Honda bucks industry trend by removing touchscreen controls | Hacker News Reader

Honda bucks industry trend by removing touchscreen controls

autocar.co.uk

1,922 pointstrenning743 comments

Look at the cockpit of any modern airliner and you will see screens, but they are never interactive. There are hardware buttons, dials and lights all over the place. A tactile interface is both more obvious, sturdy and more stable, and therefore safer. The problem that touch interfaces solve, ever since the advent of the first smart phone, is that the interface is now dynamic. You can change it without having to replace the hardware. Here's the catch: for safety critical interfaces, YOU DO NOT want the interface to change. The point is moot.

Touch screens will hopefully never make it into any critical pilot systems, because safety and stability matters to airline manufacturers, current ongoing scandals notwithstanding. I only wish automobile manufacturers took their job equally seriously.

I work for an avionics manufacturer and I can assure you most of our upcoming comercial systems (and even a healthy portion of government ones) feature touch screen inputs.

> The problem that touch interfaces solve, ever since the advent of the first smart phone, is that the interface is now dynamic. You can change it without having to replace the hardware.

I mean, the other point of a dynamic interface is that you can now have more controls than would fit on a static interface. Touchscreen fit-to-purpose controls might suck more than hardware fit-to-purpose controls, but either option is better than a single set of generic controls that control multiple systems that "should" have different control paradigms, translating to the generic controls being a compromised bad fit for any use-case.

E.g. a hardware English-language keyboard is probably better than a touchscreen English-language keyboard (though people with modern Blackberries might dispute this); but both are better than entering English text through T9 on a dial pad. And the touchscreen has the benefit of allowing you to have more keyboards (for e.g. the multiple native languages you type that use different alphabets), which wouldn't even fit on the phone as hardware keyboards.

I bring this up, because eventually you run out of space to stuff additional controls. As airplanes become ever-more advanced, their cockpits will approach that point. At that point, dynamic affordances may be necessary, just so you can have some kind of "pagination" allowing you to squeeze more controls in. (Hopefully it'd just be for the non-time-critical switches to flip.)

Too late! Boeing 777-X will have touch screen interfaces for pilots.

Personally, I think information display can and should use touch interface, but actions should be tied to physical switches or buttons.

"The US Navy will replace its touchscreen controls with mechanical ones on its destroyers

After a deadly 2017 crash between a destroyer and an oil tanker"

https://www.theverge.com/2019/8/11/20800111/us-navy-uss-john...

> Touch screens will hopefully never make it into any critical pilot systems

Touch screens in the cockpit seem like madness to me. Cockpits sometimes fill with smoke and the pilot has to be able to find and operate the controls.

Ever notice that the flap levers have little flaps on top of them? The nosewheel steering control has a little tire on the top? That's so the pilot knows without looking what his hands are on. These designs were not the result of some study group following fashion, but were the result of accidents.

This comment is off-topic: the article is about physical buttons for non-safety-critical systems. E.g. the article explicitly mentions climate control.

I am not a pilot, but it seems like MFD (multi-function displays, a sort of touch screen with touch points at the edge of the screen) are very common even in fighter planes.

Yes, the basic controls do not change. But more advanced functionality is easier presented through menus and screens which guide you through a process -- instead of adding tons of switches for every possible function.

https://en.wikipedia.org/wiki/Multi-function_display

Indeed when phones started moving over to touchscreen dial pads i always thought it was a step backward usability wise. I had one phone where I couldn't hang up calls easily as you needed to swipe in a particular direction and a number of my calls went though a VOIP number so that had a different way of hanging up. Same with whatsapp calls now answering and hanging up is different from a normal call. It's a real shitshow.

I don't think this is entirely true. The GA glass panels love touchscreens, and they are just as FAA-certified as their non-touchscreen counterparts.

You are right that they are flaky. Here's Martin Pauly (great YouTube channel!) using his touchscreen transponder and it just stops working: https://youtu.be/bopcQSJKcD8?t=732

I think that’s partially why the market for handheld game devices didn’t get crowded out by smartphones. Tactile inputs let the player’s muscle memory work as a more direct shorthand into the product for a lot of video games. It helps the repeated interactions of a video game be more accessible over a “look, then touch” method of input.

Didn't the US Navy try this and end up crashing a destroyer or something?

I'd argue that humans have evolved to physically manipulate their environment. Safety critical systems are the last place you want UI variation. Things need to be predictable and tactile to build muscle memory.

I think you're missing something pretty major -- a touchscreen can support multiple UIs, menus, and controls, with minimal hardware. If I have 20 user adjustable inputs, I would need 20 dials/buttons scattered around the cockpit. On the other hand, with a screen, I can display 5 on each page and allow the user to swap between pages.

The thing I'd like to see more of is OLED buttons (Optimus Keyboard style). Physical, clicky, mechanical buttons, but each button is a screen. MFD buttons already change their function depending on what mode the screen is in, having the button also be an icon of what it will do can be very nice.

Otherwise I agree. You need the use of the interface to be as automatic as possible, and exploiting muscle memory & tactile feedback are very important for that. Touch screens fail there.

I was about to ask if there are any pilots on this thread who could weigh in because this was my intuition as a non-pilot. Just to drive one of your points home further, you don't want the interface to change because that could confuse the pilot. Toyota's braking fiasco is an example where unfamiliarity led to mis-operation even though the hardware or the device itself was functioning "to spec".

> The problem that touch interfaces solve, ever since the advent of the first smart phone, is that the interface is now dynamic. You can change it without having to replace the hardware.

They also enable the completion of hardware design before the interface design is completed. While the plastic molds and mechanical designs are worked out, the interface and software development can continue.

Cockpit design is strictly regulated by some very risk averse and conservative people. (And not just the cockpit, the passenger space as well.)

So it's hard to tell which parts of he design stem from being conservative and which parts actually enhance safety.

(I get your overall point. But your argument isn't really a good one for the point.)

One time I had to create a button for a in-flight software which would control a lidar instrument. The button would enable you to select when to turn on and off the instrument.

We went for a few iterations where the button would light up when the laser was on. There were tons of issues--the state of the laser was not stored in the hardware, so every time the button would disconnect briefly, the state would get reset and lost. After months of back and forth dealing with algorithms to save the state and deal with spurious button presses when the device disconnected, we ended up with a completely useless button, as disconnects would happen very often inside helicopters, rendering the button useless due to the spurious presses near disconnect/reconnect.

Finally, we decided to go with a hardware switch (which I wanted from the start). In about one day the whole thing blew over.

> Touch screens will hopefully never make it into any critical pilot systems

For one, you're too late, touchscreens are prevalent in modern avionics, and unlabeled buttons on the border of a screen that change function depending on what screen you're viewing are the second most common. The more relevant part to the Honda discussion is that there are different considerations for a car and airplane interfaces, and they are so different it's not a good argument to say "planes don't/shouldn't do this so cars shouldn't either."

The issue with a car isn't the interface, it's the fact that people look inside their vehicle for too long to fiddle with the radio. Even if the buttons/knobs could be operated entirely without looking, most people would still stare at their radio while they are doing it. Pilots learning to fly are trained to look outside after pretty much any action, they shouldn't ever stare inside the plane. They do a quick instrument scan and look outside. They glance at their chart and look outside. If they need to change frequency they do it and then look back outside. It's kind of hard to break the habit of looking outside when you start instrument training.

But most car drivers don't have the same amount of training and fixate on things inside the car, like the radio or climate controls. TBH, pilots still get fixated on things, it's just that they usually snap out of it and regain situational awareness before anything bad happens because the skies are pretty spacious. But car drivers don't have spacious roads. There's another car right next to you going 75 mph and if you drift out of your lane you'll cause an accident.

The issue isn't the design of the buttons at all. That matters to a fighter pilot, but the issue for a car driver is the fact that the screen is even on and the radio is accessible while driving down the road. The real critical safety feature would be disabling the screen while driving, and either locking out controls or only allowing voice control. But people would never buy a car that doesn't let them fiddle with the radio or stare at their little screen, so the actual safety feature that needs to be implemented won't happen.

Edit: When I say "fiddle with the radio" I'm including all activities that take place in a car's center stack- audio, navigation, climate control, etc. I'm also a pilot, have designed tests for avionics upgrades for multiple fighter jets, and own my own plane. I have lamented the introduction of touchscreens into modern avionics at a professional level and the personal level. I own three cars with varying levels of touchscreen invasion. So I've thought about the issues surrounding touchscreen quite a bit, and have concluded that the interfaces in a car are so simple that the issue isn't whether you can operate it without looking, it's the fact that people aren't trained to do so.

Dynamic interface can still be useful in some cases. Cars do require the driver’s active attention, but certainly not the passenger’s, who can operate the touchscreen. Further, undoubtably part of what makes controls intuitive is familiarity, and thanks to smartphones, tablets, and to some degree even modern laptops, it’s hard to argue against it from a familiarity standpoint. I’m sure it’s been attempted but, It’s hard to imagine a good mapping interface in a car with no touchscreen at all.

I’m still glad for what Honda is actually doing, which is not unilaterally removing touch screen controls but instead moving climate control back to physical buttons. These are things a driver ought to be able to operate safely while in motion, and touch controls only ever made them more complicated I think.

> Touch screens will hopefully never make it into any critical pilot systems

The F-35 Fighter Jet has basically only touch-screens (apart from Hands-on-throttle-and-stick controls for when you are busy pulling Gs)

Depends on your definition of "critical" but I don't think this is a great example.

While not touch many Airbus aircraft feature a full keyboard and trackball for operating those screens.

The problem here is not with the technology - it's from the people implementing it. Instead of leaving the design and software to tech companies, we have car companies trying to do it on their own.

The result is terribly designed software that looks like it's from 2000.

An airplane cockpit has two pilots and literally several hundred physical buttons. No one wants to drive that car.

In cars we're mainly talking about the volume and fan speed. Personally I much prefer knobs for those. But I wouldn't call those critical safety systems.

I'm afraid I have some bad news for you:

https://youtu.be/wfOD2y_AD_w?t=190

https://www.dynonavionics.com/SkyView-HDX/

https://youtu.be/1oyCzT6sB_4

Just a few examples.

Aviation very much uses touch screens quite extensively. The difference usually is that more thought is put into when and how to use them; you're not just replacing all the buttons with a touchscreen and letting some underpaid intern design the interface for it.

All the latest Garmin avionics include touch screens. Airplane cockpits aren't the right metaphor because they are largely driven automatically.

Perhaps, but on many pilot's knees will be a digital touchscreen kneeboard which having learned with paper approach plates and map books myself was a godsend.

Sure, but there are still tons of touchscreens in the cabin -- controlling the climate, the lighting, and the playback of the safety video.

Electronic Flight Bag is one of the biggest screens, right? And that's used for navigation, which is largely strategic, not tactical.

Great points.

BUT - touch screens can be used if people are trained, the layout is rational, the device is responsive.

There are two underlying things:

1) Tactile. As you spelled out.

2) Changing interfaces. This is the real killer. 100 screens, don't know what's what, supposed to be driving.

These UIs need some thinking but I suggest that the 'knobs and buttons' can possibly be mapped to different functions depending on.

I see many people comparing automotive electronics to avionics, but there’s a big difference, in flight you can take your eyes off the “road” and in some situations even fly with 0 visibility, whereas in a car your next obstacle is only milliseconds away from you.

The f35 has a large touch screen that pilots use. Maybe the $1m helmet has a way to interact without touching.

> Touch screens will hopefully never make it into any critical pilot systems, because safety and stability matters to airline manufacturers, current ongoing scandals notwithstanding.

You might want to take a look at garmin's general aviation product lineup. They're pretty popular.

Not to mention, touchscreens really disadvantage the visually impaired.

Pretty sure this has everything to do with planes being old. Boeing pretty clearly demonstrated that planes being planes doesn't mean they're immune from bad design, corruption, or any other problem.

Check out all the latest Garmin aviation equipment. Touch is here for critical systems including GPS systems used on instrument approaches. Super fun to reprogram in turbulence.

Nope, not anymore. Honeywell is making the new Gulfstream’s Symmetry Flight Deck for the G700 and it is touchscreen enabled.

It's not useful to compare cars to planes since cars can easily pull over to the side of the road.

Look at the upcoming TEMPEST fighter to see touch screen displays in full force.

well said. I for one don't like all those touchscreen buttons. Screens should be for maps and other non-interactive stuff.

Pilots already use iPads for checklists.

I suppose you don't include the flight manuals as a critical system, since those have been touch screens for almost a decade?

Eminently sensible call by Honda. Knobs, dials and switches work well and do not demand that a driver takes their eyes off the road. I suppose that voice controls would be a reasonably safe tech option, but probably not cost-effective.

This past January, I rented an Audi Q7. It was brand new and had the Audi Virtual Cockpit which took me a full two days to grasp to the point where I could use it safely and required me to pull over several times to figure something out. Worse, the rental agency had removed the owner's manuals so I was without reference. One of the goofiest things was that the navigation moving map would mysteriously stop moving--allowing the car (me) to drive off the edge of the map--and didn't allow me to move the map, to re-center by touch, while the car was in motion, instead presenting an alert saying that the action was unavailable while the car was moving. As I was driving in an unfamiliar city, this issue forced me to stop several times so that I could see the map surrounding the car, it was simply horrible.

To complicate things, the rental agency had bound the two key fobs together with steel cable. This caused the car to arbitrarily recall the user settings from one key or the other--presumably whichever key it happened to recognize first. This had the effect of making the system behave even more randomly as it unevenly switched back and forth between the two user profiles each time I started the car. Every time I'd think I had tamed the user preferences, the "other" set would get loaded, changing the preferences. I solved this by removing the key battery from one of the keys but even with just a single profile, it was a remarkably trying experience and one that seemed to have been engineered for frivolous, show-offy, reasons rather than solid, task-solving reasons.

Mercedes-Benz has, in recent years, also made their UX very flashy and fiddly even though they are not using a touch screen. Just tuning a radio station is a multi-step process requiring the traversal of several menus but lacking the feedback in the sub-menus that is needed to let the user know they have actually made a selection. The user must come back all the way to the top of the menu structure to see if the selection they made two or three steps earlier has had any effect.

Just give me a dedicated knob.

My car has onboard voice controls, but, unlike Siri via the car's audio system, they suck. Recognition is slow and unreliable enough that I never bother, and if I did, the cognitive overhead would be unacceptably high anyway. They'd have done better not to bother implementing them at all and instead just offload that work to the driver's paired smartphone, which for a sedan that listed $32k new, it's very reasonable to assume will be present.

Voice control sucks IMO, even though a lot of people now grow up with Alexa and similar and adapt to the technology. Why does it suck, just like some "smart" buttons? Because it's not obvious what it can do. You have to learn the vocabulary it understands and its other limitations (tone etc.). For example, I tried to get my Tesla to text one of my contacts recently and failed. I didn't know whether it just didn't understand me or was unable to. Good UI are obvious and don't require you to memorize secret capabilities.

> I suppose that voice controls would be a reasonably safe tech option, but probably not cost-effective.

They all suck and they’re literally not workable for rentals. Nobody will waste an hour configuring a rental car’s VA on departure.

I drive a 2010 Honda Jazz (Honda Fit in the US I believe) and I love it.

Not only does it indeed have buttons, but good buttons as well. Clicky where needed, rubbery where appropriate, with the righ amount of resistance etc.

Just like the iPhone always has the nicest buttons of any phone. It's a little thing that makes using it a pleasure.

Thinking of voice-activation in a car in lieu of analog manual controls misses the point - perhaps you weren't advocating for it as a replacement, but the article above seems to.

There are multiple points of failures for a voice-control only interface for a high-speed moving vehicle:

1. Problems with NLP, and needing to adjust oneself to the proprietary format of this particular car's voice recognition grammar/syntax/parsing/etc. (discussed elsewhere in this thread)

2. Loud location (driving by a jackhammer)

3. Female voices [1]

4. AAVE voices [2]

5. Immigrant voices [1]

Some argument can be made that 3-5 could be improved on with larger training sets. Given that we're still working off of much of the same recording equipment that we developed in the early 20th century, this is only true up to a point in regards to #3, since it's difficult to calibrate higher pitched "noise" out and keep higher pitched voices in when they're at a similar range.

Even assuming all the above problems are solved, there are still the cases of someone losing their voice temporarily due to illness, or someone speech-disabled but otherwise able-bodied, and the cases of a computer malfunction in the car itself that causes voice recognition to break. I was in such an incident myself quite recently, where a rented Mini Cooper's onboard computer made it appear that the engine was overheating (red hot engine symbol) and that I must cease driving immediately lest the car explodes. After the car was off, the computer malfunctioned so severely that attempting to turn the car on only resulted in a light-up display of every error message the car had.

All that to say - voice controls are not a suitable replacement for standard knobs and dials, and the latter should always be available as a reliable back-up method for operating a car.

[1] - https://hbr.org/2019/05/voice-recognition-still-has-signific... [2] - https://en.wikipedia.org/wiki/African-American_Vernacular_En...

Is taking your hands off the steering wheel much better?

Ideally, all controls would be available on the steering wheel without removing either hand. Or, as you say, voice controls.

I'm still waiting for the convenient floor-mounted headlamp dimmer switch to come back.

Voice control still isn't there yet. Siri frequently thinks I'm talking to her when I say "hey [kid's name]" — despite the fact that my kid's name shares zero letters with "Siri", either in English or in IPA.

This is pretty surprising/disappointing, considering that the /s/ sound is pretty distinct, and considering that I did the "hey Siri" training when I got my phone.

Exactly - touchscreen offer no unique tactile feedback you can associate with an individual action. Also when the screen is multi use depending upon options then you really handicap the muscle memory and with that, cause distractions.

However, I do feel HUD display tech is one area in which could do with some love as would compliment and be less distracting over a screen located away from the drivers view of vision.

I like physical controls better, but I don't think that's a totally realistic contrast. Neither touchscreens nor knobs necessarily demand you look. I've known lots of people who looked while using physical buttons, and I can do my common touchscreen tasks in my car (e.g. switch to Bluetooth, hit play or pause) without looking. I suspect most people will look in both cases.

Voice controls are nice. Except when you get a sore throat. I'm not talking about disabled people, people with funny accents (everybody speaks English of course and if they don't, they at least speak one of the five other languages we proudly support) and people with non-standard timbre of voice, also.

Honda cars already have built-in voice controls, plus Siri (and I'm assuming OK Google) if your phone is plugged in.

Voice controls are a living hell with kids in the car.

My 2007 Honda already has voice controls. Never use them.

hondas have had voice controls for ~20 years. they're in e.g. the 2004 acura tl. simple voice controls are cost effective and already implemented.

I'm hoping they take up some kind of googles or periscope like interface.

I will never understand how this trend started in the first place. If you look at e.g. digital cameras, they all have touch screens these days, but they also still have buttons and dials everywhere. The one in front of me has eight physical buttons, seven dials and a power switch. The reason is obvious: when you're taking photos, you don't want to be looking down at a screen; you want to be focused on the task at hand and aware of what is happening around you. If camera designers know this, why don't automotive designers, where the task at hand is a matter of life and death?

They make cars significantly cheaper to produce - both in terms of parts and construction time. Evey button you lose is a button you don't have source and install when building the car. There's less wiring, less replacement parts to stock. It's a win in all kinds of situations for the manufacturer.

It's no coincidence that the company with most manufacturing issues - Tesla - also went with completely touchscreen based cabin with pretty much no additional cost. This is further confirmed by the fact that they didn't offset the screen issue by installing a projected HUD display (which is these days available in most 20.000$ cars) - it's complicated to install.

Touch screens are cheap and very easy to develop for. They also de-couple the software development from the device development.

It also allows updates and changes after the car is in production.

I can only conclude from the situation you describe, as well as the fact that most cameras (as cameras, not phones) still lack GPS and sometimes Wi-Fi, that a great many industrial designers are trend-following, uncreative bores.

Most modern car interior controls are horrible (Tesla’s giant laggy touchscreen included).

It started with the iPhone and the software-ization of all buttons. Buttons were seen as the enemy, inflexible blackberry relics of the 1990s. Tesla took this trend and ran with it, and many other cars, like my 2008 Prius, had already begun with their incredibly crappy touch-screen. Mine broke years ago and the audio is ENTIRELY on the front-right speaker. It can not be changed ever because ONLY the touch screen can change sound balance and touch screen's broken. Welp, at least my co-pilot has ideal audio?

I think because cars, although primarily a form of transport, are often also a fashion accessory, or a statement of who you are (or aspire to be). And touchscreens are very fashionable[1].

Cameras aren't really like that. Nowadays if you've actually bought a camera as opposed to just using the one in your phone you're probably at least a little bit into photography, and in that context the function matters more than the form and style.

[1] I'll grant that cost, along with simplicity of reconfiguration via software, may also be factors.

I just got a new camera, and the touchscreen wonderfully compliments the three dials I have. They have several physical shortcut buttons to back out of menus and to get to quick actions more quickly.

Tactile feedback and quicker response time on certain actions is amazing, and just using the camera is a blast.

My 2014 mazda has some touch functions, but the main radio controls and the entire climate interface are all still knobs and buttons. The clock is separate from the entertainment screen. I love it!

In the 90s and 2000s, cars had too many buttons and knobs. And car reviewers started using that as a criterion to judge the dashboard. So less became better. OCD!

Yeah, that’s what I hate about taking pictures with their iPhone, the lack of a tactile button for taking the shot. It’s especially bad when you have to photograph without a view of the screen.

(So that you don’t guess, I mean when photographing the back of my head or the inside of a tight crawl space when my cat got inside.)

Automotive designers are simply catering to their perceived marketplace in order to realize maximum profit extraction.

Most people who drive would prefer to be insulated from the activity as much as is legally possible. Fly-by-wire everything, touch interfaces, etc. This is how you achieve that objective. These "innovations" are obviously the antithesis of safe. But, we all know money is more important than safety, and most consumers are attracted exclusively to this kind of shiny bullshit.

I think there's too much selection bias for this analogy to be particularly strong.

Digital cameras are increasingly a niche product that exist mainly to cater to people who specifically want a the "SLR" user experience. If you just want to take decent enough photos, your smartphone's camera today is better than most prosumer DSLRs from like five years ago.

I don't think today's DSLRs have the form factor that they have because it's objectively more usable. It's just the form factor that their self-selected customers want. One piece of evidence in favor of that is that all DSLRs do have big LCD screens on the back and often require a lot of menu diving to access any functionality that didn't exist in cameras before the digital revolution.

There's still a physical dial for switching modes even though that's not something you actually change that often. Meanwhile you often have to dig into a menu or go through crappy buttons to do things like delete a photo.

I won't go so far as to describe it as fetishization, because I think that's unfairly critical. But I do think a camera UX designed from first principles purely for usability would not have the same physical controls as a typical DSLR.

Always seemed like a cargo cult: iphone good; touch good; buttons bad (steve said).

Let’s hope it’s a new trend! I love that my Audi still has tactile buttons and wheels, and was a big factor in choosing it. My previous car had a touch screen and it was so dangerous having to brace my hand for accuracy and stare at my fingers while driving.

How do you feel about the MMI scroll wheel? I don’t feel safe using that either; it still requires eyes on the screen to see what you’re selecting. Even my passengers, who can give it their full attention, are routinely stymied by the process of pairing their phones for Bluetooth audio playback.

At least I can get the volume and temperature knobs by feel.

even on phones it's "bad", it requires high focus.. it's was only good in the theory that smartphones would be used for prolonged dedicated tasks. how i miss blind use of buttons on old dumbphones

Audi may be doing some things right (I love the location of the audio volume knob), it doesn't compensate for that ridiculous touch pad which you're supposed to use to enter street names by writing the characters one by one with your finger... and many other UI warts.

The best thing about Audi's console is that fact that they support Apple CarPlay.

I always assumed companies were drawn to it not just because of the trendiness but because engineering and general development of these features are faster, cheaper, and easier to fix compared to dealing with physical parts (e.g. knobs, wiring).

It was never a cargo cult of being good... it was a cost cutting play that they correctly thought they could shove down consumers throats. Glad to see Honda acting on behalf of the driver again.

... I see your cargo cult, and I raise you Elon Musk and Tesla with gargantic touch screens in their cars. Aesthetically neat; realistically trouble.

You (or maybe the market) are misinterpreting Steve Jobs. In a phone context, he is correct.

Any monitor manufacturers want to follow suit?

What I'd give for hardware input source buttons. This is a ridiculous UX:

1. Touch button to open menu

2. Select "Input Source"

3. Click up/down (possibly multiple times) to find your source

4. Click OK

...when your monitor has TWO sources. Just give me 2 buttons: HDMI and DisplayPort.

I don't care that this would be marginally more expensive. I'd pay it.

Touch screen controls strike me as not being about cost but being about incredibly lazy UI and a good example of worrying about problems you'll never have like "what if you have 10 input sources? That'd mean 10 buttons". But YOU DON"T. You can't add an input source with software after the fact.

/rant

Touch screens make complete sense for phones. You're looking at it. There is limited screen real estate. The second is why hardware keyboards died. It's even why the Home button on iPhones died, which makes me sad because I loathe Face ID with a passion (compared to Touch ID).

Touch screens require you to look at them. This seems bad when driving. Physical controls are tactile and you can learn to use them without looking. People seem to forget how important this is.

It's also why I prefer keyboards that split function keys into groups of 4 over the terrible design of putting them all together. Function keys are really too far to touch type effectively so you need a non-visual cue of which is which.

Beyond that, every touch interface I've seen in a car is objectively terrible.

Kudos to Honda for bucking this stupid trend.

Totally agree. I have Dell monitors, which are fantastic in every way bar one - they have invisible, capacitive buttons at the bottom right of the bezel. I literally have to put my face right up beside it and squint to have any idea where the buttons are.

You can't easily see where they are, and you get no feedback at all when you press on them, and it seems to be 50/50 whether a press is registered. Horrible, horrible UX!

Dell Ultrasharp gives you shortcut keys that you can assign to any function. I have one for Displayport and a second for HDMI.

My TV has capacitive psuedo-buttons even for the power button. Of course it isn't illuminated, since that would distract from the screen, which means that turning on or off the screen when the lights are out is literally a matter of making wild stabs in the dark.

Since DisplayPort can normally be passively converted to HDMI, your easiest option is probably to get a convertor and also an HDMI switch box. It'll be external to the monitor but there are plenty to choose from that have physical buttons. There are even some with remote controls!

> Touch screen controls strike me as not being about cost but being about incredibly lazy UI

It's almost always about cost. Touch screens are shocking cheap thanks to smartphones driving the prices down and physical controls can be shockingly expensive per control.

I dunno, when it comes to configuring monitor settings, I'd trade a touch screen menu for any of the physical button configuration schemes on any monitor or TV i've ever owned.

https://www.motorauthority.com/news/1121372_why-mazda-is-pur...

A more in-depth article last year on why Mazda is doing this.

Link [1] to HN discussion around the Mazda announcement

[1]: https://news.ycombinator.com/item?id=20200335

Yep, Mazda were the most vocal about rejecting the touch screen trend, and Honda and Toyota are losing a lot of customers to Mazda who have been producing the best cars in their class in the last several years so it makes sense that they are jumping on the bandwagon.

I love the Command Center in my Mazda. I can switch modes, change content, many other things, all without looking at the knobs. Voice controls using Android Auto makes it even better, but I do not use it much since the knobs do a great job. I never use the touch screen.

> Honda has done what no other car maker is doing,

Huh? The author is not familiar with Mazda - their latest models of the CX line (possibly others) have full control of the interface with a very functional physical knob / dial along with other tactile controls.

Also, the headline is a bit misleading - they aren't removing touchscreen controls entirely - only removing them from "some" controls as their preface text indicates, which to be clear is solely A/C controls.

I have a Mazda from 2019 (not the CX), and I can do everything with a couple of knobs. Even the extra buttons you mention (volume up/down in the wheel or quick access to media/navigation/radio in the center console) are redundant and nearly useless. Just a couple of weeks after I got the car, I have never touched the screen again.

A/C controls are completely manual. Mazda got it right from the beginning.

I will defend the screen in the Tesla 3, the simple fact is that you really never have to use on your typical drive and yes muscle memory works for simple UIs like what the Tesla employs.

Compared to 40+ buttons in my previous Volt AND a screen. Many cars have that many or more buttons and this is easier? Where screen's become distracting is when the UI is shit, having more than one or two clicks to do anything, and worse duplicating features there are physical buttons for but naturally distract the driver who thinks they should use that screen.

From automatic headlamps, climate control, and wipers, I really have no reason to interact with the screen except as glances while I do the standard look around while driving.

My favorite test... put a sticky on every button and only remove one if you truly had to use the function. bonus points for not having to remove the sticky to find out what the button did.

edit: spelling error

I work in data visualization. Every programmer who gets into datavis goes through a "3D all the things" phase where they look around and realize that 3D visualizations aren't used in a lot of areas where it seems like they should. Eventually you realize that there are big tradeoffs and 3D is very difficult to get right for the human brain.

Touchscreens are similar. The appeal is obvious. Screens are space efficient, customizable, upgradeable, flashy, etc. But they simply aren't as nice as physical controls with tactile feedback. Not only should we not be forcing touchscreens into every HCI situation, I think we should be moving the other direction, adding more physical buttons, dials, sliders etc to our computers and smartphones. In school a couple years back I worked on a project[0] for adding generic bluetooth buttons on a wrist device.

Imagine if you had 4 extra physical buttons, a scroll wheel, and a slider all sitting on your wrist, and your phone and apps were designed in such a way that you could map these to whatever you wanted.

[0]: https://docs.google.com/document/d/1lTOxHxHFjwJXeCLROAPf6OJD...

Last year I bought a motorcycle (Honda CB300R) and it was a revelation.

No touch screens. No stereo. No GPS. No Bluetooth. No automatic transmission. No cupholders. Basically no "modern car features". And it's a much, much better road experience than any car I've ever been able to afford to drive.

Get rid of all the distractions at a price point under $30k with an engine that doesn't feel sluggish and maybe I'll consider buying a car again.

(Necessary disclaimer: I live somewhere where it's sunny and warm the majority of the year and where car parking is nigh impossible. Maybe I'd feel differently if I lived in rural Canada.)

I buy cheaper cars, because I don't really care about any of the high-end features that cars offer, and this has been one legitimate advantage, at least as of the last time I was shopping; cheap cars cheaped out on the display, so they still had physical controls for things like volume. I hope this trend takes off before I buy another.

I think it would be possible to make a decent touch screen, but it seems like the only people who give even a quantum of a damn about latency are the video game folks. A UI with 95th-percentile or 99th-percentile multi-second latency is an active hazard on the road, even if it is normally acceptable, which itself is fairly rare and expensive.

Interfaces to the car should have standardized use cases/requirements that are interface independent.

There should be requirements like - user should be able to rapidly change stations manually without looking at the interface.

For example I have a dial for my XM radio in my tundra, if I want to go from station 10 to station 100, I can easily do it by spinning the dial really fast two or three times without looking at the screen.

In my honda, to get to that UI where I can arbitrarily tune, I have to hit 3-4 buttons all in different places. Then I have to hit the up button 90 times.

I could imagine a digital tuner that is a bar that lets you simply touch where on the spectrum you want to select a station. But up/down arrow to tune should be banned.

Funny coming from Honda. My parents own a 2017 (I think) Civic and the only way to adjust volume is by tapping +/- on the touchscreen. It's the most unusable volume adjustment I've ever experienced. It was completely unusable with a glove in the winter.

So while most other companies only implemented touchscreen controls for secondary features Honda did it for the most widely used one. Now they are backtracking 100%, go figure.

Bout time this started happening. Now, the rest of the industry just needs to follow suit.

The first and best microwave I ever had simply had a dial, a cook button, and an open button: that's it. The more you turned the dial, the longer your food would cook, and the hotter it would get. That's all you needed to know or do. And you could even turn it back or forth while it is microwaving, to change the time after you start, or finish cooking immediately. But it's almost impossible to find one with a dial, now.

Microwaves with digital keypads are terrible, even with those "convenient" +10 sec / +1 min / hot dog / peep buttons, you still have to hit a little target you can't feel, instead to twisting a big knob you can.

Entering time by touching digital buttons isn't a rapid, reversible, and incremental action with continuous feedback -- i.e. "direct manipulation", like turning a dial is.

https://en.wikipedia.org/wiki/Direct_manipulation_interface

Great! I may actually consider a newer car for the first time in a long while.

I rent regularly and the last decade or so of car UX has been pretty terrible.

Some features are compelling, but distraction limits their effective value.

Feels a lot like early smart TVs to me. Better experiences are had by turning all the crap off.

With a car, the stuff is just there, often does not stay off, requiring constant attention to push out of the way, etc...

An older vehicle, equipped with a bluetooth capable radio, smartphone with voice is better than just about all the new car goodies nearly always.

Frankly, the metrics, mpg, other performance data and sound processing where present are great! I use them when present.

The rest is just a mess.

My other quibble is LEDs. The flicker used to prolong LED life and power use is very seriously distracting both in car, and outside to other drivers.

In car, lots of bright things will often inhibit night vision. Displays, dash lighting, indicators all increasingly bright and many flicker.

As a driver, a quick glance to and fro results in a field of dots. I have asked others about that and have done a few tests when road conditions and traffic present an opportunity.

Older car dashes are not distracting much at all. Can run very dim too. Newer dashes distract far more. Won't always dim, or worse, will dim, sans for one bright thing, usually a little display.

The always on bulbs do leave a vision trail, compared to the dots from LED lights. For many, that trail appears to be processed in a less distracting way.

LED tail lights are the big offender here.

Over time, as I land in various airports, the conversion to LED has been completed. The pattern of speckles are crazy! Various colors, and modest duty cycle rates make for a mess.

Any pilots care to comment?

Mazda also did this. Long ago.

https://news.ycombinator.com/item?id=20200335

Thank god.

I just got a 2005 Porsche Boxster with a CDR-24 radio, and one of my favorite things about it is how simple the interior controls are. Knobs, buttons, switches. The PCM radio I think might have had touch screen, but anyway...

My main point is that my 15 year old car doesn't feel anywhere near as dated as many cars from 2009, that have slow, low resolution, or awkward touch screens. The only car I've used with an acceptable (but still flawed in many ways) touch screen is a Tesla 3 or S.

An absence of features often for me turns out to be a feature. It's a car; do I even really need that many buttons or controls?

My spouse has a 2019 Honda HRV and I hate dealing with the touchscreen while driving. It's cognitively very difficult to do the degree of fine motor control and visual attention the touchscreen requires while also driving safely. It's actually a little terrifying.

I remember reading an article a while back about some manufacturer, can't remember which one, adding a clickable knob to control the "smart" functions. Since so much of what we use a touchscreen for is actually menu selection, that could cover most of what we need to "touch".

Let me give you another perspective - I own a Suzuki SX4 S-Cross that comes with a fluid, super responsive and intuitive touchscreen with voice commands in it, which can also be controlled from the steering. You can do stuff like ask to call someone's phone, or read a text message or ask you to give directions.

But the real highlight is the UI + UX. It's super intuitive in that the volume controls, although touch, are always in a fixed location and you just drag over towards the edge of the screen up or down. In addition, you also have steering mounted controls to control the volume. The whole screen is divided into 4 quadrants and clicking anywhere in those quadrants will take you to either of the main screen (Eg. Navigation or FM Radio). It doesn't take long to get used to remembering which quadrant is for which. Also, for safety, you can't use some core features of navigation while driving the car. It will ask you to stop first if you want to say, type for a certain address.

I have never seen such fantastic UX in such mid-tier cars. To me, it seem's Honda's move is backward - perhaps it could be solved by simple UX..which Suzuki and Bosch (the supplier of their infotainment units) have already solved?

I find it super convenient because I don't even need to take my eyes off the road...while enjoying the convenience of a touch screen.

A picture (or video is worth a thousand words): https://www.youtube.com/watch?v=HvERUKfjufg

> The new Audi A3’s electronics boss Melanie Limmer told Autocar recently its decision to remove some physical buttons was made as “more and more people are getting into touch functions with smartphones”

That is totally the wrong lesson to take from the popularity of touchscreens on smartphones.

Hardware buttons were eliminated from smartphones for one reason: the devices are small and screen real estate is critically important. So the screens need to take up 100% of the usable area of the front of the device, leaving no room for buttons. (The first iPhone proved how game-changing it was to be able to any web site.)

It's not because touching the screen is better than pressing a button. It's because eliminating buttons is an acceptable sacrifice for something even more important.

But in a car, of course, you don't have these constraints on size, so the reasoning doesn't translate from one type of device to another.

It's the equivalent of saying, hey, we noticed that having only one burner is overwhelmingly popular on camping stoves, and campers really seem to prefer it that way, so for all our new models of kitchen ranges, we are going to just have one burner too.

Of course, I am assuming it even is a sincere belief. It's also possible they have some other reason (maybe cheaper?) and this is just a way of justifying the choice.

I have a 2013 Accord with a touch screen (bought it used and it's what came with it) instead of button controls. The latency on the screen is terrible and my father's Acura RDX which does not have a touch panel is much more user-friendly. I would love if more manufacturers started to put more analog controls back into their cars, but kept the non-touch screens for info display/backup cameras.

There is a trend to assume new technology will always replace old technology.

During the Vietnam war the F4 ditched its cannons for the sidewinder air to air missile.

The sidewinder is fire and forget for the most part and somewhat of a no brainer whereas cannons require more skill and training in dogfighting.

It also gave the US a MAJOR advantage (in theory) vs the MiG.

The problem is that the Vietnamese picked up on this and changed their tactics.

Before this the US had air superiority and a significantly higher kill ratio.

The Vietnamese built 'hidden' airfields and would dispatch their MiGs when the F4 was directly overhead and engage them in head to head dogfights.

The F4s needed to build distance to launch a sidewinder and this dramatically rebalanced the kill ratio and during this time it came out to close to 1:1 with the US having a slight advantage.

They US realized their mistake and subsequent F4s had cannons...

Same thing here with cars. When the new screens came out everything was done through software but it's just NOT a good design.

When I'm driving I want to push a button and be done. IF just for safety.

When the UI locks up or is slow it's literally a safety risk.

Funny anecdote: when I was working on the Ultimaker S5 touch screen interface, there were lots of tests done with new users. Many people had trouble figuring out that the previous version did not have a touch screen but a tiny, tiny OLED display that was controlled by a rotating button. People poking their finger at this tiny OLED which was not giving a damn about what they did there. Really funny to look at, but made it very clear customers expected us to have a touch screen.

That allowed for a much nicer interface (we won some award with it), but the only thing that was not better with a touchscreen (IMO) was manual bed leveling, where you look at the print head and not the screen. We did our best on that and to make it as unneeded as possible. I don't have a such an S5 3D-printer and I don't work for Ultimaker anymore, so I'd be glad if anyone here could tell me how they like the bed leveling UI, if they ever used it :-)

The best "driver UX" I've ever had is on a 1995 Rover Coupe, with its dashboard and switchgear carried over from a 1988 Honda Concerto:

https://imgur.com/a/YvpSUFa

Just 3 knobs for regulating blower fan speed, blown-in heat level, and flow direction. One switch for letting/stopping incoming air flow, another for recirculating/capturing cabin air. A single on/off button for the AC.

Left steering column switch is entirely light-related, right switch is entirely dedicated to the wiper system.

Three buttons just below the dials: emergency lights, fog lights and rear screen heating.

Just the bare essentials that you'll be using 99% of the time, easily discoverable at the most sensible locations, with a solid mechanical "clicking" feedback. It's a significantly safer car simply by virtue of lacking distracting controls.

Fuck Honda. I'm never buying another "new" Honda:

While their new and "certified" inventory is rusting on dealer lots, they have declared [0][1] that peons owning Takata NADI inflators (SHRAPNEL BOMBS sitting inches from our faces) should just wait (NO FREE RENTAL) a few MONTHS (sometime in 2021) for them to redesign, test, manufacture, ship, and ultimately replace said shrapnel bomb.

[0] https://hondaairbaginfo.com/nadi

[1] https://hondanews.com/en-US/honda-corporate/releases/release...

Unrelated, but this line stood out, and now I have a question:

> Honda’s decision to return to physical controls will be popular with some - including, no doubt, its ageing owner base in the UK

Ageing? Is there some stat showing Honda owners are older than owners of other models? Or that young people aren't buying Hondas?

Excellent move. I totally support this. Touchscreens removes all the accurate tactile feedback of what you are doing to a control and that forces the user to fallback to taking a look, hence deviating the center of the sight from the ideal zone for a really unjustifiable reason.

Good. Touch screen controls are a good tool for communicating visual information while having some way to still give users some controls. Tactile controls are good tools for when the operator needs to visually focus on something else and feedback can be sensed in other ways. For cars, the only visual information it should show to the driver are either things that are controlled outside of a touch screen (speedometer, tach) or can't be controlled while the vehicle is in operation anyway (oil temp gauge, etc). Everything else can be intuitively sensed by the driver's other senses and the important part of the control is the feedback saying that you are in fact manipulating the controls properly.

> While Honda’s decision to return to physical controls will be popular with some - including, no doubt, its ageing owner base in the UK - the predicted move towards more voice-controlled actions in cars could eliminate the debate around touchscreens versus analogue controls in the future.

No, I will not be talking to my car.

Knobs and buttons with tactile feedback were great and should be used for major functionality.

My favorite fan controller of all time was a fan pull knob on a FJ40, which had physical clicks for each fan setting as you pulled it out:

https://forum.ih8mud.com/attachments/p13-jpg.509941/

It was intuitive and deeply satisfying.

The way my brains works, it’s hard for me to even read road signs when I’m driving (excepting simple ones like speed limits and one-word signs). Switching contexts is just really slow. If I try to use voice recognition I basically go blind for three seconds.

Just and knobs for everything I need when driving. But please keep touchscreen for everything complex I can do while standing still. Complex setup of display options, car setups or whatever is infuriating without touch.

This is how every other device works such as a digital camera. You can’t have an unlimited amount of knobs and buttons but you can have a dozen so you choose very carefully which functions need direct manipulation.

I don’t think it’s so bad currently in many cars to be honest. My VW does what I want, there are knobs and buttons for climate and radio, but setup and other things are touch. Only complaint is that switching source between line in and bt is touch only (and is used a lot).

Car audio systems pose greater dangers than texting, pot

https://techxplore.com/news/2020-03-car-audio-pose-greater-d...

Thank goodness. I bought a Honda quite a few years ago and got the top trim, but insisted that they remove the navigation system as just using it as a radio was far worse because it was entirely touchscreen focused. This might keep on me on Honda's for a while.

Edit: Wait... they're just talking about climate controls here. They're not adding any dials. Apparently other carmakers are going just touchscreen for everything. I guess its just gotten worse over the last few years. Tesla can sort of get away with it. But goodness a huge part of the reason I liked a lot of luxury cars was they tended to do more dials instead of touchscreens.

Who thought of the "touch-screen in cars" idea and how did he bait the entire industry into copying such a shitty idea? Touch screens in cars don't even make sense in fully self-driving cars, since you'd just lay down in the backseat on your phone on your way to work.

I still have no idea how to work the screen in my dad's old car. I would literally have to pull over and whip out my phone to Google how to use the damn thing. It made setting/changing the music way harder than it should've been. No one wants to pull over to look up a dissertation on how to use change the channel on your radio.

Did you know that in a car you can change the climate control temperate for when the system is "off"?

Try it today. In an old car, set the fan speed to off. Then adjust the temperature. Switch it back and forth a few times and test it.

In a new car you need to turn the fan on, then change the temperature and then turn it off again, in order to change the "off" temperature.

On a touch screen this task requires many seconds of eyes-off-the-road-time to accomplish.

But hey, designers need flexibility because changing temperature is a new concept and they couldn't possibly handle a deadline to work with component selection and other parts of the product team.

All they did was put the analogue A/C controls back. I don't see this as a big deal and I kind of like that some of the basic controls don't require digging through menus. One of the drawbacks of touchscreen controls is you have limited screen space to cram everything that used to cover the whole dashboard.

New features like navigation and music makes sense when moved to the touch screen, but adding things you don't need to just means I have to switch away from navigation or music controls just to change the A/C.

I think Honda is making a good choice in terms of User Interface design.

I think what Tesla did was cost cutting by putting everything in one big screen - no need to think about hardware knobs and buttons - have the screen UI and update it remotely if it is not intuitive.

I have a touch screen in my Nissan and it's almost useless when I'm driving down the road at 60 miles an hour and trying to navigate menus 8 layers deep to do something simple that used to be a button press on my old Honda.

Relatedly, the US Navy removed touchscreens from all vessels after a collision happened because someone couldn't navigate the interface in time. https://www.bbc.com/news/technology-49319450

The dirty secret of the electronics industry is that the reason for moving to sensor buttons is cost.

When first capacitive touch devices appeared around 2005-2006, everybody immediately noticed that you only need a single digitizer for the whole front panel of the device, or dirt cheap cap sens circuits for point sensors.

For as long as the sensor is just a piece of polyimide pcb you can glue to anything, but a metal surface, things get really cheap without mechanical buttons, holes in the case to cut, and PCB to hold the buttons.

Touch controls in the car should not be used for anything that is supposed to be used while driving. 1) They give no haptic feedback and require to look at them to do anything. 2) They are very finicky if the hands are wet/dirty/sweaty/cold. Knobs can be rotated and buttons pressed without looking. Physical objects can be manipulated as long as you can move your fingers.

One more reason to consider Honda when the time comes to buy a new car. Another reason is their outstanding reliability ratings.

I recall Mazda doing this a while back. [1]

[1] - https://news.ycombinator.com/item?id=20200335

Mazda have also announced the same thing. (https://www.motorauthority.com/news/1121372_why-mazda-is-pur...)

Favorite quote: “Doing our research, when a driver would reach towards a touch-screen interface in any vehicle, they would unintentionally apply torque to the steering wheel, and the vehicle would drift out of its lane position,”

I understand the problem that touch-screens solve on cell phones. How do you squeeze the most functionality onto the smallest surface? By doubling up on functions, and reusing existing functionality to be context-sensitive. The trend of moving designs from where they make sense to where they do not because the inferior design is seen as more modern since the devices with these constraints are modern has been a very annoying one. Designers have been making software and technology progressively worse. It isn't in any way surprising, it's just annoying that when a piece of equipment or technology can fully utilize the features and interfaces that cheap and mobile technology is forced to do without, they choose not to use it. It'd be like if bicycles became really popular all of a sudden, so auto industry stopped putting in seatbelts and airbags in cars, because bicycles are modern and they don't have those things in bicycles. You're building a car with a fixed set of controls for the machine you're driving in: you can afford to use a physical UI, what are you doing putting in touch-screen?!

IMHO the physical controls invoke muscle memory and haptic feedback which enables driver to keep attention on the road. Touchscreen does not give this to same/any degree. I should not have to take eyes, or attention, off the road for longer than it takes my hand to find the right control. Physical controls current significant advantage over digital is "no interpretation required" - feel or microsecond glance tells you the state. Digital, other than big dials and flashing colors, needs you to think. Those thoughts take focus away from the act of driving.

Im summary i think Honda is right - digital controls for things with no time and attention pressure (like music channel), and physical for those that need near immediate action (like defogging a window or wipers).

FTR been driving cars, trucks, motorcycles, bicycles for nearly 50 years - 2 accidents, no hospital trips, and that includes plenty of 'hooning about' well into danger-zone for conditions/environment (wont say more or might get a visit from authorities :-p )

More accurately, they're putting back dials for temperature control in one of the only cars the make that doesn't already use dials. They're not getting rid of touch screens, nor are they pushing against the industry: they're aligning temperature controls in one of their cars with their other cars and much of the rest of the industry.

I do hope other automakers follow, because why should a head-unit should be considered safe to operate while driving, while touching a smartphone display isn't.

Both lacks the tactile feedback buttons are offering, both don't work with gloves and both requires you to look at the display (and stop looking at the road) to know what action you're making.

My preference in cars is touchscreen for rarely used functionality, and then dials, buttons and sliders for everything else.

Messing with audio is a peripheral task so expecting me to take my eyes off the road etc is still unacceptable. Especially to change the track or adjust volume.

That said, we're all used to small touchscreens. These tend to require the user to maintain the current context on what is on screen. The whole screen is subject to change and this complicates / weakens muscle memory. I've used larger touchscreens where controls don't change around so much and muscle memory can easily kick in if the design is sane. The difference is tangible. Eg you can blind press using peripheral vision and it can work well.

The real answer might just be that we're in the early days of touchscreen use. The technology and application thereof is still immature. The design language, for want of a better term, is still evolving.

Not all touch controls are created equal.

I'm really happy with the interface in my Model 3 and I've found every other car interface I've used to be on the spectrum from terrible to okay (BMW, Mercedes, Mazda, VW, Porsche).

I think you can probably do either well, but it seems like the car companies (other than Tesla) just don't have this capability.

Weird how the article says that Honda is the only manufacturer doing this when I read about Mazda doing it in Feb

https://www.motorauthority.com/news/1121372_why-mazda-is-pur...

If Honda is "bucking a trend" here it just shows how much professionals, experts and decision-makers can be fooled by trends.

I speculate that Tesla made it something other luxury brands wanted too, and then those brands went bezerk: E.g. the newest Range Rover models feature multiple touch screens, they turned your AC-control-knobs into touch-screens as well, and even the once-handy controls on your steering wheel. What's the point then of still having a phisical gear shift?

I applaud Honda for making common sense an official statement again. I ditched the newest lineups of all new premium brands (also) because of that touch-screen nonsense and choosed a cheaper brand instead where it's at least just one touch device.

Premium brands still don't seem to get how un-premium a dashboard full of thumbprints and smudges looks like, even less if everything else has a piano finish.

I've designed hardware and software for a number of touchscreen devices and applications. I even invented the double-sided touchscreen, which may show up in phones in another few years.) So I love touchscreens, BUT, he lack of feel makes knobs, buttons and switches a far better choice for moving environments - I hate most new cars that lack real controls for at least commonly used functions. It's hard to beat the knobs and sliders in most 50-60 year-old cars for easy work-by-feel operation.

(FWIW, I always thought it would have been cool if someone in StarTrek TNG/DS9/Voyager had sat down at one of those smooth glass LCARS interface panels and been surprised that they could "feel" 3D controls on them. That seems to be the sort of thing that would be done given the obvious capabilities of their holo/forcefield generators...)

To be fair to touch screen controls, Honda had exceptionally poor ones. There’s always at least a second delay on brightness/volume/etc. adjustments in our 2018 Honda. I think the unit is just incredibly underpowered, because CarPlay (phone driven) doesn’t suffer from the same issues.

Not all touchscreens are same. There are few touchscreens that are good. Most Honda touchscreens are bad.

"Honda is doing what no other car company is doing..."

Except Mazda. 6 months ago. Oof.

https://www.motoringresearch.com/car-news/mazda-getting-rid-...

Hallelujah, at least one auto maker is willing to publicly come out against this insane trend of making car controls visual only. I mean, one would think that when you are inside a ton of metal speeding along the road at 60 mph, you would insist on controls that allow you to have 100% of your visual attention on the road and rely on your other senses to change the channel on the radio or adjust the AC. But somehow people think driving a car should be something like playing candy crush on your tablet. That means going forward I'll always have a car manufacturer I can rely on to not sacrifice my safety for a marketing coolness. Too bad makers like Tesla (who have a lot of leadership in other areas) completely drop the ball in this regard.

This is a great decision by Honda. Touchscreens are a terrible UX when driving. Its difficult to be accurate with the touchscreen in a car when its moving, even as the passenger. Physical controls are much easier to control, less distracting, and we develop muscle memory for knobs dials and switches easily.

As an example, I personally am not a fan of the extreme minimalism in Tesla cars. A fully touchscreen experience makes sense only after a car is capable of fully autonomous self driving. As long as a human is responsible for controlling/monitoring the car, physical controls are better than touch controls.

To be fair, touch screens are better for changing settings, monitoring the health of the car, and any other action that we do while the car is parked.

I recently purchased a 2020 Honda Odyssey. I hate the touchscreen controls, and I hate the button gear shifter (i.e. press a button to reverse...)

For example, it's so difficult to change AC controls w/o looking at the screen. It's very dangerous. I don't do it unless I'm at a light, or if on the highway, no one is around me.

Toyota Sienna still has their AC controls as buttons/knobs, still has a regular shifter, and probably more traditional controls.

But the Sienna was lacking in many other areas...so we went with Odyssey. Looks like we coulda waited a few more years!

The article also refers to "voice control". No, I don't want to yell at my car to lower the AC when my baby is sleeping.

I'm all for innovation...but please leave controls as it is!

Wasn't it the US marine that moved back from touch screens to physical controls in their ships?

While the actual reason could be a case for discussions, the decision is correct in my humble opinion. The reason was, the loss of (I think) one destroyer, that crashed with a civil ship (a big freighter or tanker), because of misinterpretations of the steering control on the bridge of the destroyer.

In my humble opinion, when it comes to mission critical and distraction free control, nothing is better than physical control systems. In a good car, not only you have physical control systems, but also buttons and switches are of different shape or size or way of control (pressing, twisting, ...), so they can be identified simply by touching them.

I very recently bought a BEV Hyundai Ioniq of the first Generation, and one of the reasons I chose the 1st gen was that it still has some knobs and dials. The new generation has almost only touch controls, and this feels kinda wrong. Notably, it’s not a pure touch screen, but the knobs and dials have been replaced by a lot of capacitive touch elements with backlights. When it’s dark outside, it would take me a few hundred ms to find the right button, and then I still have to visually guide my finger there. And then there’s no physical feedback when pressing.

There are other reasons for why I chose the 1st gen (better fast charging capability), but the touch elements were the last straw.

Me in my bit older Mercedes, my gears, lights, whipers all go automatic, im left with radio control and temperature. Audios is easy, it on the steering wheel, so is answering the phone. Two knobs for warm/cold and airco. No need for touchscreen.

In my opinion, we need a mix of both. Trying to interact with an entertainment unit with some type of abstracted/bastardised joystick or touch pad is just plain awful. Whereas interacting using touchscreen is just so much simpler, plus allows for UIs that weren’t part of the original product development (e.g Apple CarPlay).

But for other things, we need dedicated physical controls that mean we don’t have to take our eyes off the road to use; for example, climate control, volume control, cruise control, hazard lights, etc.

A good manufacturer will integrate a mix of both, and user test at each stage of the manufacturing process to ensure what they’re doing is in the best interest of the user.

My Prius prime has the worst controls possible. During spring and fall, when one part of the day is cold and the other warm, it’s annoying to turn on the car. For instance, if I commute home and used A/C, turning on the car in the morning will have the A/C blasting when it’s also cold outside. And I have to sit through the animation and boot up of the screen and then go to the climate control settings which usually is a very uncomfortable 30 seconds every time. And then, the touch is not as smooth as a modern smartphone. This is a 2019 Prius too so idk what Toyota was thinking going all touch on this one.

If my blood is on the touchscreen, it shouldn't break my car or prevent me from getting to the hospital. Who needs more controls? I want physical levers and switches on the dash. Case in point, Ford Mustangs have electronic controls to pull heat off the engine, but the valve only opens when the thermostat says it okay so when it breaks (and costs $250 to fix, and breaks regularly) you're stuck with cold air in the winter until you exceed 50MPH. Just give me a freaking lever that manually/physically opens the valve to and pull the heat off the engine. If it ain't broke, don't fix it.

Good. Those damn things were clearly never tested on real roads with bumps.

> "Oops you fat fingered the main menu icon while you were 5 layers deep! Guess you have to start over!"

Physical controls have their place, and in a car is one of them

My Mazda 3 sGT (and current year model) already does this and it's a joy to use, no touchscreen climate control!

Pics: http://www.2wired2tired.com/wp-content/uploads/2015/03/Mazda...

https://blogmedia.dealerfire.com/wp-content/uploads/sites/45...

My 2018 Opel Insignia also has mechanical dials and buttons for climate control and answering phone, setting cruise control and other “critical” things. But if you want to do detailed configuration for radio, sound balance, music etc you have a rich touch screen available with big buttons to hit. I find navigating between radio stations and navigation so much more accessible with touchscreen that I wouldn’t want these things in another way. Anything that has to be accessed using a joystick like interface should be on the touchscreen and critical one dial features on mechanical where possible.

I like touchscreen controls for my apple car play, but literally nothing else.

I can adjust the climate control temperature with dash buttons, but if I want to control which vents the air comes through, I have to use the touch screen.

Same for if I want to turn on the heated steering wheel. I don't use it too often so that doesn't bug me _too_ much, but sometimes if it's cold when I start my car, it turns it on for me. So when my hands get too hot I have to minimize apple car play, select the ford app, select climate control, and tap the tiny button for the steering wheel.

Seems way more dangerous than reading a text.

I really wish Bevi (and similar kiosk type coffee or dispensing machines) would add physical buttons for dispensing and use the screen only for flavor information. I think our machine's screen isn't properly grounded or insulated from the refrigeration motors. I'm appreciative we have one in the office, however, and look forward to getting to see it again sometime.

Definitely agree with the sentiment that touch screens have gone too far. Feels kind of like the over use of plastic as we got better at manufacturing it. Hopefully the pendulum will swing back on touchscreens too.

I don't think anybody even bothered to read the article!

It explicitly states, right below the top image -

>> The new Jazz has a touchscreen for many functions, but not temperature control

So, only the temperature control is via a dial. Everything else is still on the touchscreen.

Image - https://www.autocar.co.uk/sites/autocar.co.uk/files/styles/g...

I could only image being blind and having to navigate a world that is completely flat. Knobs and controls add something to grab onto. to me it's also safer, if I'm driving and want to change the radio channel, i can feel for the knob without having to take my eyes off the road. i can't do this with a flat, touchscreen display and I also find that voice command work most times but, on the time they don't, i get frustrated and that's not good when driving either as it can cause road rage.

innovation is wonderful, however sometimes the old ways win out.

Until voice control (or fully autonomous vehicles) is 100% awesome, nothing beats tactile, physical controls. If designed right, you never have to leave your eyes from the road (good luck doing that with a Tesla, for example) and you know 100% what you're controlling.

What also nobody discusses (except for maybe Doug DeMuro): digital screens and controls age POORLY. Any car that's 10-ish years old with tons of tech looks ancient, where an even older one with mostly-analog controls can still look sharp. This obviously mostly matters for high-end cars.

Good. There is no way to use a touchscreen effectively while going 80km/h.

It would have to be the smartest touch screen ever, and be able to make assumptions as to what i'm trying to press -- fudge a lot of precision -- and even then I wouldn't know if what I'm touching and what the machine thinks I'm touching are correct without taking my eyes off the road. Meanwhile, I can feel for the 3rd knob or 3rd button and be sure what it is without taking my eyeballs off of the bumper of the car in front of me.

My mom she has a Honda Jazz and the UI is one oft the most terrible from all cars I have seen so far. A few things at least like Audio volume, some radio channels and temperature should have hardware knobs or buttons, so I easy can change it without looking on the screen. Then also should be a place under the screen where i can rest my hand (like has something, at least in a new car I drove recently), so I can touch the screen with a single finger multiple times on the same spot without looking on the screen.

Great.

I think voice controls are obviously the theoretical best for automotive and I hope that in the near future every car will have built-in functionality - that will even work offline.

Obviously Siri can't control my AC right now but it infuriates me that CarPlay voice controls don't work without internet.

I have a hard time imagining it would be difficult to increase offline functionality when in my old 2010 Ford I had Microsoft Sync with the functionality to voice control an iPod offline and it worked pretty much perfectly.

My 30 year old Land Rover has one of my favourite user interfaces.

For example, the cooling system. It's a big lever on the dash connected directly to a big flap on front of the car. You pull down the lever and it opens the flap, and air gets rammed in through the vent. The only moving part is the lever.

Everything is just so tactile. The cooling system, the gear stick, the transfer box, the switches. The user interface goes CLUNK and you can feel the thing on the other end doing the thing you made it do.

Another fad that should go away is blue light dashboards. They give horrible afterimages in dark conditions. BMW's red-orange lights are so relaxing to look at in comparison.

Touchscreens are not the only UI problem in recent cars. I have a recent Honda, it suffers from the problem that they decided to make the crucial buttons below the touchscreen touch sensitive (with no tactile feedback). So for eg. if you wipe them should some dirt land on them, you set them all off/on by accident! What's worse, they are glossy, and hard to see with glare. It looks good though and sells cars, I call it "the iPhone" design.

I was somewhat dreading what the interface would look like on my next car purchase. On the one hand I'd love the idea that the navigational map on my phone may be displayed on the car's "big screen", but I didn't know what to make about this touchscreen stuff.

Makes a lot of sense! Right now I can adjust almost everything by touch alone: cruise control, volume, etc. I would hate for those to get buried in an inevitable submenu on a touchscreen.

I see the benefits of touch screens, but I really miss tactile controls, like the temperature controls on my Dad's 1978 GMC pickup truck, which is the vehicle I learned to drive in. Three super simple, super intuitive slider type controls for all temperature settings: https://i.ytimg.com/vi/iVtb4z2QozY/maxresdefault.jpg

I for one will be really glad if this sparks and industry wide trend to go back to buttons and shift levers rather than buttons or levers that aren't actually good.

GM had it right!

http://carphotos.cardomain.com/ride_images/2/4797/3641/24491...

https://bestcarmag.com/sites/default/files/65564525286387311...

I own a Honda HRV and I agree. The fucking touch-slider for volume control is absolutely atrocious.

There are buttons on the wheel but these are slow compared to a good old knob.

I like the UConnect system in my ‘19 Jeep Grand Cherokee. It has a mix of touch features and dials/buttons. Some features you can control with both touch and dials/buttons. Front seat warmers and coolers do no have physical buttons and I dislike it a lot.

Physical and touch controls should complement each other. Too much touch is bad and too many buttons are bad. I remember looking at Acura’s with to many buttons a few years ago.

Ha! This reminds me of Navy and it’s destroyer doing the same thing.

“The US Navy will replace its touchscreen controls with mechanical ones on its destroyers” https://www.theverge.com/2019/8/11/20800111/us-navy-uss-john...

This is fantastic news. Reminds me of this GREAT READ: https://www.amazon.com/Hand-Shapes-Brain-Language-Culture/dp...

I would not be surprised if this decision resulted in even greater brand loyalty, e.g. drivers becoming attached to the "feel" of a Honda.

I'm glad they're going in this direction. Mostly all BMWs use the screen for simply displaying informational content and for input and navigation it uses a multi-direction knob right by the arm rest and it's best most intuitive way to navigate the screen imo. Car screens are either matte and less responsive or glossy and more responsive but less visible in sunlight.

Physical devices are far superior to touch controls. Not only can you operate them much more reliably under movement and vibration, you can reliably control them without looking.

Or would you replace your (_mechanical_) keyboard with a touch keyboard of current tech?

Anyone pushing touch screens to replace physical input devices in situations where you need to be doing other things at the same time is just insane.

I've always been very happy with Honda's car interfaces. Everything from basic driving stuff to climate control and the radio seem... well, not great, but simply, fine. Most other manufacturers seem to have some trick up their sleeve, with crazy turn signal mechanisms or some other nonsense. I think they have relatively sensible people working in that department.

What I really hate about touchscreen, and love with analog input, is the "speed control" of rotary dials.

For example, if you want to change the air speed of your ventilation system down to zero from max, a quick rotation of your vent speed knob will bring it back to zero and you can rotate it back to put it to one, without looking.

With touchscreen, it's hard to do the same without looking.

The only feedback I've seen has echoed my own sentiment: "Good". What is the opposing point of view? Is there anyone out there who is willing to defend touch-screen interfaces in vehicles? What's the advantage to the driver? Is it even that much cheaper to the manufacturer? Surely plastic knobs are not a big hit to the bottom line of car makers.

Unfortunately we won't get this car in the US. I love my 1st gen Honda Fit and plan to keep it pretty much forever, they're such a great vehicle. Incredible gas mileage and they're actually fun to drive with the short wheelbase. In tune with the article I put the only Carplay receiver with a physical volume knob in it and it's perfectly easy to use.

“The new Audi A3’s electronics boss Melanie Limmer told Autocar recently its decision to remove some physical buttons was made as ‘more and more people are getting into touch functions with smartphones’ and added that the new system is as user-friendly as the previous one.” - her model is the thing that is responsible for countless distracted driving deaths... great

Now let's go back from this flat UI nonsense everywhere to the kind of solid look that Windows '95 had and make my decade, not just day!

https://socket3.wordpress.com/2018/02/03/designing-windows-9...

The only touch screen in any car should be a dumb terminal for CarPlay/Android Auto. Everything else should be dials and buttons.

As far as I'm concerned, there is nothing inherently wrong with "touchscreen controls", it's the usage and user interface assumptions.

Make a touchscreen slider and who knows, perhaps it's even more durable and reliable than a moving part. It's the whole sets of modal interfaces, and jenky controls which make this stuff unsafe, and irritating.

Touchscreens unfortunately only provide visual feedback. I can feel the wheel turn without taking my eyes off the road.

There are lots of development-related reasons touchscreens are appealing, but in situations where the user cannot reasonable look at the screen much, it's not obvious a touchscreen is a good solution. Maybe with haptic feedback of some kind.

Real buttons provide tactile feedback. Plus they don't move around for each menu screen and with software updates. They can be operated by muscle memory. Buttons also provide information about the current state (think ventilation/heat). It's not about comfort but security that is maintaining visibility while keeping eyes on the road.

I believe Mazda made this decision not too long ago.

I hope this is a sign of things to come for the automobile industry.

Essentially, why I don't like touchscreens in automobile media interfaces:

-no tactile response

-more distracting due to the increased dexterity required to get where you need and the greater necessity to focus your eyes

-less intuitive than button/wheel controls (in my experience)

-uglier interfaces (again, in my experience)

There should be UX research about an ideal combination of tactile physical interfaces (knobs, dials, push buttons, whatever) and voice-activated and voice-feedbacked control while driving.

I we're missing out on not only voice-activated commands, but AI responding and updating about the situation through speakers, not just through the screen updates.

From the article, “ Honda has done what no other car maker is doing”.

False, Mazda started this with the Mazda 3 in 2019 (https://www.motorauthority.com/news/1121372_why-mazda-is-pur...).

> We changed it from touchscreen to dial operation, as we received customer feedback that it was difficult to operate intuitively. You had to look at the screen to change the heater seating, therefore, we changed it so one can operate it without looking, giving more confidence while driving.

Finally, logic, reason, and listening to customers.

This is #1 reason why I still love my Toyota over Tesla. Physical buttons make a huge difference in driving experience.

If only someone invented a tactile touchscreen, i.e it would bulge where user can interact. like dynamic braille. Material design where different elevation levels are literally popping out of screen at different levels.

Apple should follow and ditch the Touch Bar. Also, the excellent The Best Interface is No Interface book [0] talks quite a bit about unnecessary touch screen controls in cars and related topics.

[0] http://www.nointerface.com/book/

What you need is both. The screen should still be a touch screen, and there should be buttons and toggles around it.

For example, I hate it when I can't pan/scroll through the map on the car screen for lack of touch screen. But I also hate it when I have to use touch buttons to change the music that's playing.

Is there a name for this phenomenon where manufacturers build harmful features into their products purely for the technical credibility? It's the same with motion smoothing and televisions -- the experience is massively degraded, but the technology is impressive, so presumably sales will increase.

I'd like to point out Mazda has been doing this for a while. https://www.motorauthority.com/news/1121372_why-mazda-is-pur...

> the predicted move towards more voice-controlled actions in cars could eliminate the debate around touchscreens versus analogue controls in the future.

Oh yuck, voice controls are the absolute worst. Anyone who thinks otherwise probably only drives alone (or at least without kids yelling in the background).

I find the climate control on my Honda Civic to be maddeningly confusing for reasons that aren’t really too related to the fact that it has some touchscreen elements, but I’m glad nonetheless. I feel like forcing things to physical buttons forces certain design decisions that you want anyways.

I have a 2019 civic. even with the non touch screen option, a weird glitch in bluetooth freezes the entire media center. I have to pull over and park the car and turn the engine off to reset.

Hitting the power button does not solve this. So in some sense having hardware buttons does not solve everything.

It seem like what I would really want is a big screen in the middle Tesla style and glue an xbox pad on the steering wheel. And give me a fold out keyboard if I need to type something real. And maybe a minimal hud with just the speed and top speed. That's at least what I would do.

I remember years back there was a company that had and inflatable layer ontop of touch screens for flexible "buttons". I'm sure there's ergonomic pros/cons of different kinds of physical switches but seems like we need to move on better touch screen haptics.

My parents have a Subaru with a touch screen for the radio, and I guess some apps.

For the radio, it's crap. For anything else, why would I want something built into the dashboard that'll cost 10x what it's worth to repair?

I don't understand what's so appealing about an in-dash touch screen.

Wow, this is easily the most annoying site I've come across recently. Not only does it block the whole page if I'm using adblock, but it doesn't even start working when I disable it (I'm using uBlock Origin).

We really need to figure out a proper anti-anti-adblock technology...

"Honda has done what no other car maker is doing, and returned to analogue controls for some functions on the new Honda Jazz."

https://news.ycombinator.com/item?id=20200335

Mazda began doing this last year [0]. If anything Honda are jumping on the trend they started, and I'm really happy to see it.

[0] https://news.ycombinator.com/item?id=20200335

I'm hoping to get an EV or hybrid for my next car - probably a sedan. My biggest apprehension with most of the ones I've seen is the excessive dependency on software/touchscreen based controls.

Does anyone have experience with Honda EV or hybrids? Would you recommend?

Isn't this the perfect application for ASR technology? Instead of fiddling around with buttons, or worse, a touchscreen, you could just say "Honda: heat up", "Honda: heat two down", "Honda: aircon off". I would love that in my car.

In an odd way that Douglas Adams predicted this weird evolution of controls making fun of how manual controls were eventually replaced by controls that forced a pilot to sit perfectly still so as not to change a radio station.

I applaud Honda. Touch screen is a distraction in my car.

Didn't Mazda do it first?

https://www.express.co.uk/life-style/cars/1148040/Mazda-3-to...

Why not do a standard benchmark, perform a number of actions in different conditions and configurations, and see whether humans can react faster and more precisely with 2 touchscreens (let's level the 2 hands advantage) or multiple hardware controllers...

Probably the one thing that bugs me about my M3. 90% of the time, it's fine... but that once or twice a week where I'm actively driving and need to turn on AC or whatever and have to navigate while trying to interact with the touchscreen is rough.

Tactile buttons and switches are much easier to use by muscle memory. There is very little muscle memory with touch buttons and touch screens.

Also, touch screens are much harder (thus more dangerous) to use while you are in motion, hitting bumps, going around curves.

The physical dial and nobs and the tactile feedback when human drivers get from operating them is much better than touch screen controls. I applaud Honda's the design, as well as some German automakers' design in the same vein.

Touch controls have their place in the car - they are a MUST for a quick map navigation and they are very handy for entering addresses.

Otherwise - yes, 100%, physical buttons and knobs are far superior. Especially for making adjustments without looking.

"Honda has done what no other car maker is doing" isn't correct. For the latest Mazda 3 (2019) Mazda has made the same decision. There's a non-touch screen which can only be controlled by a center knob for safety.

What physical controls are an ABSOLUTE MUST for you?

Leaving aside Legally mandated: Wipers, Headlights, Gear Selector, Brake, Accelerator, Horn, Hazards.

Me: Required: Mute Music, HVAC temperature and fan speed. Strongly preferred: All audio controls, Cruise control.

I've long wanted to create an after market midi style controller that people can put on their dash to give their car the spatial, tactile control again.

Integrating it with the vehicle though... Somewhat dubious if you're just CAN writing

Good. I recently told my local honda dealer I wouldn't be buying another of any car because of touchscreens exactly because of the need to look and lack of tactile interaction.

Clearly this change is because of me, specifically, only me. ;)

I wouldn't mind a tactile touch screen [1] with optional voice prompting, and a jog dial.

[1] https://www.youtube.com/watch?v=IiM0u79fals

> Honda has done what no other car maker is doing, and returned to analogue controls

It's what I like with the Up! no touchscreen, no stupid added function: just a good mobile phone fixation for the infotainment and good old buttons.

I bought a Kia minivan in 2018 just because it had the most real buttons and dials.

Odd article as this isn't new. Is it new in that other manufacturers have started and they have no plans to? My current Honda has a very comfortable interface, probably my favorite of any car I've driven.

Yay Honda!

I've never been in a car that had a better designed interface than my '91 Accord. I'm sure they're out there but if a budget car can have such a pleasant interface there's no reason they all can't.

I wonder if you could take existing touchscreens and just have them vibrate differently depending on where the user is touching. Or have them vibrate differently as the user's hand moves across different buttons.

I just need a basic stereo, manually cranked windows, and a 5-6 speed stickshift.

It ia a about usability and safety of operation, not reliability. Touch screens need eye contact to operate. Knobs and physical buttons can be operated on muscle memory only, so your eyes can stay on the road.

Without seeing the buttons and knobs in question it's really hard to see whether this is a good or bad move. I've been in plenty of cars with conventional controls that were horrible to use.

Edit: fixed typo.

I'm pretty happy with the newest Rams - large screen with touch controls, but things like volume and temp are physical buttons. Volume is the one thing that should NEVER be a touch control (IMO).

Tell that to my 2019 HRV. Doesn’t even have CarPlay so I can’t use Siri to do anything for me. All AC controls are touch screen so I have to look away from the road just to change the fan speed.

The headline had me alarmed. This is just moving heating and whatnot to physical controls. That's awesome. I was worried they were getting rid of the touchscreen entirely.

Android Auto has been an absolute game changer, and it works so much better with touchscreen. Voice controls actually do cover the majority of use cases, but things like being able to quickly see the next turn or zooming out on the map with one hand are simple, and can be done without really looking at the screen until you need to see the info.

Unfortunately, the car I ended up buying (Acura TLX) had Android Auto, but was controlled by a dial knob. It was really the worst of both worlds, since heating was still managed through a second screen that was a touchscreen.

A lot of mention of Mazda, but my 2018 Ford Escape has buttons/knobs for all the standard car features. I also have Apple CarPlay which is great for maps/media. Very happy with it.

New Mazda models also no longer have touchscreens. We lease one of the newer models and honestly do not miss the touchscreen at all. The rotary controls are intuitive and less distracting

Related (re Mazda) from last year: https://news.ycombinator.com/item?id=20200335

Good for Honda. Frankly, some of these touch screen features were getting out of hand - no pun intended. I miss the tactile experience of touching a knob, a dial, flip a switch.

Hopefully other manufacturers will follow suit. In my view the large touchscreen / lack of mechanical controls is the single biggest deficiency of the Model 3.

I'm baffled by the insane number of upvotes this post has. "Honda is putting knobs in their cars" is the least interesting thing posted in HN history

Mazda waves hi. Touchscreens auto- disable when driving but buttons/ knobs are still usable to control it. Last I heard they wanted to move to a HUD system.

Dear Honda, my next car is now much more likely to be a Honda. I am literally at less risk of dying with controls I don't need to look at. Thank you.

Every single car - except Tesla - does a bad job implementing touchscreens, from my experience. They are generally a pain to use and unintuitive.

At least it's good that honda has been thinking about this kinda stuff. Some of their UI's have been pretty atrocious in the past.

Their touchscreen implementation is so horrible, that this sounds like the right move. This is the only thing I hate about my 2019 Clarity.

Solution here is generic knobs that can be assigned to different functions: Temp, fan speed, volume, for those with kids fader control :-)

Best thing that ever happened in the automotive industry. Thank you, Honda. The many accident victims will thank you for the rest of us.

Honda has good safety and UX in this. Touchscreens are sexy but distracting and probably not sensible. Tesla will fight back I am sure.

Voice control maybe works for folks in silicon valley but it is very lacking in quality for smaller languages or even non-existent

A friend bought a Tesla and he was very pleased with it. He insisted I drive it. I couldn't stand it. It was obviously designed to not have a human drive it. My friend doesn't care for driving, but I love it.

I think that's where the touch screen comes in. The driver is secondary. They have to let the driver access things, but they don't have to make it safe or convenient.

Now that the self driving hype is dying down, perhaps the driver will be put first again.

Top complaint of my Civic, touchccreen controlled volume/mute. TS buttons just don't cut it for some things.

hallelujah!

the controls were a big reason i got a subaru; its a car you can drive the ____ out of with eyes on the road/dirt/gravel. the layout on the honda looks astonishingly similar. i duno how it is formalized in business terms, but it seems neat how all the japanese auto companies seem to cooperatively share tech

God bless em.

RWD electric car, and tactile controls, and all the other cool stuff they're doing? Honda is becoming cool again.

Although I do not drive around much, fully hardware buttons + Siri voice controls(maps, music) does not sound bad.

Hooray! That’s why I didn’t get the EX-L and went with the EX, didn’t want a touchscreen, give me knobs and dials!

I'm surprised there's no mention of Mazda, who started doing this a year ago in more mainstream models.

https://outline.com/VugcGV

Hurrah!

I am suppose to be looking at the road, not a screen when I want to adjust the air con.

Give me something I can feel every day of the week.

This article talks about using dials for climate control. I have a 2019 CR-V and it still uses dials for that.

Never liked the touchscreen on my car. Tactile buttons are easier to control when your eyes are on the road.

Mazda did it first :) The new Mazda 3 2019 model (launched in october 2018) has no tuchscreen control.

Yay for Honda and common sense! Touchscreens are no good when you gotta keep your eyes on the road.

I still don't get it. For me banning touchscreens from every car interface is a no-brainer.

Well done!

At the risk of sounding like a broken record, I'm going to repeat a comment I made a few times before, as I think it's relevant:

In one of the talks at Google I/O a few years ago a VP from Audi (or Volvo?) spoke in a thick and lofty German accent about how "ve haf completely oferhauled ze driver exzperienze". He played a sexy video clip showcasing their new Android infotainment system (something like https://youtu.be/h_7_fKJ0PNs), and the first thing I noticed is how they'd taken away my traditional temperature knobs and replaced them with digital touchscreen ones. They looked just like physical ones, and were in the exact same place you would expect (https://9to5google.com/2017/05/15/android-cars-audi/). So, I've gained absolutely nothing, and now I have to take my eyes off the road and look down at the stupid console just to change the temperature.

TLDR: Tacticle feedback is a Good Thing(tm) and designers should cultivate - not fight - muscle memory.

The elegance of the Apple Watch Series 5 crown with haptic feedback for scroll function.

The trend towards a touchscreen interface seems like a nod to startrek and the LCARS OS.

I think Land Rover has made one of the neatest and most balanced yet useful interior designs with Land Rover Defender 2020. https://carbuzz.com/cars/land-rover/defender

honda generally gets called out for having some of the most touch heavy interface of any cars in their reviews, im not sure if finally responding to that feedback counts as "bucking the trend"

Good, the Honda touchscreen controls in my 2016 Pilot are incredibly awful.

Mazda started doing this a couple years ago. Glad to see it spreading.

This is a thread of people arguing about touch vs physical buttons.

I love me some good switches and buttons.

Sadly cars are kinda bad at those too.

IIRC, Mazda are also pushing back against the touchscreen fad

good, can't count how many times I've had to mash a touch-screen (not in Honda, so may be different) to get something to "click."

i'm all for physical UX.

a great reason to buy a Honda... if this becomes the new normal, I wonder how much this will hurt Tesla since it'd be a pretty drastic change for them.

It takes courage to remove the touchscreen.

Larry Tesler would be pleased today.

Nice. I see a Honda in my future...

All I can say is, thank fk

these days I am having very difficult time unlocking my iphone with my facemask on

They're not wrong.

UX designer here, I approve.

This link is better: https://jalopnik.com/honda-follows-mazda-by-ditching-some-to...

Autocar requires you to disable your ad blocker in order to view the website. Jalopnik does not, and does not even show a GDPR cookie popup.

Bret Vector's classic paper "Magic Ink: Information Software and the Graphical Interface" boldly states: "Interactivity considered harmful"!

When I first read that, I was flabbergasted and offended, because I love interactivity. But he's right.

"Interaction should be used judiciously and sparingly, only when the environment and history provide insufficient context to construct an acceptable graphic." -Bret Victor, Magic Ink

This is a critically important lesson to apply to automotive user interfaces. It's much safer if, as much as possible, instead of being interactive, the user interface predicts and infers from context and history what the user wants, when or before they want it, and presents it to them without them asking, so it's there when they need it.

http://worrydream.com/MagicInk/

Interactivity considered harmful

Chris Crawford defines interaction as a three-phase reciprocal process, isomorphic to a conversation: an interactant listens to her partner, thinks about what was said, and speaks a response. Her partner then does the same. See Crawford’s book The Art of Interactive Design (2003), or his essay Fundamentals of Interactivity (1993). For manipulation software, interaction is perfectly suitable: the user views a visual representation of the model, considers what to manipulate next, and performs a manipulation. The software, in turn, inputs the user’s manipulation request, updates the model, and displays the updated representation. With good feedback and an effective means of “speaking” to the software, this process can cycle smoothly and rapidly. It mimics the experience of working with a physical tool.

Information software, by contrast, mimics the experience of reading, not working. It is used for achieving an understanding—constructing a model within the mind. Thus, the user must listen to the software and think about what it says… but any manipulation happens mentally. Except possibly for signaling a decision, such as clicking a “buy” button, but that concludes, not constitutes, a session. The only reason to complete the full interaction cycle and speak is to explicitly provide some context that the software can’t otherwise infer—that is, to indicate a relevant subset of information. For information software, all interaction is essentially navigation around a data space.

For example, Amazon’s data space consists of their catalog of items. For a yellow pages directory, the data space contains all business listings; for a movie guide, all showtimes and movie information; for a flight planner, trips to and from all airports. In all of these cases, every interaction, every click and keystroke, search term and menu selection, simply serves to adjust the user’s view into the data space. This is simply navigation.

Alan Cooper defines excise in this context as a cognitive or physical penalty for using a tool—effort demanded by the tool that is not directly in pursuit of a goal. For example, filling a gas tank is done to support the car, not the goal of arriving at a destination. Cooper goes on to assert that software navigation is nothing but excise:

…the most important thing to realize about navigation is that, in almost all cases, it represents pure excise, or something close to it. Except in games where the goal is to navigate successfully through a maze of obstacles, navigation through software does not meet user goals, needs, or desires. Unnecessary or difficult navigation thus becomes a major frustration to users. In fact, it is the authors’ opinion that poorly designed navigation presents the number-one problem in the design of any software application or system… Alan Cooper and Robert Reimann, About Face (2003), p143.

If all interaction is navigation, and navigation is the number-one software problem, interactivity is looking pretty bad already. However, when compared with the other two sources of context, interactivity has even worse problems than simply being a frustrating waste of time:

The user has to already know what she wants in order to ask for it. Software that infers from history and the environment can proactively offer potentially relevant information that the user wouldn’t otherwise know to ask for. Purely interactive software forces the user to make the first move.

The user has to know how to ask. That is, she must learn to manipulate a machine. Donald Norman’s concept of determining a user’s “mental model” has become widespread in the software usability community, and is now considered a core design challenge. See Donald Norman’s book The Design of Everyday Things (2002), p9. However, Norman described this concept in the context of mechanical devices. It only applies to software if the software actually contains hidden mechanisms that the user must model. A low-interaction, non-mechanical information graphic relieves both user and designer from struggling with mental models.

Navigation implies state. Software that can be navigated is software in which the user can get lost. The more navigation, the more corners to get stuck in. The more manipulable state, the more ways to wander into a “bad mode.” State is the primary reason people fear computers—stateful things can be broken. The only state kept by a book is which page it is open to, which is why “getting lost in a book” describes a pleasurable experience!

Beyond these cognitive problems are physical disadvantages of interaction. The hand is much slower than the eye. Licklider described spending hours plotting graphs and seconds understanding them. A user who must manually request information is in a similar situation—given the mismatch between mousing and reading speeds, most of her time may be spent navigating, not learning. Further, the user might prefer to learn information while using her hands for other purposes, such as writing or eating or stroking a cat. Each time software demands the user’s hands, this activity must be interrupted. Finally, the growing prevalence of computer-related repetitive stress injuries suggests that indiscriminate interactivity may be considerably harmful in a literal, physical sense.

Unless it is enjoyable or educational in and of itself, interaction is an essentially negative aspect of information software. There is a net positive benefit if it significantly expands the range of questions the user can ask, or improves the ease of locating answers, but there may be other roads to that benefit. As suggested by the above redesigns of the train timetable, bookstore, and movie listings, many questions can be answered simply through clever, information-rich graphic design. Interaction should be used judiciously and sparingly, only when the environment and history provide insufficient context to construct an acceptable graphic.

It is unfortunate that the communities concerned with human factors of electronic artifacts have latched onto the term “interaction.” Most professional communities and academic programs use the term Human-Computer Interaction, or HCI; the ACM special-interest group is CHI, the converse. Many practitioners, following Cooper and Bill Moggridge, refer to their profession as “interaction design.” For information software, the real issue is context-sensitivity. Interaction is merely one means of achieving that. And as long as “speaking” is constrained to awkwardly pushing metaphors with a mouse, interaction should be the last resort.

The working designer might protest that interaction is unavoidable in practice, and may even consider my ideal of interaction-free software to be a scoff-worthy fantasy. This is only because the alternatives have been unrecognized and underdeveloped. I believe that with the invention of new context-sensitive graphical forms and research into obtaining and using environment and history, the clicking and dragging that characterizes modern information retrieval will be made to seem laughably archaic. But every condonation of “interactivity,” from the annals of academia to the corporate buzzvocabulary, postpones this future.

difficult to operate intuitively is exactly right.

Welp, looks like Honda's back in the running for my next car!

There's another Hacker News discussion about a related topic, a paper Ben Shneiderman wrote in 1993, "Beyond Intelligent Machines: Just Do It".

https://news.ycombinator.com/item?id=22742100

http://www.cs.umd.edu/hcil/trs/93-03/93-03.html

>It seems that you can hardly go to a computer conference without seeing a videotape of a futuristic computer system that talks to you from a wall, desk, or some random appliance. Is this the interface of the future? Over the last decade, Ben Shneiderman, head of the University of Maryland's Human-Computer Interaction Laboratory and author of Designing the User Interface: Strategies for Effective Human-Computer Interaction (Addison-Wesley, 1992), has been the most forceful voice against anthropomorphic interfaces. He argues that users want a sense of direct and immediate control over computers that differs from how they interact with people. He presents several examples of these predictable and controllable interfaces developed in the lab at UM.

Ben Shneiderman and his colleagues like Richard Potter, Andrew Sears, and Catherine Plaisant at his HCIL lab have performed a lot of research and user interface design with touch screens, like the "lift off" high precision pointing strategy, and early touch screen keyboards.

An experimental evaluation of three touch screen strategies within a hypertext database

https://www.tandfonline.com/doi/abs/10.1080/1044731890952595...

Clocks, calendars and schedulers on a touchscreen (1988, HCIL)

https://www.youtube.com/watch?v=kzMHqF9Pkv4

Touchscreen keyboards - Andrew Sears, Ben Shneiderman. Human-Computer Interaction Lab, University of Maryland.

https://www.youtube.com/watch?v=JDRKP2ATBRg

And he just dropped by HN to announce a new paper he's just published on a similar topic, the result of years of work, which also relates to this discussion:

Shneiderman, Ben (2020). Human-Centered Artificial Intelligence: Reliable, Safe & Trustworthy, International Journal of Human-Computer Interaction,36, 6 (Published Online March 27, 2020).

https://doi.org/10.1080/10447318.2020.1741118

Human-Centered Artificial Intelligence: Reliable, Safe & Trustworthy

>Abstract

>Well-designed technologies that offer high levels of human control and high levels of computer automation can increase human performance, leading to wider adoption. The Human-Centered Artificial Intelligence (HCAI) framework clarifies how to (1) design for high levels of human control and high levels of computer automation so as to increase human performance, (2) understand the situations in which full human control or full computer control are necessary, and (3) avoid the dangers of excessive human control or excessive computer control. The methods of HCAI are more likely to produce designs that are Reliable, Safe & Trustworthy (RST). Achieving these goals will dramatically increase human performance, while supporting human self-efficacy, mastery, creativity, and responsibility.

>Introduction

>This paper opens up new possibilities by way of a two dimensional framework of Human-Centered Artificial Intelligence (HCAI) that separates levels of automation/autonomy from levels of human control. The new guideline is to seek high levels of human control AND high levels of automation, which is more likely to produce computer applications that are Reliable, Safe & Trustworthy (RST). Achieving these goals, especially for complex poorly understood problems, will dramatically increase human performance, while supporting human selfefficacy, mastery, creativity, and responsibility. This paper focuses on the three RST goals, which may help to achieve other important goals such as privacy, cybersecurity, resilience, social justice, human dignity, and environmental preservation.

>The traditional belief in computer autonomy is compelling for many artificial intelligence (AI) researchers, developers, journalists, and promoters. I assume a broad definition of AI to include automated/autonomous systems using technologies such as machine learning, neural nets, statistical methods, recommenders, adaptive systems, and speech, facial, image, and pattern recognition.

>The goal of computer autonomy was central in Sheridan and Verplank (1978) ten levels from human control to computer automation/autonomy (Table 1). Their widely cited one-dimensional list continues to guide much of the research and development, suggesting that increases in automation must come at the cost of lowering human control. Shifting to HCAI could liberate design thinking so as to produce computer applications that increase automation, while amplifying, augmenting, enhancing, and empowering people to innovatively apply systems and creatively refine them.

>Sheridan & Verplank’s ten levels of automation/autonomy have been widely influential (Sheridan, 1992), but critics suggested refinements such as the four stages of automation: (1) information acquisition, (2) analysis of information, (3) decision or choice of action, and (4) execution of action (Parasuraman et al., 2000). These stages refine discussions of each of the levels, but the underlying message is that the goal is full automation/autonomy.

>Even Sheridan (2000) commented with concern that “surprisingly, the level descriptions as published have been taken more seriously than were expected” (see Hoffman & Johnson (2019) and Kaber (2018) for detailed histories). However, in spite of the many critiques, the 1-dimensional levels of automation/autonomy, which only represents situations where increased automation must come with less human control, is still widely influential. For example, the US Society of Automotive Engineers adopted the unnecessary trade-off in its six levels of autonomy for self-driving cars (Brooks, 2017; Society of Automotive Engineers, 2014) (Table 2).

>Critics of autonomy have repeatedly discussed the ironies (Bainbridge, 1983), deadly myths (Bradshaw et al., 2013; Mindell, 2015), conundrums (Endsley, 2017), or paradoxes (Hancock, 2017) of autonomy. A common point is that humans have to spend more effort monitoring autonomous computers because they are unsure of what it will do, often leading to inferior performance (Blackhurst et al., 2011; Strauch, 2017).

[...]

Yesssssss

The Audi MMI system consisting of a central dial is still the most intuitive and least distracting of all car systems I have seen in the market. And, as a bonus, it also works really well with Android Auto and CarPlay - without touchscreen.

Finally! Bring back some sanity to machine controls.

I much prefer knobs, dials, and switches to control things inside the car. And especially if I need to modulate it while operating the car.

The steering wheel and acceleration pedals should be made into touch screens. Physical interfaces are so old school.

Oh thank goodness. Haptics are a real.

Appeal to authority is one of the informal fallacies of logic. It can/should never be used to prove a point.

Thank you, Honda!

Touch screen controls in automotive environment is such a bad idea (unless the vehicle is stationary, and even then...)

What if--and stay with me for a moment--what if it's good for different manufacturers to have different approaches to the problem, giving people the option to choose Honda for buttons or Tesla for touchscreens?

What if there are legitimate arguments for both and it's good that there's both options available?

I've said it before, I'll say it again.

Touchscreens CAN BE DONE RIGHT. (or at least better)

I have an aftermarket JVC headunit in my 2003 VW. It has one button, that's the "Menu" button on single press, or Power button when held.

The screen itself is relatively intuitive to control without looking because of a gesture feature it has as well as button placement being useful.

If I want more volume I put one finger on the screen and make a circle gesture in the clockwise direction. Down is counter-clockwise. The play/pause button is in the very bottom right so you can find the button with feel. You can shuffle songs with another gesture (though I never use it).

I really like my touchscreen. And every time I am using a work vehicle (Ford Explorers), or my wife's GMC Sierra, I hate the controls because you have to look up to see what you're doing. Thankfully their steeringwheel controls are generally good.

Reply on news.ycombinator.com