Click here to preview the new Fast Company

Want to try out the new

If you’d like to return to the previous design, click the yellow button on the lower left corner.


Designing A Sexy, Safe Touch Screen For Cars Is Harder Than It Looks

There are reasons why the latest concept dashboard UI hasn't been made a reality.

A few weeks ago, a gorgeous concept video of "a new car UI" made the rounds of the Internet, gathering praise as it went. With good reason: The video, created by product designer Matthaeus Krenn, shows a working prototype of a simple, sleek touch-screen interface that relies on glance-free gestures instead of cluttering the screen with tiny skeuomorphic buttons. The idea is simple: The less you have to look at the touchscreen in your car to effectively manipulate it, the better.

Then Apple's CarPlay system came out. What was the result of Cupertino's vaunted think-outside-the-box interface design process? A screen filled with tiny buttons. Sure, there's a lot of voice control on offer via Siri, but if you want to use that touch screen, you still have to take your eyes off the road and use them to aim your finger at haptically invisible digital controls. What's stopping designs like Krenn's from becoming a reality?

David Young, an interactive designer and former creative director at BMW Designworks USA, expressed misgivings about Krenn's concept on Twitter, so we asked him for some constructive criticism. (Krenn did not respond to interview requests.)

Young praised Krenn's design as a "beautiful, innovative, and unexpected" alternative to "the current hierarchical menu-driven interfaces, and all-buttons-at-once touch-screen interfaces we're currently seeing." However, Krenn's focus on re-creating the gestural simplicity of physical controls comes at the expense of flexibility. "Vehicles are increasingly complex systems, with lots of information to display and a wide range of customization and configuration options. Krenn's interface, however, only supports adjusting eight settings," Young says. "It's not nearly expandable enough for the complex demands of a modern vehicle." Instead of truly solving the problem of "too much information and buttons on a car's touch screen," Young suspects that Krenn's design merely avoids it.

There's also the problem that all gestural interfaces still have: They're unfamiliar and all have different rules that must be learned. Krenn's UI is admirably "logical," and "everything works fluidly," Young says, but "at a glance, it's not immediately obvious how things work." That might amount to a minor quibble on a smartphone app. "But for drivers unfamiliar with the interface—new drivers, infrequent drivers, car renters—it will be as perplexing as the icons on your clothes that give washing instructions," Young says. By re-creating an iPhone-like, icon-driven interface for CarPlay, Apple may not have wowed any futurists. But in the year 2014, pretty much anyone who gets behind the wheel of a car also knows how to operate an iPhone. That said, CarPlay is designed to augment the iPhone's display, not function independently like Krenn's. But Young's point—that in the context of safely operating a motor vehicle, usability and familiarity are nearly synonymous—is well-taken.

Finally, Krenn's UI has no haptic feedback. "So when changing a setting that doesn't give immediate feedback, such as changing a climate option, the driver is required to look at the display to see if their gesture is complete," Young says. This is less a criticism than an acknowledgement of hardware limitations—the iPad that Krenn used to mock up his concept has no vibration feature, after all. But there's nothing stopping car manufacturers from including this kind of haptic feedback into their designs. Physical knobs and dials often have catches or "detents" in their movement, which let you know that the knob has been turned sufficiently to register a change without having to look. Smartphone screens vibrate—by now, a familiar kind of haptic feedback—so why can't dashboard touch screens do the same?

This isn't to tear down Krenn's creation. Like any concept design, it exists primarily to drive constructive dialogue, and we're glad that David Young has added to the back-and-forth. Carmakers can do better—and the more voices we have telling them how to do so, the safer our dashboards will become.

[Matthaeus Krenn's touch-screen concept]

Add New Comment


  • A very inspired concept. With the inevitable rise of touchscreens in cars, we need as much creative thinking as we can on the subject.

    I too echo everyone's belief that traditional knobs and dials likely reflect one of the best solutions in the space. However, that should not stop us for exploring new concepts. I have another idea that also takes advantage of gestures, but perhaps does so in a more scalable way (and with one finger):

    Your critique is welcome.

  • Paul Ski

    I challenge anyone to drive 70mph in the rain and use this system more accurately and with less eye movement than a bunch of knobs.

    Or driving in stop and go traffic.

    Or using it at all if you don't have 5 fingers (or full use of 5 fingers).

    Or driving on a bumpy road.

    Technology is awesome but using technology for technology's sake is just dumb. We are trying so hard to turn our driver's seats into sofas, that we lose site that we are in control of 4,000 pounds of speeding metal.

  • I think voice will play an important role in the future as well. Things like choosing a song or calling a person when choosing them from a list will always be tedious in a car no matter how great the knob

  • I think this is oversimplified to the point where the novice would have no visual cue to start with. I love the principled thought behind this but I don't think it's the future.

  • Kerrigan Marois

    I think its beautiful, but its solving a problem that doesn't exist yet.

    The only other thing I'm curious/nervous about, is what happens to a screen like that in the extreme cold? There are some mornings in MN when my radio doesn't want to light up.

  • Alex Berkowitz

    That's the point, though. It's solving a problem (or attempting to solve a problem) that doesn't yet exist but will very soon. Some cars already have integrated touch screen controls and it's not crazy to think that we will be seeing a lot more of that sort of integration within the next few years. Having a proper user interface for these platforms -- ones that improperly designed could actually lead to injury and death -- is pretty important.

    The cold thing is a good point though. I'm fairly sure that modern capacitive displays can handle cold weather, but what happens if you're wearing gloves? Or what if you accidentally leave a window open; is frost going to form on your fancy touch screen? All things to consider for auto makers.

  • Henry Biggs

    Why not a system that does away with the need to interface with the screen at all? The screen still exists in order to display information but functions are controlled through hand gestures registered by a Kinect style device? Vocal confirmation of a successfully completed command is then supplied by the car.

  • Thank you for providing a critical view on this project. It seems that people jumped on the hype train and forgot their critical thinking at home. In a way I can't blame them, the automotive interfaces leave much to be desired for and there is a lot of room for improvement. But if you are going to create an improved version, at least do your research and do it right. The first thing I did when I started work on a similar project ( was to get in touch with people in the automotive industry, read guidelines and standards. Only then I was able to make informed decisions about the rules I can break and the ones I can't ignore.

    Working for the automotive industry I've been able see a lot of improvements lately. But as it is with this industry, those improvements are going to be rolled out slowly over the next 2-4 years. As for the lack of quality haptic feedback; At the moment there just aren't enough automotive grade components available on the market.

  • Kerrigan Marois

    Wow, that is interesting as hell. I'm not sure if you had inspiration from somewhere for an idea like that, but its certainly unique, haptic feedback controls that interface between your hand and the device? Also, the feedback isn't just a little buzz under your finger, its air or tilting.

    I think I will be following that concept, its very inspiring.

  • Evan Carpenter-Crawford

    David is dead-on in his crit of this concept. I was lead designer on one of the better current OEM infotainment systems, and I feel I can say with some authority that Krenn's demo as inadequate as it is beautiful. I kept wanting to ask, "yeah, but how do I input a street address?, How do I look up a phone number?, How do I tag a song I like?, How do I know the status of one feature when I'm displaying another? ...and on and on and on... The bottom line is he came up with a single elegant solution, shoehorned five or six of the most simplistic tasks into it and declared victory. Infotainment systems are just that - systems. What Krenn has proposed is a widget, not a system solution. Also, FWIW gestures in a car are kind of a non-starter - 50mph on a bumpy country road pretty much kills your finite motor control at arms length. ECC