapidly than ever we thought. But, the problem is that the engineers think they've gotten even more progress and the engineers have left out the human elements. And, sure can we do highway driving automatically? Yes, that's the easiest part because in highway driving the people are really pretty constrained. And you follow the car behind you, sometimes you change lanes but the lanes are often marked and you can pass or not pass as long as you can look around you. And we believe that sometime in the year 2017 or 2018 almost every automobile manufacturer will have their luxury vehicles do completely automatic driving on the highway. What about city streets? The city streets, no way. Because in the city streets all sorts of unexpected things happen and people do all sorts of bizarre things. And on top of that people will see these automatic cars coming and therefore they'll say, hey, I don't have to worry. I know it's programmed not to hit me, so I can just cross the street, I don't care what the traffic light says. Or I can just drive across because I know it's not going to hit me. And that's going to cause, well calamity. Can say nothing of the fact that unexpected events are going to happen and I like to tell people that we know two things about unexpected events. First is that they always happen and second when they do happen they're unexpected and when something goes wrong in a automobile how much time do you have to respond? If you're flying an airplane, and something goes wrong. This happens, by the way, it happens a reasonable number of times. So all these incidents are studied. The pilots, what's going on, they say. And they're very well trained, okay? Even so, it can take them several minutes to figure out what's going on. And to save the airplane. Because when an airplane is flying high up in the sky, what 30,000, 40,000 feet, 10,000 meters, there's a lot of time. It takes five or even more minutes before the airplane is going to be really in great danger. In the automobile, if you're traveling at 60 miles an hour, or 100 kilometers an hour, in one second, you've gone 90 feet, or 30 meters. That's a big distance. You have to respond in less than a second. And if you haven't been paying attention, if the automation has been working quite well and taking care of everything for the last 20 minutes, you're not paying attention. You can't. And so there's where the danger is. Great danger. And we've already seen a number of accidents that appear to be caused by just this. This is called vigilance. It's a field that's been studied in psychology for 50 years. We know a lot about it. And the automobile manufacturers don't seem to be aware of the fact that, hey, there are real problems here. So the hardest problem that you're going to have in this kind of automation is getting the engineering crew to understand the complexity of human behavior. I mean, I've been studying what happens. We're working with several automobile companies, and we watch. We look at people standing on the sidewalk, and their back is to the street. And so if you're driving by you assume that, well, they're not going to cross the street, I can keep going. And then suddenly without any warning, they jump in the air and they turn around, they run across the street. If you can find them and ask them they'll say, I just realized I was late for an appointment. I was talking to my friends and, my goodness, I'm late. So off they go. What does a driver do? How does a driver account for that? How does a driver predict that this might happen? These are the difficult problems. The other problem is communication. You see a driverless car coming, can you wave it out, go on, go on, or can you indicate, I want to cross? Or could it wave you on? And if it waved you on, how would you know it meant you and not the three people next to you and I didn't actually see you? Or how do you know what it's going to do? How does a driverless car tell you? What we do today is we look at eye gaze and try to find the driver and see where the driver's looking. But there is no driver. There are all sorts of interesting new issues that are going to show up. There's a trend that I find dangerous. That as companies realize they need to bring more designers into the field, they often bring the wrong kind of designers. Design is a very complex topic. And the traditional designers, and by that I mean those who go to design school, they really are craftspeople. They don't usually understand science or technology. They're really great at making beautiful looking, attractive items. Now if we ask them to do high technology items, they make them beautiful. But can we use them? So let me use as an example a company I like, I love. In fact I'm standing in front of their products right now, Apple Computer. So Apple used to make products that anybody could pick up and use without ever reading the manual. It was always obvious what was going to happen because you could always see. It's this process I call discoverability. I could discover what alternatives I had any time. And because of feedback, I could always see what had happened. And because there was a lovely command called undo, if it wasn't what I wanted, I could change it. And I soon learned to use undo on purpose. I would deliberately make a change to see if I liked it, and if I didn't like it, I knew I could always undo it. And I had a good conceptual model of what was happening. However what happened at Apple was that the designers who were trained in traditional design schools, which is focused primarily with art and appearance, took over. And so when we did the iPhone and gesture phones in general, why they were beautiful devices, they were nice to hold, they were good to feel, and the screens were very attractive. But there were no signifiers, no clues as to what operations you could do, it was not discoverable. Should I tap with one finger, or two fingers, or three fingers? Should I swipe left, or right, or up, or down, or one finger, or two fingers? One tap, double tap, long tap? How do you know? Most of the time the obscure things you learn is because someone told you. And you said I didn't know that, well that's nice. Could you remember it? Not necessarily. Apple with its mouse has this really nice thing in the control panel, a little illustrated example, of what each of the different taps and things do and it's a good way to learn. Except how come I have to go back to that over and over and over again to remind myself what all those obscure gestures are? Because the gestures that I use on my iPad are not the same as I use on my smart mouse. Which are not the same I would use on a track pad. Same company. It's crazy. And so you have to be aware of the fact that there's when you talk about design, there are some designers who are so focused upon making it fun and beautiful that they completely ignore the fact that nobody understands it. Graphic designers who want to make sure that, that ugly font, it gets in the way, so we're going to make it light gray on a dark gray background. Or they're going to make it really small so it doesn't destroy the visual appearance. Who could read it? It's crazy, and it's getting crazier, so beware of that. Your job is yes, making aesthetically delightful and pleasing and fun to use, but also, make it understandable. Make the actions discoverable. Make what happened visible through feedback and allow undo. Remember the times you say I have this great picture I want to show you. And so I swipe back and forth, here it is, and I hand you the tablet and the other person, you, touches the tablet with their hands accidentally and so the picture disappears? And you can't get it back. You have to start all over again. That's crazy. Just crazy. So, don't be crazy. Be wonderful. >> Well, thank you, Don. This has been a wonderful way to bring a close to this first course on user interface design in our sequence. And I want to thank you for joining us and remind everyone this has been an interview with Don Norman on current thoughts and trends in user interface design. And we will see you as we go forward in the rest of these courses together. Thank you, have fun folks.
