Welcome back. In today's lecture we're taking a look at some of the challenges, successes, and not yet successes, in automotive user interfaces as an example of how different use contexts leads to different user interface design decisions. So, let's start With some of the things that have been huge successes. And as we go through, I'm going to switch back and forth between my slides and the Toyota Prius images that are online at the Toyota site. I'm using these, well because I have a Prius and that's what I drive. But because it's pretty emblematic of a modern car and illustrates many of these points. So, when I look at that site, and I see a picture of the car and the controls, the first thing I see is that I know this is a car. Look, there's a steering wheel, there's a shifter. There's a gas pedal and a brake, and a parking brake down here, all the things I'm sort of used to. Now, one of the natural questions we could ask is, which way should you turn the wheel. If you want the car to go to the right? Everyone's going to say you turn it to the right. I say great. What do you turn to the right, the top or the bottom? And the answer is, well what is the answer? We know the answer is well in cars you turn the top, but why? It's not what we inherited from carriages. But we turn the top of the wheel because all the other cars turn the top of the wheel to the right to go right. There's nothing mechanical that requires it to be that way. There's nothing special that requires it to be that way. What we're seeing is the power of standards. The fact that the steering wheels top turns in the direction you want the car to go when you're going forward is something that if you learned once you'll do right every time. And indeed cars and driving have been an amazing example of standards. The brake and accelerator pedals. Why do you accelerate on the right and brake on the left, or second from the right? Because that's what every other car does. And because this is what allows somebody to go fly halfway across the world, hop into a car, and not crash it because they step on the wrong pedal. Cars are designed to make driving become an automated learned behavior where you may focus on your destination. You use some awareness around getting there safely. But you don't spend any of your cognitive energy on thinking about which pedal do I press when I want to go forward. It works pretty well, and we've seen this in the clutch. Lights are an interesting case because they've moved over time. It used to be that high beam and low beam was commonly a pedal on the bottom left where your left foot could go get it. We've moved to the point where most cars are consistent. That that's something that you can pull on the left of the steering wheel and we've adjusted to new things. Windshield washers have changed over time, but they're mostly standardized. And we do see where things start falling apart when they have very different ways of working. And the speedometer is perhaps the best example of this. That there's a fair number of varieties, but they all have a few basic features. That they show you your speed, the speed the car is actually going. They are nearly all adjustable to metric or English measures, and they're all somewhat front and center for you to take a look at. Now there are also some less than standards. If we go back to this display, you may notice that there's a clock sitting over here. And in some cars the hardest thing to figure out is how to change the clock. That's sort of a relief, because it's the thing that's probably the least critical. But it's also one of the things that's less standardized. Different cars have different models. Some of them change the clock through the radio. Some of them change it through a complex menu system. Some have a dial that you turn. Some have buttons for the hour and minute. I remember having an old car that had the property that you changed it through the radio, but only after you turned the radio off. How are you supposed to know this? You don't. You look it up in the manual. The clock's not that critical but it also hasn't been standardized. If we look at radios, if we look at climate controls, if we look at the soft consoles, all these things that are being displayed up here. These are increasingly different. Radios, when all we had was a radio were pretty simple. There was a dial, and maybe some buttons that you could push one of them in to go to a station. That was where the radio button interface came from. You could only select one at a time, and you could pull them to program a station. Then you had a volume dial. They've gotten much more sophisticated, there are soft panels. There are ones that have many buttons. Some of they you hold a button to program, some of them you use two buttons to program. Dealing with the radio has become increasingly complicated. But the thing I want to point to, and this is something that was also pointed to very early in our first course. Is that some of the things that we used to take for granted are changing as well. So, this shifter here, and I think they actually have a bigger picture of it, and notice nobody worries about the controls. Most of the pictures are about the lovely upholstery or the charger for your phone. But this shifter here is very different from a classic shifter. The classic shifter went through a progression of park, reverse, neutral, drive, and then maybe some low gears. And it stepped through each of those steps. This is a shifter that has no state. It stays in the middle when it's idle, and it has a bunch of things that are very different. One of them is, you move it and hold it in a position and let go and it shifts, and that something people learn to deal with. But the other is the two of the things that people were used to seeing are no longer there or at least not obviously. Where's low gear? Well that's what this B is? It stands for engine breaking, I believe. And Toyota made the decision that since their engine was different, they weren't going to pretend it was going into first gear. They we're just going to tell you what it was, which was using the engine to break. But that means there are a lot of people who when they first get in to this car don't know how to set that, and that could be confusing. Even trickier, I've learned for some people whose first time is inside a Prius, they rent one for instance. Where's park? It's not here. It's actually buried over there as a button that's completely separate, but I've seen people rent a Prius be perfectly happy with it. They never figured out how to put the car in park. They turned the car off if they wanted to stop the car. Suggests that we have some challenges as we're talking about standardization. Increasingly, the navigation systems have the same property. There are several of them. They have very different ways of interacting. They have a lot of controls and it's not always obvious how to deal with them. But as we start talking about navigation systems we're also talking about this brave new world of interaction with driving. There's some principles here, but they're also just some general examples and challenges. There's nothing inherently new about the idea that you need to interact with your car while you're driving, right? We steer when we're driving. People have for a long time adjusted the radio volume while they're driving. It's just that the types of things that people are interacting with are becoming much more, one potentially more distracting, and, two, complex. So, telling a navigation system where to go to isn't something you can do with a quick glance. It requires a sustained interaction, and many cars actually have lockouts that say you can't set it while your moving. Because we're afraid that you're going to crash right into the car in front of you while you're in the process of trying to tell the car where you're going. With other things, there were Touch Controls that helped somebody. Most cars today that are made, at least most middle to high end cars, interact with your cell phone. If you want to answer the phone, there's a button, usually on the steering wheel, that you can press and that you can detect entirely by feel to pick up the phone, hang up the phone, or adjust the volume. You don't have to change where you are looking. Increasingly phones and cars are adding voice control and voice output. Getting directions in a map you have to focus on is not always going to solve your problem. If you don't know where you're going, you don't want to keep looking down at the map. Voice Output that says turn right on Main Street in 500 feet, allows you to keep looking at the street while you hear the directions. In some cases, voice input, still more common on the phones than it is in the cars themselves, might allow you to set something up without distracting yourself. Something like, give me directions to the nearest gas station, then it can just go do that for you. We don't see as much of this, but cars have started to incorporate heads-up displays. The idea that displaying things down here or here may not be as useful as displaying them projected sort of into the space where you're looking. Those heads-up displays I think the jury is still out as to whether people find them easy to use or not, and substantial work is going into making them better. The other thing we're seeing is increased use of Automated and Context-Sensitive Displays, so perhaps the classic example of this. Let's get to, the right place in this one. Well this is as good as any. I believe in this car, this is where your backup camera would be displayed. And you can't even control that in general. You put the car into reverse, and the picture is there. That's what you backing up to. Put the car in drive, you don't need to see that anymore. One less thing for you to worry about, it's just always there when you need it. There are versions of these cars that put the backup image up here. Because they know this is where you tend to look when you want to know what's going on behind you, also a useful idea. Understanding that context of driving. As cars get smarter and realize, uh-oh, you're getting out of your lane, changing the display to make it clear what's going on can be really helpful. Letting you know what's going on when the car starts to break, because it senses you're getting too close to the car in front of you, is another example where we can be Context-Sensitive. That doesn't mean we don't have challenges. One of the great challenges is the lack of standards and particularly standard iconography. I don't even have a great picture with all of the different icons that show up, but I can tell you that you drive a new car. And a light goes on, and you're looking, and you're looking, and you're wondering is this something I have to worry about? And so, many of them are things you don't really know. Is this I need maintenance soon, or is this my tire is deflated and I need to pull over right away? And right now, the answer is, these aren't standardized and well known enough for them to be useful for anything but stopping, pulling over, getting a manual out, and matching them up against a manual. There's a lot of effort to change that, that's going to make these automotive interfaces a lot easier. Part of the reason that all of these are icons, of course, is internationalization. The attempt is to get rid of words. Words would work pretty well. It would be nice if it said alert, your tire has no air, but you have to write that in a hundred and some odd different languages. And having an icon makes that easier if people understand the internationalization, which we're not there yet. Two last points, the Driver vs Passenger Control issue. It's an interesting issue that few cars have yet figured out a way to recognize, the passenger's trying to set the map or whatever it is. I guess I can allow that while we're moving without also allowing the driver to do things that we might not want a driver to do while moving. That concept is certainly going to have to come into the cars as we move towards self-driving cars, which is a whole different issue. And last, the whole concept of distraction. How do we deal with the fact that people are only good at paying attention in certain ways. And the area where human factors, and therefore user interface design, makes this most obvious is in the really challenging circumstance of the mostly self-driving car. Some of you I'm sure have read about these. There are test going on from many companies, from the test that Google and Uber are doing to the features that Tesla has deployed. And right now most of these require a standby driver who's there, and ready, and can take over on basically a moments notice. If the computerized driver figures out it doesn't know what it's supposed to do now. What we know about human capacity is that people are no good at that. One of the worst things you can give somebody is a job that says, just sit there, and be alert and ready, in case one day in the next couple of months I have to ask you to do something. People get distracted. Their mind wanders. Other things go on, and then when the system finally comes in and says, hey, we need you now. How can you quickly get up your alertness, gain situational awareness, and actually take over in a meaningful way? This is a huge challenge, and it's a huge user interface design challenge that still needs to be solved. And is going to be key to these automotive interfaces working. So, this is been just a brief introduction to automotive user interfaces. You'll notice we've left out things like what kind of web browser belongs in a car, because I think many people would say maybe none. And i've left out many of the things that are really about phone interfaces being used in a car, like how much bouncing can you do, because we've covered those elsewhere. But as we focus on the basic interface for the car, there's some successes there are some challenges. And there are some really interesting challenges as we go forward into the future, an area I hope you may take a look at. See you next time.
